Slim I know 100% the GTX has aluminum top..still uncertian is if fusion block is aluminum or copper. I think it's copper.
Printable View
Is not that I dont want to listen to you Skinnee, but I want to give that block a try. Changing Heatsink City would cost some money and not really into spending money on mosfet coolers and SB heatsink if its not neccesary.
Is your Rad and pump ok?
No new Bioses yet?
Well ASUS just released two and they are not like pmp that bring us new BIOSes every two weeks.
http://img.techpowerup.org/080713/fast.jpg
I think I like the new bios:)
On other notes I have ran the block with a maze4 with out those problems.
I have a 120GTX with it's own pump ready to be installed.
Been waiting on a res, and a few other things. But I think I'm going to rig it out anyway
http://img.techpowerup.org/080713/pump.jpg
DD-CPX1 12V Pump--Small pump
DD12V-D5 Pump Fixed Speed by Laing
Danger Den Single 5 1/4" Bay Reservoir
DangerDen - 9800GTX blocks
I had absolutely no issue with the Maximus SE block... think this is just a mixup of bad fluids and components, think very little to do with the fuzion block as we would have had many more reports from day one if this happenend...
Finally I started OCing on my new build (see my sig for spec)
This is my first OC project on the Intel camp, I was on the AMD camp for the last 2 build. So I am pretty much still on the learning curve now
First thing I notice is that this board have soo many options in BIOS and is pretty difficult to start with. But with the hint from Anandtech (Their S2E review comes with a OC tutorial for this board, I don't know why they did that but it is a godsend for me), I have a good starting point.
I started with RAM first, I cranked up FSB to 1600 while lowered multiplier to 6x. Ran memtest86 v3.4 for a round, seem ok.
I cut a corner a bit and try to boot into Windows for a Orthos run. but I was greeted immediately with BSOD :P I Increased CPU Voltage to 1.2v and that solved the problem. Ran Orthos for an Hour without issue, then I went back to BIOS for more.
I set to 1800 FSB, multiplier still at 6x. Run 3 round of memtest and seem ok. I decided to take this, don't wanna stress all the way to 2000 MHz at this moment. Then I crank multiplier back to 9.5x and intent to stabilize on this setting (CPU @4275 MHz, RAM @ 1800 MHz 9-9-9-28).
Tried to boot into Windows, greeted with BSOD again. Pump CPU voltage to 1.3v, boot is ok but Orthos failed immediately. Pump again to 1.35v and 1.4v, Orthos could run for 30 minutes before Core #1 gave in.
Tried several small pump and I found that I need 1.43v to have a Orthos-stable setup. Run for 10 hours and I was satisfied with it. But I see that CPU Core temp was as high as 72c, and the max CPU temp reported by the chipset was at 65c, is it ok??? MCP temp is good, it pretty much stayed at 57c as I have a fan blowing to the SB, not from the top but from the side, i think it is a good setup and I will show some photo once I finish OCing.
Then I tried to run 3dmark06 with Orthos in the background. Seem good but it lasted for several round only. It failed when it try to start a test and showed a black screen.
Then I tried to run 3dmark06 alone and I hit the same issue. Hmm, finally saw something challenging in this OC project :P
Now I dropped to 9x to see what will happen. If the same thing occur then I can safely bet it is a RAM issue. Let's see how it goes :)
Thats what all of this is...live and learn! I am just trying to pass along what I know about WC'ing. I know more about WC'ing than I do OC'ing...thats for certain!
IIRC (I'll try to find the article), the Fusion Heatpipe system on the SIIE is copper plated aluminium. I remember reading that somewhere, now I have to find it.
I completely understand where you're coming from Slim, I'm just trying to help you spend your money wisely, don't go through the mistakes other have had to endure.
If you are going to run the Fusion block, I would run it in a seperate loop, don't run it in your CPU or GPU loop, you can't avoid galvanic corrosion, but you can keep it away from other components.
Ugh!!!!!
