ASUS GeForce GTX 590 Video Card Review @ [H]

Yakk said:
HOLY CR@P!

Did you see that thermal image for the 590?

Compare to thermal image for the 6990.

You don't need translation to see that! I was a little nervous seeing the 6990 thermals, the 590 wins HANDS DOWN here!

Y.

Mind you both are up there...
Click to expand...

Yikes- I guess we can see at least part of why #1- its quieter and #2 why they're burning lol :-p
 
Wow, it seems that even quiet came at a price. No surprise that they're burning up left and right!
 
Interesting info as nvidia has made a official statement regarding the cards dying

Link blocked will post article info below (a few posts down)

Apparently in the reviewers kit they told reviewers to use the new driver and not the bata, the beta was to test early engineering samples, and they also listed a set of voltages they recommended staying within, You had to specifically ignore both bullet points for this card to fry.

Anyway on a side note, I'm sure people will really be considering a factory overclocked model of these cards as the risk of a frying card that is covered under warranty due to factory overclocking sounds like a good proposition with all the current turn of events. Here is a card that's on the way http://www.nordichardware.com/news/...verclocked-geforce-gtx-590-ultra-charged.html
 
Last edited:
Lord_Exodia said:
Interesting info as nvidia has made a official statement regarding the cards dying

http://www.nordichardware.com/news/...verclocked-geforce-gtx-590-ultra-charged.html

Apparently in the reviewers kit they told reviewers to use the new driver and not the bata, the beta was to test early engineering samples, and they also listed a set of voltages they recommended staying within, You had to specifically ignore both bullet points for this card to fry.

Anyway on a side note, I'm sure people will really be considering a factory overclocked model of these cards as the risk of a frying card that is covered under warranty due to factory overclocking sounds like a good proposition with all the current turn of events. Here is a card that's on the way http://www.nordichardware.com/news/...verclocked-geforce-gtx-590-ultra-charged.html
Click to expand...

Ignoring bullet points is exactly why enthusiast hardware sites exist :)
 
With a successful launch of their flagship card today, March 24th, 2011, the NVIDIA GeForce GTX 590 was met with positive feedback…for the most part.

It appears that some of those who were out to test the limits of the behemoth, managed to capture the $700 card going up in smoke, literally. Sweclockers, among others, reported and video taped that the GTX 590 was in fact so powerful that the card killed itself during overclocking. While using the beta drivers which were not intended for release, some reviewers willingly took the voltages of their GTX 590s far past NVIDIA’s limits or recommendations. It should be noted, that a quick sampling of many of the web’s premier review websites were using driver version 267.71 in their reviews while others were using slightly older AiB-supplied drivers which did not implement the over current protection properly.

“The few press reports on GTX 590 boards dying were caused by overvoltaging to unsafe levels (as high as 1.2V vs. default voltage of 0.91 to 0.96V), and using older drivers that have lower levels of overcurrent protection. Rest assured that GTX 590 operates reliably at default voltages, and our 267.84 launch drivers provide even more additional levels of protection for overclockers. For more information on overclocking and overcurrent protection on GTX 590 please see our knowledge base article here: http://nvidia.custhelp.com/cgi-bin/nvidia.cfg/php/enduser/std_adp.php?p_faqid=2947.”

In addition, here is an extract from ASUS’ GTX 590 reviewer’s guide:

It is not advised to exceed the 1.050 to 1.065 vcore range as this begins to meet the limits for the OCP/OVP mechanism on the card. Exceeding these values without disabling OCP/OVP or having superior cooling could affect the lifespan and functionality of the card/gpu.

The lesson of the day is this folks: ignoring manufacturer’s recommendations and overclocking your card way beyond its limitations (especially when using beta drivers and beta software) can in fact lead to the unfortunate killing of hardware. Fancy that eh?
 
Do they think this card was treated any different than all the others people have tested? I've never seen anything die on as many reviewers as the GTX590.

Not to mention crucial protection is implemented in software... WTF. I guess you can do that when you have such reliable and great drivers, eh?
 
I finally got around to reading this, and I have to say that it is yet another nice review from you guys. If this is truly a limited edition card, then i don't see $700 being a bad value despite the fact that the 6990 appears to be a better option for performance. I cannot wait to read the followup you guys will do as far as overclocking is concerned on the 590, as it just seems to be a shame to use GPUs with a full 512 cuda cores clocked this low. If near GTX 580 SLI performance can be achieved from overclocking, then this would go from being a meh value to being a great value considering that the cheapest 580 SLI setup is still $1000.

What I find that was totally AWESOME about this review, was that you guys ended it with a Duke Nukem one-liner on the same day that we just learned that DNF was going to be DELAYED :mad:
 
Not really surprising that the card couldn't keep up. While the GTX 580 is still the top GPU, it isn't that efficient and the performance per watt of the 6970 is much better. With the dual card solutions being power limited by the PCIe spec, performance per watt is much more important than the cards raw performance and Nvidia just had to downclock the card too far to fit within 375W.

That said, the GTX 590 may be a very interesting card when water cooled. For the price of a pair of GTX 580s, you could buy a GTX 590 and add an entire water cooling setup for it. If the power regulators are good enough, the card once overclocked would be faster than a pair of GTX 580s.
 
Lord_Exodia said:
The lesson of the day is this folks: ignoring manufacturer’s recommendations and overclocking your card way beyond its limitations (especially when using beta drivers and beta software) can in fact lead to the unfortunate killing of hardware. Fancy that eh?
Click to expand...

You're acting like 1.05 to 1.065 is monumental or uncommon for the 580. Since this is an enthusiast card, how many people do you think are going to go for 580 clocks on this 590? I'd say damn near all. Those voltages fall quite close to standard 580 voltages so most people will think those voltages should be acceptable. The truth of the matter is that this card is made for low temps and low voltage/power which is great ..........if it was being marketed this way. Instead many sites are talking about how "we just can't wait to get it to 580 speeds" as if this is guaranteed without upping the voltage. It's not. Contrast this to 6990 whereby flipping a switch pretty much brings you pretty damn close to 6970 speeds.

Nvidia is trying to hide the per per watt disadvantage here while inviting people to overclock like there's no tomorrow. You can't have both in any great quantity when the architecture itself can't do it in great quantity on lesser models.
 
They didn't turn on the HD 6990's switch in this review, I'm sure 99% of people who buy a HD 6990 will turn on the switch if they know about it, which makes the 590 look even worse :p
 
chanchan said:
They didn't turn on the HD 6990's switch in this review, I'm sure 99% of people who buy a HD 6990 will turn on the switch if they know about it, which makes the 590 look even worse :p
Click to expand...
and the same type of person would simply oc the gtx590...
 
Lord_Exodia said:
The lesson of the day is this folks: ignoring manufacturer’s recommendations and overclocking your card way beyond its limitations (especially when using beta drivers and beta software) can in fact lead to the unfortunate killing of hardware. Fancy that eh?
Click to expand...

When did this become oft forum? People here overvolt and overclock all the time. There have been very few cards known to actually go up in flames even when pushed past spec. The 590 happens to be one of those. Hopefully those 'drivers' fix the problem but it sounds like a hardware issue to me. I think we see now why Nvidia had to downlock the 590 so severely. This video card fails as enthusiast hardware if it can't take performance to enthusiast levels before burning itself up.
 
I don't have a TechPowerUP username so please bear with me. One of the things that I like, which I totally forgot to look at until now, is perf/watt. I thought that since the 590 will have more work units and lower clocks, it would have a relatively low voltage, so the perf/watt would be good. That said, look at TechPowerUP's perf/watt measurements.

perfwatt_2560.gif


The 6990 gets 30% more FPS per watt than the 590. If the slightly better performance wasn't one thing, the perf/watt at stock clocks is another one entirely. Even the 580, which is running ~20% faster, has better perf/watt.
 
how does the 5450 top that chart since I cant think of any game it could even run at that resolution? and certainly not at the settings they use in their reviews.
 
kac77 said:
You're acting like 1.05 to 1.065 is monumental or uncommon for the 580. Since this is an enthusiast card, how many people do you think are going to go for 580 clocks on this 590? I'd say damn near all. Those voltages fall quite close to standard 580 voltages so most people will think those voltages should be acceptable. The truth of the matter is that this card is made for low temps and low voltage/power which is great ..........if it was being marketed this way. Instead many sites are talking about how "we just can't wait to get it to 580 speeds" as if this is guaranteed without upping the voltage. It's not. Contrast this to 6990 whereby flipping a switch pretty much brings you pretty damn close to 6970 speeds.

Nvidia is trying to hide the per per watt disadvantage here while inviting people to overclock like there's no tomorrow. You can't have both in any great quantity when the architecture itself can't do it in great quantity on lesser models.
Click to expand...

