GeForce GTX 285 On Water Cooling: Zotac's Infinity Edition

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Toggle sidebar Toggle sidebar
Status
Not open for further replies.
[citation][nom]neiroatopelcc[/nom]If even a noob can build a great water cooling setup, how come you couldn't? Just wondering. If you know about water cooling, why still do it wrong? I've only got 1 water cooling setup, so I'm no expert, but even I can see flaws...[/citation]

And you wonder about the short responses. Your use of the word "Couldn't" is beneath me. If you were to ask something inteligent, like "didn't", you might elicit a more favorable response.
 
I used the correct word for the purpose intended in my sentense. If I'd use didn't I'd imply I had confidence in your water cooling skills, which honestly I do not. You know a shitload about other things, but appearently not this subject. Sorry if my honesty offends you.
 
[citation][nom]neiroatopelcc[/nom]I used the correct word for the purpose intended in my sentense. If I'd use didn't I'd imply I had confidence in your water cooling skills, which honestly I do not. You know a shitload about other things, but appearently not this subject. Sorry if my honesty offends you.[/citation]

Those comments do not call into question my water cooling knowledge, but your perspective. You're not questioning a level of knowlege, but instead a level of intelect. Its not your honesty that's offensive, it's your stupidity.

Having said that, a smart person would have asked "Why did you use that configuration" to which there are some responses in previous posts. The card wasn't plumbed into the big cooler for a couple reasons: It didn't come with 1/2 inch barbs, and the spare barbs from previous Swiftech water block kits did not thread easily into the block. Rather than force things (something you seem to prefer), this was used as a second reason not to plumb it into that loop.

What was the first reason? It was considered to be an unnecessary intrusion on a perfect-working system that would have also reduced CPU cooling performance. After all, there are a couple smaller water coolers here.

Of course, all the smaller water coolers on hand were old. Of these, the Gigabyte unit had already proven its moderate capabilities in Core 2 Duo overclocking. Since it hadn't been tried with current hardware, this was seen as an opportunity to at least find out if it was "good enough" for the card.

Now, the determination of what's "good enough" is not directly related to actual temperature, but temperature relative to overclocking capability. IE, if you need more cooling to get a better overclock, you use a bigger cooler. In that respect, it was good enough.

How can it be good enough if the card couldn't go past its factory overclock? Simple: It reached the same stable speeds in both short tests (54C) and long test (74-78C). It would crash within minutes at 732 MHz (still in the 50's), but it would run for hours at 722 MHz (reaching the upper 70's). An unexperienced overclocker might not see the problem, but experienced overclockers would say "Needs more voltage".

Of course, more voltage would have caused more heat, changing the clock ceiling from a voltag issue to a heat issue. But here's where everything comes together: WE don't voltage mod for graphics overclock testing.

If the overclocking limit HAD been caused by inadequate cooling, rather than "needs more voltage", the article would have been delayed while we waited for a new cooling system to ship. Fortunately, "needs more voltage" issue meant that we'd already reached the card's limit at stock voltage, which is how we test cards. The reason this is fortunate is that the article didn't need to be delayed. Zotac wanted this published right away.

So, if overclocking is the reason to get better cooling, and the card was already at its limit due to other issues, no better cooling was needed. Why is it that you cannot understand this?

IN summary, if you wanted to ask a question, you used the wrong word. If you wanted to make a statement and mislead others by putting a question mark at the end, your deception cannot stand.
 
[citation][nom]Crashman[/nom]Those comments do not call into question my water cooling knowledge, but your perspective. You're not questioning a level of knowlege, but instead a level of intelect. Its not your honesty that's offensive, it's your stupidity.[/citation]
Clearly you've misinterpreted my responses. If you understood it as an insult to your interlect, you're mistaken. I'm sorry if I couldn't formulate myself in a matter you understood. But until someone is able to explain to you what I mean, I see very little reason to respond to the actual contents of your last post. At the end of the day, you wouldn't understand me anyway. You're still just trying to defend a choice you made, which incidently some people consider a wrong choice.

With regards to your summary though.
"if you wanted to ask a question, you used the wrong word. "
I assume that is regarding "how come you couldn't?" which questions the fact that obviously you could not built a great loop, even though you claim any neophyte can. That is in fact a question, and it is worded right. Granted it assumes you cannot do it, and not that you actively chose not to, but since that appears to be correct I don't see anything wrong with the wording. Understand that, and understand what I am on about in general, and you'll see I am not stupid, but merely not speaking your simplistic language.
 
[citation][nom]neiroatopelcc[/nom]Clearly you've misinterpreted my responses. If you understood it as an insult to your interlect, you're mistaken. I'm sorry if I couldn't formulate myself in a matter you understood. But until someone is able to explain to you what I mean, I see very little reason to respond to the actual contents of your last post. At the end of the day, you wouldn't understand me anyway. You're still just trying to defend a choice you made, which incidently some people consider a wrong choice.With regards to your summary though. "if you wanted to ask a question, you used the wrong word. "I assume that is regarding "how come you couldn't?" which questions the fact that obviously you could not built a great loop, even though you claim any neophyte can. That is in fact a question, and it is worded right. Granted it assumes you cannot do it, and not that you actively chose not to, but since that appears to be correct I don't see anything wrong with the wording. Understand that, and understand what I am on about in general, and you'll see I am not stupid, but merely not speaking your simplistic language.[/citation]

"Simplistic language" yeh, there we go again, so again I'll make this simple: You already knew that we COULD have built a better cooler, if we'd had the parts in hand. But you asked why we couldn't, rather than why we didn't.

The proper ANSWER would be why we didn't. And there are proper reasons why we didn't.

"Defending a choice"? The choices were the one used, something even less powerful, or beg and wait. The choice not to do the article wasn't available.

So, with no BETTER choice than the Gigabyte cooler, and the only possible alternative being to tell Zotac to go forth and multiply, the Gigabyte cooler turned out to be the superior choice. A bigger cooler was not one of the available choices.

Or, to put it in terms you might better understand: If you were living on the street and had been unable to afford food for several weeks, you might have the choices of eating from the trash or the soup kitchen. You would not have the choice of dining in a fine restaurant, even if you knew that a large amount of money was going to be direct-deposited into your bank account in another month.

Taking that a bit farther, say your background had been in culinary arts...and someone saw you eating in the soup kitchen. Would they be stupid enough to say "If you're eating here, you must not know how to cook"?
 
[citation][nom]neiroatopelcc[/nom]I see very little reason to respond to the actual contents of your last post.[/citation]

You mean the post that proves you're wrong, correct? Yeh, it's hard to respond to something like that. This is no longer a matter of opinion, just facts:

1a.) We admitted that the cooling system wasn't adequate for achieving the temperatures many water cooling officianados would prefer to see.
1b.) You won't admit that the problem was caused by lack of equipment.

2a.) We provided sufficient evidence that the card's overclocks were limited by non-thermal factors.
2b.) You won't admit that the only reason we would have NEEDED additional cooling is if the card's overclock had been limitted by thermal factors.

3a.) I showed how a neophyte could go in blind and build an adequate single-component water cooler simply by choosing bigger parts.
3b.) You won't admit you were wrong to insist that some kind of advanced knowledge would be required to build a single-component loop.

4a.) We proved that we had more than that level of knowlege but were limited to the parts we had on hand.
4b.) You keep insisting that we were somehow not limited to parts on hand and must therefor lack the advanced knowledge that you already know we didn't need...but had anyway.

So just admit that you're wrong on all counts. You see, fairness and honesty are more important to me than kindness.
 
[citation][nom]Crashman[/nom]You mean the post that proves you're wrong, correct? Yeh, it's hard to respond to something like that. This is no longer a matter of opinion, just facts:1a.) We admitted that the cooling system wasn't adequate for achieving the temperatures many water cooling officianados would prefer to see.1b.) You won't admit that the problem was caused by lack of equipment.2a.) We provided sufficient evidence that the card's overclocks were limited by non-thermal factors.2b.) You won't admit that the only reason we would have NEEDED additional cooling is if the card's overclock had been limitted by thermal factors.3a.) I showed how a neophyte could go in blind and build an adequate single-component water cooler simply by choosing bigger parts.3b.) You won't admit you were wrong to insist that some kind of advanced knowledge would be required to build a single-component loop.4a.) We proved that we had more than that level of knowlege but were limited to the parts we had on hand.4b.) You keep insisting that we were somehow not limited to parts on hand and must therefor lack the advanced knowledge that you already know we didn't need...but had anyway.So just admit that you're wrong on all counts. You see, fairness and honesty are more important to me than kindness.[/citation]


You say you were limited to parts on hand. We'll thats your root problem. Your Toms Hardware, you are worth millions as a company. You can't afford or don't have the clout to get with the manufacturers to get good gear samples? I can't believe that.

It's a lack of effort or a lack of upper management to support proper, accepted (getting respect) testing.

Your probably too busy to do it all and with the economy and advertising cutting back, you can't hire a Q WC person or provide him with a office and a budget. Just stop the hack WC reviews till you can get respect.
 
[citation][nom]Conumdrum[/nom]You say you were limited to parts on hand. We'll thats your root problem. Your Toms Hardware, you are worth millions as a company. You can't afford or don't have the clout to get with the manufacturers to get good gear samples? I can't believe that.It's a lack of effort or a lack of upper management to support proper, accepted (getting respect) testing.Your probably too busy to do it all and with the economy and advertising cutting back, you can't hire a Q WC person or provide him with a office and a budget. Just stop the hack WC reviews till you can get respect.[/citation]

We had put in a request for additional cooling components, but it hadn't arrived yet. This article was time sensitive, so a decision had to be made.

We made the right one, to go along with the manufacturer's desire to get the review up quickly. That it was the "right decision" was determined by the finding that the card didn't overclock better in the mid-50's than it did in the upper-70's. For Tom's Hardware, better overclocking was THE ONLY REASON to consider additional cooling.

So there's the other side, "We don't like those temperatures regardless of whether or not the GPU was stuck". That's OK, it's at least as important as sayihg "it's the wrong color" but not worthy of several pages of discussion.

The important thing is that we admitted the cooler was inadequate for achieving the desired temperatures, but went ahead anyway since the card didn't overclock better in the mid 50's than it did in the upper 70's.

Of course you can look forward to eventually seeing that new cooling kit we'd requested in action. But the "GPU Stuck" meant that there wasn't an adequate reason to wait for it to show up.
 
[citation][nom]Conumdrum[/nom]You say you were limited to parts on hand. We'll thats your root problem.[/citation]

BTW, the water never got hot. The radiator never got hot. And the copper portion of the card's top only got warm. Most likely reason for the high temperature reading is a bad sensor or software glitch, since we can't believe Zotac would send a hand-assembled card with the water block improperly installed.

Because the water block was much warmer than the radiator, we decided to got ahead and get a better cooling system anyway.

You won't like what we picked.
 
I just have to quote you on this part.

"We made the right one, to go along with the manufacturer's desire to get the review up quickly."

When doen the manufacturer ever dictate to the tester that it has to be done now?

This says your reviews are based on a business model and not science. You should on never ever said that. Ever.

I give up and will take all of the TH reviews with a grain of salt tossed in your general direction, forever more.

I'll stick with other reviews for any final decisions.

And I won't twitter your words to the rest of the testing world.

You might as well just stop postin, your digging a deep deep hole for TH.
 