I just went to give the board another OC attempt. Now, I power the thing on, it runs for 3-4 seconds and powers off, LCD poster does not go past CPU Init.
Bad CPU or bad board? HELP!!!
OK, I dropped to 9x multiplier and seem the problem is solved... 3dmark06 seem to be able to run on and on, and CPU voltage can be lowered to 1.35v also, temps are much acceptable, more core temp is around 60c.
But just when I thought it is a job well done, I hit another issue. Real Player had problem playing video, sometimes the whole screen (not only the Real Player window) would freeze for several seconds. I stopped RealPlayer and try WMP, my screen flickered like a dicso and my PC hung.
I cranked up my CPU voltage again all the way back to 1.43v, but still hit the same problem....
Now I just dropped to 8x multiplier , let's see how it goes...
I also think it was a bad mix of fluids. That Pure PC liquid doesnt sound good at all, but I cant say anything since I never tried it. I wonder why he didnt just bought a bottle of UV Green coolant :confused:
Im planning on using Feser One UV Blue Liquid on my setup, hopefully this is safe for my CPU, NB and both GFX cards.
And I really appreciate your comments here. When is your first time water cooling, you really want to know all the risks to be carefull.Quote:
Thats what all of this is...live and learn! I am just trying to pass along what I know about WC'ing. I know more about WC'ing than I do OC'ing...thats for certain!
I also read this a second ago lol. Hopefully using UV Blue Feser coolant wont have any of this kind of effects.Quote:
IIRC (I'll try to find the article), the Fusion Heatpipe system on the SIIE is copper plated aluminium. I remember reading that somewhere, now I have to find it.
Like I said, I really appreciate the help :up:Quote:
I completely understand where you're coming from Slim, I'm just trying to help you spend your money wisely, don't go through the mistakes other have had to endure.
Unfortunately, I cant afford two pumps and two rads :(Quote:
If you are going to run the Fusion block, I would run it in a seperate loop, don't run it in your CPU or GPU loop, you can't avoid galvanic corrosion, but you can keep it away from other components.
Try removing battery and clearing CMOS.
Sure to feel for folks struggling with this board. Its a good thing Asus is continually tweaking the bios, but soooo many problems :(
Seem ok at 8x, but I really want to achieve my target, which is 4275 MHz.
I think it is a matter of stabilizing my CPU with higher voltage. But since I am not very familiar with C2D CPU yet, I am a bit hesitate.... What is the safe limit that I should never exceed on this E8500? I heard it is about 1.45v for us average people, but it is the value taken in CPU-Z or set in BIOS?
Tried the battery and CMOS trick, nope. I have also reseated the CPU and block, no dice. We had an electrical storm the other night...wonder if the CPU got fried during that, my own fault for not having a UPS.
Its passed my 30 days with NewEgg...wonder if they'll do anything or I just need to buy another Proc.
Feser One is fine, I do not know if it has corrosion blocker in it...check that. I only used the Feser UV Dye in my loop, although on my next build, I won't use UV dye either, just UV reactive tubing and run distilled water and some biocide.
No problem, I've become a little over-passionate about warning people on galvanic corrosion since I made the mistake myself.
If the Feser has corrosion blocker, this will delay the corrosion from taking over, won't stop it...only slow it.
NP! :up:
I wish I had one of each to send you...but I don't at the present moment, a friend of mine needed them, so he got them first.
I cannot for the life of me get my rig stable even @ stock speeds(prime fails after 4 hours on one core with 0901) with the OCZ RAM. I am gonna try putting my gskills on and see.
Use either Probe II or Everest. BIOS shows voltage at idle only.
Did the lightning shut your pc down?Quote:
Tried the battery and CMOS trick, nope. I have also reseated the CPU and block, no dice. We had an electrical storm the other night...wonder if the CPU got fried during that, my own fault for not having a UPS.
I know for CPUs is only 7 days...Quote:
Its passed my 30 days with NewEgg...wonder if they'll do anything or I just need to buy another Proc.