In case you missed it, those weren't my words, I was posting a short article with nvidia's official response to that site and the article writers sentiments.

a. The Asus review kit specifically said not to use the old driver and use the new one

b The asus review kit specifically said the range of voltages one could use to overclock the card as they had tested them and found them to be the safe zone. You can indeed overclock and overvolt just not all the way to 1.2V on the stock cooler, not with the stock fan setting, and especially not with the driver they told you to avoid in the first place LOL

vjcsmoke said:
When did this become oft forum? People here overvolt and overclock all the time. There have been very few cards known to actually go up in flames even when pushed past spec. The 590 happens to be one of those. Hopefully those 'drivers' fix the problem but it sounds like a hardware issue to me. I think we see now why Nvidia had to downlock the 590 so severely. This video card fails as enthusiast hardware if it can't take performance to enthusiast levels before burning itself up.
Click to expand...


Well for now what it being explained and widely accepted seems to be that the issue is driver related and no more. The asus review kit said the safe voltages to use with the correct driver the other driver was specifically mentioned to not use for the review. Furthermore the voltage recommendations were a range and recommended only if you were adjusting the fan and if going over... then getting a better cooler for the card as to avoid damaging the product.
 
Why are the settings not the same? It seems to me you'd want to have the settings for each card identical not just 'highest settings' the card will play. You're comparing the cards to each other correct? Maybe I'm just not getting it, but in testing I've done in other areas of research (school and my job) we wanted the control to be the same for every test especially if we are comparing two things. Just my 2 cents.
 
gman said:
Why are the settings not the same? It seems to me you'd want to have the settings for each card identical not just 'highest settings' the card will play. You're comparing the cards to each other correct? Maybe I'm just not getting it, but in testing I've done in other areas of research (school and my job) we wanted the control to be the same for every test especially if we are comparing two things. Just my 2 cents.
Click to expand...

If your referring to the article then it's highest playable settings possible. Obviously you have 1 main thing to look at there. 1. The card allowing the highest playable settings while staying above the playable fps line wins.

Also they have apples to apples which is same settings. Is this the first graphic card review you've read here in about the last year or is this the first time you asked about how they do their reviews?
 
Yup, another Nvidia DUal GPU fail. Even with GF110 being better Heat and power wise than the GF100 they still can't get those issues under control without downclocking.

Great article.
 
cannondale06 said:
how does the 5450 top that chart since I cant think of any game it could even run at that resolution? and certainly not at the settings they use in their reviews.
Click to expand...

Yup they need to use a reasonable cutoff point for low end performance. I think they feel they need to use 30 cards to make their review look more impressive.
 
Ultima99 said:
Yup they need to use a reasonable cutoff point for low end performance. I think they feel they need to use 30 cards to make their review look more impressive.
Click to expand...

They just have a suite, they go through alot more hardware than hardocp.
hardocp tells us more about the card itself while techpowerup usually have more neat things like bios things about the cards, while hardocp actually doesnt have that.
Alltho hardocp finds where the card stops chewing, cause of memory etc etc.
Techpowerup is better than many many sites out there, and one of the few that runs a performance test on 5450 etc.
but perf/watt for 590 and 6990 is pretty useless at 2560x1600...
 
Lord_Exodia said:
The lesson of the day is this folks: ignoring manufacturer’s recommendations and overclocking your card way beyond its limitations (especially when using beta drivers and beta software) can in fact lead to the unfortunate killing of hardware. Fancy that eh?
Click to expand...

Yes, but usually the damage due to overvolting/overclocking show up a few years down the road, not a few minutes.
 
I will wait for DX12 cards to come out, then i will buy the top DX11 card and i will get a great card for a great price with solid proven drivers. Should be plenty of DX11 games by then ;)
My 4870 Dark Night 1gig still rockin it.
 
Lord_Exodia said:
In case you missed it, those weren't my words, I was posting a short article with nvidia's official response to that site and the article writers sentiments.

a. The Asus review kit specifically said not to use the old driver and use the new one

b The asus review kit specifically said the range of voltages one could use to overclock the card as they had tested them and found them to be the safe zone. You can indeed overclock and overvolt just not all the way to 1.2V on the stock cooler, not with the stock fan setting, and especially not with the driver they told you to avoid in the first place LOL
Click to expand...

