AMD Radeon 7990 reviews thread

[parvadomus]

They didn't use the driver because it would not be indicative of the expected performance and usage that someone buying the card gets. You can't get that driver, so when you buy it you don't get the fixes.

This is a case of you not even bothering to read. They actually showed the new drivers performance in their testing on a separate page and explained why they didn't use it. Put simply they didn't want people to think that's how it works out of the box, it doesn't.

I read every review, its a beta driver, and as every beta driver will go out very soon, thats why its 13.5 (May beta driver). Ofc, final version will come maybe june or july.
For me the most interesting part of this launch was the driver thing (i will never buy a 7990, but may crossfire my 7870 someday), and not showing it just plain sucks, well I could not expect much from that garbage site anyway.

 

[3DVagabond]

I agree with Chiropteran. I'm willing to bet though that it was far less nefarious than he is assuming. I'll bet that they had already run much/all of the benches with the new driver before they ran BF3 and would have had to rerun everything to revert back to the old driver. It's been stated that both companies dropped the new drivers on the reviewers late. Which seems to be getting more and more common, unfortunately.

I think they should have left BF3 as a fail, though. Not allowed for different drivers for that game. Also, it doesn't seem like any other sites had issues with drivers and BF3. Likely not nVidia's fault but something with the test setup. A lot of sites still used games that crossfire doesn't work in and left the results. Or, FC3 not running in surround, as I saw in one review. None of the results were omitted though.

 

[boxleitnerb]

Nice that you point to the most NV biased site out there (ask yourself why they didnt use the driver provided by AMD). Why not talk about for example, tomshardware finding? The beta driver removes runt frames on every game. :awe:
But I know biased sites are the favourite for trolls like you :biggrin:

You should maybe actually read the review and think before participating here and calling people trolls. AMDs driver is alpha, it's a work in progress. This is the reason why it was not tested thoroughly (but it was looked at in the article, nonetheless - which you overlooked). So you cry about nothing.
Just because a site writes something a fanboy doesn't like, it is biased? Right :whiste:

http://www.pcper.com/reviews/Graphi...lta-Gets-Frame-Rated/AMD-Frame-Pacing-Prototy

AMD at the last minute provided me with a prototype driver with a completely from the ground up frame pacing algorithm in an attempt to fix the runt frame issues we have been seeing with CrossFire for nearly a year. However, this driver is not nearly close to end-user availability so we were not comfortable reviewing a product with it. I don't expect this driver until the late-June or July time frame

Clear now? Do you have any idea how much work such a review is? Do you expect the reviewer to repeat all his work within a few hours maybe? AMD had almost 10 years to fix their CF. They should not be rewarded for being late (again) to this review with an unfinished product.

 

[Lonbjerg]

It doesn't matter how many times you post the same thing.

I'll keep posting untill anandtech changes their useless GPU-reviews.

They need a new GPU-reviewer as the current tool is to stuck in the past and insists on using a useless single FPS numbers.

No min/avg/max FPS number, no FPS graphs, no frametimes, no runt frames...NO relevant information...just a useless review that only feeds false data to fanboys...the worst 7990 review I have read so far.

 

[3DVagabond]

I'll keep posting untill anandtech changes their useless GPU-reviews.

They need a new GPU-reviewer as the current tool is to stuck in the past and insists on using a useless single FPS numbers.

No min/avg/max FPS number, no FPS graphs, no frametimes, no runt frames...NO relevant information...just a useless review that only feeds false data to fanboys...the worst 7990 review I have read so far.

You are not important enough to get them to change anything. Get over yourself.

 

[boxleitnerb]

But he is right. Criticism should always be welcome as it leads to improvements. Why do you defend AT? Do you work for them?