It has corrosion protection against Copper, Aluminum, Brass and another one I cant rememberQuote:
Feser One is fine, I do not know if it has corrosion blocker in it...check that. I only used the Feser UV Dye in my loop, although on my next build, I won't use UV dye either, just UV reactive tubing and run distilled water and some biocide.
So you are saying corrosion like Triggers will happen no matter what? If so, how long should I expect it to happen?Quote:
If the Feser has corrosion blocker, this will delay the corrosion from taking over, won't stop it...only slow it.
Is there anything that wont make this happen?
I was told Feser liquid by itself wouldnt do anything like what happened to Triggers as long as I dont mix the coolant with anything else.
aaah! :rolleyes:Quote:
I wish I had one of each to send you...but I don't at the present moment, a friend of mine needed them, so he got them first.
Sorry to hear that Tek
Maybe RAM incompatibility like you said?
Yup, just found that out about the Egg. I think the lightning and the crappy power situation in my office killed it.
I don't know if you will see the same results as Trigger (his are quite bad), but yes galvanic corrosion will eat through the metal (Copper, Alu and Brass) in your loop.
You can flush your loop every 2-3 months, thats about the only way to extend the life of your loop. For this, I recommend distilled water and biocide, add some Feser UV Blue Dye (same as my CPU and Chipset loop)if you want the color. But flush and replace the liquid every 2-3 months.
was -it OEM CPU - RMA to Intel FTW - very fast (within a week)
so Galvanic corrosion happens no matter if you use distilled water, coolants, anticorrosion liquids etc etc??? There is no "safe" combination?
Ok so if I decide to use just Feser UV Blue Coolant and clean my loop every 3 months, Im risk free?
With everything you told me, Im concluding that everybody who has water cooling should have galvanic corrosion happening inside their loops, am I wrong?
Retail Box, I am going to order another and send this dead one to Intel.
That was the old way, the Zerex and Hydrx only give a little corrosion protection, thats all they do. The best way to run a loop now (through lots of testing) is with distilled water and biocide (Petra's PT_nuke). Over the last year and a half there has been tons and tons of testing, the stickies in the LC section are horribly out of date.
Galvanic Corrosion happens no matter what when you run mixed metal loops, ie: Aluminium and Copper in a loop. If you run an all copper loop, no problem. It all has to do with the metals in the loop.
No, but you are lengthing the life of your loop by flushing and getting fresh coolant in the loop often. This won't prevent it, but will extend the life of your loop and blocks.
Only the loops with mixed metals (Aluminum in the loop mostly), Copper, brass, nickel and silver can all exist in one loop and be fine. Aluminium is the catalyst (improper word usage I know) to galvanic corrosion.
Ok Skinnee, Tyke and everybody else here who supported the idea to remove the heatsink. You guys won :bows:
I guess running with that fusion block will bring more problems than I will be able to tolerate so I will just get Type 4 for mosfets, Enzotech for SB and EK block for NB. I will get rid of that heatsink city to avoid corrosion problems. With a $200 rad I dont think I would be strong enough to see if full of galvanic corrosion. Wow, Prometheus Project is went from $400 to almost $800 :eek:
Rather than flood this topic with more WC stuff...Slim, send me a PM with your parts list, I'll help you scale back the funds while still keeping performance the highest priority!
PM sent
Lets keep the WC stuff on PM
Now back on topic :)
Well this thread is almost dead anyways. Im not going to say anything.
LOL @ the spider pig.. nice find
is the bandwidth drop off over 475FSB only limited to the s2e or is it with all 790i boards?
The bandwidth drop is due to the chipset
I agree. As long as WC is specific to this board it is on topic. I have just ordered a third GTX 280 and will be playing about with a tri-sli watercooled setup this week!! I may have to revert to air cooling on the SB, I have ordered an Enzotech CNB-S1L and Antec spot fan just in case.
