GTX 590 reviews

[halfwaythere]

Its faster than most people will ever need problem is most reviews still use stock cpus which can't take full advantage of these babies.

I think its more a problem for folks expecting this to crush the radeon than a fail on nvidias part. Fact of the mater is none expected amd to use full cayman chips on their card.

[[XC] Synthetickiller]

Its faster than most people will ever need problem is most reviews still use stock cpus which can't take full advantage of these babies.

I think its more a problem for folks expecting this to crush the radeon than a fail on nvidias part. Fact of the mater is none expected amd to use full cayman chips on their card.

What speed do you need to not bottle neck? quad @ 4.5ghz?

[halfwaythere]

Check out the lab501 article about cpu frequency scaling.

[Pantsu]

Disappointing in terms of OC ability. Hopefully the drivers really were the problem with the blown up cards. It would certainly be a bummer to brick a 700$ card.

I'd say Nvidia did a tremendous job in making the card in the first place, considering the power issues it has had with fermi. The card seems like a decent pick for those who want SLI performance in one slot and use the reference cooler. But that's a really small niche I'd say, and pretty much in all other circumstances it's better to go with SLI/Crossfire or 6990.

[highoctane]

Hardwarecanucks
http://www.hardwarecanucks.com/forum...gb-review.html

Hey OTH we need you to work your magic and make a nice clean consolidated thread for reviews!

[SKYMTL]

Predictably few Dragon Age 2 benchmarks around.

I wonder why?

Personally, I don't include it due to the number of bugs it exhibits. There's nothing like a crash to desktop when you're in the middle of a benchmark run.

In addition, most sites use the Civ benchmarking tool which I find IS NOT representative of the in-game performance. So we don't bench Civ either.

[jimbo75]

Personally, I don't include it due to the number of bugs it exhibits. There's nothing like a crash to desktop when you're in the middle of a benchmark run.

In addition, most sites use the Civ benchmarking tool which I find IS NOT representative of the in-game performance. So we don't bench Civ either.

That's fair enough so long as a bit of consistency is shown. Anandtech for example has been benching the original DA:Origins since it was released, and also Civ 5 since release...but not DA 2 here?

I bet Crysis 2 is benched asap though with little regard given to the state of crossfire drivers.

[SubZero.it]

HW Canucks GTX 590 vs 6990 -1% in avg fps, +4% in min. fps. So dunno, what are you seeing.

Agree but that little less performance combined with power consuption and PRICE doesn't make it a nice offer for the customer.
In my opinion nvdia would have done a better job with investing more money in 28nm and get rid of their monolitich technology.It's at the end here.
Though it's really really really quiet compared to 6990, but with 600mhz........

[bamtan2]

the people nit picking about performance differences arent serious buyers. they're just spectators. 6990 and 590 perform so similarly you can't tell the difference by looking at them. the only difference you can tell is a) the noise from 6990 and b) the lower speed of 590 in select games at triple monitor resolutions.

people with ears shouldnt buy 6990. people with 3x monitors who play metro 2033 shouldnt buy 590.

[hurrdurr]

A less silly version of 6990, still not good enough to buy.

[BeepBeep2]

the people nit picking about performance differences arent serious buyers. they're just spectators. 6990 and 590 perform so similarly you can't tell the difference by looking at them. the only difference you can tell is a) the noise from 6990 and b) the lower speed of 590 in select games at triple monitor resolutions.

people with ears shouldnt buy 6990. people with 3x monitors who play metro 2033 shouldnt buy 590.

Performance per watt on the 590 is 20% less than 6990 according to TPU, and two of their game tests are nvidia biased games (in which a 570/580 would outperform the 6990 for example)

The 6990 is loud (not like we don't know how to adjust fan speeds) but in a lot of games it will perform as well or better than the 590 I think.

[saaya]

Its faster than most people will ever need problem is most reviews still use stock cpus which can't take full advantage of these babies.

I think its more a problem for folks expecting this to crush the radeon than a fail on nvidias part. Fact of the mater is none expected amd to use full cayman chips on their card.

xbitlabs used a 4.5ghz 6core
helps the 590 pull ahead...

i still think 6990 and 590 are too close to call a winner...
i guess if there had to be one, it would be the 590... but yeah, both can handle anything you throw at them...

i dont like nvidias way of handling ocp and power management... and their pwm seems to be weaker than that of a 6990 while they draw slightly more power... which is a shame as it probably limits the 590 when overclocking it...
if it were just up to the gpus, then a 590 would be the better overclocking card, but with this pwm its not that clear it seems...

if i could have either one for free, id pick a 6990 since im worried about those blow-up and ocp problems of the 590...
if that gets sorted, and there is an open power limitation implementation you can disable, like on the 6990, and it turns out the pwm can handle the two gpus when overclocking without popping, then id probably chose the 590 as it would clock better...

EDIT: but really... if somebody would offer me to get either a 6990 or 590 for free... id ask him "cant i have a single 580 instead?"

[damtachoa]

One word for "GTX 590": Disappointed!

