GeForce 8800 GT Mega Heatsink Roundup @ [H]

Status
Not open for further replies.
FrgMstr

FrgMstr

Just Plain Mean
Staff member
Joined
May 18, 1997
Messages
55,598
GeForce 8800 GT Mega Heatsink Roundup - Many of you know that NVIDIA's 8800 GT can run rather hot under load. Our GTs ran at over 90C. That spells disaster in a hot case. Let HardOCP show you the way to cool down that white hot GT while not breaking the bank and getting great performance.

We saw a lot of coolers, ran some tests and got quite a bit of data. But what does it all mean? Does declaring one heat sink the 'winner' give you the entire picture? We don't believe it does. In fact we think that it causes more confusion and spreads misinformation then it serves to help you with your buying decision. So rather than just point out who had the lowest temperatures, which anyone can do by looking at the graphs, we are going to break it down into separate sections.
Click to expand...
 
I have to concur. The S1 cooler seems very popular in a lot of the 8800gt threads.

This was a wonderful roundup. I really wish now that I got the Thermalright HR-03. I already bought a Zalman VF900 for my GT but the figures on the the Thermalright are so much better!
 
also, why aren't the prices of these units listed? a vf1000 may be a so-so buy at $45, but a great buy if you can get it at $30. without the prices, the price/performance chart is useless. :confused:
 
Glad you brought that up.

We requested a sample from AC before we even started testing. Their initial reply was that it was not compatible and as such would not even allow us to try and fit it on our card. (Who turns down free testing?).


During testing we received notification that it would fit but by that time it was to late to obtain a sample and finish this article in a timely manner.


A future article perhaps!
 
Just to add, I did call and talk to AC directly here in the States and when we explained exactly what we wanted to do with the S1, they explained to us that it would not work and no sample would be provided. At that point we moved on with our testing.
 
Regarding prices;

if I find the cooler at one site and you find it for a different price which one is correct??
Do we include rebates in our calculations?

In the end it is designed to give you a rough idea about the cooler performance.
Obviously if you find one cooler for a steal then its value would be calculated differently.
 
Were any of those coolers single slot? I was pretty excited to see the article as the stock cooler on my 8800gt is louder than I'd like, but I'm limited to a single slot and it doesn't look like any of those coolers would be an option. Oh well...
 
bobzdar said:
Were any of those coolers single slot? I was pretty excited to see the article as the stock cooler on my 8800gt is louder than I'd like, but I'm limited to a single slot and it doesn't look like any of those coolers would be an option. Oh well...
Click to expand...

All of the coolers tested block the adjacent slot at the very least. This allows them to keep the temps low at an acceptable noise level. Sorry. :(
 
RamarC said:
also, why aren't the prices of these units listed? a vf1000 may be a so-so buy at $45, but a great buy if you can get it at $30. without the prices, the price/performance chart is useless. :confused:
Click to expand...


If you find our value charts to be useless you to, I can appreciate that. Surely there is enough data on performance to help you make a buying decision of your own. Hopefully that will elivate your confusion.
 
That was a great review/comparison of the heatsinks

the graphs with the idle and load temperatures on the same plane were a good idea since it makes it easy to compare..

thanks for the effort since installing all those heatsinks and reading temperatures must have been a hassle
 
Aztlan said:
That was a great review/comparison of the heatsinks

the graphs with the idle and load temperatures on the same plane were a good idea since it makes it easy to compare..

thanks for the effort since installing all those heatsinks and reading temperatures must have been a hassle
Click to expand...


Thank you for the comments.

Installing all the heatsinks gives me a MUCH deeper appreciation of heatsinks that use a simple and efficient design. :D

I think I was just more worried about mishandling the card over and over again as I swapped out so many sinks.
 
Absolutely great job guys and all inclusive for the most part. I've recommended the VF900-CU to several people for use on their GTs and now I have the data to back it up. Thanks for doing all the work. :D As far as the HR03GT goes, it does an incredible job but it's just sooooo damned big. I'd rather have a smaller solution with a bit less cooling. But that's just me...
 
honestjohn said:
Absolutely great job guys and all inclusive for the most part. I've recommended the VF900-CU to several people for use on their GTs and now I have the data to back it up. Thanks for doing all the work. :D As far as the HR03GT it does an incredible job but it's just sooooo damned big. I'd rather have a smaller solution with a bit less cooling. But that's just me...
Click to expand...

You mean you DON'T want to block the adjacent THREE slots with this cooler and fan?? :D
 
The VF900 is probably the best decision I've made regarding video cards as a whole, hah. Does it fit the HD 38xx series as well?
 
That VF900 is the biggest work horse I have ever seen.

It just keeps going and going. A true testament to Zalman.
 
excellent article, good temperature tests and noise readings /me like. Maybe a chart or table combining noise & temp data to balance the two out more clearly or putting those charts on the same page, makes it easier to reference :)

In the future might consider testing inside a case with 120mm exhaust, no in-take, most popular config for 3rd party cases and it will benefit the passively cooled cards massively and be more accurate real world wise for those results.

Concerning Accelero S1, it beats all, and for the price, it very hard to recommend any other product for VGA cooling (if it fits your card).

compared to HR-03 and others: http://www.abchw.com/node/19
my own tests comparing 8 different VGA coolers: http://www.madshrimps.be/?action=getarticle&articID=593
 
Personally I rather have a stock cooler that actually works instead of having to pay an additional $40. That $40 can sometimes mean the difference between buying 2 cards
 
GREAT JOB on the roundup guys...

I have a feeling Newegg is about to sell out of their current stock of HR03's :D

Gonna post a news link from gamebeat.net if you don't mind. :D
 
Interesting info. on not including AC's S1... perhaps they were afraid that it would fail your rigorous "overclocking" tests when run in passive mode? On their website, new support for 8800GT cards is actually proclaimed (see here:

http://www.arctic-cooling.com/vga2.php?idx=138&data=7&disc=

Even on the main page they include 8800GT and HD3870 cards in the compatibility list.

Even Thermalright's HR03 GT actually failed in passive mode when oc'ed! I could see AC sighing a huge relief when they refused to provide you with a free sample. Well, nonetheless, I would not be surprised if the S1 did pass your o/c tests at lower than 90 degrees with "normal" case air flow.

Here are a couple questions:
Regarding the HR03-GT, what fan did you use? Fan size and RPM, please. Did you test passive mode with the heatsink behind the card (next to the CPU cooler) or in front of the card?
 
I thought I was gonna be the only one - yeah what are the 2 fans you guys used (importantly: what was the "Fan" (#1?) ?
 
Marc Adams and Kyle Bennett, thanks for this awesome round-up review! I bet it was fun doing this one, right? For me, it certainly would be much more fun than doing a review on a boring mid-range GPU.

Funny thing is I just posted a top list of best GPU air coolers before this review. And this review was in accord with my list:

http://www.hardforum.com/showthread.php?t=1249876

:D and :cool:
 
Deusfaux said:
I thought I was gonna be the only one - yeah what are the 2 fans you guys used (importantly: what was the "Fan" (#1?) ?
Click to expand...

If you read the HR03-GT page, you'd find out.. it's the fan position (#1 being front and #2 being rear).
 
JMke said:
Concerning Accelero S1, it beats all, and for the price, it very hard to recommend any other product for VGA cooling (if it fits your card).
Click to expand...

Doesn't the S1 have height issues that some users need to be worried over, depending on their case? I am pretty sure I read that somewhere (perhaps around here somewhere). I'm sure that would have been a issue to contend with vs. the HSF's tested here.

Also, Marc, did you have any issues at all with the ram sinks falling off or not wanting to stick? I've been reading of a lot of issues with the ram sinks that come with the S1. I presume you didn't use any 3rd party stuff to attach them?
 
ir0nw0lf said:
Also, Marc, did you have any issues at all with the ram sinks falling off or not wanting to stick? I've been reading of a lot of issues with the ram sinks that come with the S1. I presume you didn't use any 3rd party stuff to attach them?
Click to expand...

I think a lot of that comes from people who don't clean the BGA memory modules properly before trying to apply the ramsinks. The thermal pads that NVIDIA uses on the memory are greasy as hell and they leave a residue that doesn't come off without chemical warfare. Swabbing the modules with alcohol until they 'look' clean isn't going to do the job because the ceramic of those memory modules soaks up so much of that greasy residue. People need to be using something like Xylene (Goof-Off!) or the 2-part cleaning kit from Arctic Silver.
 
ir0nw0lf said:
Also, Marc, did you have any issues at all with the ram sinks falling off or not wanting to stick? I've been reading of a lot of issues with the ram sinks that come with the S1. I presume you didn't use any 3rd party stuff to attach them?
Click to expand...

I made sure to clean the memory chips thouroughly before installing the aftermarket ones. Once I did that it was much harder to remove the ram sinks than it was getting them to stick.:)

Deusfaux said:
I thought I was gonna be the only one - yeah what are the 2 fans you guys used (importantly: what was the "Fan" (#1?) ?
Click to expand...

Only one fan was used. #1 and #2 refer to the position it was mounted in.
THIS is the fan that was used for testing.
 
Now that I think about it more.. especially with AC's officially declared support for 8800GT and HD3870 cards for a while since (at least a couple weeks now), I really do believe that AC's S1 would definitely not have any problems passing HardOCP's rigorous passive mode testing with little air flow. If TT's HR03-GT could pass the default mode while being passive, AC's S1 could probably do it at least 20 degrees better with those widely-spaced fins (and perhaps a tad bit more total surface area).

The only reason I could think of is that HardOCP asked AC about it a long time ago and then turned a cold shoulder to AC's initial response without checking their website updates. I know for sure that support for 8800GT's was listed on their website for at least two weeks now.
 
Wow so you guys couldn't pool $25 to pick up an Accelero S1 for this article?
 
IsLNdbOi said:
Wow so you guys couldn't pool $25 to pick up an Accelero S1 for this article?
Click to expand...

I don't think it was a question of money. Think about it. If a company is unwilling to send you a sample and support the article why should you buy the cooler and give them what amounts to free publicity?
 
Great article, very much needed by the community as there has been a lot of speculation about GT cooling, bit of a shame there is no Accelero S1 as this has been favored by many.
 
Marc Adams said:
Only one fan was used. #1 and #2 refer to the position it was mounted in.
THIS is the fan that was used for testing.
Click to expand...

For convenience's sake, this is an Enermax 92mm fan that runs at up to 2400-2500 rpm, fully adjustable.. I still have this one and loved it over the years past. We could assume that HIGH is at 100% speed and that LOW is at maybe 50%, I guess (LOL)..
 
The Gonz said:
I don't think it was a question of money. Think about it. If a company is unwilling to send you a sample and support the article why should you buy the cooler and give them what amounts to free publicity?
Click to expand...

IMHO, it is because doing a service to your readers, who are after all picking up the Accelero S1 anyway with or without a comparative review of it, is more important than a measly $25 or spiting a company for not being cooperative. No disrespect intended, it was still a great review! :D
 
If the site did a little bit of research, you can find that AC offers fan modules for the S1 cooler, making it a active solution. It's only $8 + shipping from Newegg. S1 is a popular choice for after market cooler for 8800GT and it's a shame that HardOCP didn't order such cheap part and not included the S1 in the round up. You guys can do better than that. You've done system reviews that ordered PCs that costed $2000-3000. Why would you need to get review sample for a measly $20 GPU cooler?

Also, HR-3, Zerotherm 8800 might not be a viable solution for some. It looks like HR-3 with fan will block 3 slots and Zerotherm 8800 will eat up 2 more slots. This might be a critical issue for some as free PCI slots are precious thing these days.
 
IsLNdbOi said:
Wow so you guys couldn't pool $25 to pick up an Accelero S1 for this article?
Click to expand...

This article was put into motion in the first week of November. AC said it did not work on the GT. NOT they did not know, they explained it was not compatible. We moved on. Keep in mind that the level of testing we do is a bit more strict that what you might be doing at home and simply "adding" another cooler at any given time may not be so easy. Two weeks ago when AC sent us the news we had moved far enough forward in our work that adding it would have been a headache and likely put us off beyond Christmas, and I did not want to push the publication schedule back.

So it is not as simple as "pooling $25." I would suggest the amount of money we have spent on hardware this year would astound you. Money had nothing to do with the decision.
 
And for those of you that have further issue with no ACS1 being included, I understand, but you need to gripe at AC, not HardOCP. We were given bad information by AC directly. I am not sure how you can fault us for that. If we spent all our time trying coolers that mfgs said did NOT work, we would have had little time for anything else.

Thanks for the feedback. The point is fully noted. All other posts on AC will be deleted. Let's talk about what we do have rather more posts about what we do not.
 
Status
Not open for further replies.
Back
Top