Some peeps were not happy about the WC discussion, I'm a nub around here and didn't want to ruffle any feathers worse than I already had.
You'll have to go back to air on the SB for Tri-SLI...:up: on the Enzotech, Was messing around with it before I got my Chipset loop running, you'll be very happy.
Bandwidth drop happens when you go over 425Mhz FSB afaik. P1 & P2 fix this problem, but you would need to test stability again if you enable this two.
Well, I never saw it as off topic anyways :D
Tyke, whats up with all this cards, you are going to end up needing a nuclear plant to power your pc!. Btw, if the Enzotech picture is right, it might work with the little fan on and Tri-SLI right?
It should fit no problems, I havent read about uncompatibility so far.
I can vouch for this. Running 445x9 and with p1 & p2 enabled get around 12000mb/s read bandwidth, with it disabled, about 9500mb/s. The problem is getting it stable. Thought I had it... 14hrs prime stable etc but the pc kept cutting out (no BSOD or screen freeze, just a reboot) at random times - i could play games for hrs, but occasionally it would do this. Spent the last few days testing and I think its when P1 & P2 are enabled.
Problem is im already running 1.6vnb, and 2.0v on the memory so I really cant push it up any higher (im assuming anything over 2.0v isnt too clever on DDR3 for 24/7 running). If i work something out i'll let ya no. I should also have the nb on water by the end of the week, not sure if this will help!
well, my PLL is at 1.54v in bios, but according to everest it overvolts to 1.57v so i dont really wanna push any higher. Ive tried VTT up to 1.36v
(as read by everest)... no joy :( could be that my mem just doesnt like it! I have just got myself a DMM today, so im gonna see exactly what voltage is going through the board. Ive noticed quite a few descrepencies over the last couple of weeks.
I needed 1.58 to be stable (1.56 in BIOS)
Are you using 1T? Because 2T is what you are suppose to use with this board.
Running an Evercool behind my tri-sli here.
Perfect solution for cooling the SB, where the standard rollercoaster couldn't fix the job (SB got 90+ `c and shut down the board)
What blocks are you getting on your cards? EK?
And if i'm right it should be possible to use a southbridge waterblock when using EK waterblocks for the cards.
On the other hand is WC'ing the SB a waste of money and flow in your setup.
Evercool will do the job :up:
ok, i might move the PLL up a notch then and see what it does. I def dont wanna go above 1.6v on it though. Yes i am running 1T, but largely because of reading through the anandtech tweak guide/review of the board where they seemed to think 1T was perfectly happy with 2gb up to around 1830mhz. im running 1780mhz so figured it should be ok. Will give 2T a shot after ive tried the PLL and see what happens.
Oh, and btw... thanks for the tips slim :)
Maybe PLL is not the responsable after all. 1T is a nightmare on nvidia chipsets. I dont remember when was the last time I saw an NVIDIA chipset with 1T stable so Im pretty sure that is whats causing your instability.
I also recall many saying in here that Anandtech had Hand picket sticks so maybe that had something to do with them been able to run at 1T?
NP btw :up:
Hey Justifire
Could you post those pictures of your build again? (lost the page). I want to see the Type 2 how they look on the S2E. Will have to get those no matter what lol.
Also, shorter side goes to mosfets right?
Tyke what do you do? I want your job.... nice hardware cant wait to see some scores... I think you have the fastest setup in here.
He does for sure. I wish my 8800GT's could become GTX 280 from night to morning xD
I am lucky in that both my wife and myself have good jobs. I have two great interests in my life: my PC and my car:
http://www.ccook.pwp.blueyonder.co.uk/535d_1.jpg
http://www.ccook.pwp.blueyonder.co.uk/535d_2.jpg
I am at home recovering from a spine operation at the moment so I am very bored, and when i am bored I play games on my PC. Then I see my Crysis frame rates and I want a faster computer!!