#1. Actually you quoted partially from Hardware Canucks and the ASUS reviewer guide. Anyway Asus is recommending you stay below 1.065 (obviously also below 1.2). The problem as I stated before, 1.065 falls WITHIN the range of variance of a normal 580 which most people are going to strive for on a $700 video card. It's going to happen.

#2 Now in terms of drivers, when was the last time a video card has fried due to drivers? How many of us have used Beta drivers before? I know I have. I'm going to go on a limb here and say just because a driver is beta that shouldn't mean that my whole rig is in danger of catching fire.

Finally, I'm pretty sure that most hardware reviewers can read and probably got shipped the drivers WITH the card. If ASUS (see #2) sent a driver with the card only to send another later and says "don't use that one use this one because the old driver could cause a meltdown" I think we would still have a problem because it doesn't take away from the fact that something below par was sent to reviewers. In addition now you're talking about how a simple driver update could cause your perfectly stable overclock to not only become unstable, but to put your $700 video card at risk of catching fire within the case.
 
Nope I've read plenty of reviews here and have always wondered about it. Not sure I agree with how it's done, but it's not my site. And no they are not always the same settings.

If your referring to the article then it's highest playable settings possible. Obviously you have 1 main thing to look at there. 1. The card allowing the highest playable settings while staying above the playable fps line wins.

Also they have apples to apples which is same settings. Is this the first graphic card review you've read here in about the last year or is this the first time you asked about how they do their reviews?
Click to expand...
 
gman said:
And no they are not always the same settings.
Click to expand...

The apples to apples are ALWAYS the same settings as that is the very definition of Apples to apples.

You'll see the highest playable settings which will be different, then just below that you will usually see another chart where both cards are running the same exact settings (apples to apples) to show how they both compare with exact same settings.

This should be right up your alley as this is exactly what your asking for. Note though that sometimes they apples to apples are at the high settings the weaker card can't run at a playable fps so watch that carefully, Either way it gives you a great idea how much more powerful one card is over the other. Hell they actually do the math for you 99% of the time by saying The Radeon 6990 is 23% faster than 570 SLi here etc...
 
Creig said:
Yes, but usually the damage due to overvolting/overclocking show up a few years down the road, not a few minutes.
Click to expand...

Again a driver issue is supposedly causing the instant meltdown. If you overclock/overvolt with the right driver, it will take time to damage the card. Asus review kit said if you are overclocking and overvolting you need to adjust the fan profile for higher cooling. If overvolting beyond their recommended range you may need a better cooler otherwise it will degrade the reliability of the product... not instant meltdown.

kac77 said:
#1. Actually you quoted partially from Hardware Canucks and the ASUS reviewer guide. Anyway Asus is recommending you stay below 1.065 (obviously also below 1.2). The problem as I stated before, 1.065 falls WITHIN the range of variance of a normal 580 which most people are going to strive for on a $700 video card. It's going to happen.
Click to expand...

If that is the case then it's a PEBKAC issue. Your not going to compare a single gpu card and it's voltage tollerances and apply that automatically to a dual gpu videocard. They may be using lower voltage chips etc.. They may have more/less pwm to gpu ratio etc.. That is why you should be looking at reviews or overclocking carefully and scaling things up little by little while monitoring. If someone ignores all reviews, gets a card that they know nothing about the design, and card is a dual gpu card and they crank the voltage to the max then... PEBKAC

kac77 said:
#2 Now in terms of drivers, when was the last time a video card has fried due to drivers? How many of us have used Beta drivers before? I know I have. I'm going to go on a limb here and say just because a driver is beta that shouldn't mean that my whole rig is in danger of catching fire.
Click to expand...

Sadly nvidia has released drivers to the retail market or available for download that have cause issues, fan stopping etc... this is a little different from what I gather. This was a early driver to test engineering samples for sake of simple functionality and application testing. Nvidia is making it seem like this driver should never have been burned and added to a CD for retail use. Not sure how much of that is a B.S. damage control excuse or if it holds any water but that is the part that makes me refuse to join any fanboy bandwagon and jump to conclusions without information to go by.

kac77 said:
Finally, I'm pretty sure that most hardware reviewers can read and probably got shipped the drivers WITH the card. If ASUS (see #2) sent a driver with the card only to send another later and says "don't use that one use this one because the old driver could cause a meltdown" I think we would still have a problem because it doesn't take away from the fact that something below par was sent to reviewers. In addition now you're talking about how a simple driver update could cause your perfectly stable overclock to not only become unstable, but to put your $700 video card at risk of catching fire within the case.
Click to expand...

