Swiftech Releases Komodo HD5800 F/C Waterblock For ATI

[gabe]

We are pleased to announce that the HD5800 FC waterblock for AMD/ATI 5850 and 5870 is now released, and will ship to the channels next week.

Construction is Black Delrin housing, and Chrome Plated Copper base. Fastening to the board is by means of stainless steel standoffs, and pre-applied thermal pads for memory and mosfets facilitate installation.

Here are some pics:

http://www.swiftnets.com/assets/images/products/Komodo/Komodo-5800x800.jpg

The back plate shows the three alternate standoff locations allowing the change from HD5870 (stock config) to HD5850.

http://www.swiftnets.com/assets/images/products/Komodo/Komodo-5800-baseX800.jpg

Cooling of the GPU is enhanced by Komodo's standard 0.6mm pin matrix:

http://www.swiftnets.com/assets/images/products/Komodo/PIN%20MATRIXX400.jpg

MSRP is $114.95, including a set of 1/2" chrome plated barbs.

Let me also take this opportunity to update members about a few more things:

- HD5850 and 5870 Heatsinks are due for release within a couple of weeks
- MSRP Price reductions are now in effect for nVidia F/C blocks, including Komodo series, and Epsilon.
- The AM2/AM3/XEON Mounting plates for Apogee XT are here. We just need a few more days to package them.

[millertime359]

Wow, That is very cool Gabe. I love how you guys think about the end user when designing your blocks. Make it as easy as possible to install. Great work Gabe. :up:

[gabe]

Wow, That is very cool Gabe. I love how you guys think about the end user when designing your blocks. Make it as easy as possible to install. Great work Gabe. :up:

Thanks.. We do that whenever it's technically possible. Unfortunately, we couldn't for the heatsinks though.. but the heatsinks are inexpensive, so I guess we'll have to live with two separate models..

[WaterFlex]

Good job, Swiftech. Really nice block.
All manufactures have made their 5870 blocks now. It`s time for some numbers :up:

[millertime359]

Good job, Swiftech. Really nice block.
All manufactures have made their 5870 blocks now. It`s time for some numbers :up:

SKINNEE!!!!!!!!!!!!!!!!! :sofa:

[gabe]

Good job, Swiftech. Really nice block.
All manufactures have made their 5870 blocks now. It`s time for some numbers :up:

I published some performance numbers here: http://www.xtremesystems.org/forums/showpost.php?p=4065510&postcount=198

[bentleya]

Good job, Swiftech. Really nice block.
All manufactures have made their 5870 blocks now. It`s time for some numbers :up:

I've been thinking about it, But it would rely on every manufactures support, as the card is pretty expensive itself before adding block prices. And thats before begin able to get hold of one.

[Mekrel]

Gabe - fantastic design, love the cross compatibility. Other manufacturers of full cover solutions should take a leaf out of your book.

I would go as far to say that Swiftech now manufacture the best full cover blocks on the market:


Great performance
Ease of install with the standoffs
Great looks!

[Mech0z]

HD5850 and 5870 Heatsinks are due for release within a couple of weeks

Is that the "unisinks" for MCW60?

[gabe]

Gabe - fantastic design, love the cross compatibility. Other manufacturers of full cover solutions should take a leaf out of your book.

I would go as far to say that Swiftech now manufacture the best full cover blocks on the market:


Great performance
Ease of install with the standoffs
Great looks!


Wow, thank you! The compliment certainly hits a cord here!

[Mazlov]

Looks awesome!

The combined mount for 5850 and 5870, and the pre-applied pads are really interesting. Don't they influence production costs much?

[Boulard83]

hummmmm, maybe ill buy one for my Ordered HD5850 :)

[skinnee]

I've been thinking about it, But it would rely on every manufactures support, as the card is pretty expensive itself before adding block prices. And thats before begin able to get hold of one.

That is the problem with GPU blocks, its just extremely expensive without manufacturer support, both blocks and a card now and then. Takes a long time to build relationships like that.

[Mekrel]

Wow, thank you! The compliment certainly hits a cord here!

Well it's true, I loved the look of the Epsilon block and that would have been my choice if I had bought a 295GTX.

I've never heard of any cases of a Epsilon/Komodo block not fitting.

I own a EK8800GTX block which doesn't have standoffs and I've fitted this block 100's of times since doing my PC343B build.

When I last redid the loop, I must have done something wrong and the card was getting artefacts.

Three remounts later and all is well - but it goes to show the error margin is quite small!

I always recommend standoffs now.

[gabe]

http://www.swiftnets.com/assets/images/products/Komodo/HD5870-komodo%205800-INSTALLEDX800.jpg
HD5870 installed -Top side

http://www.swiftnets.com/assets/images/products/Komodo/HD5870-komodo%205800-backside-INSTALLEDx800.jpg
HD5870 installed - back side

[iandh]

You may have me sold on one. I was going to machine my own but considering this is interchangeable to 5870 and I planned on switching back to a 5870 eventually, it's sort of irresistable. I'm too damn busy lately anyways.

[Feklar]

Looks great. Is a 5970 block coming soon as well?

[Utnorris]

Looks good Gabe. A+++ for Swiftech. A lot of great products coming from you guys lately.

[woffen]

There has been a lot of talk about water over the vrms lately. How does this card fair in that case? Maybe there's been pictures of it on this forum but can't remember seeing any of this block.

[JaD]

Happy to hear it's finally out. Keep up the great work with your Komodo line, definitely the best full cover solutions available. Good looks, high materials & finishing quality, excellent GPU cooling, an eye to VRM cooling (since before the recent fuss about it), all at a price competitive to clearly lesser quality blocks. :toast:

@woffen: iirc, water flows almost directly above VRMs.

[_G_]

http://www.swiftnets.com/assets/images/products/Komodo/HD5870-komodo%205800-backside-INSTALLEDx800.jpg
HD5870 installed - back side

Wow, you where even able to reuse the spring loaded ati screws, nice touch. :up:

[Hondacity]

gabe

do you have the picture of the underside of the pom top? i wanna see hoe the water flows thru the pin matrix...i'm guess its like the single 295 block that you have...

[Manicdan]

There has been a lot of talk about water over the vrms lately. How does this card fair in that case? Maybe there's been pictures of it on this forum but can't remember seeing any of this block.

i would like to see the difference between water over, or nearby the VRMs, i cant see 40C vs 60C vrms making too much of a difference

[Arni]

Nice job swiffy :)

[Sadasius]

Very nice! Fantastic job Gabe!:up:

[CthulpiSS]

Gabe,
could you please give me accurate dimension of this part:

http://img80.imageshack.us/img80/7200/sizekomodo.jpg

I'm afraid it may be hard to fit this nice block in my cube case
(I have PSU hanging just above graphic card)

[gabe]

Gabe,
could you please give me accurate dimension of this part:

http://img80.imageshack.us/img80/7200/sizekomodo.jpg

I'm afraid it may be hard to fit this nice block in my cube case
(I have PSU hanging just above graphic card)

0.97" (24.7mm)

[Philwong]

Gabe

First off, thumbs up for the excellent service rendered by your staff, Michelle, on my recent purchase of the Komodo 5800.

I would like to share some feedbacks on the actual product.

http://img.photobucket.com/albums/v97/tmwong/komodo3.jpg
I love the chrome plating for that polished look. Factory installed standoffs and thermal pads are additional pros for hassle-free installation.

http://img.photobucket.com/albums/v97/tmwong/komodo4.jpg
http://img.photobucket.com/albums/v97/tmwong/komodo5.jpg
That said, the finishing of my block can definitely be better. For starters, the base is not as smooth as I hope. Furthermore, one of the inlet/outlet holes is chipped with loose bits of acetal. Lastly, you may like to consider throwing in the allen key in the package as standard, too.

This is about the first time I encountered such an issue after using Swiftech products since the Apogee days. Hopefully, this is just an isolated case.

I do sincerely hope that the cost cutting measure of shifting production to China will not hurt the quality, as well as reputation of your offerings. Please take my feedback positively as I really like Swiftech watercooling components.

Phil

[Main]

This looks like the Chinasyndrome Philwong - seen it happen on numerous occassions before when the western world companies moves their production to third world countries. These countries always use time to sort out their qc sorted out - meanwhile the products are beeing shipped out and the owners are sitting there with crossed fingers hoping things is ok.

This whole mess is caused by customers that want to pay as little as possibly for the products, shareholders that wants to earn money and so on. It's pretty complex.

I would easily pay 2-3 times more for my cooling gear if I knew it was produced in a western country and beeing "Rolls Royce" products.

I would like to see watercooling blocks for dual GPU cards with possibility for separate loops, likewise small effective pumps aka C Systems pumps for these setups. And 140 mm quad,triple and quad radiators with four inlets and outlets - this will make my day :D

EDIT: I don't belive that the pinmatrix in the picture below is produced in a high end cnc mill :p

http://i50.tinypic.com/mb74ut.jpg

[warriorpoet]

...
EDIT: I don't belive that the pinmatrix in the picture below is produced in a high end cnc mill :p

http://i50.tinypic.com/mb74ut.jpgYeah, I was kinda surprised to see that photo up on their site.

[Main]

Yeah, I was kinda surprised to see that photo up on their site.

jupp I will call it a headshot :) - I work with these things and know wery well the Chinasyndrome.

But as said - Swiftech is one of my favorite producers - I would hate to see them go down the drain. IT MUST NOT HAPPEN !

[gabe]

Gabe

First off, thumbs up for the excellent service rendered by your staff, Michelle, on my recent purchase of the Komodo 5800.

I would like to share some feedbacks on the actual product.

http://img.photobucket.com/albums/v97/tmwong/komodo3.jpg
I love the chrome plating for that polished look. Factory installed standoffs and thermal pads are additional pros for hassle-free installation.

http://img.photobucket.com/albums/v97/tmwong/komodo4.jpg
http://img.photobucket.com/albums/v97/tmwong/komodo5.jpg
That said, the finishing of my block can definitely be better. For starters, the base is not as smooth as I hope. Furthermore, one of the inlet/outlet holes is chipped with loose bits of acetal. Lastly, you may like to consider throwing in the allen key in the package as standard, too.

This is about the first time I encountered such an issue after using Swiftech products since the Apogee days. Hopefully, this is just an isolated case.

I do sincerely hope that the cost cutting measure of shifting production to China will not hurt the quality, as well as reputation of your offerings. Please take my feedback positively as I really like Swiftech watercooling components.

Phil

I do appreciate the criticism, it helps us do better, but I'd also like to get some operational feed-back. Have you tested the board yet? Let's see some temps and overclocks!

Base finish: I've asked our CNC guy to modify the tool path on the surface finish operation over the GPU to a circular pattern. Hopefully we can start implementing this in the next production batches.

Thread burs: totally concur. Been asking my ppl to be careful with this, and it looks like I'm going have to yell louder this time.

Allen wrench: OK

Cost cutting measures: these days, most if not all of the components in our PC's are made somewhere in Asia. If we want to stay in business, we have no choice but to make things over there. The toughest part for small American businesses when shifting production to Asia is to create a quality culture within the organization. As usual, it boils down to picking the right individuals to do the job, train them right, and keep pushing for perfection.

[warriorpoet]

I do appreciate the criticism, it helps us do better, but I'd also like to get some operational feed-back. Have you tested the board yet? Let's see some temps and overclocks!

Base finish: I've asked our CNC guy to modify the tool path on the surface finish operation over the GPU to a circular pattern. Hopefully we can start implementing this in the next production batches.

Thread burs: totally concur. Been asking my ppl to be careful with this, and it looks like I'm going have to yell louder this time.

Allen wrench: OK

Cost cutting measures: these days, most if not all of the components in our PC's are made somewhere in Asia. If we want to stay in business, we have no choice but to make things over there. The toughest part for small American businesses when shifting production to Asia is to create a quality culture within the organization. As usual, it boils down to picking the right individuals to do the job, train them right, and keep pushing for perfection.
gabe,

I love what your company does. Y'all innovate like mad, listen to your customers and adjust according to end-user feedback in a way few others do. I also appreciate the realities of the modern design and manufacturing business.

Here's the way I see the perceived quality issue: you operate in a unique, high-expectation marketplace where perception is reality. to compound the problem, most water-coolers do it as a hobby or view it as a luxury. The problem is, even if your micropin bases perform like crazy (and all testing I've seen confirms this), the visual flaws bring up quality concerns that may or may not be warranted from a performance standpoint.

Where far-East manufacturing is concerned, I've seen some AMAZING machining done, so the country of origin is absolutely a non-issue. I've seen about a dozen companies move to Korean or Chinese manufacturing, and they all start the same way- substandard visual defects. Where they go from there, and the level of success ultimately attained is generally directly related to the amount of defects the design firm is willing to accept. The successful ones I've seen solve the visual issue within one or two product shipments.

Best wishes to you and Swiftech on your transition. I hope to see the rapid improvement I'm sure you expect, and look forward to purchasing more product from you within the next generation.

[gabe]

gabe,

I love what your company does. Y'all innovate like mad, listen to your customers and adjust according to end-user feedback in a way few others do. I also appreciate the realities of the modern design and manufacturing business.

Here's the way I see the perceived quality issue: you operate in a unique, high-expectation marketplace where perception is reality. to compound the problem, most water-coolers do it as a hobby or view it as a luxury. The problem is, even if your micropin bases perform like crazy (and all testing I've seen confirms this), the visual flaws bring up quality concerns that may or may not be warranted from a performance standpoint.

Where far-East manufacturing is concerned, I've seen some AMAZING machining done, so the country of origin is absolutely a non-issue. I've seen about a dozen companies move to Korean or Chinese manufacturing, and they all start the same way- substandard visual defects. Where they go from there, and the level of success ultimately attained is generally directly related to the amount of defects the design firm is willing to accept. The successful ones I've seen solve the visual issue within one or two product shipments.

Best wishes to you and Swiftech on your transition. I hope to see the rapid improvement I'm sure you expect, and look forward to purchasing more product from you within the next generation.

you are right on the button. excellent comments. the human factor is the most critical. how many times do i have to tell ppl to pay attention to details? on the other hand, I push for high production to satisfy demand. so it's a balancing act.

[warriorpoet]

...the human factor is the most critical. how many times do i have to tell ppl to pay attention to details? on the other hand, I push for high production to satisfy demand. so it's a balancing act.
I imagine that can get pretty frustrating, especially with something so low-tolerance.

[Philwong]

I do appreciate the criticism, it helps us do better, but I'd also like to get some operational feed-back. Have you tested the board yet? Let's see some temps and overclocks!

Base finish: I've asked our CNC guy to modify the tool path on the surface finish operation over the GPU to a circular pattern. Hopefully we can start implementing this in the next production batches.

