R9 290X reviews

[marten_larsson]

No one speaks about that!!! we talked about this particular site
as in the previous time.......you have a problem in understanding our words next time try google translate.

Could you please show the proof of kitguru being funded by AMD? I agree their articles are often tilted towards AMD but with so many sites tilted towards Nvidia I don't see the problem in that. Most of their methodology was presented in their review although not if it was time demo or just randomly played in-game (but which review out there would be accepted in the scintific world? Where are the power calculations, the p values, the standard deviations?!). Their results aren't obviously skewed in my view (compared to other reviews).

[thatdude90210]

Seriously, "AMD funded sites"? You would think the news section should be above that sort of video card warriors stuff.

[BababooeyHTJ]

Fraps is not really the good to use for test it ( but i think you are like me, you dont want invest 5000$ just for run Fcat lol ).

I dont know for Crysis, but look like framepacing is woking not so bad. ( 7990 vs 690, BF4 )

Different games, they're also not forcing SSAA at 2560x1440 in that review.

I noticed microstutter so I did a fraps capture. Looks like it did a pretty good job capturing it to me. Hell its what techreport has been doing for quite some time. I honestly think that it does a pretty good job relaying what I see most of the time.

I'm not even sure that FCAT alone explains microstutter. How do you explain microstutter when vsync is enabled? It can't be runts and dropped frames.

I do agree that AMD is doing far better but they still have a little ways to go. I can't imagine that 4k performance is as good as stated if I can force microstutter at 2560x1440 with SSAA. 4k with no aa is more demanding if I'm not mistaken.



I'm also going to bash techpowerup while I'm at it. Far Cry 3 still to this day performs terribly in crossfire. Performance is as choppy as I remember at launch when I was using 7950 crossfire. You would never get that impression from reading their review. Thats one of the reasons that I blame TPU for AMD going so long before finally addressing microstutter.

[STaRGaZeR]

How do you explain microstutter when vsync is enabled?

Vsync only guarantees that a complete frame is delivered for each monitor refresh. It doesn't control the actual content of the frame, that's entirely up to the game engine. It's perfectly possible to have constant X fps and for it to be a stutter mess. For a very easy example see the 3:2 motion judder you NTSC people are so used to.

[BababooeyHTJ]

Vsync only guarantees that a complete frame is delivered for each monitor refresh. It doesn't control the actual content of the frame, that's entirely up to the game engine. It's perfectly possible to have constant X fps and for it to be a stutter mess. For a very easy example see the 3:2 motion judder you NTSC people are so used to.

I agree with you thats probably exactly the issue. My gripe is the comments that runts and dropped frames are the only cause of microstutter.

[GoldenTiger]

Seriously, "AMD funded sites"? You would think the news section should be above that sort of video card warriors stuff.

In every industry imaginable there are plenty of shills, it's hardly unique to video cards . Yes, AMD probably has shills. I'd bet nvidia does too. General Motors probably does too on car advice sites, as would Chrysler likely. Heck, amazon has tons of shills/fake reviews to pump up products on it of all sorts of things by "viral marketing" companies. I would have thought the people here would be above not knowing basics like that... LEGAL DISCLAIMER This is not a statement of fact on my behalf that any particular company is shilling, names are used as example possibilities only.

[Final8ty]

Move aside TITAN SLI, Radeon R9 290X CrossFire just took over as the fastest 2-way video card performance in gaming across the board, period. For the price of one TITAN Radeon R9 290X CrossFire provides the most impressive performance we've seen to date, and beats not just one, but two TITANs.

http://www.hardocp.com/article/2013/.../#.UnNIZBBOoxc

[kadozer]

In Eyeinfinity 5760x1200:

77% scaling in Crysis 3
88% scaling in Far Cry 3
75% scaling in Metro Light
86% Tomb Raider

So much talk about smoothness, frametime, and the actual experience of multi-GPU video cards while gaming has been discussed lately. In the past, AMD has been highly criticized of having a sub-par CrossFire experience, and rightly so. There were major issues with smoothness where games would stutter or feel choppy, even though the framerates looked good. We've been telling our readers for years that CrossFire just didn't feel as good as SLI while gaming.

Those times have changed, at least on the new Radeon R9 290/X series. The new CrossFire technology has improved upon the CrossFire experience in a vastly positive way. Playing games on the Radeon R9 290X CrossFire configuration was a smooth experience. In fact, it was smoother than SLI in some games. It was also smoother on the 4K display at 3840x2160 gaming, and it was noticeably smoother in Eyefinity at 5760x1200.
It is safe to say, from a real-world gaming perspective, stuttering is no longer an issue one has to worry about. Low framerates don't feel "stuttery" anymore, you no longer have to achieve the highest framerates to get a smooth experience. We were pleasantly surprised how smooth the gameplay was on the 4K panel with CrossFire. We can only hope this technology evolves to future products, and AMD stays on its feet and never lets a bad CrossFire experience happen again.

Sounds like all the criticism about microstutter and choppy gameplay really hit AMD when they worked on Hawaii.
Its a no brainer what to get for multi gpu.

[MaddMuppet]

Impressive results for the 290x. But a shame they didn't run the same settings for all cards. Metro no PhysX on the AMD card, they should have disabled it on the Nvidia cards as well. it is a huge performance hog. Also on some of the games, 290x was using MSAA and the Nvidia cards SMAA. Still a nice quick review. Loved the Crossfire scaling with only just released hardware. Very impressive from AMD.

[boxleitnerb]

AMD cards running in uber mode.
Nvidia cards running in default mode.

Yeah...right. There goes the objectivity of another site.

[boxleitnerb]

PLease think about it just little bit. Are you asking the reviewers to run the nvidia card at 90C when it was design to run at 80C? I don't remember any reviews like that from before, just maybe there is a good reason for that.

I'm asking to run both cards either at default settings (which is quiet mode for the 290X). Or at modified settings. You cannot mix and match.

[tajoh111]

After reading the review, I wonder if he set the fans to 100% fan cap. From the sounds of it, he definitely didn't leave it on quiet mode and recommended putting a 100% fan can on it.

He also neglected to mention anything about noise except it doesn't always run at 100% fan when you set it to that level unless you need it. Uber sounded like the very minimum he set it at but it also sounds like he set it to 100%.

With Nvidia's higher end cards, they do seem to have problems with scaling lately.

[tajoh111]

Could you please show the proof of kitguru being funded by AMD? I agree their articles are often tilted towards AMD but with so many sites tilted towards Nvidia I don't see the problem in that. Most of their methodology was presented in their review although not if it was time demo or just randomly played in-game (but which review out there would be accepted in the scintific world? Where are the power calculations, the p values, the standard deviations?!). Their results aren't obviously skewed in my view (compared to other reviews).

Seriously, "AMD funded sites"? You would think the news section should be above that sort of video card warriors stuff.

AMD definitely puts it's money out there more now.

I cringed when I saw that Anandtech has an AMD center section now that officially sponsored by AMD. It the first thing you see now when you enter anandtech.com now. At least they aren't being transparent about it and do mention the section is sponsored by AMD. What I don't like is that they made almost their entire videocard benching suite gaming evolved games.

[blindbox]

After reading the review, I wonder if he set the fans to 100% fan cap. From the sounds of it, he definitely didn't leave it on quiet mode and recommended putting a 100% fan can on it.

http://www.hardocp.com/article/2013/.../#.UnNpPvmnoQw

Says here at the last paragraph at the bottom: all benchmarks are done on Uber.

[tajoh111]

http://www.hardocp.com/article/2013/.../#.UnNpPvmnoQw

Says here at the last paragraph at the bottom: all benchmarks are done on Uber.

Your right. I was just looking at the conclusion.

[boxleitnerb]

With Nvidia's higher end cards, they do seem to have problems with scaling lately.

Not that I know of. I could open another thread and run some benchmarks on my Titan SLI. Usually scaling is just excellent. I rather think due to the increased heat in SLI mode the cards clock lower than in single card configuration, thus the illusion of bad scaling. My cards are on water and a rocksteady 1006 MHz under any load.

[To(V)bo Co(V)bo]

Wow, AMD must of did something right with the 290X to see so many people whining about a great product that's priced fair. As far as I am concerned, 290X just made ALL the high performance video cards in closer reach for the average consumer. If people want to pretend that stock performance should be 10% less because it makes X brand seem better. It is just plain denial.

As for reviews, all reviews are only comparable to results obtained from their own site. It doesn't really matter what the testing methodology is, as long as it is consistent from test to test. If you were to make special considerations because one card does this and that one doesn't, you would be altering results. As such, UBER mode is a factory setting available to anyone who purchases the 290X. If you were to change power targets on the Nvidia hardware there would be no way to guaranty performance from one person to the next, creating a special considertion that does not guaranty consistent performance.

[boxleitnerb]

As for reviews, all reviews are only comparable to results obtained from their own site. It doesn't really matter what the testing methodology is, as long as it is consistent from test to test. If you were to make special considerations because one card does this and that one doesn't, you would be altering results. As such, UBER mode is a factory setting available to anyone who purchases the 290X. If you were to change power targets on the Nvidia hardware there would be no way to guaranty performance from one person to the next, creating a special considertion that does not guaranty consistent performance.

No. What would you say if hypothetically all AMD cards were tested at 1600p and pitted against Nvidia cards running only 1080p? If the reviewer did this for all of his reviews, they would be consistent...consistently unfair.

The power and temperature targets are also available to anyone who purchases an Nvidia card. It is also a factory setting if you will - the maximum of 95C and 106% is the same on every Nvidia card; you just need to set is just like you need to set the uber mode. And even in Uber mode and with those maxed out targets, there is no guarantee for both about consistency. Sometimes AMD cards will still throttle in uber mode (see those reviews that tested with 100% fan as well where the cards clocked higher thann in Uber mode), sometimes Nvidia cards will throttle with maxed out targets. Say goodbye to fixed clocks or any guarantees about them. That doesn't exist anymore.

So everything you said is wrong. The point of a review is to make a fair and unbiased comparison. Most of the 290X reviews I've seen aren't.

[STEvil]

Impressive results for the 290x. But a shame they didn't run the same settings for all cards. Metro no PhysX on the AMD card, they should have disabled it on the Nvidia cards as well. it is a huge performance hog. Also on some of the games, 290x was using MSAA and the Nvidia cards SMAA. Still a nice quick review. Loved the Crossfire scaling with only just released hardware. Very impressive from AMD.

In fact they do have an "applies-to-apples" comparison actually.

I find their wattage numbers a bit suspect. My system currently has a 2600K @ 4.6 (with HT turned off no less) and a GTX670 (1241/6ghz auto-boost) and I can break 500w load. Their 780 SLI system is only using 91w more than me..??

edit

I'm glad to see a review site FINALLY mentioning SLI's multi-display issues.

[STEvil]

I'm asking to run both cards either at default settings (which is quiet mode for the 290X). Or at modified settings. You cannot mix and match.

Give the 780x a switch.

oh wait, that would be a 780Ghz (ti) 2.

[boxleitnerb]

Give the 780x a switch.

oh wait, that would be a 780Ghz (ti) 2.

It has one...a software switch. People just choose to ignore it for whatever reason.

[MaddMuppet]

In fact they do have an "applies-to-apples" comparison actually.

I find their wattage numbers a bit suspect. My system currently has a 2600K @ 4.6 (with HT turned off no less) and a GTX670 (1241/6ghz auto-boost) and I can break 500w load. Their 780 SLI system is only using 91w more than me..??

edit

I'm glad to see a review site FINALLY mentioning SLI's multi-display issues.

In Farcry and Tomb Raider, yes they do have an apples to apples comparison. But in Crysis and MetroLL, the settings are different, so aren't really what you would call an apples to apples comparison.

Looks like I was reading the highest in game settings. So my mistake

[blindbox]

In Farcry and Tomb Raider, yes they do have an apples to apples comparison. But in Crysis and MetroLL, the settings are different, so aren't really what you would call an apples to apples comparison.

What are you talking about? I just checked Crysis 3 and Metro: Last Light on their article. Their apples to apples comparisons have the same settings. Can't we all check our facts clearly before pressing that post button?

http://www.hardocp.com/image.html?im...5fM18zX2wuZ2lm
http://www.hardocp.com/image.html?im...5fNl8zX2wuZ2lm

Respective links, scroll down to the apples-to-apples section.

http://www.hardocp.com/article/2013/...3#.UnN_gPmnoQw
http://www.hardocp.com/article/2013/...6#.UnN_QfmnoQw

[MaddMuppet]

What are you talking about? I just checked Crysis 3 and Metro: Last Light on their article. Their apples to apples comparisons have the same settings. Can't we all check our facts clearly before pressing that post button?

http://www.hardocp.com/image.html?im...5fM18zX2wuZ2lm
http://www.hardocp.com/image.html?im...5fNl8zX2wuZ2lm

Respective links, scroll down to the apples-to-apples section.



http://www.hardocp.com/article/2013/...3#.UnN_gPmnoQw
http://www.hardocp.com/article/2013/...6#.UnN_QfmnoQw

What I am talking about is in the Crysis comparison the 290x is running 4xMSAA and the Titan and GTX780 are running SMAA MGPU 2x. In the Metro LL the Titan and GTX780 are running PhysX and the 290x obviously can't run PhysX. That is not apples to apples.

[MaddMuppet]

Ok, I maybe wrong there. I was reading the Highest in game playable settings. My bad, by the looks of it.