 

[parvadomus]

You should maybe actually read the review and think before participating here and calling people trolls. AMDs driver is alpha, it's a work in progress. This is the reason why it was not tested thoroughly (but it was looked at in the article, nonetheless - which you overlooked). So you cry about nothing.
Just because a site writes something a fanboy doesn't like, it is biased? Right :whiste:

http://www.pcper.com/reviews/Graphi...lta-Gets-Frame-Rated/AMD-Frame-Pacing-Prototy
Clear now? Do you have any idea how much work such a review is? Do you expect the reviewer to repeat all his work within a few hours maybe? AMD had almost 10 years to fix their CF. They should not be rewarded for being late (again) to this review with an unfinished product.

Yes thats why the other sites could do it, it just proves how biased they are
PS: I dont cry about garbage sites not posting everything, I just hate haters =D

 

[Lonbjerg]

But he is right. Criticism should always be welcome as it leads to improvements. Why do you defend AT? Do you work for them?

I guess improvements are only okay, when they do not put AMD in a bad light :hmm:

 

[3DVagabond]

But he is right. Criticism should always be welcome as it leads to improvements. Why do you defend AT? Do you work for them?

I'm not defending AT. He's made his critique. Repeating it ad nauseam isn't needed, or contrary to what he believes, of any use. AT have made their position clear on using FRAPS frame times and FCAT at this point in time.

Edit: No I don't work for AT. You're being ridiculous.

 

[3DVagabond]

I guess improvements are only okay, when they do not put AMD in a bad light :hmm:

You can guess anything you want to. As usual, you are wrong.

 

[Zanovar]

You are not important enough to get them to change anything. Get over yourself.

That comment makes me wanna puke.i like you 3d,.but that shit is wrong!

 

[boxleitnerb]

Yes thats why the other sites could do it, it just proves how biased they are
PS: I dont cry about garbage sites not posting everything, I just hate haters =D

You have not read anything I have written. How should he have done it if AMD has sent out the driver so late? Do you know when exactly which site received this prototype driver? Enlighten us.
Aside from that it is commendable that PCPER didn't test with such an unfinished driver. Even betas are questionable imo. Why should a company be rewarded for such a sloppy launch without polished drivers?

I'm not defending AT. He's made his critique. Repeating it ad nauseam isn't needed, or contrary to what he believes, of any use. AT have made their position clear on using FRAPS frame times and FCAT at this point in time.

Edit: No I don't work for AT. You're being ridiculous.

Sometimes things have to be repeated in order to be heard. You getting personal doesn't help things at all, criticism might. The problem with AT reviews is, that their graphs suggest AFR is equal to non-AFR. It isn't. They should at least convey their subjective impressions about stuttering if they don't feel comfortable measuring anything. But as it stands, their MGPU reviews are misleading, big time.

 

[cmdrdredd]

Maybe people would be more happy to just see fps charts with no info on drivers or issues. Give them that and see if they want to know what's going on and read a little bit about the review testing.

 

[3DVagabond]

That comment makes me wanna puke.i like you 3d,.but that shit is wrong!

So you think they are going to change their reviewing practices for a single poster? Come on! Especially since they've explained their position. They've said that they do plan on using the tools once they are sure that they are giving a fair assessment. I don't feel like going back and rereading the article and pulling out quotes for the relevant parts, but they made their position clear. Someone trolling the forum isn't going to influence that.

To be clear, saying it once is fine, but to repeat it over and over and in the tone he's using isn't going to accomplish anything.

 

[Lonbjerg]

Maybe people would be more happy to just see fps charts with no info on drivers or issues. Give them that and see if they want to know what's going on and read a little bit about the review testing.

Maybe people want more than a useless single FPS number....ESPECIALLY on multi-GPU reviews...

 

[cmdrdredd]

Maybe people want more than a useless single FPS number....ESPECIALLY on multi-GPU reviews...

That's what I am saying. Any info is good info when trying to make a consumer informed on the options. More is always better IMO.

 

[3DVagabond]

Maybe people want more than a useless single FPS number....ESPECIALLY on multi-GPU reviews...