Thread burs: totally concur. Been asking my ppl to be careful with this, and it looks like I'm going have to yell louder this time.

Allen wrench: OK

Cost cutting measures: these days, most if not all of the components in our PC's are made somewhere in Asia. If we want to stay in business, we have no choice but to make things over there. The toughest part for small American businesses when shifting production to Asia is to create a quality culture within the organization. As usual, it boils down to picking the right individuals to do the job, train them right, and keep pushing for perfection.

Thanks for listening, Gabe.

I'm still recovering from a nasty jet lag and will keep you folks posted on the performance soon. By the way, the inlet/outlet hole is chipped as well and is a bit out of shape at the opening so take note of that, too.

Actually, most products are made in China nowadays. But with tighter QC, the quality can be on par with just about any developed countries. This is a kink that could have been resolved by Swiftech before the blocks are shipped. You'll just have to crack the whip especially when it comes to these low-cost Chinese factories. That's the way things work there at the moment with no racism intended. After all, I'm also a Chinese.

Phil

[Philwong]

http://img.photobucket.com/albums/v97/tmwong/rig6.jpg
This is my current rig and the following is the revised loop configuration.

ST MCP655 -> ST GTZ (i7@4Ghz) -> ST MCW-NBMAX (NB) -> BI GTS120 -> ST Komodo 5800 (5870) -> ST MCW30 (SB) -> ST MCR220 -> BI GTS120 -> ST MCP655.

Fans will be swapped from Noctua P12 to AC F12 PWM thermal controlled via Rampage II Gene.

http://img.photobucket.com/albums/v97/tmwong/komodo1.jpg
The Komodo is replacing my MCW60 rev. 2 due to high PWM temp during gaming.

Phil

[Jamesrt2004]

Hi gabe any news on the heatsink covers for the 5 series as ill purchase it + mcw60 + my 5850/70 when it's out =)

[Philwong]

Gabe

I'm facing the same no display problem encountered by another XS member here;
http://www.xtremesystems.org/forums/showthread.php?t=244152

It seems like the stock GPU cross bracket might be shorting the card without the springs as suggested by Swiftech's installation manual. Either way, the instruction "fasten gradually in a cross pattern until each leg of the bracket bottoms out" is obviously wrong according to your suggestion in that post.

I hope you're already investigating this issue and will be issuing an advisory on a permanent solution soon. Will the workaround solution suggested in the other post compromise the CPU and block contact and cooling performance, too?

http://img.photobucket.com/albums/v97/tmwong/newloop.jpg
I just spent two days installing my tight loop and it's nothing short of a nightmare to correct this problem. Not only does this waste my precious time, my new coolant is probably going down the drain as well. Sigh.

Phil

[iboomalot]

you can access the screws from the backside of the card just back them off 1-2 turns this solved my problem.

Good luck and stellar install

[Philwong]

you can access the screws from the backside of the card just back them off 1-2 turns this solved my problem.

Good luck and stellar install

I tried, the screws are too tight. More importantly, there aren't sufficient space to even reach them.

How's your GPU temp after the correction? Thanks anyway.

Phil

[Philwong]

http://img.photobucket.com/albums/v680/elpibe10/5870GPUbracket.jpg
One of my friends snapped this photo of the cross bracket. As you can see, it does come with a thin layer of isolation film. That eliminates the short circuit theory, but I'm still not quite convinced the springs should be removed. They're there for a reason and possibly to prevent over tightening.

Phil

[iboomalot]

the difference in height is 1mm thus 2 turns

This equals the distance at stock tension.

When fully tightened it puts pressure on the GPU or some other part.

All i know is loosen your screws and it should work.


Might try a wrench on the end of the scewdriver to give you leverage to break them loose.

[Philwong]

the difference in height is 1mm thus 2 turns

This equals the distance at stock tension.

When fully tightened it puts pressure on the GPU or some other part.

All i know is loosen your screws and it should work.

Might try a wrench on the end of the scewdriver to give you leverage to break them loose.

Will consider, thanks once again for the suggestion.

Phil

[BulldogPO]

Where to buy one in EU?
Or, Gabe can I buy one directly from you?

[gabe]

Gabe

I'm facing the same no display problem encountered by another XS member here;
http://www.xtremesystems.org/forums/showthread.php?t=244152

It seems like the stock GPU cross bracket might be shorting the card without the springs as suggested by Swiftech's installation manual. Either way, the instruction "fasten gradually in a cross pattern until each leg of the bracket bottoms out" is obviously wrong according to your suggestion in that post.

I hope you're already investigating this issue and will be issuing an advisory on a permanent solution soon. Will the workaround solution suggested in the other post compromise the CPU and block contact and cooling performance, too?


I just spent two days installing my tight loop and it's nothing short of a nightmare to correct this problem. Not only does this waste my precious time, my new coolant is probably going down the drain as well. Sigh.

Phil

Technical Bulletin - Komodo HD5800 - Reprint from HD5800 Product Page (news and technical bulletins section) and Swiftech forums.

Date: 2-2-10

Issue:
No graphics display after installation of the Komodo HD5800 (black screen). This is a random occurence, and no permanent damage to the card is expected.

Cause:
Overtightening of the 4 screws fastening the stock spring-backplate.

Field/user resolution:
Uninstalling the block is not necessary.
Backing off the 4 screws by a minimum of 1 turn, and not to exceed 2 turns (an equal amount of turns for each screw is essential) resolves the issue, and restores graphics functionality.

Scope:
This issue may affect all water-blocks shipped between December 1st, 2009 and January 31st, 2010.

Production Resolution:
Effective February 1st, 2010, inclusion of a nylon washer between board and spring back-plate prevents the user from overtightening of the spring back-plate. Installation guide was revised accordingly (http://www.swiftnets.com/products/installation_guide_KOMODO-HD5800-Multilangue.pdf)

[Philwong]

Technical Bulletin - Komodo HD5800 - Reprint from HD5800 Product Page (news and technical bulletins section) and Swiftech forums.

Date: 2-2-10

Issue:
No graphics display after installation of the Komodo HD5800 (black screen). This is a random occurence, and no permanent damage to the card is expected.

Cause:
Overtightening of the 4 screws fastening the stock spring-backplate.

Field/user resolution:
Uninstalling the block is not necessary.
Backing off the 4 screws by a minimum of 1 turn, and not to exceed 2 turns (an equal amount of turns for each screw is essential) resolves the issue, and restores graphics functionality.

Scope:
This issue may affect all water-blocks shipped between December 1st, 2009 and January 31st, 2010.

Production Resolution:
Effective February 1st, 2010, inclusion of a nylon washer between board and spring back-plate prevents the user from overtightening of the spring back-plate. Installation guide was revised accordingly (http://www.swiftnets.com/products/installation_guide_KOMODO-HD5800-Multilangue.pdf)

Those nylon washers should solve this problem. Thanks for following up and offering a permanent solution.

Phil

[RoadconeTuning]

i think gabe just told everyone to stop going hulk status on their backplates lol

[irenedakota]

Damn, I wish I could test my Komodo.

It is mounted on my GFX card in my case, but I can't do anything until I've received my AM2 mounting kit for the Apogee XT (Was shipped by Performance-PCs on the 5th of Jan, I think it is lost).

[Philwong]

Damn, I wish I could test my Komodo.

It is mounted on my GFX card in my case, but I can't do anything until I've received my AM2 mounting kit for the Apogee XT (Was shipped by Performance-PCs on the 5th of Jan, I think it is lost).

Did you bottom out all 4 screws for the GPU cross bracket according to the original Swiftech instruction?

Phil

[irenedakota]

I did, but after reading this and the other thread I have back turned by 1.5 turns. I think I'm going to unmount it and put some washers there (I have some lying around)

[Philwong]

I did, but after reading this and the other thread I have back turned by 1.5 turns. I think I'm going to unmount it and put some washers there (I have some lying around)

Don't. I believe your card should be fine after the correction.

Remounting the block might affect the performance of your thermal pads since they're already compressed. This should not be a problem unless you can achieve the exact screw pressure again.

Phil

[irenedakota]

Don't. I believe your card should be fine after the correction.

Remounting the block might affect the performance of your thermal pads since they're already compressed. This should not be a problem unless you can achieve the exact screw pressure again.

Phil

Forgot about the thermal pads, thanks. Will have to wait for my CPU mounting kit to arrive to test it though :(

[Babrbarossa]

I unfortunately did not find this thread until I already installed my Komodo on my 5870, and I too ran into the same problem. I recognised the problem with the bracket myself once my initial installation came up dead because of the short, and backed it off- now the monitor gets a signal (no image- the monitor just picks up a signal for a few seconds and then goes black again). I am very confident I installed it properly and I didn't get anything wet so the most likely scenario is that this backplate short did permanent damage. I even tried taking the backplate right off and still dead.

I swore I'd never buy another swiftech block after the seal on my Apogee GTZ started leaking one day and killed two 260s- I don't know what I was thinking getting another one.

Very ticked off right now at Swiftech's shoddy execution of their products.

[Philwong]

I unfortunately did not find this thread until I already installed my Komodo on my 5870, and I too ran into the same problem. I recognised the problem with the bracket myself though and backed it off- now the monitor gets a signal (no image- the minitor just picks up a signal) for a few seconds and then goes black again. I am very confident I installed it properly and I didn't get anything wet so the most liekly scenario is that this backplate short did permanent damage. I even tried taking the backplate right off and still dead.

I swore I'd never buy another swiftech block after the seal on my Apogee GTZ started leakeding one day and killed two 260s- I don't know what I was thinking getting another one.

Very ticked off right now at Swiftech's shoddy QA/QC.

You should try to reinstall the stock HSF and give your 5870 another go. Be prepared to RMA the card if necessary. Good luck.

Phil

[Boulard83]

Juste ordered a Komodo. Hope ill not have any issue !

[Philwong]

Juste ordered a Komodo. Hope ill not have any issue !

You may like to check whether the package include the new GPU plastic washers.

Phil

[Boulard83]

I already have these on hand ;)

[Philwong]

I already have these on hand ;)

Have fun then...

Phil

[dejanh]

Can somebody post a pic of what is supposedly being shorted? Is it on the backside of the card, where the X bracket sits? If it is how come it does not short by default then? I'm not sure I understood the problem from this thread and I'd be curious to see it before I decide to cool my cards with this block.

[gabe]

Can somebody post a pic of what is supposedly being shorted? Is it on the backside of the card, where the X bracket sits? If it is how come it does not short by default then? I'm not sure I understood the problem from this thread and I'd be curious to see it before I decide to cool my cards with this block.

here is the explanation:

the stock heatsink standoffs go thru the pcb, and stick out on the back side of it by 1mm. when you tighten the stock X spring bracket, it sits on the standoffs and is therefore 1m above the pcb.

Since we are not using standoffs that go thu the pcb, when we tighten the x bracket, it comes to mate with the pcb backside. it's not a problem to touch the pcb, since these spots are grounded. but since we are bending the arms of the X spring bracket a tiny little bit more, the end of the arms might be sometimes slightly touching some surface mount components that are close to the grounded holes. it is very subtle, and doesn't happen systematically. It really depends on the height tolerances of said surface mount components. For example, we never had the problems during our validation testing, on several different ATI boards, including 5870, 5850's and 5970's.. this is the reason why it never caught our attention to begin with.

The fix to this problem was very simple: include washers that go under the X spring bracket so that it stays sufficiently above the board to never touch anything. the same result is achieved (without washers) by simply backing off the screws by a turn or two..

[dejanh]

here is the explanation:

the stock heatsink standoffs go thru the pcb, and stick out on the back side of it by 1mm. when you tighten the stock X spring bracket, it sits on the standoffs and is therefore 1m above the pcb.

Since we are not using standoffs that go thu the pcb, when we tighten the x bracket, it comes to mate with the pcb backside. it's not a problem to touch the pcb, since these spots are grounded. but since we are bending the arms of the X spring bracket a tiny little bit more, the end of the arms might be sometimes slightly touching some surface mount components that are close to the grounded holes. it is very subtle, and doesn't happen systematically. It really depends on the height tolerances of said surface mount components. For example, we never had the problems during our validation testing, on several different ATI boards, including 5870, 5850's and 5970's.. this is the reason why it never caught our attention to begin with.

The fix to this problem was very simple: include washers that go under the X spring bracket so that it stays sufficiently above the board to never touch anything. the same result is achieved (without washers) by simply backing off the screws by a turn or two..

Got it. Thanks Gabe. I have a lot of spare nylon washers anyway so if I do decide to go down this route I should be good then. Still waiting for my second HD 5850 :(

[Boulard83]

Just installed my komodo. Using MX-3 as thermal paste.

GPU temp : 39°c under OCCT gpu test after 30+ minutes

My Vreg are at ~70°c but i keept the stock thermal pad. I didnt removed it to use paste as contact and i think the temp are normal as this with thermal pad for the Vreg.

[Badfastbusa]

Just installed my komodo. Using MX-3 as thermal paste.

GPU temp : 39°c under OCCT gpu test after 30+ minutes

My Vreg are at ~70°c but i keept the stock thermal pad. I didnt removed it to use paste as contact and i think the temp are normal as this with thermal pad for the Vreg.

I read at the end of this thread that people were removing the thermal tape and using TIM and getting excellent results:
http://translate.google.com/translate?hl=en&ie=UTF-8&sl=de&tl=en&u=http://extreme.pcgameshardware.de/wasserkuehlung/78095-extreme-review-das-grosse-wakue-roundup-hd-5870-oc.html&prev=_t&rurl=translate.google.com

[Razrback16]

Just installed my komodo. Using MX-3 as thermal paste.

GPU temp : 39°c under OCCT gpu test after 30+ minutes

My Vreg are at ~70°c but i keept the stock thermal pad. I didnt removed it to use paste as contact and i think the temp are normal as this with thermal pad for the Vreg.

70*C load temps on the vregs are still incredibly good temps. Have you voltmodded the card yet?

[dejanh]

Bah I need my second HD 5850 to be shipped out and then I can put everything together and see what temps I would get on the VREG with the paste instead of the pads.

By the way, Gabe, the first Komodo I received is brand new but kinda beat up. There are scratches on the acetal, loose pieces of metal from machining, and part of the base is not even chrome plated. I do not have a camera handy right now to take a pic but that's pretty iffy QC. :shrug:

I've already cleaned up the loose metal pieces and the scratches on the acetal are not the end of the world, but the plating, I'm wondering how that one got missed.

[Boulard83]

70*C load temps on the vregs are still incredibly good temps. Have you voltmodded the card yet?

Currently running at 950/1300 with 1.175v. Furmark takes my Vreg a lil high as i told but its still far from bad !

After playing GTA4 for several hours :

Single loop with Dual MCP655, 1x TFC360 and 3x SplitStream shrouded at ~1200rpm.

CPU : ~45°c ( coretemp )
GPU : ~37°c ( AfterBurner )
Vreg : ~55°c ( GPUZ, i think, ill check againg soon )

Ill be doind more testing but my PSU is currently RMAed so im stuck with an old CoolerMaster 650w.

[gabe]

Bah I need my second HD 5850 to be shipped out and then I can put everything together and see what temps I would get on the VREG with the paste instead of the pads.

By the way, Gabe, the first Komodo I received is brand new but kinda beat up. There are scratches on the acetal, loose pieces of metal from machining, and part of the base is not even chrome plated. I do not have a camera handy right now to take a pic but that's pretty iffy QC. :shrug:

I've already cleaned up the loose metal pieces and the scratches on the acetal are not the end of the world, but the plating, I'm wondering how that one got missed.

We re-machined one of the cavities in a post production operation, reason why it is not plated.

BTW, I found a great product to erase minor surface scratches from the Delrin housing: brake cleaner.. it ever so slightly etches the skin of the Delrin, and gives it a very nice and smooth finish.

[EnJoY]

We re-machined one of the cavities in a post production operation, reason why it is not plated.

BTW, I found a great product to erase minor surface scratches from the Delrin housing: brake cleaner.. it ever so slightly etches the skin of the Delrin, and gives it a very nice and smooth finish.

Brake cleaner is powerful stuff. For anyone who tries this, be sure not to get any on the rubber o-rings or any rubber seals. It's only safe on metals and strong plastics, such as Delrin.

[gabe]

Brake cleaner is powerful stuff. For anyone who tries this, be sure not to get any on the rubber o-rings or any rubber seals. It's only safe on metals and strong plastics, such as Delrin.

:nuts: am using brake cleaner now... :nuts: should also mention that ... :nuts: good room ventilation is needed... :nuts:

just kidding.. but here is the technique: spray some Brake Cleaner on cloth, and then wipe the Delrin the surface...

[Boulard83]

Heres my block.

http://www.infodupat.com/Cosmos/Komodo004.JPG

[EnJoY]

:nuts: am using brake cleaner now... :nuts: should also mention that ... :nuts: good room ventilation is needed... :nuts:

just kidding.. but here is the technique: spray some Brake Cleaner on cloth, and then wipe the Delrin the surface...

Haha, yes. And you might want to wear gloves to as it can irritate your skin and you definitely do not want it in your eyes either. Then again, I probably have about a gallon of it in my blood stream from working on cars over the years.

[gabe]

Heres my block.

http://www.infodupat.com/Cosmos/Komodo004.JPG

..and you can see the cavity we re-machined and that is not plated on the left.

[Boulard83]

Im very happy with it Gabe. Install was EZ and straight foward. Ill try it with paste on the Vreg on my next Flush/Refill.

If i do this, i only remove the 2 tape on the extrem left and use paste ? ( MX-3 in my case )

[gabe]

Im very happy with it Gabe. Install was EZ and straight foward. Ill try it with paste on the Vreg on my next Flush/Refill.

If i do this, i only remove the 2 tape on the extrem left and use paste ? ( MX-3 in my case )

that's right. it's also documented in the install guide..

[Boulard83]

that's right. it's also documented in the install guide..

Just wanted to be sure. The contact gona be OK without the tape ( thickness ) ?

[gabe]

Haha, yes. And you might want to wear gloves to as it can irritate your skin and you definitely do not want it in your eyes either. Then again, I probably have about a gallon of it in my blood stream from working on cars over the years.

almost forgot one more: don't smoke when you are spraying or :explode2: :zombie:

[gabe]

Just wanted to be sure. The contact gona be OK without the tape ( thickness ) ?

yes, according to our tests it is much better.

[Boulard83]

Thanks Gabe !

[dejanh]

Thanks for the heads up. I was wondering how that one got missed :p:

[Philwong]

Time to report Komodo's performance

Idle (rad fans throttling @ 1.1 to 1.2K)
GPU/MEM/Shader/VRM: 40c/40c/44c/44c

Furmark (rad fans @ 1.3 to 1.5K)
GPU/MEM/Shader/VRM: 52c/60c/65c/86c

I'm still using thermal pad for VRM because it's too troublesome for me to switch over to paste.

The high temp is not exactly a big deal since the VRM usually hovered around 60c when I play MW2 and Dirt 2.

Phil

[Boulard83]

I have similar VRM temp with the pads. But i barely brake ~41°c under furmark. Ambiant ~25°c and water ~31°c after 10 min of furmark at 950/1300 1.175c.

[Philwong]

I have similar VRM temp with the pads. But i barely brake ~41°c under furmark. Ambiant ~25°c and water ~31°c after 10 min of furmark at 950/1300 1.175c.

My ambient and water temps are 30+c and 42c, respectively. The cooling performance is also dependent on individual setup. Mine is far from optimal due to space and layout constraints.

Phil

[Badfastbusa]

Got my Komodo in today... Man what a sexy block!!!

[dejanh]

Hey Gabe, I have a question for you...what length of Bitspower direct connect do I need for connecting two of these in CrossFireX on EVGA Classified E759 (NF200 version)? I measured things and it looks like I need the 25mm connector (http://www.sidewindercomputers.com/bishg14simid.html), but I am not 100% sure. This also matches what you have printed here http://swiftnets.com/products/Komodo-gtx285.asp.

Thanks in advance.

Edit: Well, I just put in an order for the 25mm direct connect fittings. I read your page, measured, then measured again, then calculated, then calculated again and I always come out to ~25mm. I really hope I got this right.

[EnJoY]

almost forgot one more: don't smoke when you are spraying or :explode2: :zombie:

I disagree. The fire is part of the fun. ;)

[dejanh]

yes, according to our tests it is much better.

Hmmm....well...

I recently bought two Komodo blocks and installed one on one of my HD 5850 card to do some overclock testing while I am waiting for my second HD 5850.

I went straight for the Advanced Thermal Utilization step seen on page 5 here http://swiftnets.com/products/installation_guide_KOMODO-HD5800-Multilangue.pdf because I did not even want to bother with the thermal pads but I am getting awful VRM temps and I think I know why...the block does not seem to be making adequate contact with the VRMs at all, but there is no way to screw in the block any further without cutting the fasteners to shorten them.

At 1.225V and 1000/1250 overclock running Furmark the VRM temperatures go over 100C. At 1.187V and 975/1250 overclock the VRM temperatures hit high 90s. I did not dare let it run to see what the max temperatures would come to, but all I know is that they kept on rising when I stopped the test. The stock air cooler kept the VRM temperatures to ~80C at these voltages and clocks. GPU temps themselves are excellent topping out in mid 30s.

Any ideas what to do? As it is right now I have no choice but to back down the overclock to get the temperatures on the VRMs within working range (i.e., below 100C).

[Badfastbusa]

Hmmm....well...

I recently bought two Komodo blocks and installed one on one of my HD 5850 card to do some overclock testing while I am waiting for my second HD 5850.

I went straight for the Advanced Thermal Utilization step seen on page 5 here http://swiftnets.com/products/installation_guide_KOMODO-HD5800-Multilangue.pdf because I did not even want to bother with the thermal pads but I am getting awful VRM temps and I think I know why...the block does not seem to be making adequate contact with the VRMs at all, but there is no way to screw in the block any further without cutting the fasteners to shorten them.

At 1.225V and 1000/1250 overclock running Furmark the VRM temperatures go over 100C. At 1.187V and 975/1250 overclock the VRM temperatures hit high 90s. I did not dare let it run to see what the max temperatures would come to, but all I know is that they kept on rising when I stopped the test. The stock air cooler kept the VRM temperatures to ~80C at these voltages and clocks. GPU temps themselves are excellent topping out in mid 30s.

Any ideas what to do? As it is right now I have no choice but to back down the overclock to get the temperatures on the VRMs within working range (i.e., below 100C).

Something is definitely wrong, you might try a reseat or something...

My best I have got so far stable is 1081/1306 at 1.35v... My core maxed out at 35*C and my hottest VRM got to 65*C... I think I will stick with this overclock, I don't want to push my luck with BC2 coming out tomorrow hehe, its stable, and its a pretty respective overclock IMO.

It still isn't breaking a sweat yet, so i'm sure it will go further. :)

BTW, this is at 75.8*F ambient temps, and with my san ace 1011 fans at almost silent RPM's, but I have a pretty overkill watercooling setup.

[dejanh]

Something is definitely wrong, you might try a reseat or something...

My best I have got so far stable is 1081/1306 at 1.35v... My core maxed out at 35*C and my hottest VRM got to 65*C... I think I will stick with this overclock, I don't want to push my luck with BC2 coming out tomorrow hehe, its stable, and its a pretty respective overclock IMO.

It still isn't breaking a sweat yet, so i'm sure it will go further. :)

BTW, this is at 75.8*F ambient temps, and with my san ace 1011 fans at almost silent RPM's, but I have a pretty overkill watercooling setup.

I took down the loop, removed the block and I was right, the contact with the VRMs is terrible. Only 1 VRM seems to have made good contact and all of the ones on the back (rear) of the card (5 or so) did not make any sort of contact really. I re-did all of the block fasteners/bolts to tighten them as much as physically possible, re-applied the paste, put everything together again and did not get any improvement in VRM temperatures at all. Running Furmark 1.8.0 in windowed mode at 1920x1200 with 8x AA on loads the VRMs to 100C in ~2 minutes. This is brutally bad. Check out the screenshot below...

@2 minutes hottest VRM is at 97C and climbing steadily...

http://img.techpowerup.org/100302/vrm_temps_after_remount.jpg

Gabe?

[Badfastbusa]

Did you clean the tape off really well off of the waterblock, along with the clear gooey leftover?

[dejanh]

Did you clean the tape off really well off of the waterblock, along with the clear gooey leftover?

Yep all of the pad, all of the tape, all of it period. I've done it twice over, mounting problems are out of the question now. There is physically no way the block is gonna get any tighter to the graphics board.

I really want to get Gabe's comment on this but most of all I want to see what happens with the incoming Asus EAH5850. I am currently running MSI R5850. I know that there has been issues in the past with HD5970 cards where certain components on the cards sat higher/lower than usual. These did not happen to be MSI cards but still, I am not excluding anything at this point. I cannot believe that Gabe would have missed this in design so I want to get his comment...then again, who knows. I will have to measure the thickness of both blocks now too, the one that is mounted and the one that is in the box still.

I wonder if anyone else with the Komodo and paste instead of the tape would be willing to do some Furmark 1.8.0 action at 1920x1200 with 8x AA. Just to point out too that running Kombustor instead of Furmark is not the same thing. The electrical current (amp) load on the VRMs is much higher with Furmark 1.8.0 and above settings than that of Kombustor. At least that's what I have observed.

Edit: I powered down the system and measured the thickness of the mounted block with the standoff vs. the thickness of the block with the standoff that is still in the box. There is a total of 0.35mm-0.4mm difference in thickness between the two blocks. This could be the reason for such awful VRM temps. The mounted block is ~15.55mm thick measured with a digital caliper from the top side of the block to the top of the standoff, and the one that is still in the box measures ~15.2mm from the top side of the block to the top of the standoff. Considering that we are dealing with very fine measurements here this could make all the difference between the VRMs cooling or not. Also note that I re-did the measurements several times and they always come out to within 0.05mm so the difference of >0.3mm is definitely there.

[gabe]

Yep all of the pad, all of the tape, all of it period. I've done it twice over, mounting problems are out of the question now. There is physically no way the block is gonna get any tighter to the graphics board.

I really want to get Gabe's comment on this but most of all I want to see what happens with the incoming Asus EAH5850. I am currently running MSI R5850. I know that there has been issues in the past with HD5970 cards where certain components on the cards sat higher/lower than usual. These did not happen to be MSI cards but still, I am not excluding anything at this point. I cannot believe that Gabe would have missed this in design so I want to get his comment...then again, who knows. I will have to measure the thickness of both blocks now too, the one that is mounted and the one that is in the box still.

I wonder if anyone else with the Komodo and paste instead of the tape would be willing to do some Furmark 1.8.0 action at 1920x1200 with 8x AA. Just to point out too that running Kombustor instead of Furmark is not the same thing. The electrical current (amp) load on the VRMs is much higher with Furmark 1.8.0 and above settings than that of Kombustor. At least that's what I have observed.

Edit: I powered down the system and measured the thickness of the mounted block with the standoff vs. the thickness of the block with the standoff that is still in the box. There is a total of 0.35mm-0.4mm difference in thickness between the two blocks. This could be the reason for such awful VRM temps. The mounted block is ~15.55mm thick measured with a digital caliper from the top side of the block to the top of the standoff, and the one that is still in the box measures ~15.2mm from the top side of the block to the top of the standoff. Considering that we are dealing with very fine measurements here this could make all the difference between the VRMs cooling or not. Also note that I re-did the measurements several times and they always come out to within 0.05mm so the difference of >0.3mm is definitely there.

I am out of the country for 2 weeks travelling.. Will try to get back to you asap in between two meetings..

[dejanh]

I am out of the country for 2 weeks travelling.. Will try to get back to you asap in between two meetings..

Thank you sir :up:

[Boulard83]

Im also waiting for this Gabe ... im on thermal tape and the temp are so-so on the VRM ... i wish to switch to Paste but im afraid now ...

[dejanh]

Im also waiting for this Gabe ... im on thermal tape and the temp are so-so on the VRM ... i wish to switch to Paste but im afraid now ...

Don't do it. My card is now toast too. I'll show tomorrow that this does not work 100%(explanation + pics).

[Philwong]

Don't do it. My card is now toast too. I'll show tomorrow that this does not work 100%(explanation + pics).

Sorry to hear that, but the new 5850 and 5870 are supposed to have built in overheating protection against killer software such as Furmark. Plus honestly speaking you are pushing the card to its limit at such high resolution and AA settings.

One of the workarounds is to file the standoff to improve contact, but then it might warp the card in the long run as well. There's a reason why most full face blocks use thermal pad and that's to buffer for the different height tolerances.

Phil

[dejanh]

Plus honestly speaking you are pushing the card to its limit at such high resolution and AA settings.

What is this, http://www.moderatesystems.org?

I doubt the card has protection in that sense since I doubt VRM would be allowed to hit 110C+. Anyway, what killed the card is the mounting, and remounting, and remounting in hopes to get data for Gabe and to solve this problem. Besides, nothing that is in the installation manual is out of line in terms of proper installation. Unfortunately, I have two Komodo blocks and now a dead card for which I can show that paste method does not work. The uneaven thickness of the block, height of stanoffs, etc., all play a part and when you are dealing with measurements as small as fractions of a millimeter something being off by 0.1mm or 0.2mm or more is significantly off. I will have the pictures up very soon with some data. It's unfortunate I had to sacrifice a card to get this. :shakes:

[gergregg]

Bummer about the card, but good information to know.

[Badfastbusa]

What is this, http://www.moderatesystems.org?

I doubt the card has protection in that sense since I doubt VRM would be allowed to hit 110C+. Anyway, what killed the card is the mounting, and remounting, and remounting in hopes to get data for Gabe and to solve this problem. Besides, nothing that is in the installation manual is out of line in terms of proper installation. Unfortunately, I have two Komodo blocks and now a dead card for which I can show that paste method does not work. The uneaven thickness of the block, height of stanoffs, etc., all play a part and when you are dealing with measurements as small as fractions of a millimeter something being off by 0.1mm or 0.2mm or more is significantly off. I will have the pictures up very soon with some data. It's unfortunate I had to sacrifice a card to get this. :shakes:

Dude, that freaking sucks to hear man, that is a ton of money wasted, and I am sure you are very unhappy, especially seeing as BFBC2 is out now and is epic and you can't even play it... However, I haven't had any troubles out of mine, and getting absolutely great temps... I followed the instructions to a T and everything has worked flawlessly... I ran furmark with the exact specs you wanted me to run it and my temps didn't ever get out of hand, so I don't know what to tell ya man...

It would be cool if gabe would throw his insight in on this and maybe give you a helping hand.

[Philwong]

What is this, http://www.moderatesystems.org?


Unfortunately, the Komodo is not the most "extreme" block as far as VRM temp is concerned. It's all about expectation and you ought to be looking at EK for the lowest possible temp. Like I mentioned earlier, if Swiftech can bring down the temp by such a huge degree just by using paste, there's no reason why EK will not benefit from this tweak as well.

Phil

[dejanh]

Unfortunately, the Komodo is not the most "extreme" block as far as VRM temp is concerned. It's all about expectation and you ought to be looking at EK for the lowest possible temp. Like I mentioned earlier, if Swiftech can bring down the temp by such a huge degree just by using paste, there's no reason why EK will not benefit from this tweak as well.

Phil

My comment was only intended as a rebuttal to your statement of pushing the card to the extreme. I hardly pushed the card to the extreme, and personally I even stated that I do not think the card died because of VRM temperatures but rather all of the mounting and remounting. Anyway, the main reason I took everything down again last night is so that I can get data for Gabe to show him the problem. I got the data, but unfortunately I also got a dead card now. I had no choice anyway though but to take down the loop again last night since the VRM temperatures were out of control. I was intending to replace the paste with pads again but once everything was done the card would no longer start. I'm loading the images and doing the write up now, they will be up shortly.

Edit: @Philwong, I forgot to mention something. Furmark got nerfed by ATI in their drivers after I reported a problem with it overheating the VRMs at launch of HD 4800 series cards several years ago. I realize that Furmark is not representative of normal load however it is a great test tool and shows problems quickly. After the VRM debacle that happened with the HD 4800 series cards I did not expect a repeat of the problem, and as you can see below I was right. Furmark or not, the block is not making contact as it should be so the VRMs burn, burn, burn :D :up:

[dejanh]

Ok, I sorted through the photos and my notes and I think this one illustrates the best the lack of contact. I needed to put on a lot of paste to get anything to come across onto the VRM cooling area. The block was compressed as much as it can be to the graphics card.

http://img.techpowerup.org/100303/P3020553_small.jpg

Notice that the GPU die and the single VRM in the lower right corner have good compression, but the VRMs at the rear of the card barely make contact, and definitely do not have any compression.

And another screenshot following maximum compression of block to card, this time with abut 1/2-3/4 of the thermal paste; notice one of the VRMs has no more contact at all. Also notice the scratches around standoffs as a result of trying to tighten them down as much as possible to get the block to make proper contact in vain.

http://img.techpowerup.org/100303/P3020555.jpg

Now, I have measured the PCB height, VRM height, standoff height when fully screwed in and standalone, VRM plateau height, and determined that the VRMs cannot be cooled with the paste method on either of my Komodo blocks with this card. All measurements have a margin of error +/- 0.05mm due to human error and were done with a digital caliper.

PCB, VRM dimensions:

PCB thickness: 1.55mm - 1.6mm
VRM (rear) thickness: 0.85mm
VRM (front, single) thickness: 1.1mm

Block dimensions, around rear VRM area:

Top-left corner + standoff: 15.30mm
Bottom-left corner + standoff: 15.1mm
VRM plateau height (average): 14.15mm
Difference: 0.95mm - 1.15mm

As you can tell from these dimensions it is not possible to make good contact with the VRM using paste only as the difference between the VRM thickness and where the plateau sits once mounted ranges from 0.1mm to as much as 0.3mm.

The second block is more or less the same, having dimensions 15.05mm/14.75mm/13.9mm (avg); net difference 0.85mm - 1.15mm. Furthermore, on the second block even the single VRM would not make proper contact as the net difference in height between the standoff and plateau height is 1.2mm - 1.3mm, larger than 1.1mm height of that VRM.

Furthermore, the screw-in depth for standoffs and standoffs themselves are not equal. They range in height from 3.85mm to 4.1mm in no practical order. The screw in depths appear to do the same and are uneven. I took a 4.1mm standoff and placed it into the center hole by the DRAM cooling pad near the inlet/outlet and the PCB would still warp here because the screw in depth was deeper by about 0.3mm than other areas. This would happen every time but if the fastener was loosened to eliminate the warping insufficient contact with the DRAM would be made. Finally, around the single VRM area the board warps significantly once the card is mounted in the case as the perpendicular forces exerted by the weight of the block pulling down and away from the mounting slots on the PCB.

Two blocks later I would say that using this block without thermal pads and instead using thermal paste is strongly not recommended, at least with MSI R5850. I will test things out on the Asus EAH5850 as well once it is here in a few more days. The MSI R5850 is now off for an RMA and I really, really hope I actually get the replacement.

[Badfastbusa]

I would take mine apart and see what my results would be but I would be afraid that I would never get it back as good as it is now... Maybe we got blocks from different batches of blocks and your batch was put into a jig wrong or something... Sucks either way man, and hopefully they make it right with you.

[dejanh]

I would take mine apart and see what my results would be but I would be afraid that I would never get it back as good as it is now... Maybe we got blocks from different batches of blocks and your batch was put into a jig wrong or something... Sucks either way man, and hopefully they make it right with you.

I can see the operator in China right now going "Eh, it's off by 0.2mm, nobody will notice" :cool:

Two blocks bought at very different times both have same issues though more less, that's the really puzzling part. So either it will turn out that the MSI cards are different by design or there is a very low level of accuracy when manufacturing these in which case using paste is out of the question. Funny thing is had I not opted to follow the "advanced thermal utilization" I would have an alive card now :shakes:

In terms of taking yours apart, I strongly advise against it. One you are risking damaging the card, and two you may not get good contact again, though it does seem that your block fits better. Maybe it is because you are using HD 5870, I have no idea.

At least I managed to get all the relevant info (I think) for Gabe before the card died. That's something :wave:

[Badfastbusa]

I can see the operator in China right now going "Eh, it's off by 0.2mm, nobody will notice" :cool:

Two blocks bought at very different times both have same issues though more less, that's the really puzzling part. So either it will turn out that the MSI cards are different by design or there is a very low level of accuracy when manufacturing these in which case using paste is out of the question. Funny thing is had I not opted to follow the "advanced thermal utilization" I would have an alive card now :shakes:

In terms of taking yours apart, I strongly advise against it. One you are risking damaging the card, and two you may not get good contact again, though it does seem that your block fits better. Maybe it is because you are using HD 5870, I have no idea.

At least I managed to get all the relevant info (I think) for Gabe before the card died. That's something :wave:

I anxiously await gabes response to this. :yepp:

[gergregg]

I have a question for you. The differences in the blocks is definitely not acceptable, but if your going to remove thermal pads, wouldn't you want to remove the rest and use TIM there also? By leaving the other ones on there you are widening the distance from the block to the VRM's by the thickness of the pads.

The odd thing is looking at your picture the pad above the VRM's has almost no mark on them, so even it wasn't really making good contact. No pressure on that end of the block. Card or block that is the million dollar question?

[Philwong]

My comment was only intended as a rebuttal to your statement of pushing the card to the extreme. I hardly pushed the card to the extreme, and personally I even stated that I do not think the card died because of VRM temperatures but rather all of the mounting and remounting. Anyway, the main reason I took everything down again last night is so that I can get data for Gabe to show him the problem. I got the data, but unfortunately I also got a dead card now. I had no choice anyway though but to take down the loop again last night since the VRM temperatures were out of control. I was intending to replace the paste with pads again but once everything was done the card would no longer start. I'm loading the images and doing the write up now, they will be up shortly.

Edit: @Philwong, I forgot to mention something. Furmark got nerfed by ATI in their drivers after I reported a problem with it overheating the VRMs at launch of HD 4800 series cards several years ago. I realize that Furmark is not representative of normal load however it is a great test tool and shows problems quickly. After the VRM debacle that happened with the HD 4800 series cards I did not expect a repeat of the problem, and as you can see below I was right. Furmark or not, the block is not making contact as it should be so the VRMs burn, burn, burn :D :up:

If running Furmark at such extreme settings is not pushing your card to the limit with all the warnings of average VRM temps delivered by the Komodo, I don't know what else to say. I will be pissed too if I zapped my card so I do feel for you. But what else can we Komodo owners do when the coolant is not flowing anywhere close to the VRMs? It's not some rocket science that the Ek shines in this area, while the Swifty aced in the GPU department with its micro pins.

Phil

[dejanh]

I have a question for you. The differences in the blocks is definitely not acceptable, but if your going to remove thermal pads, wouldn't you want to remove the rest and use TIM there also? By leaving the other ones on there you are widening the distance from the block to the VRM's by the thickness of the pads.

The odd thing is looking at your picture the pad above the VRM's has almost no mark on them, so even it wasn't really making good contact. No pressure on that end of the block. Card or block that is the million dollar question?

Pads do not control the distance, standoffs do. The block was fully compressed down until it hit the standoffs on every mount. The result you see is after maximum compression. What you are saying would have made sense if the standoffs are not being employed, but they are :)


If running Furmark at such extreme settings is not pushing your card to the limit with all the warnings of average VRM temps delivered by the Komodo, I don't know what else to say. I will be pissed too if I zapped my card so I do feel for you. But what else can we Komodo owners do when the coolant is not flowing anywhere close to the VRMs? It's not some rocket science that the Ek shines in this area, while the Swifty aced in the GPU department with its micro pins.

Phil

Reading comprehension problem? I am pointing out that Komodo paste method is not safe as it may appear and does not work on every card. My card dying was an unfortunate part of the discovery process. Yes, I am pissed about it, but clearly not as upset as you :rolleyes:

[Philwong]

Reading comprehension problem? I am pointing out that Komodo paste method is not safe as it may appear and does not work on every card. My card dying was an unfortunate part of the discovery process. Yes, I am pissed about it, but clearly not as upset as you :rolleyes:

There's nothing for me to be upset about and the Komodo works within my expectations even on thermal pads. You're missing my point so far, but to each of his own.

Phil

[dejanh]

There's nothing for me to be upset about and the Komodo works within my expectations even on thermal pads. You're missing my point so far, but to each of his own.

Phil

What is your point exactly, because I think everyone is missing your point...

Let's reiterate...

1. Komodo is not the best block for cooling VRM in default configuration - fact.
2. According to Swiftech Komodo can experience 20C - 30C improved VRM temperatures by switching from pad to paste - fact.
3. If (2) is true then Komodo can actually cool VRMs pretty well. This has luckily been confirmed by at least some people (e.g., Badfastbusa) - fact.

Now, why would I want to use pads then that get sub-par performance and not go straight away to paste, after all, it is right in the installation instructions for Komodo, you know, the ones that came printed with the block and that are posted on Swiftech website? Yes, those ones. Well that is were the problems started because as I have now shown beyond any doubt this cannot be done on every card and/or with every Komodo block as the dimensions of block/card/standoffs simply do not allow it. Unfortunately for me while I was troubleshooting this my card died, quite possibly from mounting the block over and over and over again combined with excessive VRM temperatures experienced while the thermal interface was malfunctioning.

So again, please enlighten me, what is your point :shrug:

[Boulard83]

I fell bad for you dejanh :( Very bad :(

Here is a shot of a 3min Furmark.
Asus EAH5870. Komodo with Thermal pads on VRM. ( well thighten VRM bolts. )
http://www.infodupat.com/pics/Furmark_950_1300.jpg

[Philwong]

1. Komodo is not the best block for cooling VRM in default configuration - fact.
2. According to Swiftech Komodo can experience 20C - 30C improved VRM temperatures by switching from pad to paste - fact.
3. If (2) is true then Komodo can actually cool VRMs pretty well. This has luckily been confirmed by at least some people (e.g., Badfastbusa) - fact.


Exactly. We all know the Komodo is just a mediocre performer for cooling VRM. Yet, you still insist on stress testing your card with the latest "enhanced heat output" Furmark at such dangerous level. Let me reiterate, that's 1,900 x 1,200 with 8x AA. This has nothing to do with your overclocking in this case.



After all, it is right in the installation instructions for Komodo, you know, the ones that came printed with the block and that are posted on Swiftech website?

Can we take everything at face value liberally just because it's documented in the user manual? The original instructions caused the infamous black screen issue for some of the early adopters including myself.

Plus 20 to 30c reduction at what kind of card and simulation settings? Typical or extreme? Just look at the frame rate of the screen caps from Gabe, other Komodo users and yours for starters.

http://www.swiftnets.com/assets/images/products/komodo/KOMODO-HD5850-2.jpg
Gabe's testing

http://img.techpowerup.org/100302/vrm_temps_after_remount.jpg
dejanh

Here's the thread on in-house testing performed by Swiftech.
http://www.xtremesystems.org/forums/showthread.php?t=243227&page=3&highlight=5970


Quite possibly from mounting the block over and over and over again combined with excessive VRM temperatures experienced while the thermal interface was malfunctioning.

The extreme Furmark's setting probably has a hand in causing the malfunction as well. How many times did you install the block, too? 10, 20 or 30? I seriously doubt the failure is due to repeated installations.

I know you're upset, but it will help if you can analyze the whole incident calmly. The mounting problem is a genuine concern which Gabe has to address, but stress testing the block at such extreme level is not exactly going to prevent your card from frying. :rolleyes:

Phil

[dejanh]

Phil, listen, I do not know how to put it to you more simply...

I found that paste/grease/whatever-you-want-to-call-it method does not work "as advertised". With some blocks/cards/combination there of it does not work period. I do not know how else to put this to you so that you will understand what I am trying to show Gabe.

If you still feel like pushing whatever indeterminate point you are pushing then please stop, I am not interested in hearing your opinion any more. At this point unless you want to do the same tests as me and show that the paste method works perfectly for you with stress test screenshots and block pics and all, or not, please refrain from posting back. Why did you bother commenting on the shorting issue then? Or are you suggesting we should just keep quiet when a problem is found. The problem with the paste method is crystal clear, from pictures, from measurements, from explanation.

I'll leave it up to Gabe now to comment.

As far as my card goes, remounting was done about 5 times total. Did it have something to do with the card dying, probably. Am I blaming the block for "killing" the card directly, no. At no point in time did I come into this thread whining how "omfg, my Komodo killed my card". I would almost 100% certainly say it is a combination of the stress the card underwent while diagnosing this problem and mounting and remounting. Either way at the end of the day I found a problem with the Komodo block (possibly in combination with at least some cards) and if I can help one person think twice about using the paste or understand that using paste may not get them the desired effect then I have done my part for the community. After all, you yourself said instructions are not foolproof.

[Badfastbusa]

Phil, listen, I do not know how to put it to you more simply...

I found that paste/grease/whatever-you-want-to-call-it method does not work "as advertised". With some blocks/cards/combination there of it does not work period. I do not know how else to put this to you so that you will understand what I am trying to show Gabe.

If you still feel like pushing whatever indeterminate point you are pushing then please stop, I am not interested in hearing your opinion any more. At this point unless you want to do the same tests as me and show that the paste method works perfectly for you with stress test screenshots and block pics and all, or not, please refrain from posting back. Why did you bother commenting on the shorting issue then? Or are you suggesting we should just keep quiet when a problem is found. The problem with the paste method is crystal clear, from pictures, from measurements, from explanation.

I'll leave it up to Gabe now to comment.

As far as my card goes, remounting was done about 5 times total. Did it have something to do with the card dying, probably. Am I blaming the block for "killing" the card directly, no. At no point in time did I come into this thread whining how "omfg, my Komodo killed my card". I would almost 100% certainly say it is a combination of the stress the card underwent while diagnosing this problem and mounting and remounting. Either way at the end of the day I found a problem with the Komodo block (possibly in combination with at least some cards) and if I can help one person think twice about using the paste or understand that using paste may not get them the desired effect then I have done my part for the community. After all, you yourself said instructions are not foolproof.

Unfortunately, some people here, no matter how many times or different ways you say something, just dont "get it" for some reason... In my worklog thread, I had to say about 6 different times that my overclocks were just for fun and not to be meant for 24/7 stable clocks, and one after another, they each wanted a ton of stability tests LMAO

In conclusion, there are some really smart people here:up:, along with some real idiots:down:. :yepp:

[Philwong]

Phil, listen, I do not know how to put it to you more simply...

I found that paste/grease/whatever-you-want-to-call-it method does not work "as advertised". With some blocks/cards/combination there of it does not work period. I do not know how else to put this to you so that you will understand what I am trying to show Gabe.

If you still feel like pushing whatever indeterminate point you are pushing then please stop, I am not interested in hearing your opinion any more. At this point unless you want to do the same tests as me and show that the paste method works perfectly for you with stress test screenshots and block pics and all, or not, please refrain from posting back. Why did you bother commenting on the shorting issue then? Or are you suggesting we should just keep quiet when a problem is found. The problem with the paste method is crystal clear, from pictures, from measurements, from explanation.

I'll leave it up to Gabe now to comment.

As far as my card goes, remounting was done about 5 times total. Did it have something to do with the card dying, probably. Am I blaming the block for "killing" the card directly, no. At no point in time did I come into this thread whining how "omfg, my Komodo killed my card". I would almost 100% certainly say it is a combination of the stress the card underwent while diagnosing this problem and mounting and remounting. Either way at the end of the day I found a problem with the Komodo block (possibly in combination with at least some cards) and if I can help one person think twice about using the paste or understand that using paste may not get them the desired effect then I have done my part for the community. After all, you yourself said instructions are not foolproof.

Since when have I contest the mounting problem or have you been reading between lines? Like it or not this is still an open forum and you're most welcome to ignore my posts if you desires. For those who prefer to leave sarcastic remarks, pls go ahead and stress test your Komodo at high reso and AA, then whine that temps are bad for all you want . S-h-i-t happens when you play with fire. Remember to do your homework and buy the right block next time. It's not as if there's no obvious hint that the Komodo does not stack up well in VRM cooling. Period.

Phil

[Boulard83]

Unfortunately, some people here, no matter how many times or different ways you say something, just dont "get it" for some reason... In my worklog thread, I had to say about 6 different times that my overclocks were just for fun and not to be meant for 24/7 stable clocks, and one after another, they each wanted a ton of stability tests LMAO

In conclusion, there are some really smart people here:up:, along with some real idiots:down:. :yepp:

:clap:

:clap: :clap:

[gabe]

Pads do not control the distance, standoffs do. The block was fully compressed down until it hit the standoffs on every mount. The result you see is after maximum compression. What you are saying would have made sense if the standoffs are not being employed, but they are :)


sorry for not being able to get back to you faster, but I am in China right now, and facing very hectic schedule.

You are essentially correct: the key here is standoff height. So please make sure that all standoffs are completly fastened tight, particularly in the VRM's area. One or two more turns of fastening the standoff will make the difference between poor contact and good contact. We are indeed dealing with very small distances here, of the order of 0.1 to 0.2mm. anmd since we do not want to warp the PCB in excess, tolerances are tight indeed.

I officially implemented the "no pad" tweak after successfully testing it myself with 5850, 5870 and 5970, so I felt very comfortable recommending it.

BUT making sure that all standoffs are fastened tight is something that I do automatically without even thinking twice about it, and in hindsight I should have added this in the instructions. Using pads gave us sufficient margin to allow for larger tolerance gaps, but not using pads is obviously reducing the margin of tolerance that we have, and therefore absolute repect of the specified standoff height relative to PCB becomes obviously critical.

The conendrum we find ourselves into is this: using standoffs to prevent excessive warpage of the PCB, in other words user safety, and the desire to provide a perfect TIM between VRM's and copper plate. In theory, these two goals are mutually incompatible. In practice, they are possible, but reserved to advanced users who have a solid understanding of the underlying difficulties associated with this kind of mechanical challenge. This is the very reason why I included the tweak in an "advanced users" section of the installation guide.

Other vendors (who do not use standoffs) do provide a better TIM, but this is at the cost of safety. The users of these competing solutions typically qualify themselves as advanced, and so they do not really mind taking the risk. We on the other hand, want to provide safety as well as performance, and it would appear that we still have either some tweaking or better maybe explaining to do in order to accomplish this.

Doing a hard mount (without pads) is quite possible with the Komodo's, I have done it, and I'll stick to my guns about it. Performance is exceptional, and pcb warpage was minimal. Should I revise my standoff heights and therefore reduce the tolerance gap without pads is something I now need to consider carefully.

[gabe]

It's not as if there's no obvious hint that the Komodo does not stack up well in VRM cooling. Period.

Phil

Your final statement either shows a poor understanding of the thermal and mechanical requirements to cool these vrm's or an unabated support for your favorite brand, both of which are perfectly understandable and acceptable I should say. I would hope nonetheless that my response above will shed some more light for you on the problem at hands.

As vendors, we make choices as to what we want to accomplish with respect to our users. I already said and demonstrated in our lab that our solution can cool the VRM's just as well as any other extreme solution. The reality is, the heat loads for these components are such that just about ANY full-cover solution providing a decent TIM will cool them enough to be substantially below operating specs. The difference as I pointed above, is that we try really hard to do it safely.

[jaredpace]

http://img.techpowerup.org/100303/P3020553_small.jpg

Hello,

The indention you have marked, "Bad contact, no compression" with a red line with two arrows on each end, can you remove the pad sitting inside that indention to give your block more height/clearance to successfully cover the components on the pcb marked by the white sticker, "CPL2-3". Then you may have greater compression/pressure on the 5 VT1165 slaves to the left which are not being touched by your cooling block. Also, can you make your cooling block mount tighter to that area of vrm slaves?

Why not use a highly conductive thick thermal pad of good quality to mount on those 5 components. These 5 slaves need *MUCH* more cooling than the parts under the white sticker CPL2-3, which require little cooling, and are not even touching the stock cooler.

[jaredpace]

I doubt the card has protection in that sense since I doubt VRM would be allowed to hit 110C+. Anyway, what killed the card is the mounting, and remounting, and remounting

The card does indeed have throttling protection to prevent overheating, overvolting, and overcurrent. The slaves throttle at 150C, however are recommended to run <105C, and should avoid ~120C. The cooler the better. I also read that philwong recommended you to file off the "standoffs" to create a tighter connection to the pcb. This may work, AND, I have another suggestion - could you take two cut sections of the thermal pads, stack them, and firmly mount the cooler to create solid compression? Check your temps then (if your card is still working)
:eek:

[Philwong]

Your final statement either shows a poor understanding of the thermal and mechanical requirements to cool these vrm's or an unabated support for your favorite brand, both of which are perfectly understandable and acceptable I should say. I would hope nonetheless that my response above will shed some more light for you on the problem at hands.

As vendors, we make choices as to what we want to accomplish with respect to our users. I already said and demonstrated in our lab that our solution can cool the VRM's just as well as any other extreme solution. The reality is, the heat loads for these components are such that just about ANY full-cover solution providing a decent TIM will cool them enough to be substantially below operating specs. The difference as I pointed above, is that we try really hard to do it safely.

Gabe, the EK is superb in VRM cooling but is quite a hassle to install. Whereas the Komodo is quite the opposite with the added advantage of better GPU cooling. The difference here is EK is channeling coolant over the VRM, while Komodo uses micro pins for the GPU. Like I mentioned earlier, the thermal paste tweak should work for EK and its already low VRM temp as well. To me, it is a matter of personal priorities until a near-perfect block appears. For the record, I'm not affiliated with EK and neither do I own any of their products.

May I suggest you contact the German reviewer to carry out third party testing between your block and EK's using thermal paste for the VRM? Could you also share with us the resolution and AA settings used during your in-house testing and whether the heat output increases appropriately with higher Furmark settings?

In addition, it may be advisable to warm your customers that there may be contact issue when using paste instead of pads regardless of whether they are seasoned or advanced users. At least could bench or evaluate the card cautiously after installation.

Lastly, if you could revise the block and force the coolant to flow closer to the VRM, that could be the safest solution to enhance Komodo's performance.

Phil

[gabe]

Gabe, the EK is superb in VRM cooling but is quite a hassle to install. Whereas the Komodo is quite the opposite with the added advantage of better GPU cooling. The difference here is EK is channeling coolant over the VRM, while Komodo uses micro pins for the GPU. Like I mentioned earlier, the thermal paste tweak should work for EK and its already low VRM temp as well. To me, it is a matter of personal priorities until a near-perfect block appears.

May I suggest you contact the German reviewer to carry out third party testing between your block and EK's using thermal paste for the VRM?

In addition, it may be advisable to warm your customers that there may be contact issue when using paste instead of pads regardless of whether they are seasoned or advanced users. At least could bench or evaluate the card cautiously after installation.

Lastly, if you could revise the block and force the coolant to flow closer to the VRM, that could be the safest solution to enhance Komodo's performance.

Phil

Thank you for your suggestions, but your comments confirm an incorrect thermal and mechancial assessment. Again I do not blame you for such, as you lack some basic information, sot I will try to it provide below.

1/If you take the time to compare the distance between the heat sources and the coolant between said competing solution and Komodo's, you might be surprised to find that the coolant might actually flow closer to the hot spots in the Komodo.

http://www.swiftnets.com/assets/images/products/komodo/komodo transparency.jpg

2/ Nonetheless, irrespective of the preceding statement, it really doesn't matter because the heat load doesn't warrant immediate coolant proximity to do the job well above and beyond what is necessary (which I would point out completely redeems said competing solution from my previous statement)

3/ The key element in the whole affair is not thermal but mechanical: the TIM joint is the key element here, not the thermal design.

So now you should understand why a solution favoring TIM at the detriment of safety presents an advantage. Again, as cooling solution designers we all have the freedom to make a choice in our design considerations, and as a matter of design philosophy, I choose safety first, AND performance second.

[Philwong]

http://www.swiftnets.com/assets/images/products/komodo/komodo transparency.jpg


Thanks for clarifying.

http://www.ekwaterblocks.com/uploads/images/EK-FC5870_detail-600(1).jpg
Here's a closeup of the EK block. Check out the "step" visible from this particular angle.

Phil

[gabe]

Here's a closeup of the EK block. Check out the "step" visible from this particular angle.

Phil It is difficult to tell without a transparent view in overlay on the board, but the step I see there appears to be for clearance of the inductors, which would place the vrm's East of that, thus at the very edge of the flow path. But really it doesn't matter one way or the other. The flow path in these, and other blocks is close enough to provide more than ample cooling to the VRM's.

The difference in performance is directly linked to the TIM joint, and the quality of the TIM joint is directly linked to the gaps that are been allowed by design. I will go one step further, if you reduce tolerance gaps with the VRM's, you take the risk to kill the GPU TIM.

[Boulard83]

gabe, im 100% satisfied with my KOMODO, but im afraid to try paste on VRMs ...

[Badfastbusa]

I am 100% happy with my Komodo as well, and I am using the paste on the VRMs with excellent results... Even at full load with furmark or MSI Kombustor my loaded temps never get over 32*C on the core and 52*C on the VRMs running it at 1181/1274 with 1.35v... In game (BFBC2 and COD4) my VRMs never get over 42-44c, and my core never gets over 30*c.

[Boulard83]

I am 100% happy with my Komodo as well, and I am using the paste on the VRMs with excellent results... Even at full load with furmark or MSI Kombustor my loaded temps never get over 32*C on the core and 52*C on the VRMs running it at 1181/1274 with 1.35v... In game (BFBC2 and COD4) my VRMs never get over 42-44c, and my core never gets over 30*c.

Room ?

I guess ~20

[Badfastbusa]

75*f

[dejanh]

sorry for not being able to get back to you faster, but I am in China right now, and facing very hectic schedule.