BTW Have you seen that the next Intel (Nehalem) chipset will support Crossfire and SLi!!!!
http://techreport.com/discussions.x/15113
No, it is a Sport model - sport suspension, body kit etc. I have had the engine remapped to 334BHP. from 274BHP - 506lb/ft torque.
http://www.evo.co.uk/carreviews/evoc...w_5series.html
I did and planning to get two of 'em in crossfire X with an x48 board.
Wow, I hope it's an AWD(The review does not say anything about it). Rear wheel drive+that kind of torque = frequent tire rotation and/or tire change(Mustang cobra anybody).Ok, I will keep my mouth shut now as we are going WAY off topic.
Hi guys my OC project hit some problem and I think it is time to ask for help :P
Right now I think I have a stable CPU and RAM overclock. It is either 450 x 9 @ 1.35v or 450 x 9.5 @ 1.43v, RAM is L&S in both case. They are both Orthos stable.
But I hit all kind of graphic problem in this overclock setting. RealPlayer would crash my PC, 3DMark06 stop after 3 hours, GRID crashes almost immediately, and it can't complete a 99 loops Crysis Benchmark run.
the GRID issue is worth mentioning. At 450 x 9 @ 1.35v, it crashed at the opening moive. But if I increase to 1.375v, it was much better, but seem it was still not completely stable.... then I tried a 6 Hour GTR2 run, seem ok, but when I go back to main menu it crash immediately, screen distorted.
Also, reducing my graphic card overclock seem to have some help to this situation, but this overclock was tested and confirmed stable before I start my CPU / RAM overclock....
I think that other than CPU voltage there is another thing that I am missing. Is it NB and SB voltage?
And I have SLI. Is it the SLI instability problem you guys talked about?
I pulled the trigger on the OCZ DDR3 4GB kits for $369 after MIR. I hope with this kit, I hope be able to hit atleast 1800 in sync and hope my memory problems go away.
Once you get that TRI SLI going can you do some Vantage for me? Im curious to see some scores with 1 2 and3 cards.
ummm I had my eye on those aswell, I was thinking 8GB. The details on OCZ's web site made me hold off until I see someone get results on the ASUS 790i...here's the blurb
Please post your results!!!
"* Qualified on XFX and EVGA 790i motherboards.
All Modules are tested in two module kits. Due to additional bus load, four module configurations may have to be run at lower frequency and relaxed latencies for stable operation. Test specifications are based on dual core CPUs only. If quad core CPUs are used, higher bus thresholds and core arbitration may necessitate lowering the memory frequency below tested specs."
http://www.ocztechnology.com/product...atinum_edition
Man, a nice PC and with such car, who isnt happy? :cool:
Im hearing so much about X58. It will support SLI but only on 1136 sockets, then that it will support 1366 sockets too. I dont hows this gonna end up yet.
Yeah but Im stuck with a Striker II loooool! :D
Nehalem sounds promising... I would only need to change motherboard and CPU haha, but I think Im keeping my MB for a couple of years. Will have to see something really impressive to upgrade again.
OFFTOPIC! j/k :rolleyes: I dont mind this at all ;)
I hate that it will all happen on the NF200 chip Grrr... this will make it feel like a 780i (on the SLI side only of course ;) )
Tyke
Somebody told me about this block, do you think it will work with Tri-SLI?
http://www.koolance.com/water-coolin...product_id=425
I think its 3/8 but anyways, for anybody nsisting on wcing the SB.
I am! :D
But seriously my cards are ok, I will change them once GTX 280 go below the $400 mark. Will probably get the BFG single slot so I can have space for sound cards, and maybe a RAID (Ok that is too much dreaming :rolleyes:)
Here ya go (clickable):
http://bandwidth.se/thumbs/nouser/img080047c7ab84.jpg
http://bandwidth.se/thumbs/nouser/img08053f52b858.jpg
Yes, short side on mossfet and S-type on top :)
I'm with ya Tyke...PC's and sports cars...my two fav things next to the ladies. I've moded my MDX with HID fogs..all interior lights are hi-intensity LED, 19's etc.
