R9 290X review (hardware.info)

[Grooveriding]

I don't even pay attention to 1080p any more is why I see it the way I do. 1440p, 1600p, multi-monitor etc. is all I care about. I of course saw the 1080p scores though. And I think if your plan is to play Crysis 3 or Metro:LL at 1080p with one of these cards, you should buy nvidia

 

[96Firebird]

Fair enough, I think we are all patiently waiting for the 1440/1600p reviews, which appear to be lacking.

Perhaps the multi-monitor results mean the 290X will be the better single card for future games at 1080/1200/1440/1600?

But all in all, you'd need more than a single card for many games at multi-monitor resolutions.

 

[OILFIELDTRASH]

Looks like a beast. Definitely will sell out at $650 with a copy of bf4. Tbh they could price it at $700 and sell them all. Unfortunately for those of us that don't have $700 in the budget for a card it sucks.

 

[railven]

I don't even pay attention to 1080p any more is why I see it the way I do. 1440p, 1600p, multi-monitor etc. is all I care about. I of course saw the 1080p scores though. And I think if your plan is to play Crysis 3 or Metro:LL at 1080p with one of these cards, you should buy nvidia

Starting to feel like this now. I was hoping to see more 1440p leaks. The few I saw are interesting but without 7970 CFX in them I don't have much to compare.

Wait and see what else gets leaked.

 

[Cloudfire777]

Metro Last Light Very High 1080p:
R9 290X: 60FPS
GTX 780: 62FPS (+3.3%)
GTX Titan: 68FPS (+13.3%)

Metro Last Light Very High 5760X1080
R9 290X: 29FPS
GTX 780: 27FPS (-6.9%)
GTX Titan: 31FPS (+6.9%)

Crysis 3 1080p Very High 4XAA
R9 290X: 45FPS
GTX 780: 46FPS (+2.2%)
GTX Titan: 48FPS (+4.3%)

Crysis 3 5760X1080 Very High 4XAA
R9 290X: 20.5FPS
GTX 780: 16.5FPS (-19.5%)
GTX Titan: 17.4FPS (-15.2%)

Battlefield 3 1080P Ultra 4XAA DOF 90
R9 290X: 108FPS
GTX 780: 95FPS (-12%)
GTX Titan: 103FPS (-4.7%)

Battlefield 3 5760X1080 Ultra 4XAA DOF90
R9 290X: 46FPS
GTX 780: 38FPS (-17.4%)
GTX Titan: 41FPS (-10.9%)

 

[Teizo]

hardware.info could NEVER properly bench multi-GPU. Some graphs are also mislabelled, for example in some the 280X beats the 290X, lol.

So take those results with caution.

780 SLI is only 1-2 frames faster than a 780 I noticed in that first benchmark. Definitely odd.

I think it is rather obvious though the 290X in Uber Mode is pretty fast (that is faster than any single card at this moment). I wonder about the noise of the fan though.

 

[Sureshot324]

Looks like 280x has 75% the performance at less than half the price. I think that's what I'm going to go with.

 

[Fastx]

http://us.hardware.info/productinfo/200741/amd-radeon-r9-290x-uber-mode
or
http://us.hardware.info/productinfo/200741/amd-radeon-r9-290x-uber-mode#tab:testresults

Seems on average 10% faster than the Titan, power consumption a tad higher, about 30W more than reference 7970 GHz. Not bad!

Edit:
You have to click your way through it, sometimes it doesn't work, I don't know why.

Box maybe also put this link in your first post.
http://videocardz.com/47106/amd-radeon-r9-290x-review-charts-leak

 

[boxleitnerb]

Thx, that's better!

 

[Silverforce11]

So, its average faster than Titan, for a tiny increase in system power. 9W out of 392W when the entire system is running faster suggests its on-par with Titan for power use. Outside of Furmark its extremely efficient on perf/w and perf/mm2, well done to AMD.

Voltage unlocked, heatsink/fan seems to have ample room left to tolerate a good OC. Time to get a R290 and OC it to the max like the 7950!!

 

[Crap Daddy]

I have never ever seen so weird benchmark charts. Don't believe a word. Guess we are just a few days away from real reviewers numbers.

 

[boxleitnerb]

So, its average faster than Titan, for a tiny increase in system power. 9W out of 392W when the entire system is running faster suggests its on-par with Titan for power use. Outside of Furmark its extremely efficient on perf/w and perf/mm2, well done to AMD.

Voltage unlocked, heatsink/fan seems to have ample room left to tolerate a good OC. Time to get a R290 and OC it to the max like the 7950!!

You should correlate the power measurements with the appropriate benchmark. Different benchmark, different power consumption.

Metro LL, 1080p, Very High:
Titan: 68 fps, 383 W
290X: 60 fps, 392 W

 

[Elfear]

hardware.info could NEVER properly bench multi-GPU. Some graphs are also mislabelled, for example in some the 280X beats the 290X, lol.

So take those results with caution.

Agreed. The charts look like they have never used multi-GPU before.

 

[DiogoDX]

I have never ever seen so weird benchmark charts. Don't believe a word. Guess we are just a few days away from real reviewers numbers.

Weird numbers indeed.

 

[lavaheadache]

I agree with what is being said. Those graphs are all over the place and barely make any sense.

 

[Black Octagon]

I have never ever seen so weird benchmark charts. Don't believe a word. Guess we are just a few days away from real reviewers numbers.

How many days exactly? Since the 15th, I've lost track of when the full NDA lifts

 

[Cloudfire777]

1920x1080

Elder Scrolls High 4XAA
R9 290X: 179.6FPS
GTX 780: 146.8FPS (+22.3%)
GTX Titan: 157.5FPS (+14%)

Battlefield 3 Ultra 4XAA
R9 290X: 108FPS
GTX 780: 95FPS (+13.7%)
GTX Titan: 103FPS (+4.9%)

Dirt Showdown Ultra 4XAA
R9 290X: 119FPS
GTX 780: 104FPS (+14.4%)
GTX Titan: 99.6FPS (+19.5%)

Far Cry 3 4XAA
R9 290X: 59FPS
GTX 780: 55.8FPS (+5.7%)
GTX Titan: 59.6FPS (-1.0%)

Hitman 4XAA
R9 290X: 74.8FPS
GTX 780: 62.8FPS (+19.1%)
GTX Titan: 61FPS (+22.6%)

Crysis 3 Very High 4XAA
R9 290X: 45FPS
GTX 780: 46FPS (-2.2%)
GTX Titan: 48FPS (-6.2%)

Tomb Raider Ultra 4XAA
R9 290X: 68.2FPS
GTX 780: 56.9FPS (+19.9%)
GTX Titan: 63.4FPS (+7.6%)

Metro Last Light Very High:
R9 290X: 60FPS
GTX 780: 62FPS (-3.3%)
GTX Titan: 68FPS (-11.8%)


5760x1080

Battlefield 3 Ultra 4XAA
R9 290X: 46FPS
GTX 780: 37.8FPS (+21.7%)
GTX Titan: 41FPS (+12.2%)

Dirt Showdown Ultra 4XAA
R9 290X: 63.7FPS
GTX 780: 48.6FPS (+31.1%)
GTX Titan: 50.8FPS (+25.4%)

Far Cry 3 4XAA
R9 290X: 26FPS
GTX 780: 24.5FPS (+6.1%)
GTX Titan: 26FPS (0%)

Hitman 4XAA
R9 290X: 34.8FPS
GTX 780: 27.6FPS (+26.1%)
GTX Titan: 30.1FPS (+15.6%)

Crysis 3 Very High 4XAA
R9 290X: 20.5FPS
GTX 780: 16.5FPS (+24.2%)
GTX Titan: 17.4FPS (+17.8%)

Tomb Raider Ultra 4XAA
R9 290X: 25.3FPS
GTX 780: 20.7FPS (+22.2%)
GTX Titan: 22.9FPS (+10.5%)

Bioshock infinite Ultra DOF
R9 290X: 36.5FPS
GTX 780: 37.8FPS (-3.5%)
GTX Titan: 39.8FPS (-8.3%)

Metro Last Light Very High
R9 290X: 29FPS
GTX 780: 27FPS (+7.4%)
GTX Titan: 31FPS (-6.5%)

Summary:

In average over 8 games, R9 290X is 11.2% faster than GTX 780 in 1920x1080, and 16% faster in 5760x1080.

In average over 8 games, R9 290x is 6.3% faster than GTX Titan in 1920x1080, and 8.4% faster in 5760x1080.

 

[wand3r3r]

1920x1080

Elder Scrolls High 4XAA
R9 290X: 179.6FPS
GTX 780: 146.8FPS (+22.3%)
GTX Titan: 157.5FPS (+14%)

Battlefield 3 Ultra 4XAA
R9 290X: 108FPS
GTX 780: 95FPS (+13.7%)
GTX Titan: 103FPS (+4.9%)

Dirt Showdown Ultra 4XAA
R9 290X: 119FPS
GTX 780: 104FPS (+14.4%)
GTX Titan: 99.6FPS (+19.5%)

Far Cry 3 4XAA
R9 290X: 59FPS
GTX 780: 55.8FPS (+5.7%)
GTX Titan: 59.6FPS (-1.0%)

Hitman 4XAA
R9 290X: 74.8FPS
GTX 780: 62.8FPS (+19.1%)
GTX Titan: 61FPS (+22.6%)

Crysis 3 Very High 4XAA
R9 290X: 45FPS
GTX 780: 46FPS (-2.2%)
GTX Titan: 48FPS (-6.2%)