IF and that's a big IF they sent a reviewer kit with specific instructions along with a retail box product as a end user would get then there are 2 factors to consider here. 1. Reviewers fault to read the review kit and see that they should download a specific driver instead of using the retail disk. If they go ahead and use the driver specifically told by the manufacturer to avoid then PEBKAC.. Yes even a reputable reviewer like W1zard can have a PEBKAC issue. 2. Asus or Nvidias fault for allowing the driver to make it on a retail box that a end user can get that will melt the card down. See my reply to point 1, the jury is still out on this one.

On a side note keep in mind that some of these reviews were done in the UK, Germany etc.. so they may have another manufacturers product and who knows what their reviewers kit was saying or recommending.
 
Last edited:
Lord_Exodia said:
With a successful launch of their flagship card today, March 24th, 2011, the NVIDIA GeForce GTX 590 was met with positive feedback…for the most part.

It appears that some of those who were out to test the limits of the behemoth, managed to capture the $700 card going up in smoke, literally. Sweclockers, among others, reported and video taped that the GTX 590 was in fact so powerful that the card killed itself during overclocking. While using the beta drivers which were not intended for release, some reviewers willingly took the voltages of their GTX 590s far past NVIDIA’s limits or recommendations. It should be noted, that a quick sampling of many of the web’s premier review websites were using driver version 267.71 in their reviews while others were using slightly older AiB-supplied drivers which did not implement the over current protection properly.

“The few press reports on GTX 590 boards dying were caused by overvoltaging to unsafe levels (as high as 1.2V vs. default voltage of 0.91 to 0.96V), and using older drivers that have lower levels of overcurrent protection. Rest assured that GTX 590 operates reliably at default voltages, and our 267.84 launch drivers provide even more additional levels of protection for overclockers. For more information on overclocking and overcurrent protection on GTX 590 please see our knowledge base article here: http://nvidia.custhelp.com/cgi-bin/nvidia.cfg/php/enduser/std_adp.php?p_faqid=2947.”

In addition, here is an extract from ASUS’ GTX 590 reviewer’s guide:

It is not advised to exceed the 1.050 to 1.065 vcore range as this begins to meet the limits for the OCP/OVP mechanism on the card. Exceeding these values without disabling OCP/OVP or having superior cooling could affect the lifespan and functionality of the card/gpu.

The lesson of the day is this folks: ignoring manufacturer’s recommendations and overclocking your card way beyond its limitations (especially when using beta drivers and beta software) can in fact lead to the unfortunate killing of hardware. Fancy that eh?
Click to expand...


Subjective opinion follows :

All this means to me is that nvidia upper management seem to have taken the decision to use up practicaly all the engineering safety overhead (read: Overclocking headroom) probably against the wishes of their engineering department (I would have REALLY liked to see THAT discussion!) to try & match the 6990 stock performance for their flagship product. And then software driver programers scambled to fix their code to match, hence driver delays.

It's a fine, and dangerous, line when you start using up your safety margin. We are just talking about a graphics card here, but the same safety margins apply to all types of engineering.

Good thing nvidia doesn't engineer bridges or planes... :eek:


Y.

Flabergasted...
 
Yakk said:
Subjective opinion follows :

All this means to me is that nvidia upper management seem to have taken the decision to use up practicaly all the engineering safety overhead (read: Overclocking headroom) probably against the wishes of their engineering department (I would have REALLY liked to see THAT discussion!) to try & match the 6990 stock performance for their flagship product. And then software driver programers scambled to fix their code to match, hence driver delays.

It's a fine, and dangerous, line when you start using up your safety margin. We are just talking about a graphics card here, but the same safety margins apply to all types of engineering.

Good thing nvidia doesn't engineer bridges or planes... :eek:


Y.

Flabergasted...
Click to expand...


Thats a good point.

I think the take home message and its even a message Nvidia has kindof brought up is that the GTX590 doesn't seem to be geared towards the enthusiast market.


I am hoping that someone will eventually do a Postmortem interview with Nvidia, maybe one of Nvidia's engineers on the Fermi lineup
 
Lord_Exodia said:
Again a driver issue is supposedly causing the instant meltdown. If you overclock/overvolt with the right driver, it will take time to damage the card. Asus review kit said if you are overclocking and overvolting you need to adjust the fan profile for higher cooling. If overvolting beyond their recommended range you may need a better cooler otherwise it will degrade the reliability of the product... not instant meltdown.
Click to expand...

Nvidia needs to implement, in hardware, certain minimum/maximum safety thresholds that no amount of software can exceed. If they don't, scenarios like this can, and will, happen again.
 
Ultima99 said:
Yup they need to use a reasonable cutoff point for low end performance. I think they feel they need to use 30 cards to make their review look more impressive.
Click to expand...

No, they are one of the few sites that have this info, which is greatly appreciated. Every now and then I have to select a lower end card for something, and info about what's the smart lower end buy is scarce.
 
Creig said:
Nvidia needs to implement, in hardware, certain minimum/maximum safety thresholds that no amount of software can exceed. If they don't, scenarios like this can, and will, happen again.
Click to expand...


Threshold is already there.
It's not like the card got overvolted from 0.91V to 1.2V by itself, or via fluctuations of current.

And there will allways be a way to pass manufacturers limit. Like with Afterburner that got locked at +100mV. So I immediately found unlocker. Or when they introduced power limiter, and ppl still disabled it and ran Furmark and whatnot.

And if they do it what should Nvidia do against hardware moding? Seal the card in steel container?
I prefer having an overvolt option few clicks away. Besides, no manufacturers warranty covers ANY overvolting.

In the web release driver of GeForce GTX 590, we have added some important enhancements to our overcurrent protection for overclocking. We recommend anyone doing overclocking or running stress apps to always use the latest web driver to get the fullest protection for your hardware. Please note that overcurrent protection does not eliminate the risks of overclocking, and hardware damage is possible, particularly when overvoltaging. We recommend anyone using the GTX 590 board with the reference aircooler stick with the default voltage while overclocking, and avoid working around overcurrent protection mechanisms for stress applications. This will help maintain GTX 590's great combination of acoustics, performance, and reliability. NVIDIA has worked with several watercooling companies to develop waterblocks for GTX 590, and these solutions will help provide additional margin for overclocking, but even in this case we recommend enthusiasts stay within 12.5-25mV of the default voltage in order to minimize risk.

These are guidelines only - any overclocking/overvoltaging can void your manufacturer's product warranty.
Click to expand...
 
Actually the cooling is pretty amazing.

Using Vantage as an example, we also underclocked and overclocked our GTX 590 and GTX 580 SLI pair..
For overclocking, we did not raise the GTX 590′s voltage nor did we alter the fan profile. It overclocked to a maximum of 690/1825 from the reference clocks of 605/1707. It only got slightly warmer and the fan tended to come on sooner and remain on longer but the sound was never annoying. In fact, it is only slightly louder than a GTX 580 and rather on the same sound level as a single HD 6970!
 
WaltC said:
I've seen this comment a couple of times in this thread--I think that it would certainly be possible to down clock the 6990 (just as nVidia has done with the 590) and set a suitable but lower fan speed via the CCC--if indeed noise is your highest priority. But asking for the top end of "performance" and "silence" at the same time is asking for an oxymoron, imo.
Click to expand...

Um, watercool. Then they're all the same noise level. If you're really playing in this high end space and looking for max performance with minimal noise, you should already have a watercooled rig, so the noise factor of a stock card should not play into your decision at all.

In any case, SLI'd or CFX'd cards will be more quiet anyway, unless you're planning to SLI or CFX one of these monsters. In which case, you'll probably get a Christmas card from the power company as a side bonus.
 
You are all missing a point here...the Mayans Predicted this! :D

I was considering the 590 but not so sure now. I will wait a bit until all this is cleared out. I think I am better off waiting for a 570 with more memory or so before jumping in to do a SLI setup.
 
it would be interesting if someone would post results of gtx 590 sli vs gtx 580 tri sli.....really curious myself. looking @ other websites and this review it seems obvious that if your thinking of any of these sli setups you'd better have @ least a 30 inch monitor or multi monitor setup.


UPDATE:
i just checked someone who seemed to do a review...guru 3d and compared to an earlier tri-sli review....gtx 580 tri sli....is faster except in dirt 2 bench and just barely. i think this is interesting especially since many people....**cough** myself included have tri sli. in 30" montior it makes a difference. [H] has summed it up quite accurately....its just gtx570 sli performance.
 
Last edited:
You must log in or register to reply here.
Back
Top