[citation][nom]Conumdrum[/nom]I just have to quote you on this part."We made the right one, to go along with the manufacturer's desire to get the review up quickly."When doen the manufacturer ever dictate to the tester that it has to be done now? This says your reviews are based on a business model and not science. You should on never ever said that. Ever. I give up and will take all of the TH reviews with a grain of salt tossed in your general direction, forever more.I'll stick with other reviews for any final decisions.And I won't twitter your words to the rest of the testing world.You might as well just stop postin, your digging a deep deep hole for TH.[/citation]

Well then I have startling news for you: All "product launches" are time sensitive and every reviewer faces the same decisions process from time to time:
1.) Is there a problem?
2.) Is the problem significant enough to force a delay?

Anyone who says they've never faced such a decision is lying, so if anything the disclosure makes this site even more trustworthy.

If the answer to #2 is no, you move on. This editor provides a degree of disfavoritism and lack of corporate bias that you won't find at many other sites, and if this editor is ever forced out of the business it will be for not "towing the corporate line". Read the most recent motherboard comparison for an example.

So why would anyone chose to "tow the line" sometimes and not others? Easy: The problem wasn't significant enough to force a delay. Heat wasn't the cause of the overclocking limitation, thus it didn't affect benchmarks and only showed up as a glitch in the peak temperature chart.

Some sites would have simply left the temperature chart out because the card was nowhere near peak tempertature when its overclocking problem was discovered. Perhaps you'd prefer that, but this is a site with integrity.

So on to your assertion that reviews are based on a business model and not science: We scientifically determined that the high peak temperature reading was not significant and went on with the review. What this says is that your comments are arbitrary and not based on science.
 
[citation][nom]neiroatopelcc[/nom]I thought that too, until you published this article![/citation]

LOL, admitting that the cooler was slightly underpowered and publishing the temperatures anyway is integrity. A site with less integrity could have easily neglected such dislosures. Some sites would have even rigged the test to favor the liquid-cooled card...possibly by choosing a less-capable air-cooled card sample.

You didn't like the reported peak temperatures, we didn't like that the card showed up with fingerprints oxidized into the surface. Both had the same level of effect on GPU overclocking and benchmark performance, which were THE test criteria.
 
This is proffesional reviews. A small company way below your budget.
http://cde.cerosmedia.com/the-overclocker/1K49d60f88110a8958.cde
You have been owned, you got lots to gain. But times are a changing fast.

They have the web ability, show good stuff, amazing web programming ability, amazing web publishing. You want to even think of capturing the market, these guys need to be beat.

http://cde.cerosmedia.com/the-overclocker/1K49d60f88110a8958.cde

Good luck with your pointy headed Dilbert boss. Your just to big to compete with this kind of quality unless you buy them outright.

Anyway enough of you admitting your review was limited and of no real value no matter your manpower costs to test and write it. It was a poor review. By now instead of consistently rehashing the mistakes, you should of pulled the review.

Your pride is getting the best of you. Quit justifying a faulty review, get over it. It's a job. Not your ability in the sack size issue.
 
[citation][nom]Conumdrum[/nom]This is proffesional reviews. A small company way below your budget.http://cde.cerosmedia.com/the-over [...] 0a8958.cdeYou have been owned, you got lots to gain. But times are a changing fast.They have the web ability, show good stuff, amazing web programming ability, amazing web publishing. You want to even think of capturing the market, these guys need to be beat.http://cde.cerosmedia.com/the-over [...] 0a8958.cdeGood luck with your pointy headed Dilbert boss. Your just to big to compete with this kind of quality unless you buy them outright.Anyway enough of you admitting your review was limited and of no real value no matter your manpower costs to test and write it. It was a poor review. By now instead of consistently rehashing the mistakes, you should of pulled the review.Your pride is getting the best of you. Quit justifying a faulty review, get over it. It's a job. Not your ability in the sack size issue.[/citation]

No, I'm admiting only two things:
1.) The cooler wasn't big enough to provide optimal temperatures.
2.) You're wrong about everything else.

We could have justified delaying the article if better cooling was required, but a number on a temp chart wasn't enough to convince us when everything else pointed in other directions.

Quit justifying a faulty opinion. The peak temperature reported is simply an irritant for nitpickers.

And like I said, you won't like the new LC component cooling system either.
 
[citation][nom]Crashman[/nom]No, I'm admiting only two things:1.) The cooler wasn't big enough to provide optimal temperatures.2.) You're wrong about everything else.[/citation]
You can't admit that someone else is wrong even in the extremely unlikely event that you are right.

[citation][nom]Crashman[/nom]Quit justifying a faulty opinion.[/citation]
You'd be wise to follow your own suggestion mate. Even if one would argue that an opinion can't be faulty, but merely different from another. Facts can be correct or incorrect, but you'd not know anything about that ofcourse.
 
[citation][nom]neiroatopelcc[/nom]Facts can be correct or incorrect, but you'd not know anything about that ofcourse.[/citation]

1.) I've determined that the cooling was adequate because the card's limitations were similar or identical at both moderate and high temperatures.
2.) You've determined that the cooling was inadequate because the temperatures were higher than expected.
3.) I agree with that assessment.

So what's the problem? The problem is that even though I conceeded your point, you failed to conceed mine. And we both know that both points are valid. Which says a lot about our perspectives.
 
[citation][nom]Crashman[/nom]So what's the problem? The problem is that even though I conceeded your point, you failed to conceed mine. And we both know that both points are valid. Which says a lot about our perspectives.[/citation]
I answered that question before you even asked it. On the first page actually. The article is completely useless as it doesn't come to any conclusion other than that this card overclocks less good than an unspecified other card (specified in comments by you later). But people don't care about that, they care about wether or not this is the solution to running sli without overheating.

[citation][nom]neiroatopelcc[/nom]I wonder why you bothered doing this article? As we can all agree this card only makes sense if you plan on putting it in a very tight place, or have 2-3 of them in a chassis. So why didn't you request 3 cards and compare it to a standard 3 card setup?[/citation]
 
[citation][nom]neiroatopelcc[/nom]I answered that question before you even asked it. On the first page actually. The article is completely useless as it doesn't come to any conclusion other than that this card overclocks less good than an unspecified other card (specified in comments by you later). But people don't care about that, they care about wether or not this is the solution to running sli without overheating.[/citation]

The card didn't overheat, and that was one of the concepts mentioned in the article as well. If you had 3x the cooling for 3 cards, those 3 cards wouldn't overheat either.

It's too bad we couldn't just pick up three GTX 285 air-cooled cards to show the air-cooling limits of 3-way SLI. Doing so would have cost too much.
 
Get em from newegg like you do with smb's ... or find another sponsor for that purpose. I'm sure xfx or evga would want to prove that their air cooled cards are better value than another companies water cooled. But it doesn't really change anything, cause you didn't try that. You merely tested the oc limit, which frankly I don't think is very important. Chances are in 3 way sli - which undoubtfully would be the only application for a water cooled 285 - your cpu can't deliver enough data to satiate 3 cards anyway. But we'll not know that ofcourse, cause you weren't able to provide any such setup to test on. Your article could in fact be reduced to a one page news article simply stating that zotac's got a new water cooled 285. It performs like any other 285 oc card, and apart from noise reduction you can't see any benefit in a single card setup.
 
[citation][nom]neiroatopelcc[/nom]Get em from newegg like you do with smb's ... or find another sponsor for that purpose. I'm sure xfx or evga would want to prove that their air cooled cards are better value than another companies water cooled. But it doesn't really change anything, cause you didn't try that. You merely tested the oc limit, which frankly I don't think is very important. Chances are in 3 way sli - which undoubtfully would be the only application for a water cooled 285 - your cpu can't deliver enough data to satiate 3 cards anyway. But we'll not know that ofcourse, cause you weren't able to provide any such setup to test on. Your article could in fact be reduced to a one page news article simply stating that zotac's got a new water cooled 285. It performs like any other 285 oc card, and apart from noise reduction you can't see any benefit in a single card setup.[/citation]

Thanks for the summary! When you're straightforward like that, you actually start to sound like an editor...PM me.
 
Status
Not open for further replies.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom