New Swiftech Stuff

[Angry_Steel]

http://uk.theinquirer.net/?article=40993

Swiftech Noobification. Now you can RMA the whole thing when the pump dies. Hopefully this wont be the wave of the future for Swiftech.

[SiGfever]

Finally a starter solution from a company that can deliver a fine setup.

[Patriote]

Yeah. Is nothing Xtreme but is still better than the all-in-one solutions we can get from companies like TT and others.

Looking forward for some reviews!

[nikhsub1]

http://swiftech.com/products/APOGEEDRIVE.asp

[Aldy402]

one of those self driven blocks with a pa120.1 or 160.1 should be really interesting

[Pete]

WTF, DDC's i though can be mounter verticuly upside down!!

Let alone how does any water get over the copper base of the block.

I want some real open unit photo off Gabe

Maybe really really small factored cases can be WC's!

Any plans to do them for GFX as well?

[Rob_B]

[IMG]Let alone how does any water get over the copper base of the block.

Like this?

[dnottis]

Finally a starter solution from a company that can deliver a fine setup.

Thats what I was thinking too. After the crap Coolermaster and TT put out... lets do an all-in-one the right way. I'd be leary of the 350 tho, since I had a few die.

[MrToad]

I'm looking forward to some reviews as well.

On paper it looks promising. I hope it delivers and provides a great market growth for swifty ^^

[NaeKuh]

"The nice thing about these parts are they make an all internal water cooling kit, so there are no untidy things hanging out of your box. You can get a mild cooling setup with the 120 and a little more heat dissipation with the 220, but both are mounted in the same way. With large slow fans, Swiftech is claiming 40dBA with an overclocked quad core CPU and two 8800GTXs."


Uhhh...... how does a 220 handle an overclocked quad + 2 8800GTX?

This is enough to bring a MCR320 crying. Infact i recomend members to split the loop up because the 2 x 8800GTX will impact the CPU performance quite noticibly.

Andy i cant believe your not saying anything about this. You have the config and so do i which is required for a dual 8800GTX quad platform.

Scott you going to sit calmly thought this statement as well?


Just some numbers for reference.

An overclocked quad will put out around 200W of heat. A GTX each will put out about 210W of heat. x 2 = 420w + 200W = 620W of heat you need to get rid of.

There is no chance unless you live in hell and it froze over, that a poor 220 will handle this heat load. Gabe needs to have them fix this statement.

[gabe]

There is no chance unless you live in hell and it froze over, that a poor 220 will handle this heat load. Gabe needs to have them fix this statement.

There is no "them". I am making these statements. I never claim anything that I cannot back-up from verified experience. We published some 3D marks 06 a while back, with an A.D. abbreviation which stood for Apogee Drive, and have now edited the title http://service.futuremark.com/compare?3dm06=2054649

I personally measured this system in my P180 case with all system fans on low at 37 dBA (with a room ambient noise of 27 dBA)

Was this system hot? yes but still within specs, the same way a large OEM would setup a high-end liquid cooled system.

I have this system at home with the dual rad version (except for the quad that went back into my office system), E6600 at a conservative 3.4Hgz for hot california summer stability, both Stealth Rev2 cooled GTX's running at 648/1053 no vmod, and I'm in the low 50's in orthos, mid 40's under furry cube load for the GTX's, and all at a quiet 40 dBA (dual fan adds 3 dBA compared to single).

Anyways, now that the cat is out of the bag , you can actually use our CPU temp calculator located on the H20-120-compact kit page to predict with fairly good accuracy (+/- 1C) what your CPU temps will be with this kit version

[MrToad]

Sorry to drift slightly (ever so slightly, as this could well be under "New Swiftech Stuff") off-topic, but...

both Stealth Rev2 cooled GTX's

You really know how to bring that perplexed look up. Five weeks since the official launch of the product and a Rev2 in the oven?

[nikhsub1]

Sorry to drift slightly (ever so slightly, as this could well be under "New Swiftech Stuff") off-topic, but...



You really know how to bring that perplexed look up. Five weeks since the official launch of the product and a Rev2 in the oven?

Yeah I wouldn't be happy if I bought rev 1. At all. Makes me think rev 1 wasnt properly thought out.

[NaeKuh]

There is no "them". I am making these statements. I never claim anything that I cannot back-up from verified experience. We published some 3D marks 06 a while back, with an A.D. abbreviation which stood for Apogee Drive, and have now edited the title http://service.futuremark.com/compare?3dm06=2054649

I personally measured this system in my P180 case with all system fans on low at 37 dBA (with a room ambient noise of 27 dBA)

Was this system hot? yes but still within specs, the same way a large OEM would setup a high-end liquid copoled system.

I have this system at home with the dual rad version (except for the quad that went back into my office system), E6600 at a conservative 3.4Hgz for hot california summer stability, both Stealth Rev2 cooled GTX's running at 648/1053 no vmod, and I'm in the low 50's in orthos, mid 40's under furry cube load for the GTX's, and all at a quiet 40 dBA (dual fan adds 3 dBA compared to single).

Anyways, now that the cat is out of the bag , you can actually use our CPU temp calculator located on the H20-120-compact kit page to predict with fairly good accuracy (+/- 1C) what your CPU temps will be with this kit version

Gabe you should know first hand, i value your imput a lot.

However, i dont think this statement would hold valid. i have members on anandtech complain about the weak temps a MCR220 radiator would bring on 2 gtx's alone. In fact a lot of people split the loop up because the system would require a massive radiator to in-line it.

The lowest radiator for such a config, overclocked quad + 2 gtx itself merits the need for a PA120.3 series radaitor.

Adding the statement of Large slow fans, your going to confuse the public. In fact im not saying its not possible, you would need super high velocity fans for the cooling system to keep up.


Dont mistaken me. I love your products, i use a lot of them in my rig and other rigs. I went though almost all the waterblocks you released, and i think your company is great.

However, that statement i bolded is not something you should promise. The heat load is far too great for a MCR220. And if the user isnt going to get watercooling expected temps, his migration will lead to bitter complaints.

Scott, please help with your imput. If you honestly think this system will work, show me how it would work. Also andy show me that this system will work, and the reasons why we split our loops up wasnt just for dual color coolants and bling factor.

[nikhsub1]

Scott, please help with your imput.

Is this directed at me?

[NaeKuh]

Is this directed at me?

yes... or should i just call you by your handle?

[OPPAINTER]

Gabe you should know first hand, i value your imput a lot.
However, i dont think this statement would hold valid. i have members on anandtech complain about the weak temps a MCR220 radiator would bring on 2 gtx's alone. In fact a lot of people split the loop up because the system would require a massive radiator to in-line it.

I'm curious about this statement above.

These guys getting bad temps, what blocks do they use? You could have 5 radiators, but if your block sucks with bad contact your temps will still suck.

Besides, I made those 06 benches, that Gabe posted, myself, so I know for a fact what the little kit can do.

OPP

[gabe]

Gabe you should know first hand, i value your imput a lot.

However, i dont think this statement would hold valid. i have members on anandtech complain about the weak temps a MCR220 radiator would bring on 2 gtx's alone. In fact a lot of people split the loop up because the system would require a massive radiator to in-line it.

The lowest radiator for such a config, overclocked quad + 2 gtx itself merits the need for a PA120.3 series radaitor.

Adding the statement of Large slow fans, your going to confuse the public. In fact im not saying its not possible, you would need super high velocity fans for the cooling system to keep up.


Dont mistaken me. I love your products, i use a lot of them in my rig and other rigs. I went though almost all the waterblocks you released, and i think your company is great.

However, that statement i bolded is not something you should promise. The heat load is far too great for a MCR220. And if the user isnt going to get watercooling expected temps, his migration will lead to bitter complaints.

Scott, please help with your imput. If you honestly think this system will work, show me how it would work. Also andy show me that this system will work, and the reasons why we split our loops up wasnt just for dual color coolants and bling factor.

You have formulated your objection politely enough and I appreciate that. But the facts remain. The high Junction temps we see on the quads may not "feel" comfortable, but these chips are spec'd to run reliably at such temps. Read my statements precisely, and you will see that there is absolutely nothing misleading in what I have posted. It's all true.

The key is in keeping the voltage as low as possible. I am not claiming extreme O/C anyways. Read the O/C meter in the H20-120-compact page. It says it all.


Oh, and one last thing: these products are not due for release today or tomorrow as stated in the Inquirer. They are due for release end of the month.

[NaeKuh]

I'm curious about this statement above.

These guys getting bad temps, what blocks do they use? You could have 5 radiators, but if your block sucks with bad contact your temps will still suck.

Besides, I made those 06 benches, that Gabe posted, myself, so I know for a fact what the little kit can do.

OPP

The bench people were using was a MCR220, with 2 MCW60's on a D5 platform.

That used to be what i recomended, until people werent happy with the high delta temps of the 2 x G80 from idle to load. I first assumed it was mounting error, and asked them to remount. Later on, i started digging more indept on this matter, and found the radiator to be lacking.

Then most of them stepped up to a MCR320, and problem was solved.

@gabe

When people think watercooling + overclocking, they intend to push the cpu at or near 1.4Vcore real to about 1.5Vcore real. Otherwise they question the water migration. The setup you listed would come on par with a Ultra120 Extreme and 2 HR-05's on the GTX's in a well vented case.

