[THW] Review Samples vs Retail R9290x boards(large differences)

[tential]

http://www.tomshardware.com/reviews/radeon-r9-290-driver-fix,3666.html

We first observed differences between the Radeon R9 290X cards that AMD sent out for review and the ones being sold online just before our R9 290 coverage went live. After additional testing, we have answers, feedback from AMD, and more questions.

Leading up to the Radeon R9 290X launch, I had two cards in the lab, both supplied by AMD. They ran at slightly different core frequencies in single-card configurations. No big deal, right? That was just PowerTune doing its job. But paired up in CrossFire, even with plenty of space between them, heat build-up forced the pair to drop to even lower clock rates. Right away, I was able to establish that AMD’s Hawaii GPU operates within a range of clock rates, determined by a number of variables. For R9 290X, that range starts at 727 and ends at 1000 MHz. As an aside, I do have an issue with vendors simply advertising this as a 1000 MHz GPU.

The discussion that so badly requires clarification, however, is a derivative of Hawaii’s inherent behavior. Because the GPU is always trying to run as fast as possible, and then adjusting down to obey certain power, temperature, and fan speed settings, every card is going to be just a little bit different. This is expected, and applies also to Nvidia’s GPU Boost-capable cards. But when I got my hands on a retail card purchased from Newegg, it was way slower. Like, 20% slower in many cases.


So that left me wondering: were the press boards just lower-leakage parts able to sustain higher clocks? Without a definitive answer, and with AMD insistent that my results couldn’t be right, I approached the Radeon R9 290 review much more cautiously, presenting data from both the sampled and store-bought 290X cards.

A Break In The Story

Then, on Tuesday, I received word from AMD that it had a smoking gun.

All of my numbers were generated using Quiet mode—AMD’s name for its 40% PWM ceiling, designed to keep acoustics in check. That’s a signal representing a percentage of maximum voltage into the cooling fan. So, from one card to another, 40% should yield roughly the same fan speed. Variance in fan speed is limited to the mechanical components, and should be very small.

But as it turns out, 40% on the cards AMD sent and 40% on the retail cards do not equal the same fan speed. To confirm, I compared the press card to a retail boards from Sapphire and Asus, using our Battlefield 4 benchmark at 2560x1440 and Ultra detail settings.

That's as much of the article I'll quote. Obviously the issue was "fixed" you can read up more on how AMD fixed it.

I just want to get people's opinions on the matter and how they feel and whether they feel mislead by AMD? Also, because of these new drivers, basically all reviews will now have numbers that will be slightly off since these new drivers further increase default fanspeeds.

So what you think after reading the article?
(First thread quoting an article let me know if I messed up anything)

 

[exar333]

Great silicon but crappy cooling? I don't get who would want this stock cooler. You gotta either wait for custom boards or replace the cooler with a custom one immediately.

A number of other reviews have showed a good 10-15% decrease in fps after a 15 min warm-up period as well. This thing reminds me of the 3.8ghz P4 with the acoustics of the infamous Geforce FX 5800 'Dustbuster' Edition.

 

[Elixer]

They saw a problem in the field, and they fixed it.
If it was fixed with the new drivers, then what is the big deal ?

 

[tential]

They saw a problem in the field, and they fixed it.
If it was fixed with the new drivers, then what is the big deal ?

Roughly translated, this means that retail and press cards are being altered via software to operate at the same rotational speed under load. This would be a great thing if AMD was standardizing on the press card's fan, since that's the board we reviewed and praised. However, it's using this opportunity to put an additional 150 RPM through the 290X's fan, presumably to sustain higher clock rates for longer, improving performance. The side effect, of course, is higher power consumption and increased noise.

It really means though that all the initial reviews are going to be off.

The cards will in fact make even more noise than when initally reviewed. I just want to see what people think of it though. should I add this quote to the OP?

 

[Silverforce11]

It really means though that all the initial reviews are going to be off.

The cards will in fact make even more noise than when initally reviewed. I just want to see what people think of it though. should I add this quote to the OP?