[Lanek]

HW Canucks GTX 590 vs 6990 -1% in avg fps, +4% in min. fps. So dunno, what are you seeing.

This is just the % of the conclusion page... Lostplanet 2 in the bench list make the % difference is not what the difference in other games show ( when you have +20fps on the Nvidia one in this game. so a ~ +20%, this change many thing on final % difference ) ..

As expected BC2 and Dirt2 are win for the 590... ( it's maybe time for AMD to watch what happend on thoses games, there's no reason to get such difference in thoses games ) .. But all other are in favor of AMD..

...........

On other way, i agree with Sayaaa, the perfs are so close it's really hard to say one is a winner over the other..



Anyway, thanks again to Lab501, to make a review with Overclocked vs overclocked and with a good 2600K - 5ghz CPU chips ...

[DarthShader]

Why does it get called "inovative"? Nothing new there, except an OCP that not always works.

[SKYMTL]

Why does it get called "inovative"? Nothing new there, except an OCP that not always works.

VERY quiet
Not all that bad on power consumption
Low temperatures

These three things together take it a few steps beyond past dual GPU cards.

Not everything is about performance....

[CCC]

been considering an upgrade so ive been waiting on the 590 reviews. it still seems to me the best "bang for your buck" is 2 x 6970's in crossfire... or maybe 3 x 6970's if you want something really powerful for the least amount of cashola

and mine would have full cover blocks so i dont really care about the noise. i know that brings the price up but no matter what vc's i buy they will be watercooled.

[highoctane]

On other way, i agree with Sayaaa, the perfs are so close it's really hard to say one is a winner over the other..

I agree, it's a titan war, they each have their advantages while neither is decisively dominant over the other.

A good slug fest...

[flippin_waffles]

VERY quiet
Not all that bad on power consumption
Low temperatures

These three things together take it a few steps beyond past dual GPU cards.

Not everything is about performance....

Really? in the high end ultra performance category, not everything is about performance? strange. so lower performing, higher power consumption and quite lower performance/watt doesn't count, and being a little quieter trumps all else! gotta love these ever moving goal posts....

[[XC] hipno650]

overall I think the GTX 590 is a real mixed bag.

-The minimum Frame rates are really good
-The temps are mind blowingly good (no one could have called this...)
-The power consumption is ok
-The physical size is impressive
-The noise output is really good....

But the overall perf is lower then a 6990 on average. I would say an average of about 5-10%. The thing that makes that really hard is that in some games it performs really really well in others not at all. Also it the high res high settings performance is all over the map as well. the 6990 beats it in Just Cause 2 at 2560 pretty decently but then the GTX 590 under surround.... but close to the opposite is true in AvP. Performance seems to be game Dependant.... the big thing for me tho is how little these seem to OC on stock voltage... they really dumped the voltage off to get the power under the 6990 and it seems to hurt the OC of the card a bunch. however I would like to see some OC results with some extra core voltage in afterburner.... that could make it far more attractive considering that you could get away with it on the stock cooling without getting crazy loud or hot....

it comes down to price I think which puts it neck and neck with the 6990....

I would say it is a more "N00B friendly" card considering that it's smaller and cooler and whatnot so for non enthusiast I would say its the better buy. for most XS users tho the 6990 seems to be your better bet considering the better performance. however added voltage in OCing could change all that.

[SKYMTL]

Really? in the high end ultra performance category, not everything is about performance? strange. so lower performing, higher power consumption and quite lower performance/watt doesn't count, and being a little quieter trumps all else! gotta love these ever moving goal posts....

No, it doesn't. But regardless of who you are, having a card that's easier to live with is a big step IMO. Saying gamers completely don't care about noise production is just ignorant. NVIDIA has shown that ultra high end performance DOESN'T NEED to come with a tradeoff in terms of high acoustics or high core temperatures.

No one is saying it trumps all else; I have yet to read a review that praised the GTX 590 through and through.

[Olivon]

http://www.tomshardware.com/reviews/...0,2898-17.html

Hahah, it's ridiculous how loud the Radeon 6990 is. Move to the four minute mark, start the 590 clip, then the 6990 clip

What a noise coming from HD 6990 !!!

http://youtubedoubler.com/PNtr

[matose]

Any test with 11.4 R2?

Yes, both lab501 reviews are done with Catalyst 11.4 RC2

GTX 590 vs HD 6990 | lab501.ro translation

GTX 590 vs HD 6990 - CPU scaling | lab501.ro - translation

[Olivon]

Nice popamiro's

[flippin_waffles]

No, it doesn't. But regardless of who you are, having a card that's easier to live with is a big step IMO. Saying gamers completely don't care about noise production is just ignorant. NVIDIA has shown that ultra high end performance DOESN'T NEED to come with a tradeoff in terms of high acoustics or high core temperatures.

No one is saying it trumps all else; I have yet to read a review that praised the GTX 590 through and through.

Instead they make a tradeoff of lower performance and higher power.
Yup, if you can't make it faster and the design is at it's limits, just pretend it was a design decision, and let marketing take over! Of course the few that nv actually makes will find a home anyway, so it's all moot.