This is the bug and itch i have with this statement. The Air solution would also be quieter.

Its just concerns i have, because on anandtech, i am looked highly upon in watercooling. The first question i always ask, Budget, and what do you expect or hope for idealistically.

Then i help them plan the system around what they want. I never do it the other way, of asking them just to buy brand X and Y and Z and mesh it together and see what you get.

Anyhow, i'll check out that calculator gabe. But my brow is still lifting up on this statement.

[OPPAINTER]

NaeKuh,

If those boys running the MCW60s want 2c better on those 8800GTX cards they have got try the new Stealth REV2. Un-beatable core temps

http://www.xtremesystems.org/forums/...d.php?t=151502

OPP

[gabe]

When people think watercooling + overclocking, they intend to push the cpu at or near 1.4Vcore real to about 1.5Vcore real. Otherwise they question the water migration. The setup you listed would come on par with a Ultra120 Extreme and 2 HR-05's on the GTX's in a well vented case.

This is a question I ask myself all the time. Then I'd typically ask Stephen to put a high end heat pipe in the chamber for baseline testing, and always end up in favor of W/C by a long shot. At equal low noise level and a conservative O/C, a top-end heat pipe might get you close to the W/C CPU temps but you will never drop your GPU temps by the 20 to 30C like we see in a W/C loop. My EVGA was reaching 80C with the stock cooling and dumping massive amounts of heat in the case. Now I'm hovering in the mid 40's with case temps close to ambient. What do you make of that?

Water cooling is no longer just CPU. It's a system, dropping ALL the critical component temps. That's how it has to be looked at.

[gabe]

Could ALWAYS go naked cores on the MCW60s! :-p........................

Yeah, yeah, were'all still wondering why you wouldn't baseline your system before and after eh

[NaeKuh]

This is a question I ask myself all the time. Then I'd typically ask Stephen to put a high end heat pipe in the chamber for baseline testing, and always end up in favor of W/C by a long shot. At equal low noise level and a conservative O/C, a top-end heat pipe might get you close to the W/C CPU temps but you will never drop your GPU temps by the 20 to 30C like we see in a W/C loop. My EVGA was reaching 80C with the stock cooling and dumping massive amounts of heat in the case. Now I'm hovering in the mid 40's with case temps close to ambient. What do you make of that?

Water cooling is no longer just CPU. It's a system, dropping ALL the critical component temps. That's how it has to be looked at.

i agree... :T However, i just cant see it from a physical standpoint on how such a config would work. Because idealistically, were trying to get rid of 620W Peak, of heat out though a MCR220 radiator. Bill Adams tests showed to be effective around 300-400W of heat disapation. (please correct me if im wrong)

However, i think i'll wait for Nikhsub1's comment. If he can show me how it would work, then i'll follow. But from the many posts and comments ive been getting regarding radaitors. I always understood that this wasnt possible.

If marci will also addon about how this would work, i will also feel a bit relieved. I tend to trust marci's posts more deeply when it has matters of radiators and capacity. :T


Another person who i admire and look up to is MAX racer. A forum member is trying to cool a dualcore with a SINGLE G80.

http://www.xtremesystems.org/forums/...d.php?t=125693

Even max recomends the MCR320.

[Senater_Cache]

would the pump vibration have any significant impact on the mount? my ddc-2 vibrates quite a bit.

[nikhsub1]

Was this system hot? yes but still within specs, the same way a large OEM would setup a high-end liquid cooled system.

I have this system at home with the dual rad version (except for the quad that went back into my office system), E6600 at a conservative 3.4Hgz for hot california summer stability, both Stealth Rev2 cooled GTX's running at 648/1053 no vmod, and I'm in the low 50's in orthos, mid 40's under furry cube load for the GTX's, and all at a quiet 40 dBA (dual fan adds 3 dBA compared to single).

NaeKuh, I have no reason to doubt what Gabe is saying, LOOk at what he wrote... the system is hot COMPARED to our xtreme water systems yes, of course but the temps are well within spec. Spec to most around here is 5x to hot...

Scott, please help with your imput. If you honestly think this system will work, show me how it would work. Also andy show me that this system will work, and the reasons why we split our loops up wasnt just for dual color coolants and bling factor.

I see no reason why it wouldn't work - the air/water delta maybe 20C but so what? As long as the 2 GPU's and the CPU are 'within' operating temp spec there is no issue... I know my system doesnt put out the most heat (6700 @ 3.85 w/ 1.55v + X1950XTX) but my 120.3 keeps air/water within 2c. I am not really qualified to answer really as I dont have a quad and 2 GTX's to run with a dual fan rad... but the question at hand would be what is the air/water delta with such a setup? This would be the most informing piece of info we could gather as it will tell us how things will perform with more/better rads.