Tomb Raider Ultra 4XAA
R9 290X: 68.2FPS
GTX 780: 56.9FPS (+19.9%)
GTX Titan: 63.4FPS (+7.6%)

Metro Last Light Very High:
R9 290X: 60FPS
GTX 780: 62FPS (-3.3%)
GTX Titan: 68FPS (-11.8%)


5760x1080

Battlefield 3 Ultra 4XAA
R9 290X: 46FPS
GTX 780: 37.8FPS (+21.7%)
GTX Titan: 41FPS (+12.2%)

Dirt Showdown Ultra 4XAA
R9 290X: 63.7FPS
GTX 780: 48.6FPS (+31.1%)
GTX Titan: 50.8FPS (+25.4%)

Far Cry 3 4XAA
R9 290X: 26FPS
GTX 780: 24.5FPS (+6.1%)
GTX Titan: 26FPS (0%)

Hitman 4XAA
R9 290X: 34.8FPS
GTX 780: 27.6FPS (+26.1%)
GTX Titan: 30.1FPS (+15.6%)

Crysis 3 Very High 4XAA
R9 290X: 20.5FPS
GTX 780: 16.5FPS (+24.2%)
GTX Titan: 17.4FPS (+17.8%)

Tomb Raider Ultra 4XAA
R9 290X: 25.3FPS
GTX 780: 20.7FPS (+22.2%)
GTX Titan: 22.9FPS (+10.5%)

Bioshock infinite Ultra DOF
R9 290X: 36.5FPS
GTX 780: 37.8FPS (-3.5%)
GTX Titan: 39.8FPS (-8.3%)

Metro Last Light Very High
R9 290X: 29FPS
GTX 780: 27FPS (+7.4%)
GTX Titan: 31FPS (-6.5%)

Summary:

In average over 8 games, R9 290X is 11.2% faster than GTX 780 in 1920x1080, and 16% faster in 5760x1080.

In average over 8 games, R9 290x is 6.3% faster than GTX Titan in 1920x1080, and 8.4% faster in 5760x1080.

So the 290x allegedly (barely) beats the titan. I don't know what AMD was thinking with their announcements.

The 7970 will be the fastest card this year. Then even more recently, they said something about they are not aiming to compete with the titan. Yeah maybe they meant the pricepoint but why state something at all. (I don't recall the exact wording but it stuck me as if this card might not quite catch the titan.

 

[Vesku]

It's not as if AMD would know exactly where 290X retail batches would fall in relation to the Titan, 3 or more months ago. Notice how coy AMD has been now that 290X seems to have turned out quite nicely for them.

In addition, if they declared a specific Nvidia card as their target they hand over quite a bit of potential press to their competition. Imagine how much greater the 780 Ti speculation would be if AMD had declared their intention to "beat the Titan".

 

[Erenhardt]

So the 290x allegedly (barely) beats the titan. I don't know what AMD was thinking with their announcements.

The 7970 will be the fastest card this year. Then even more recently, they said something about they are not aiming to compete with the titan. Yeah maybe they meant the pricepoint but why state something at all. (I don't recall the exact wording but it stuck me as if this card might not quite catch the titan.

AMD said they are not targeting titan because of the ridiculous price range, not because of performance. They said 780 is the card to beat when it comes to performance. As we see the gap between 780 and titan is smaller than titan to 290x.

 

[RussianSensation]

So the 290x allegedly (barely) beats the titan. I don't know what AMD was thinking with their announcements.

The 7970 will be the fastest card this year. Then even more recently, they said something about they are not aiming to compete with the titan. Yeah maybe they meant the pricepoint but why state something at all. (I don't recall the exact wording but it stuck me as if this card might not quite catch the titan.

I think AMD stated that they don't intend to compete with NV at the $999 price level as it's too niche. I do not recall them stating that they won't compete with the Titan on performance though. Strategically it turned out pretty smart since it kept NV in the dark.

Not sure if these are real but the specs on the 780Ti may not be enough to beat the R9 290X.

 

[boxleitnerb]

True. This will perform like a Titan basically. A full GK110/180 with ad tad higher clocks would be required to beat the 290X.

 

[RussianSensation]

True. This will perform like a Titan basically. A full GK110/180 with ad tad higher clocks would be required to beat the 290X.

I was honestly hoping for at least a Titan with higher clocks, and on the really optimistic side a 2880 SP @ 902mhz aka K6000. Instead they seem to have delivered neither of those.

 

[DownTheSky]

K6000 are cherry-picked cores and would make no sense to sell them for consumer/gaming market at 1/3 - 1/4 the price.

 

[el etro]

True. This will perform like a Titan basically. A full GK110/180 with ad tad higher clocks would be required to beat the 290X.

I don't think they will clock a full enabled GK110 higher than they clock their SKUs now.

K6000 are cherry-picked cores and would make no sense to sell them for consumer/gaming market at 1/3 - 1/4 the price.

Maybe not. I think it will consume the same as Titan does today.