• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Extremely lackluster EXOS results

L

Llathos

Limp Gawd
Joined
Mar 16, 2005
Messages
151
So I had the GPU-180-L06 on. The little heatsinks were kind of sad to look at, and when things got intense in 3D, they would get too hot to touch.

ATiTool determined my max OC to be 425/510 up from the stock 400/490. Needless to say...crappy.

So, I saw the fancy schmancy (ridiculously expensive) EXOS GPU waterblock out for the ATi x800 line. $115. The claim to fame is that it cools both sides of the card at the same time...all memory surfaces and the core with direct water cooling. Sounds good huh?

Well, I installed it carefully, applying new thermal paste, following the instructions, ensuring the best possible surface contact.

Good news is, the installation went flawlessly (despite being a mess draining the EXOS and removing the old GPU waterblock). Bad news is, it was basically pointless.

New ATiTool overclock: 430/520.

$115....for 5 mhz core and 10 Mhz memory. What a giant waste of money.

So, my next instinct was the bump up the voltage on the card like I did back in the day on my 9700 Pro. Then I realize I don't see any place to do that in my BIOS. Am I nuts, or is there no direct control over the PCIx voltage?

Current rig:

Athlon 64 3500+ @ 2.5Ghz
1GB (2x512) OCZ Gold VX DDR500 2-2-2-5
ATi Radeon x800 XL 256 @ 430/520
DFI Lanparty Ultra-D w/ 3-31 BIOS
Koolance EXOS I w/ CPU-H06 and VID-AT1-V06 waterblocks
 
Don't trust Ati tool.
Show me your temps, there is one problem with ati tool which is too long to explain.
 
If I had an XT instead of an XL, I might be able to show you the temps....

Just frustrated I blew the money on something I didn't need.
 
HUmmm,lets see now, I've got a video card that doesn't OC worth a damn! :)
I'm too damned good for it to be my fault and the Video gods would never let me get a mediocore card, so what????
Geez I know, lets blame Koolance, yea that'll work. It's koolance's fault!!

Gimme a break !
 
WCing doesnt automatically increase your OCs, it just lets you keep things cooler so you can potentially raise clocks more. If Atitool doesnt let you OC any higher, and you feel things are stable, try Powerstrip, no real limits beside the card and your willingness to test for stability.

As for why you spent $115 on that gpu cooler, cooling the back of the card is pretty useless, and you could have gotten just as good cooling with a plain core block & some ramsinks. If you think it looks good though....
 
Yeah, you're right, it's not Koolance's problem. After purchasing the GPU block, it simply became clear to me that the problem was with the card and I bought and I wasted my money on that block. THAT was my point...not that Koolance sucks.
 
Well perhaps you should post in the Video Card forum and perhaps someone might be able to help out. :)
 
Ok, so maybe I won't walk away sulking from the watercooling side of this.

I just installed the beta version of ATiTool 249. It unlocks the temperature reading capability. I'm running idle at 35C and hitting load temps of 60-61C. These don't sound like good GPU watercooled temps to me.

I don't know what else to do though...I've set and reset this new block a couple times and used the "manual"'s recommended technique of checking the marks from the thermal grease to ensure proper contact. I've got solid contact and all screws are down tight...

Anyone else with this waterblock able to comment on its performance?
 
You realize air cooled temps on that card can get over 80C with ease. Also the Koolance is pumping that air out of the case instead of the stock cooler which would be pumping it into the case.
 
That's bizarre..just saw someone above tell me that his Zalman air-cooled card idled at <34 and loaded up to 55c. Air cooled. That's a load temperature differential of 25 degrees between what he's claiming, and what you're saying.
 
Well, I thought that for awhile....but these temperature readings kinda suck...

Unless you guys are trying to tell me that 63C under load is supposed to be a good temp for an x800 XL? Doesn't seem to add up though when people are doing better than this with conventional air cooling.

Hence....lackluster koolance VID-AT1-V06 results.
 
Llathos said:
That's bizarre..just saw someone above tell me that his Zalman air-cooled card idled at <34 and loaded up to 55c. Air cooled. That's a load temperature differential of 25 degrees between what he's claiming, and what you're saying.
Click to expand...

My advice would be to return the block and buy a zalman then. Problem solved :cool:
 
Llathos said:
Well, I thought that for awhile....but these temperature readings kinda suck...

Unless you guys are trying to tell me that 63C under load is supposed to be a good temp for an x800 XL? Doesn't seem to add up though when people are doing better than this with conventional air cooling.

Hence....lackluster koolance VID-AT1-V06 results.
Click to expand...

Yes, your temps are very very poor. Even my X800 Pro with all 16pipes and teh core clocked at 550MHz doesnt go that high. My temps dont even hit 60C in rthdribl. This is with an AC silencer.
 
ikellensbro said:
WCing doesnt automatically increase your OCs, it just lets you keep things cooler so you can potentially raise clocks more. If Atitool doesnt let you OC any higher, and you feel things are stable, try Powerstrip, no real limits beside the card and your willingness to test for stability.

As for why you spent $115 on that gpu cooler, cooling the back of the card is pretty useless, and you could have gotten just as good cooling with a plain core block & some ramsinks. If you think it looks good though....
Click to expand...

QFT!

Those video card coolers are made to exhaust the heat from your case and cut down on noise. And I bet it did those two things very well.
 
DarkBahamut said:
Yes, your temps are very very poor. Even my X800 Pro with all 16pipes and teh core clocked at 550MHz doesnt go that high. My temps dont even hit 60C in rthdribl. This is with an AC silencer.
Click to expand...

Another big myth in the cooling community is that like chipsets running at the same voltage produce the same amount of heat. This is fully not the case as they can deviate greatly. That and the fact that you are far from working with calibrated equipment as well.
 
pigpen said:
My advice would be to return the block and buy a zalman then. Problem solved :cool:
Click to expand...

He is not going to get any better results with the Zalman. His problem does not lie in his cooling solution....unless installed or operating incorrectly.
 
Koolance tech has agreed that my temps basically suck and suggested I clamp off a tube, disconnect and remove the cooler and remount it to ensure the best possible contact.

I guess I'll do it again...but this is my third remounting of the cooler and in accordance with their guidelines, all memory chips and the core have a solid "mark" of thermal grease showing good contact and the middle screws are down as tight as they go...

I'll work on it and post with results...if anyone gives a sh$t.

:)
 
Dude, I definitely care to hear how it works out, b/c I am looking at getting the same block for my x800xl. I've been trying to find someone that has one of these blocks to see how it performs. One suggetion I would make is if you can find a way to test your water temperature, see if its maybe too warm. Maybe test and make sure there is no air stuck in the block somehow. Definitely want to get something other than the software to read your temps though. Please do post what you find out though :D
 
Llathos said:
middle screws are down as tight as they go.
Click to expand...

That might be something you want to pay attention too. tightening screws as tight as the go will sometimes create a tension that causes separation instead of better contact. Anyways, good luck.
 
Just so you know on my stock cooler, I idle at 45 C and I stay at 75 C under load. Pretty crazy high temps for no OC at all on my card. Probably partially due to the fact that it doesnt exhaust the hot air out the back and I have a SFF.
 
Water temps under load are approximately 32C.

Thanks for the comparison on the temps...that's extremely high. I'm surprised your card is stable at 75C. Mine starts coming unhinged at ~63C...that's where I start getting video corruption and errors detected in ATiTool.

I'm idling at about 38c or so...loading up to 60-63.

Again, I'll take a look at it tonight and see if I can remount the video card with minimal mess (will have to disconnect) and let you guys know how it goes.
 
Wow, water is only 32 C on load? That is still quite a large temp delta. There really must be bad contact. Have you checked to make sure the metal contact is truly flat and its not warped? Sometimes it can be really hard to situate the block properly, especially on models like that Koolance, which is made to be modular enough to fit a bunch of different video card renditions.

But ye, 75 C sucks, and that is with the fan on 100%. I am thinking about emailing powercolor to ask them about that, see if maybe my cooler is defective or something.
 
Thermal grease...really...the stuff that came from Koolance? Get some real Artic Silver 5 or Ceramique and clean it off really good before you put the stuff back on. Which EXOS version do you have?
 
It's an EXOS 1. Yeah, I idle water temp at 25-26C, and load up to about 32-33C or so.

Didn't get a chance to work on it last night :-/ I'll let everyone know how it goes.
 
Ok, I bound the tubing with rubber bands, crimped it (temporarily) at the suggestion of a Koolance tech, and gave a remounting a shot.

This time I checked every surface and tightened down every screw as tight as I could get it without stripping the screwhead. I checked all fittings and reinstalled the card.

Long story short....no real difference. I'm not shocked.

Overclocked to 440/520, I idle at 34c or so and load to 56-57c range. <shrug> guess that's as good as it's going to get.

So in short, I'm not terribly impressed and the product certainly isn't worth the cost. My reasons? I've seen/heard of people doing better with conventional air cooling. In addition, the damn thing is a pain in the ass and it's extremely easy to make a giant mess when tinkering with the block (especially for a remounting operation). My drama is hereforth completed...thanks for listening.
 
Top Nurse said:
Thermal grease...really...the stuff that came from Koolance? Get some real Artic Silver 5 or Ceramique and clean it off really good before you put the stuff back on. Which EXOS version do you have?
Click to expand...

Actually Koolance usually ships AS products; my CPU-300-V10 had AS5 with it.
 
Llathos said:
Ok, I bound the tubing with rubber bands, crimped it (temporarily) at the suggestion of a Koolance tech, and gave a remounting a shot.

This time I checked every surface and tightened down every screw as tight as I could get it without stripping the screwhead. I checked all fittings and reinstalled the card.

Long story short....no real difference. I'm not shocked.

Overclocked to 440/520, I idle at 34c or so and load to 56-57c range. <shrug> guess that's as good as it's going to get.

So in short, I'm not terribly impressed and the product certainly isn't worth the cost. My reasons? I've seen/heard of people doing better with conventional air cooling. In addition, the damn thing is a pain in the ass and it's extremely easy to make a giant mess when tinkering with the block (especially for a remounting operation). My drama is hereforth completed...thanks for listening.
Click to expand...

56-57c on load sounds pretty good. You keep comparing the 60c temps you were getting to the temps of people who have an x800 pro. There is a huge difference. x800xl is a low k process, meaning it doesn not dissipate heat well (i.e. low k factor) while the x800 pro is much better at self-dissipating its heat. The fact that you are getting 20c degrees lower than me, and I am using air cooling is convincing me that I need to get that cooler ;) . Also, if you want more out of your OC, you can't just WC and expect it to happen, you are going to have to volt mod that card in order to squeeze more juice out of it, b/c you can't up the voltage like you used to be able to on AGP's.

Another consideration is that if you want to shed more heat, create a greater temp differential, by getting a larger radiator perhaps, or a larger resevoir. Either way, I would trade your temp results for mine any day.
 
The XL should rul alot cooler than an XT PE does. Its uses a 0.11 micron process and puts out MUCH less heat.

My Pro @ XT PE (550 core) runs at 56-57C load most the time on air cooling, theres no way an XL should run at the same temps with 120MHz less core clock, a small process and water cooling.
 
DarkBahamut said:
The XL should rul alot cooler than an XT PE does. Its uses a 0.11 micron process and puts out MUCH less heat.

My Pro @ XT PE (550 core) runs at 56-57C load most the time on air cooling, theres no way an XL should run at the same temps with 120MHz less core clock, a small process and water cooling.
Click to expand...

Did you people not read about the x800xl before it was released?? It is a 0.11 micron process, but it is a LOW K PROCESS. Therefore an XL will NOT run cooler than you XT-PE.
 
Your the one whos clearly not read up on the cards. The R420 and R480 both use 0.13 Micron Low-K processes, and the R430 (XL) uses a 0.11micron non low-k process. You do know that Low-K is designed to reduce heat and power usage right?

But all this said, less heat output is less heat output, simple as that. An XL still outputs a greatly reduced amount of heat compaired to the 0.13 micron cards.
 
DarkBahamut said:
Your the one whos clearly not read up on the cards. The R420 and R480 both use 0.13 Micron Low-K processes, and the R430 (XL) uses a 0.11micron non low-k process. You do know that Low-K is designed to reduce heat and power usage right?

But all this said, less heat output is less heat output, simple as that. An XL still outputs a greatly reduced amount of heat compaired to the 0.13 micron cards.
Click to expand...

"(XL) uses a 0.11micron non low-k process" + "Low-K is designed to reduce heat and power usage" = you just contradicted yourself.

But yes, I had fact that the XL had non-low K and the other had Low K switched up.
 
You must log in or register to reply here.
Back
Top