http://www.acuramdx.org/forums/attac...&postid=404718
http://www.acuramdx.org/forums/attac...&postid=404719
If I'm not messing with my PC I'm messing with my truck! :up:
Got my fingers crossed...ALL my EK blocks should arrive on friday. Going to redo the whole WC loop, even splurged on the new special bitspower 1/2 compression fittings for Tygon R3603(should look sweet with no ugly zip ties)
With PC probe which chip is it reporting for MB? Southbridge or northbridge? Im water cooling with the Fusion block and my Motherboard temp shows about 32degrees full load.
Hey guys
I need a favor. Could you tell me what would be considered a "newegg" on Europe? I need to know a good and safe store in Europe that is able to ship internationally.
Any idea?
Slim
There are only two companies that I use for all my stuff, and they are both excellent.
http://www.overclockers.co.uk
http://www.scan.co.uk
I know Overclockers do International, not sure about Scan.
We talking about this one right
http://www.performance-pcs.com/catal...ducts_id=23200
Also, where did you order your TFC 480? Is it in stock?
Thats great, thanks tyke. None of them have the 480 :(
OK, while you guys are busy with watercooling stuffs, I kind of solved my problem
It turned out to be my RAM. I dropped back down to 1600 and guess what, 100% stable, even with SLI.
I wonder what setting I missed, my RAM is a OCZ 2000MHz module, I think it should make 1800MHz without much trouble... Seem I need to do more homework.
May be disable P1 / P2, or drop to 2T? any hints?
Anand article said tRFC is the key to 1800+ with high density modules. Try increasing that. What timings were you running for 1800?
I can't even post T1 with my OCZ...2T for sure if your 4GB
Some ppl have linked random crashes..especially at desktop to LLC enabled. try it off if you have it on now.
It is a 2 x 1GB kit, 9-9-9-28 (see my sig). I have set these timings in BIOS, but all other values are in auto.... seem I need to play with tRFC.... and try 2T also
And LLC is disabled. Actually I followed most of the suggestion in Anandtech's review when I start.
No actually, it is this one I have. Same thing I think without the fan holes - you can't fit a fan with tri-sli anyway. I will report on temps later today.
http://www.enzotechnology.com/cnb_s1l.htm
Tyke, please check ur PM man and see what you did:D
I ran Memtest for several hours before I start the overall overclock. I did it with mulitpler set at 6x to make sure I am testing the RAM only. Both tests for1600 MHz and 1800 MHz run fine for 4 hours, with 9-9-9-28 1T 1.84v
and in Windows I run Orthos with multipiler set back to 9 and 9.5. At 9x VCore is 1.35v and at 9.5x VCore is 1.4325v. They can run as long as I like.
The RAM I am using is OCZ3P20002GK, 1 x 2G Module.
OK, I push the limit a bit further, now I am at 425 x 9.5 = 4038 MHz @ 1.35v, DDR3 1700 @ 9-9-9-28 1T 1.8v. Still testing now, but at least GRID seem stable. I will test with my c0cktail test (3Dmark06 + Orthos + BitTorrent) tonight.
If it is stable then I will be satisfied for now, breaking into 4GHz is always my goal with this new build.
Im tellin you, I dont think there is anybody here able to run 1T. You wont get a huge drop when using 2T, but you gain a lot more stability. Give 1600Mhz 2T a try.
Do you have P1 and P2 enabled? If you do, you would need a little more tan 1.86 I believe.
Man, I need pictures tyke. I seriously didnt think about it but I then realized the fan holes are totally a mess on the striker. One on top and the other half way on the middle making it almost impossible to fit a normal heatsink.
Well thats what I thought :rolleyes:
Btw, my board gets here friday :wiggle:
Slim
I have put a picture of the Enzotech heatsink on my site. It gives about 2mm clearance from the graphics card.
http://www.ccook.pwp.blueyonder.co.uk/enzo.jpg