You are essentially correct: the key here is standoff height. So please make sure that all standoffs are completly fastened tight, particularly in the VRM's area. One or two more turns of fastening the standoff will make the difference between poor contact and good contact. We are indeed dealing with very small distances here, of the order of 0.1 to 0.2mm. anmd since we do not want to warp the PCB in excess, tolerances are tight indeed.

I officially implemented the "no pad" tweak after successfully testing it myself with 5850, 5870 and 5970, so I felt very comfortable recommending it.

BUT making sure that all standoffs are fastened tight is something that I do automatically without even thinking twice about it, and in hindsight I should have added this in the instructions. Using pads gave us sufficient margin to allow for larger tolerance gaps, but not using pads is obviously reducing the margin of tolerance that we have, and therefore absolute repect of the specified standoff height relative to PCB becomes obviously critical.

The conendrum we find ourselves into is this: using standoffs to prevent excessive warpage of the PCB, in other words user safety, and the desire to provide a perfect TIM between VRM's and copper plate. In theory, these two goals are mutually incompatible. In practice, they are possible, but reserved to advanced users who have a solid understanding of the underlying difficulties associated with this kind of mechanical challenge. This is the very reason why I included the tweak in an "advanced users" section of the installation guide.

Other vendors (who do not use standoffs) do provide a better TIM, but this is at the cost of safety. The users of these competing solutions typically qualify themselves as advanced, and so they do not really mind taking the risk. We on the other hand, want to provide safety as well as performance, and it would appear that we still have either some tweaking or better maybe explaining to do in order to accomplish this.

Doing a hard mount (without pads) is quite possible with the Komodo's, I have done it, and I'll stick to my guns about it. Performance is exceptional, and pcb warpage was minimal. Should I revise my standoff heights and therefore reduce the tolerance gap without pads is something I now need to consider carefully.

Hi Gabe, thanks for taking the time to quickly reply. Unfortunately I have automatically done exactly what you have suggested. All of the observations and measurements that I have provided have been taken after tightening the standoffs as much as physically possible on the whole block, and as I have indicated previously, even replacing "longer" standoffs for "shorter" standoffs in critical areas by swapping around standoffs on the block. There is simply no way to make the VRM contact the plateau with paste. Furthermore, maximum compression of the block to the card was done in that there were no gaps at all between any standoff and the PCB, or simply put the card was as flush to the Komodo block as it can physically get.

I agree completely that you should try to revisit the standoff heights to lower the tolerance gap without pads. As it is now you have a fair amount of tolerance, at least in my testing. Furthermore, I would ensure that the machining height deviation tolerances on standoffs and VRM plateau areas be virtually eliminated as not doing so results in potential problems with contact due to the fine measurements we are dealing with here. Basically, you should strive to have all standoffs be the exact or as close to exact height that they need to be as possible as well as ensuring even and correct height of the VRM cooling plateaus of each block.

I am waiting for the RMA MSI R5850 to come back as well was for an Asus EAH5850 card to arrive. I will repeat my testing with those cards as well to ensure that this was not just a rare occurrence with some MSI R5850 cards. However, having said that this should not be taken as a reason to not address this at all. I realize that you have successfully tested this method in house, but both you and I know that not all cards will be identical with identical heights and to recommend a blanket method based on a gross assumption that it would work in every case is unrealistic at best. Even with reduced tolerances there should be a notice indicating that the performance may not be as desired.

Finally, one option that you do have is to include some longer fasteners that could be used to mount the block without standoffs and indicate that this is considered advanced installation with higher risk of damaging the card but would still leave the option for those people who may want to mount the whole block without pads and standoffs to do so. After all, in terms of its thermal design Komodo is really a great block and can pretty well be a top performer if it were not for a few of these issues.

It is imperative as you already know to be extremely careful here, no matter which path is chosen, as crushing digital VRMs is very easy to do and that's the last thing you want people to do after trying to get better temperatures out of their VRMs.


Hello,

The indention you have marked, "Bad contact, no compression" with a red line with two arrows on each end, can you remove the pad sitting inside that indention to give your block more height/clearance to successfully cover the components on the pcb marked by the white sticker, "CPL2-3". Then you may have greater compression/pressure on the 5 VT1165 slaves to the left which are not being touched by your cooling block. Also, can you make your cooling block mount tighter to that area of vrm slaves?

Why not use a highly conductive thick thermal pad of good quality to mount on those 5 components. These 5 slaves need *MUCH* more cooling than the parts under the white sticker CPL2-3, which require little cooling, and are not even touching the stock cooler.

I think you misunderstood what the issue here is actually. Komodo has no problem contacting the VRMs with the stock pad or with any other pad thicker than (I believe) about 0.3mm-0.5mm. However, as of recently Swiftech has revised their mounting instructions to include an advanced step that involves removing the thermal pads from the VRM cooling areas and replacing the same with paste for better thermal interface. However, unlike in Swiftech testing, I have not found it possible to use this method with at least one HD 5850 and two Komodo blocks as the dimensions of the standoffs do not allow for proper contact to be made between the block and the card even if fully compressed.

In terms of removing pads to give more clearance, I have already responded to this earlier, and then Gabe responded to it as well. The clearance on Komodo is not controlled by pads, it is controlled by standoffs. The only way to make the block more flush to the card is to use smaller standoffs or to file down the existing ones, both of which requires a physical change to the current design.

[gabe]

Finally, one option that you do have is to include some longer fasteners that could be used to mount the block without standoffs and indicate that this is considered advanced installation with higher risk of damaging the card but would still leave the option for those people who may want to mount the whole block without pads and standoffs to do so. After all, in terms of its thermal design Komodo is really a great block and can pretty well be a top performer if it were not for a few of these issues.


Excellent suggestion, I will look into it. I am still quite puzzled by your problem. Do you want a set of reduced height standoffs? I can't do it til i get back next week, but it's feasible. email me at gabe@swiftech.com if you do.

[dejanh]

Excellent suggestion, I will look into it. I am still quite puzzled by your problem. Do you want a set of reduced height standoffs? I can't do it til i get back next week, but it's feasible. email me at gabe@swiftech.com if you do.

Email sent. In terms of the problem it is very well possible that this was, is, and will be one of those rare outlying cases, that the card was just not quite the same as other reference HD5850 cards. There are variations between manufacturers sometimes as was for example the case with HD5970 cards. Some of those had some components higher on some PCBs than others creating a contact issue. I will redo all the tests as soon as I have the new card(s) in hand to verify the problem with other cards. In the meantime, those standoffs would be helpful since it would (a) allow me to do a proper mount even if I still have the same problem, and (b) directly compare the result with stock standoffs to reduced standoffs.

[hoodlikegaza]

any updates?

[dejanh]

any updates?

Are you asking me? If so, not yet. I am supposed to receive another MSI R5850 end of this coming week and the other card is still not certain when it will arrive.

I will definitely post updates as soon as I have cards in hand to test. Hopefully Gabe gets me some shorter standoffs or something like it in the meantime as well so that I can try that too.

:)

[hoodlikegaza]

Are you asking me? If so, not yet. I am supposed to receive another MSI R5850 end of this coming week and the other card is still not certain when it will arrive.

I will definitely post updates as soon as I have cards in hand to test. Hopefully Gabe gets me some shorter standoffs or something like it in the meantime as well so that I can try that too.

:)

yeah, lol. Thanks, I'm waiting to see what your results are before I buy my waterblock.

I have the older reference asus 5850.

[dejanh]

yeah, lol. Thanks, I'm waiting to see what your results are before I buy my waterblock.

I have the older reference asus 5850.

One new card should be delivered today, next one Friday or Monday. I will do the tests as the cards come. :)

Edit: New card delivered. Testing already done and the results are the same. I will post details in a few minutes. I have another card coming in 2-4 days but I won't bother redoing the tests for the 3rd time with that card. This is now two cards, two blocks, all randomly selected both with the exact same issue.

Edit 2: "Few minutes" will be a bit longer. I am in the middle of doing some testing on the Cat. 10.3a release.

Edit 3: Here are the pics. I did not bother circling anything on them this time. It is pretty obvious that no contact is made. I accidentally scratched one of the pads on one block but that's not really relevant :)

No contact with the main VRM...

http://img.techpowerup.org/100318/P3170572.jpg

Single VRM, contact is there as was the case with the other card...this is the only VRM that makes contact with paste and it shows how contact should look like. First the card...

http://img.techpowerup.org/100318/P3170574.jpg

...then the block

http://img.techpowerup.org/100318/P3170577.jpg

And now the second block...again no contact with the main VRM...

http://img.techpowerup.org/100318/P3170578.jpg

And as anticipated before, second block does not even make contact with the single VRM (see my earlier post(s) where I said this will be the case)...first the card...

http://img.techpowerup.org/100318/P3170581.jpg

...and then the block...

http://img.techpowerup.org/100318/P3170583.jpg

[dejanh]

Taking into account the second set of tests and the fact that these were randomly sampled cards and blocks and that there is no way to make either fit the other properly when using paste I can confidently conclude that the paste method does not work for MSI R5850 and quite possibly other HD 5850 cards. I cannot comment on HD 5870 as I do not have any to test though if somebody wants to send me some to test I would gladly do it :) Therefore, I have to make a personal recommendation against using paste with the Komodo and HD 5850 cards until an adequate solution is found.

As I have identified before, the problem are basically the standoffs, they are simply too high/not accurately enough machined. Reducing standoff size may also require slightly shorter fasteners as they need to get bolted in fully. As it stands right now it is impossible for the VRM to make contact with the block properly.

I will keep everyone updated if I have any new results, new tests, etc., to report.

[Boulard83]

Thx dejanh ! Great post :)

[hoodlikegaza]

looks like i'll be looking at a different block.... Anyone know if the dangerden or ek blocks have this same problem?

[gergregg]

You can mount the EK block with no stand-offs at your own risk and use the paste method.

[Boulard83]

looks like i'll be looking at a different block.... Anyone know if the dangerden or ek blocks have this same problem?

My VRMS hit ~70°c (950/1300 1.175v) under Furmark with the stock thermal pads. Its something with the MSI/KOMODO combination that dont work. Maybe with other cards.

[dejanh]

looks like i'll be looking at a different block.... Anyone know if the dangerden or ek blocks have this same problem?

They may all exhibit the same type of problems unless you do not use standoffs or standoffs are machined more accurately. This is not necessarily a Swiftech only issue and it is also not necessarily an issue with every card/block combination. I made my recommendation primarily for MSI R5850 and as a secondary note suggested that problems may occur with other HD 5850 cards as well so users should be aware of it.

Note that paste installation is an advanced installation method and this block works great with the pads as well, and most importantly it does an exceptional job cooling the actual GPU core.

[Church]

Hmm. I have simple idea. - orderable standoff sets customised for those a bit different PCB cards. Can it work? Should be _way_ cheaper customising block. Or at least instructions of measurements about how much one should dremel from height of shipped ones?

[dejanh]

Hmm. I have simple idea. - orderable standoff sets customised for those a bit different PCB cards. Can it work? Should be _way_ cheaper customising block. Or at least instructions of measurements about how much one should dremel from height of shipped ones?

I'm going to be testing reduced standoffs in the next week to two weeks. I will post results back as I have them. :)

[Prometheus2010]

It's way too late now. I found this thread after I installed everything "advance users" way. :(

I'm having the same exact bad VRM temp problem during overclock.

Asus 5850 Voltage tweak edition

http://i577.photobucket.com/albums/ss219/thunderbluff_prom/WC003.jpg (http://s577.photobucket.com/albums/ss219/thunderbluff_prom/?action=view&current=WC003.jpg)

http://i577.photobucket.com/albums/ss219/thunderbluff_prom/WC009.jpg (http://s577.photobucket.com/albums/ss219/thunderbluff_prom/?action=view&current=WC009.jpg)

@ Dejanh

Are you getting the shorter standoffs from Gabe or DIY?

[Badfastbusa]

maybe its just 5850's???

[johnnymossville]

I've got a 5870 at home and a komodo waiting to go on it. I've been reading this thread with a lot of interest, to put it mildly.

I've got some more parts coming for this build so it'll be a couple weeks before I can find out how it works.

Good thread guys.

[Boulard83]

WIch brand is your HD5870 ?

Your going to try the Paste mount or not ?

[johnnymossville]

WIch brand is your HD5870 ?

Your going to try the Paste mount or not ?

Powercolor 5870, and I'm going to Paste mount it.

[Boulard83]

Thanks. Come back with results :)

[dejanh]

It's way too late now. I found this thread after I installed everything "advance users" way. :(

I'm having the same exact bad VRM temp problem during overclock.

Asus 5850 Voltage tweak edition

@ Dejanh

Are you getting the shorter standoffs from Gabe or DIY?

Well, I am getting them from Gabe but they are only a test run. I will actually be testing several different sizes. I will post back results once I have them in hand and a chance to try. I think that if my tests are successful they will probably be offered mainstream :)


maybe its just 5850's???

Could very well be, though I am not sure how much difference between the layouts there is...I never owned a reference HD5870. :shrug:


Powercolor 5870, and I'm going to Paste mount it.

Curious to hear the results. Make sure that you actually check for proper compression (pics would be good like the ones I posted in this thread) and watch your VRM temps carefully :up:

[johnnymossville]

dejanh, Thanks for the tips. I'm going to be very methodical when doing this and will post my findings here.

[dejanh]

Received the standoffs. One standoff had a tapping fault and broke one of my fasteners and another standoff came loose when I was trying to remove the block after initial testing and knocked off a corner of one of the parts but no contacts were broken so the card still works...sucks though :( The rest seem fine.

Currently trying 0.5mm reduction on main VRM area and 0.25mm reduction on the single VRM at the front of the card. The temperatures are more-less under control, however by no means great. I am not seeing consistent improvement as the contact is not consistent. At 1.262V the VRM temperatures range from ~70C (decent contact) to ~100C (poor contact) So far I am seeing two issues...

(1) The fasteners cannot be fully screwed into the standoffs as they are too long for the standoff thread depth now that I am using shorter standoffs around VRM area(s).

(2) The other standoffs, particularly the one near the rear main VRM area do not allow the PCB to be compressed much further to the block resulting in uneven block contact with the VRMs.

[Prometheus2010]

@Dejanh, Thanks for the update! It looks like my last option will be replacing the thermal pads on these VRMs. I just bought EK thermal pads .5mm and 1mm thickness, not sure what's the original's, but hopefully this will bring the VRM temp back to normal. If not, then I'll have a very expensive paperweight :shakes:

[ibuycheap]

@Dejanh, Thanks for the update! It looks like my last option will be replacing the thermal pads on these VRMs. I just bought EK thermal pads .5mm and 1mm thickness, not sure what's the original's, but hopefully this will bring the VRM temp back to normal. If not, then I'll have a very expensive paperweight :shakes:

Lmao I just did the same thing. I'm not using the komodo but the swiftech hd5850 heatsink and it has the same issue. The standoffs are too high. 2 of the ram chips make little contact and the vrms barely any contact.

[dejanh]

Testing on hold again...turns out the card I just received is defective and needs to go back. It constantly keeps locking up under 3D load. Got to wait until Monday to see about getting it replaced.

This is so damn annoying now...I am getting quite irritated with all of the assembling and dissasembling. :shakes: I guess another round of taking down the loop is coming up :(

[gabe]

@Dejanh, Thanks for the update! It looks like my last option will be replacing the thermal pads on these VRMs. I just bought EK thermal pads .5mm and 1mm thickness, not sure what's the original's, but hopefully this will bring the VRM temp back to normal. If not, then I'll have a very expensive paperweight :shakes:

NO you won't.. you can always call us for support you know?

[gergregg]

NO you won't.. you can always call us for support you know?

That right there is great customer service!

[dejanh]

Look guys, let me make this very clear...

Pads work just fine on this block. The VRM temperatures with pads are far from optimal, but they are under control and can hold a very decent OC just fine. Having said that, on to my results...

Mounting the block with paste at VRM areas instead of pads does significantly reduce the VRM temperatures as stated by Gabe, up to 20C in fact. The standoff reduction that is needed around main VRM area is 0.5mm, and around the single VRM at the front of the card minimum 0.25mm. It also requires full compression of the PCB to the standoffs. However, this is not without issues.

I reported earlier that my card appears to be locking up and I thought that the card was faulty. However, it turns out that was not the case and in fact it was the reduced standoffs that were causing the problem, specifically, it was the board warping that happens as a result of the use of reduced standoffs that was causing random lockups on 1 out of 2 cards and was resulting in no video post or loss of video in some cases.

Compressing the PCB all the way down to the standoffs is a must if you want to make proper contact with the block when using paste on VRMs and this causes significant warping of the board both lengthwise and widthwise. In fact, widthwise the PCB starts to take on a bow-shape. As a result, only the outer VRMs make good contact while the ones in the center (two large and one small) make poor contact again. There is no warping around the single VRM at the front of the card. The warping is the result of two things. First, the rest of pads on the block used for memory, that concave cut near the VRMs, etc. get in the way and need to be removed and replaced with paste as well. Second, the rest of the standoffs are too tall resulting in warping and need to be equally reduced by 0.5mm as removing the rest of the pads also means that the rest of the board components will not make contact any more with the block.

So, what does this all mean? Well, it really means only one thing. If you want to use paste two things need to happen...

1. All pads need to be removed on the whole block and must be replaced with thermal grease (paste).
2. Gabe supplies all new set of standoffs for the whole block reduced by 0.5mm and supplies a set of new mounting screws to go with the reduced standoffs OR you reduce your own standoffs by 0.5mm which I do not really recommend as inaccuracies can cause more problems than good.

As it is right now, this block is not reliably and consistently suitable for use with HD5850 cards when using paste method and this method should therefore not be used.

If Gabe supplies me with standoffs reduced as per my findings above I will attempt to redo the tests again and post back the results. For now I am back on pads.

[johnnymossville]

I just got my rig together tonight. I have some cleaning up to do inside the case and some tweaking to the bios, etc,... but I thought I'd run a furmark test to see how the komodo with my 5870 is doing. This is my first attempt. It's still running, over 10 minutes now, and the vrm's haven't gone over 59c.

I'm very happy with it for now, and know it's still got more left in it after some minor tweaking.

http://i39.photobucket.com/albums/e194/xnodesign/furmark_grab.jpg

[Philwong]

I just got my rig together tonight. I have some cleaning up to do inside the case and some tweaking to the bios, etc,... but I thought I'd run a furmark test to see how the komodo with my 5870 is doing. This is my first attempt. It's still running, over 10 minutes now, and the vrm's haven't gone over 59c.

I'm very happy with it for now, and know it's still got more left in it after some minor tweaking.

Are you using pad or grease?

Phil

[johnnymossville]

Are you using pad or grease?

Phil

I'm using thermal grease. I also ran into the shorting/no vga output problem on the back of the card. I did the back-off solution Gabe mentioned and that worked.

[dejanh]

I just got my rig together tonight. I have some cleaning up to do inside the case and some tweaking to the bios, etc,... but I thought I'd run a furmark test to see how the komodo with my 5870 is doing. This is my first attempt. It's still running, over 10 minutes now, and the vrm's haven't gone over 59c.

I'm very happy with it for now, and know it's still got more left in it after some minor tweaking.

http://i39.photobucket.com/albums/e194/xnodesign/furmark_grab.jpg


Are you using pad or grease?

Phil

Use FurMark 1.8, Xtreme Burning Mode, set the resolution to at least 1680x1050, and crank that voltage up to at least 1.25V with a decent overclock. I am curious to see what happens then. Those VRM temps you are posting are not that impressive considering you are running stock voltages.

Having said that, I have been testing yet another proposed solution from Gabe and I am happy to report that the results are looking good. I am going to let Gabe know today what the results are and then I will post them here together with the solution. While my VRM temperatures are still high they are under control and, I can confirm that paste works for me now as well since I have dropped an average of 20C from the peak VRM temperatures with the new test setup.

I will post the details and requirements of the new setup, i.e., what you need to make it work later today.

:)

[johnnymossville]

Will do dejanh, and good to hear you are getting those vrm temps under control. you had some scary numbers going there.

I'm going to have to get a different bios from Powercolor, all they let me do with the installed bios is 900/1300. More than enough for me, but I do want to push it and refine it a bit more before I settle into an economical 24/7 setup.

I haven't played with overclocking other than to set it to 4.2ghz and set the memory to DDR2000/CAS8. It's been running since I first powered it yesterday.

I'm a 3D designer/animator, so realtime graphics performance takes a second seat to Render speed for me. I'm no overclocker that's for sure. haha

having fun though!

[dejanh]

Will do dejanh, and good to hear you are getting those vrm temps under control. you had some scary numbers going there.

I'm going to have to get a different bios from Powercolor, all they let me do with the installed bios is 900/1300. More than enough for me, but I do want to push it and refine it a bit more before I settle into an economical 24/7 setup.

I haven't played with overclocking other than to set it to 4.2ghz and set the memory to DDR2000/CAS8. It's been running since I first powered it yesterday.

I'm a 3D designer/animator, so realtime graphics performance takes a second seat to Render speed for me. I'm no overclocker that's for sure. haha

having fun though!

Fur sure :) I have had my share of stress though as well. Testing this when you are paying out of your own pocket is no joke. Parts die and nobody finances my replacements.

Having said that, I would be really curious to see your results with FurMark 1.8 using Xtreme Burning Mode, if you can 1920x1200 resolution, ~1.3V on the card with 1000MHz core and 1250MHz-1300MHz on the memory.

[johnnymossville]

...I would be really curious to see your results with FurMark 1.8 using Xtreme Burning Mode, if you can 1920x1200 resolution, ~1.3V on the card with 1000MHz core and 1250MHz-1300MHz on the memory.

I might work my way up to that as I gain some courage. Raw performance is cool and Dirt2 totally kicks arse on this machine, but I think the nicest part of water cooling the gpu is how quiet it is compared to the stock cooler. That thing sounds like a jet taking off!

And yes, failure can be expensive if things go bad. I haven't built up a pc in a few years before this one, and I definitely had a few moments of stress when things weren't going right along the way. All kinds of thoughts run through your head when things go bad!!! LOL

[Xtrykr]

Look guys, let me make this very clear...

Pads work just fine on this block. The VRM temperatures with pads are far from optimal, but they are under control and can hold a very decent OC just fine. Having said that, on to my results...

Mounting the block with paste at VRM areas instead of pads does significantly reduce the VRM temperatures as stated by Gabe, up to 20C in fact. The standoff reduction that is needed around main VRM area is 0.5mm, and around the single VRM at the front of the card minimum 0.25mm. It also requires full compression of the PCB to the standoffs. However, this is not without issues.

I reported earlier that my card appears to be locking up and I thought that the card was faulty. However, it turns out that was not the case and in fact it was the reduced standoffs that were causing the problem, specifically, it was the board warping that happens as a result of the use of reduced standoffs that was causing random lockups on 1 out of 2 cards and was resulting in no video post or loss of video in some cases.

Compressing the PCB all the way down to the standoffs is a must if you want to make proper contact with the block when using paste on VRMs and this causes significant warping of the board both lengthwise and widthwise. In fact, widthwise the PCB starts to take on a bow-shape. As a result, only the outer VRMs make good contact while the ones in the center (two large and one small) make poor contact again. There is no warping around the single VRM at the front of the card. The warping is the result of two things. First, the rest of pads on the block used for memory, that concave cut near the VRMs, etc. get in the way and need to be removed and replaced with paste as well. Second, the rest of the standoffs are too tall resulting in warping and need to be equally reduced by 0.5mm as removing the rest of the pads also means that the rest of the board components will not make contact any more with the block.

So, what does this all mean? Well, it really means only one thing. If you want to use paste two things need to happen...

1. All pads need to be removed on the whole block and must be replaced with thermal grease (paste).
2. Gabe supplies all new set of standoffs for the whole block reduced by 0.5mm and supplies a set of new mounting screws to go with the reduced standoffs OR you reduce your own standoffs by 0.5mm which I do not really recommend as inaccuracies can cause more problems than good.

As it is right now, this block is not reliably and consistently suitable for use with HD5850 cards when using paste method and this method should therefore not be used.

If Gabe supplies me with standoffs reduced as per my findings above I will attempt to redo the tests again and post back the results. For now I am back on pads.

Just out of curiousity, is it possible to sand down the standoffs or do they crack/break?

[gabe]

Having said that, I have been testing yet another proposed solution from Gabe and I am happy to report that the results are looking good. I am going to let Gabe know today what the results are and then I will post them here together with the solution. While my VRM temperatures are still high they are under control and, I can confirm that paste works for me now as well since I have dropped an average of 20C from the peak VRM temperatures with the new test setup.

:)

good to hear from you, and that the new screws worked out well. looking forward to see your new test results.

[dejanh]

good to hear from you, and that the new screws worked out well. looking forward to see your new test results.

Sorry, I've been completly busy and unable to prepare results. I did test though already just quickly for myself as soon as I received the parts from you and everything worked out. I would suggest that you include the revised mounting with all the blocks. I can tell you that tightening the screws 1/2-1 turn more can result in 10C-15C reduction in temperatures. I originally thought that it was not working again on one card but then I started tightening the blocks further and achieved very decent results.

The block can actually hold 1.35V now and >1000MHz overclock on both cards in CrossFireX.

:up:

[hoodlikegaza]

so what's the verdict? Is the komodo going to start shipping with these standoffs and screws?

[gabe]

so what's the verdict? Is the komodo going to start shipping with these standoffs and screws?

There is no change in the standoffs. Just the screws. We will send replacement screws upon request for any product purhcased in the channels that are using the old screws, and roll-out the new screws for the new shipments.

[dejanh]

I did not include a screenshot here because FurMark crashed just as I was about to take the screen but I can tell you that at 1.35V, 1030MHz/1250MHz with two HD5850s hooked up in CrossFireX with parallel water flow the peak temperatures on the VRM are 98C on card #1 (top slot), and 90C on card #2. The temperatures range a lot, with the hottest temperatures being observed on the middle VRM of the 3 since this is the hardest one to make contact with as there are no fasteners around the middle of the card width-wise. The two outer VRMs are both lower by 10C - 15C. Therefore, range is from 70C to 98C on all of the VRM on both cards. Tightening the screws futher does help lower the temperatures a bit still, but this is okay since the temps are completely under control now. There are no more runaway thermals.

All of the tests were done using FurMark 1.8 Multi-GPU, Xtreme Burning Mode, at 1920x1200 with no AA, Post Processing enabled. The load was consistent at 97% on each card, peak power draw was ~90 amps/120W, average was ~85 amps/115W. This power load will never be observed under normal operating conditions and under normal conditions the VRM temperatures never exceed 55C-60C from what I noticed.

[johnnymossville]

Dejanh: What kind of frame rates were you getting with those settings and with crossfire?

John

[dejanh]

Dejanh: What kind of frame rates were you getting with those settings and with crossfire?

John

Pretty low, I think about 40FPS.

[Philwong]

All of the tests were done using FurMark 1.8 Multi-GPU, Xtreme Burning Mode, at 1920x1200 with no AA, Post Processing enabled. The load was consistent at 97% on each card, peak power draw was ~90 amps/120W, average was ~85 amps/115W. This power load will never be observed under normal operating conditions and under normal conditions the VRM temperatures never exceed 55C-60C from what I noticed.

So far, only Crysis is capable of heating up my 5870's VRM to 90c+ at 1,920 x 1,200 reso. For most games such as Dirt 2, COD:MW2 and CNC4, the temp is around 65c peak. Nevertheless, good job on all the beta testing. :up:

Phil

[Philwong]

There is no change in the standoffs. Just the screws. We will send replacement screws upon request for any product purhcased in the channels that are using the old screws, and roll-out the new screws for the new shipments.

Gabe, I don't quite understand how the screws can remedy the contact issue instead of the standoffs. Could you please elaborate?

By the way, Michelle will be sending me some missing screws and spacers for the Apogee GT/AM2 plate I ordered from PPCS. Can I request the revised screws for my Komodo 5800 to be shipped over as well? Thanks.

Phil

[gabe]

Gabe, I don't quite understand how the screws can remedy the contact issue instead of the standoffs. Could you please elaborate?

By the way, Michelle will be sending me some missing screws and spacers for the Apogee GT/AM2 plate I ordered from PPCS. Can I request the revised screws for my Komodo 5800 to be shipped over as well? Thanks.

Phil

We shipped with spring screws (nvidia style). Tension is limited.

Regular screws have no such limitation, thus allowing better contact.

[Philwong]

We shipped with spring screws (nvidia style). Tension is limited.

Regular screws have no such limitation, thus allowing better contact.

Thanks for the clarification.

Phil

[johnnymossville]

Tonight I ran some tests with the komodo using Furmark 1.8 at 1920x1080 and found out a few things about the 5870 in my machine. At least in Furmark, upping the voltage and clock speeds way over 1000core 1325mem doesn't give much more in the way of fps performance. 1 or 2fps isn't worth the extra overclock for me on this gfx card. 975x1325 is where I'm going to keep it for the time being anyway. Oh, and I'm happy with these temps. The Komodo just laughs at games now, even at the highest settings. Furmark cranked up to 1920x1080 heats it up a bit though.


Furmark 1.8 Stability Test,
1920x1080,
Xtreme Burning Mode,
Displacement mapping, Post FX on.

http://i39.photobucket.com/albums/e194/xnodesign/furmark_975_1325.jpg

[PaganII]

If there is too much gap above the VRMs for optimum cooling with thermal pads and way too much gap for paste then why not shim the area (or each VRM) with metal. Stick the shim on with Ceramique. I've modified CPU coolers this way. Jewelers supply will have thin gauge copper and silver. Probably order it online. The Bezel (99.9% pure silver) material may be perfect and can also be used for "kill coils". Comes in various widths. Better than warping/ bending the card.

[Philwong]

Tonight I ran some tests with the komodo using Furmark 1.8 at 1920x1080 and found out a few things about the 5870 in my machine. At least in Furmark, upping the voltage and clock speeds way over 1000core 1325mem doesn't give much more in the way of fps performance. 1 or 2fps isn't worth the extra overclock for me on this gfx card. 975x1325 is where I'm going to keep it for the time being anyway. Oh, and I'm happy with these temps. The Komodo just laughs at games now, even at the highest settings. Furmark cranked up to 1920x1080 heats it up a bit though.

Furmark 1.8 Stability Test,
1920x1080,
Xtreme Burning Mode,
Displacement mapping, Post FX on.

Wanna try benching with 8x MSAA and check out the VRM temp?

Phil

[Gimmpy224]

Question Gabe: does this block support Non-Reference Cards?

[Philwong]

Question Gabe: does this block support Non-Reference Cards?

Answer for him. No.

What card are you using? MSI Lightning?

Phil

[dejanh]

If there is too much gap above the VRMs for optimum cooling with thermal pads and way too much gap for paste then why not shim the area (or each VRM) with metal. Stick the shim on with Ceramique. I've modified CPU coolers this way. Jewelers supply will have thin gauge copper and silver. Probably order it online. The Bezel (99.9% pure silver) material may be perfect and can also be used for "kill coils". Comes in various widths. Better than warping/ bending the card.

This is a good general suggestion. :)

[Circaflex]

my block has the spring screws who should i email to get the replacement ones? havent installed the block yet but have removed the pads. have been waiting for my 2nd card to come.

[Philwong]

my block has the spring screws who should i email to get the replacement ones? havent installed the block yet but have removed the pads. have been waiting for my 2nd card to come.

Michelle
Michelle@swiftech.com

Phil

[johnnymossville]

Wanna try benching with 8x MSAA and check out the VRM temp?

Phil

I'll give that a try and let you know. have you tried it?

[Philwong]

I'll give that a try and let you know. have you tried it?

Yes, and monitor your VRM temp very closely.

Phil

[Circaflex]

Michelle
Michelle@swiftech.com

Phil

thanks phil just shot her an email :up:


On a side note, are we still recommended to use grease on vrms now that we have these new screws? What about new standoffs? I have already taken then pads off my block, whats the new general consensus on this and what setup is optimal.

[dejanh]

thanks phil just shot her an email :up:


On a side note, are we still recommended to use grease on vrms now that we have these new screws? What about new standoffs? I have already taken then pads off my block, whats the new general consensus on this and what setup is optimal.

#1 You do not need new standoffs.
#2 You should use new screws without springs for mounting as you can apply significantly more pressure for better contact.
#3 Paste is better than pads with this option.

[Gimmpy224]

Answer for him. No.

What card are you using? MSI Lightning?

Phil

Sapphire HD 5850.

http://www.newegg.com/Product/Product.aspx?Item=N82E16814102884

[johnnymossville]

Yes, and monitor your VRM temp very closely.

Phil


It looks like 8x on the MSAA doesn't effect fps much if at all, but it does warm it up another 3-5c across the board. I think if I had a better radiator I could get better temps, but I'm not complaining.

http://i39.photobucket.com/albums/e194/xnodesign/furmark1920MAX5Min.jpg

[johnnymossville]

And as you see here, less volts and a bit less core and memory speed have almost no effect on fps (at least on powercolor 5870 cards), but temps go down quite a bit.

http://i39.photobucket.com/albums/e194/xnodesign/furmark1920SubMAX5Min.jpg

[johnnymossville]

and a random idle temps screengrab. :)

http://i39.photobucket.com/albums/e194/xnodesign/idle_temps2.jpg

[Philwong]

Not bad at all..

Phil

[dbase88]

what is the best way to remove the pads & the sticky tape from the block?

I have removed all pads as advised and put articsilver Céramique on the vrm's & ram & as5 on the gpu. Is this correct?

with the new screws the ram is still not touching the plate of the komodo.

any ideas? also my card doesnt seem to work after i did all this, i was extremely careful when replacing it.

edit: just read about the x bracket. i completely removed this and just used 4 screws with the plastic washer. still nothing
does this mean my card is stuffed/short circuited?

i doubt my retailer is going to rma it after I took the stock hsf off :(

[Johnny87au]

nice mate, loving the swiftech 5800 Series Gpu block, hi-5

[Sadhana]

Can someone post the picture of old screws and new screws? I'm from Indonesia and bought Komodo from PPCS about 2 months ago when I realized that Komodo have the "new" screws regarding Gabe's statement.
Komodo still in its package because I haven't instaled yet in my 5850, I'm on duty in other province far away my home.

[Philwong]

Can someone post the picture of old screws and new screws? I'm from Indonesia and bought Komodo from PPCS about 2 months ago when I realized that Komodo have the "new" screws regarding Gabe's statement.
Komodo still in its package because I haven't instaled yet in my 5850, I'm on duty in other province far away my home.

The new screws are slightly less than 6mm tall.

Phil

[gabe]

Can someone post the picture of old screws and new screws? I'm from Indonesia and bought Komodo from PPCS about 2 months ago when I realized that Komodo have the "new" screws regarding Gabe's statement.
Komodo still in its package because I haven't instaled yet in my 5850, I'm on duty in other province far away my home.

all you got to do is email michelle@swiftech.com, and she'll mail you a set anywhere you are, except other planets in the solar system of course ;)

[Philwong]

all you got to do is email michelle@swiftech.com, and she'll mail you a set anywhere you are, except other planets in the solar system of course ;)

Excellent after sales service!!

Phil

[Circaflex]

all you got to do is email michelle@swiftech.com, and she'll mail you a set anywhere you are, except other planets in the solar system of course ;)

Shes amazing! i have had to contact her twice and everytime its been 2nd day air to me. Top notch service, i will always buy swiftech!

[gabe]

Shes amazing! i have had to contact her twice and everytime its been 2nd day air to me. Top notch service, i will always buy swiftech!

I know, that's why I got so pissed off the other day when a lunatic called me to tell me that she was the worse employee because she hadn't overnighted something he needed, and as I took her defense, started cussing me at me.

People make Companies, and she's among our best assets.

[Sadhana]

all you got to do is email michelle@swiftech.com, and she'll mail you a set anywhere you are, except other planets in the solar system of course ;)

Hahaha... yes, I am in the same planet as you all in this forum :ROTF:

Thanks, Gabe. I will send mail to Michelle.

[Alexandr0s]

all you got to do is email michelle@swiftech.com, and she'll mail you a set anywhere you are, except other planets in the solar system of course ;)

She doesn't ship to Uranus? :eek:

Sorry bout that one :D.

[Grinder]

A set of screws are on their way to me, thanks to Michelle (very quick response).

I haven't installed my card yet, a Sapphire 5850. What is the current wisdom for the advanced install? i.e. remove ALL thermal pads, or just those for the VRM's? This is what is described in the current installation guide, but from the discussion here it sounds like all the pads need to be removed.

Also, do the screws need to be tightened completely to the standoffs?

[dejanh]

A set of screws are on their way to me, thanks to Michelle (very quick response).

I haven't installed my card yet, a Sapphire 5850. What is the current wisdom for the advanced install? i.e. remove ALL thermal pads, or just those for the VRM's? This is what is described in the current installation guide, but from the discussion here it sounds like all the pads need to be removed.

Also, do the screws need to be tightened completely to the standoffs?

1. All pads *except* those covering the memory modules should be removed.
2. You basically tighten all the way down to the standoffs but do it carefully. If you know that there is too much resistance I would stop. You do not want to crush any of the digital VRMs (very easy to crush them). This should not happen, but a cautionary note is worth while nonetheless.

[Grinder]

1. All pads *except* those covering the memory modules should be removed.
2. You basically tighten all the way down to the standoffs but do it carefully. If you know that there is too much resistance I would stop. You do not want to crush any of the digital VRMs (very easy to crush them). This should not happen, but a cautionary note is worth while nonetheless.

Leave the deeply recessed one in place as well, then (whatever that chip is, doesn't look like ram). Basically the same as your pics back on page 6 or so?

Thanks for all your hard work and testing btw :up:

[dejanh]

Leave the deeply recessed one in place as well, then (whatever that chip is, doesn't look like ram). Basically the same as your pics back on page 6 or so?

Thanks for all your hard work and testing btw :up:

Np :) :p: I'm currently on to other toys...playing with a i3 530 on the Gigabyte GA-H55N-USB3 mITX board. I'm up to 4.53GHz right now LinX burning it in 27C heat with an H50 all mounted in a tiny Lian Li PC-QO8 B case :p:

And no, the pad in the recessed area goes as well. Only memory pads stay (two long pads near the GPU).

[Grinder]

Np :) :p: I'm currently on to other toys...playing with a i3 530 on the Gigabyte GA-H55N-USB3 mITX board. I'm up to 4.53GHz right now LinX burning it in 27C heat with an H50 all mounted in a tiny Lian Li PC-QO8 B case :p:

And no, the pad in the recessed area goes as well. Only memory pads stay (two long pads near the GPU).

Heathen!:devil::devil::devil:

I haven't done Intel for a while lol. Sounds like fun in such a small package. I'm sure you'll post some pics, if you haven't already.

Thanks for the heads up :rocker:

I'm on the other coast btw. :)

[Circaflex]

And no, the pad in the recessed area goes as well. Only memory pads stay (two long pads near the GPU).

I think you are mistaken/incorrect on the recessed area pad. According to a PM from Gabe along with the manual he has linked me to many times, it says to only remove it from 2 spots, leave behind the memory and the recessed pad.
http://www.swiftnets.com/products/installation_guide_KOMODO-HD5800-Multilangue.pdf

Page 5 clearly states only those two pads should be removed.


Im also on 5870's and not 5850's if that matters, i also USED the xbracket, am i supposed to not use it with the new screws?

Hopefully Gabe can come in and let us know.

[dejanh]

Heathen!:devil::devil::devil:

I haven't done Intel for a while lol. Sounds like fun in such a small package. I'm sure you'll post some pics, if you haven't already.

Thanks for the heads up :rocker:

I'm on the other coast btw. :)

Fun?! It's a blast :D These little guys and this board are absolutely amazing. I was all AMD for many, many years, but with the arrival of Nehalem...well...I needed a change lol. I want AMD to start bulldozing so I can start pushing some of their parts again :p:


I think you are mistaken/incorrect on the recessed area pad. According to a PM from Gabe along with the manual he has linked me to many times, it says to only remove it from 2 spots, leave behind the memory and the recessed pad.
http://www.swiftnets.com/products/installation_guide_KOMODO-HD5800-Multilangue.pdf

Page 5 clearly states only those two pads should be removed.


Im also on 5870's and not 5850's if that matters, i also USED the xbracket, am i supposed to not use it with the new screws?

Hopefully Gabe can come in and let us know.

Trust me, I've spent enough time working with Gabe on this to know what needs to go. I did all the testing ;) I am not speaking about 5870s, only 5850s which Gabe and I were testing. All pads go except for memory. I see no reason why it would be any different however on a 5870 *if* you are having the same type of contact problems.

The X-bracket is used. New screws are only used to be able to provide better tension (or compression really) of the PCB to the standoffs and hence to the VRM.

[Circaflex]

How are these temps in your guys opinion.

TIM hasnt settled yet, minor overclock of 975 on the core, memory at 1400, voltage at 1.2.

After 3 runs of vantage, 3 runs of 06, and some msi kombuster my temps maxed out at 45 on GPU(according to MSI Afterburner). GPUZ says max temps on my GPU were 44/45, GPU Temp 2 Max was 47, GPU Temp 3 was 43. 2nd card had similar results. Are these in line with what other are seeing?

Im using the komodo without pads on VRM area.


EDIT: Just ran a few hours of Furmark Stability test and i maxed out at 56 under load.

[dejanh]

How are these temps in your guys opinion.

TIM hasnt settled yet, minor overclock of 975 on the core, memory at 1400, voltage at 1.2.

After 3 runs of vantage, 3 runs of 06, and some msi kombuster my temps maxed out at 45 on GPU(according to MSI Afterburner). GPUZ says max temps on my GPU were 44/45, GPU Temp 2 Max was 47, GPU Temp 3 was 43. 2nd card had similar results. Are these in line with what other are seeing?

Im using the komodo without pads on VRM area.


EDIT: Just ran a few hours of Furmark Stability test and i maxed out at 56 under load.

Mine peak out at about 38C - 40C with approx. 25C ambient temperatures on two 5850s in X-fire connected in parallel.

It depends on your ambient temperatures, though you will definitely be running hotter than me because of extra shaders.

[Grinder]

Is it just me, or do the leaked photos of the 6970 pcb look like a komodo might fit? That would be cool, since I now own two...

And thanks to Michelle for a second set of screws. :)

[Church]

It's hard to tell from low quality pics if some chip is placed 1mm higher or deeper. Without actual cards on hands nothing can be said about compatibility for sure.

[dejanh]

Is it just me, or do the leaked photos of the 6970 pcb look like a komodo might fit? That would be cool, since I now own two...

And thanks to Michelle for a second set of screws. :)


It's hard to tell from low quality pics if some chip is placed 1mm higher or deeper. Without actual cards on hands nothing can be said about compatibility for sure.

That would be absolutely F-ing AWESOME if it fit. But, experience teaches not to get my hopes up too high :p:

Anybody care to test and see?

[Grinder]

It's hard to tell from low quality pics if some chip is placed 1mm higher or deeper. Without actual cards on hands nothing can be said about compatibility for sure.

Agreed. But I can hope :p:


That would be absolutely F-ing AWESOME if it fit. But, experience teaches not to get my hopes up too high :p:

Anybody care to test and see?

Yeah, it would be great news. But I just bought a used ref 5850 and second komodo (not even installed yet!) to add to my first, so it won't be me testing on a 6970 for a while unless they are giving them away!

It does look like the right layout, if the pic is accurate. But churchy is right, it wouldn't take much to screw it up. I suppose we'll find out soon enough.

[Grinder]

On closer examination of some of the pics, I don't think the block will fit :(