FPS are not useless. I'll grant you that they don't tell the whole story. Especially on multi GPU setups. We still need to know the FPS info though.

 

[96Firebird]

Is this card available at retailers yet? If not, I don't see a problem in reviewing with the BETA drivers as long as they are available upon actual release. If they are available now, review with the latest drivers available to the public.

I don't see much on overclocking with this card, can anyone link me to a review that dives into OC? I saw TPU's overclocking page, but it didn't really make much of a difference and the temperature throttling clocked it down anyways...

 

[3DVagabond]

Is this card available at retailers yet? If not, I don't see a problem in reviewing with the BETA drivers as long as they are available upon actual release. If they are available now, review with the latest drivers available to the public.

I don't see much on overclocking with this card, can anyone link me to a review that dives into OC? I saw TPU's overclocking page, but it didn't really make much of a difference and the temperature throttling clocked it down anyways...

So far any O/C testing I've seen is pretty disappointing. The throttling seems to be a major issue. It appears that AMD is so concerned with power usage they have really crippled this card. They must really be feeling pressure about Tahiti's lack of efficiency. At least in relation to the competition. Everything from the original Tahiti release clocks to not releasing a 7990 for so long, and now this cards power characteristics all seem to point to that.

Here's another O/C at Legit Reviews.

You can see that the performance improvement is almost non existent.

Something isn't right here! The score went from P15225 to P15242 with a 100MHz overclock on the core? We let AMD know this and the had no answer.

They don't seem to know why. Likely that the card is throttling.

 

[Face2Face]

Good looking card and cooler, but this is kind of a pointless $1000 card when two $300 7950's with a small overclock will perform better. I am more excited about an improved driver over the this card

 

[Souv]

@3DV,
yes that's the power that throttling that card,not temparature I believe as gpu can go upto 95c.As my card also reaches 86c with max OC and significant performance gain without any throttling...

 

[Stuka87]

You should maybe actually read the review and think before participating here and calling people trolls. AMDs driver is alpha, it's a work in progress. This is the reason why it was not tested thoroughly (but it was looked at in the article, nonetheless - which you overlooked). So you cry about nothing.
Just because a site writes something a fanboy doesn't like, it is biased? Right :whiste:

http://www.pcper.com/reviews/Graphi...lta-Gets-Frame-Rated/AMD-Frame-Pacing-Prototy
Clear now? Do you have any idea how much work such a review is? Do you expect the reviewer to repeat all his work within a few hours maybe? AMD had almost 10 years to fix their CF. They should not be rewarded for being late (again) to this review with an unfinished product.

No, the driver is in BETA, not ALPHA. And the Beta driver is available to everybody, there is ZERO reason to not use it. Every other site did.

 

[Lonbjerg]

FPS are not useless. I'll grant you that they don't tell the whole story. Especially on multi GPU setups. We still need to know the FPS info though.

So what does the single FPS number in AT's review tell?

Please share...because the numbers are useless....not matter how you try and spin this to be about me...it isn't....so drop the FUD.

 

[sontin]

Back to AMD's lies:

Above is our run-through at "max settings" at 3840x2160. You can clearly see that performance is very poor on the AMD Radeon HD 7990. The minimum framerate is 10 FPS, and the average is 19.4 FPS, well under 30 FPS. This level of performance is not playable in Crysis 3.
[...]




This claim is clearly invalid.

http://www.hardocp.com/article/2013/04/24/amd_radeon_hd_7990_video_card_review/4#.UXfuPqwfnYw


They must be really desperate to put out such slides.

And for the users who love OC: There is none with the 7990:
http://www.guru3d.com/articles_pages/radeon_hd_7990_review,28.html

 

[railven]

Welps, AMD totally shot themselves in the foot with the $1,000 price tag. This product is now irrelevant. Kudos AMD, you continue to surprise me with your stupidity.