You didn't read the review carefully enough, Tom found retail cards had less RPM then press, so they would be quieter than the press sample..

Now its been fixed, it will run at the same RPM and achieve the same noise, thermal and performance as the press sample.

 

[tential]

You didn't read the review carefully enough, Tom found retail cards had less RPM then press, so they would be quieter than the press sample..

Now its been fixed, it will run at the same RPM and achieve the same noise, thermal and performance as the press sample.

I think you're just misunderstanding my wording really I understand that.

However, you did read the quote I just posted right? It states that AMD didn't take this opportunity to just FIX it to run at the same RPM. They also boosted that RPM by 150.
It's in the quote...

 

[3DVagabond]

Oh no, the other thread about this has fallen off of the front page! Let's start another.

Let me add this. Just in case anyone wonders where this all started.
Guru3D

Over the past few days there have been some claims that retail samples of the Radeon R9 290X are clocking down once they hit the 90~95 Degrees C temperature margin. Now first let me eliminate a rumor, this might have been reported first by Tom's Hardware, but they didn't discover it. This was actually mentioned in an NVIDIA presentation and from there on some editors rightfully picked up on it.

nVidia marketing strikes again.

Warning issued for thread crapping.
-- stahlhart

 

[desprado]

Oh no, the other thread about this has fallen off of the front page! Let's start another.

Let me add this. Just in case anyone wonders where this all started.
Guru3D

nVidia marketing strikes again.

that why they are on top they quickly identify there competitors mistakes and take advantage of it.

 

[SlickR12345]

Wow what a mess by AMD. I was really excited about these cards even with their [less than optimal] cooler, but now it just confirms that if you want a 290 or 290x you need to wait for a custom cooler.

These type of variations and high temperatures and loud noise are really bad, not to mention the power draw.

I think its more important than ever to just wait for custom boards with custom cooling solutions. I mean there are some custom cards with their cooling solutions that are super quiet and at the same time very effective.

Profanity isn't allowed in VC&G per forum rules
-Elfear

 

[Abwx]

that why they are on top they quickly identify there competitors mistakes and take advantage of it.

They can expect their viral marketing to backfire on them
since the slides they fed the press with also aknowledge
for whom can catch the hint that the 290X has about 20%
better perf/MHz than their 780ti wich doesnt stand a chance
against a stock 290X that would have a slightly better cooler.

 

[amenx]

Oh no, the other thread about this has fallen off of the front page! Let's start another.

Let me add this. Just in case anyone wonders where this all started.
Guru3D

nVidia marketing strikes again.

And which AMD confirmed and acknowledged. Which negates your point... which seems to be that Nvidia being the source, therefore it must be disinformation or false.

 

[VulgarDisplay]

And which AMD confirmed and acknowledged. Which negates your point... which seems to be that Nvidia being the source, therefore it must be disinformation or false.

Nvidia needs to stop. They keep pointing out AMD's issues so they can fix them faster. They should be ignoring the issues, and letting them persist.

It was brilliant of them to push all the review sites into their crusade against crossfire. Who's laughing now that crossfire is better than SLi (mostly)?

 

[Will Robinson]

Oh no, the other thread about this has fallen off of the front page! Let's start another.

Let me add this. Just in case anyone wonders where this all started.
Guru3D

nVidia marketing strikes again.

This totally.

Now we have another thread on this subject?

Guru:

Quote:
Over the past few days there have been some claims that retail samples of the Radeon R9 290X are clocking down once they hit the 90~95 Degrees C temperature margin. Now first let me eliminate a rumor, this might have been reported first by Tom's Hardware, but they didn't discover it. This was actually mentioned in an NVIDIA presentation and from there on some editors rightfully picked up on it.

See the quote above.

 

[amenx]

Nvidia needs to stop. They keep pointing out AMD's issues so they can fix them faster. They should be ignoring the issues, and letting them persist.

It was brilliant of them to push all the review sites into their crusade against crossfire. Who's laughing now that crossfire is better than SLi (mostly)?

Sadly it looks like a good part of the laugh may be on AMD, as it does make them seem manipulative in that their press cards performed better than retail. So a slight PR dip for them here, even though the issue is largely corrected (with a couple db noise penalty perhaps).

 

[R0H1T]

Sadly it looks like a good part of the laugh may be on AMD, as it does make them seem manipulative in that their press cards performed better than retail. So a slight PR dip for them here, even though the issue is largely corrected (with a couple db noise penalty perhaps).

Can you say the same about Nvidia when they've sent binned chips to reviews like these ~

www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_780_Ti/30.html

Double standards I guess are a common feature of Nvidia's viral marketing & also widely practiced by their supporters

 

[3DVagabond]

And which AMD confirmed and acknowledged. Which negates your point... which seems to be that Nvidia being the source, therefore it must be disinformation or false.

Never said it was false.

 

[amenx]

Can you say the same about Nvidia when they've sent binned chips to reviews like these ~

www.techpowerup.com/reviews/NVIDIA/GeForce_GTX_780_Ti/30.html

Double standards I guess are a common feature of Nvidia's viral marketing & also widely practiced by their supporters

You are reaching that conclusion, not the reviewer in the article. You need to provide a few retail samples to compare it to for your argument to have bite. By the same token, how do we know that retail 290x's dont OC abysmally vs the press samples? whooops... :wub:

This is not to say that both sides dont send their better chips, but if your saying that one side does this to the exclusion of the other, wake up and smell the coffee.

 

[amenx]

Never said it was false.

Well then your point was basically lost.

 

[3DVagabond]

Sadly it looks like a good part of the laugh may be on AMD, as it does make them seem manipulative in that their press cards performed better than retail. So a slight PR dip for them here, even though the issue is largely corrected (with a couple db noise penalty perhaps).

Only if you ignore the large number of reports by people, other sites, and retailers, including members here, that the retail cards they've experienced all perform like the review samples. So, contrary to what you are trying to say "their press cards perform better than retail" does not hold true as a general statement.

 

[R0H1T]

You are reaching that conclusion, not the reviewer in the article. You need to provide a few retail samples to compare it to for your argument to have bite. By the same token, how do we know that retail 290x's dont OC abysmally vs the press samples? whooops... :wub:

This is not to say that both sides dont send their better chips, but if your saying that one side does this to the exclusion of the other, wake up and smell the coffee.

Nope but extending your analogy to the 780Ti review done by TPU, one shouldn't expect such a hefty overclock on retail cards(unless you're inferring that ~30% overclocks would be common place once 780Ti hits the retail market in large numbers) ergo the anomaly in THG's review/retail cards isn't that big a deal. Hope the issue is cleared now cause its been blown way out of proportion!

 

[ShintaiDK]

Now its been fixed, it will run at the same RPM and achieve the same noise, thermal and performance as the press sample.

Its still not the same performance.

And thread should be merged with this one I guess:
http://forums.anandtech.com/showthread.php?t=2351916

Unless the topic is changed to still not fixed.

 

[Abwx]

Only if you ignore the large number of reports by people, other sites, and retailers, including members here, that the retail cards they've experienced all perform like the review samples. So, contrary to what you are trying to say "their press cards perform better than retail" does not hold true as a general statement.

The whole story is that they took retail cards whose
fan speeds where set out of spec by the manufacturers,
read fan running 6.5% slower than in ref cards , and then
presented it as fully specced cards not performing accordingly.

 

[amenx]

Only if you ignore the large number of reports by people, other sites, and retailers, including members here, that the retail cards they've experienced all perform like the review samples. So, contrary to what you are trying to say "their press cards perform better than retail" does not hold true as a general statement.

Well I stand corrected if thats the case. But AMD did seem to think it generally widespread enough to issue a driver fix.

 

[Abwx]

Its still not the same performance.

This graph is totaly irrelevant because it assume that the retail
and ref cards are equaly speeced , read my post above...

 

[R0H1T]

Its still not the same performance.

And thread should be merged with this one I guess:
http://forums.anandtech.com/showthread.php?t=2351916

Unless the topic is changed to still not fixed.

You mean to say that the 6% performance dip is worth noting D: