Thursday, December 7th 2017

AMD Issues Official Statement Regarding RX 560 Silent Downgrade

AMD has come forward with a statement regarding the state of things as it purports to the recent RX 560 downgrade issue, which has been covered extensively by media outlets just this week. The issue stems from the fact that there was no clear differentiator between two different versions of AMD's RX 560 graphics card, which could ship with two different levels of performance: one with 14 CUs (Compute Units) enabled (896 stream processors) or 16 (1024 stream processors, the original specification for the card).

"It's correct that 14 Compute Unit (896 stream processors) and 16 Compute Unit (1024 stream processor) versions of the Radeon RX 560 are available," stated a company representative. "We introduced the 14 CU version this summer to provide AIBs and the market with more RX 500 series options. It's come to our attention that on certain AIB and e-tail websites there's no clear delineation between the two variants. We're taking immediate steps to remedy this: we're working with all AIB and channel partners to make sure the product descriptions and names clarify the CU count, so that gamers and consumers know exactly what they're buying. We apologize for the confusion this may have caused."
While it's true that AMD is now dealing with the situation openly, the company itself isn't completely blameless on the issue. AMD's own RX 560 product page lists the two possible configurations (14 and 16 CU) absent of any further differentiation in product name. At the same time, AMD could have better enforced differentiation opportunities with its AIB partners upon supplying them with both versions of the GPU. As it stands, some users have been unwittingly buying a lower performing part than what was initially advertised by AMD. While this new transparent stance from AMD is refreshing, it doesn't do much regarding already affected customers. Of course, if users suspect or want to check whether or not they have purchased the higher performance version of the card (16 CUs), they can perform a GPU verification with TechPowerUp's GPU-Z tool.Source: Tom's hardware
Add your own comment

94 Comments on AMD Issues Official Statement Regarding RX 560 Silent Downgrade

#1
Vayra86
100% fail and zero points gained. Nicely done AMD.
Posted on Reply
#2
RejZoR
Is it really so hard to call it RX 555? It's not RX 550 and it's not RX 560. Would be a natural course of things...
Posted on Reply
#3
punani
oh AMD's so sorry... as who could have known identical names on different products could cause confusion :shadedshu:
Posted on Reply
#5
medi01
nVidia doing shit is justified by us knowing Huang goes as low as it gets.
AMD doing shit is not justified by that, clear let down, boo, AMD...
Posted on Reply
#6
Fluffmeister
And there is the token "yeah but Nvidia" post.

AMD you were the light, the shining beacon of goodness! Noooooooo!
Posted on Reply
#7
sutyi
RejZoR said:
Is it really so hard to call it RX 555? It's not RX 550 and it's not RX 560. Would be a natural course of things...
Technically it is and should be called RX560D... but AIB partners are not giving a single duck about naming schemes.
Posted on Reply
#8
Dj-ElectriC
john_ said:
AMD did what Nvidia is doing for ages. Let me be clear from the beginning.
THEY SHOULD HAVE NOT DONE SO.





Here is a list with articles about Nvidia's misleading branding: Nvidia GT 730 articles
And here is Nvidia's apology: Nvidia explains the three models of GeForce GT 730
Let's all ignore the fact that in many, possibly most cases, Nvidia's cards actually get stronger with updated architectures.
Posted on Reply
#9
RejZoR
NVIDIA always didn't do nuffin, but as soon as AIB's (not even AMD themselves) screws users over, whole internet implodes in rage over AMD. So typical in last years...
Posted on Reply
#10
Sasqui
john_ said:
AMD did what Nvidia is doing for ages. Let me be clear from the beginning.
THEY SHOULD HAVE NOT DONE SO.





Here is a list with articles about Nvidia's misleading branding: Nvidia GT 730 articles
And here is Nvidia's apology: Nvidia explains the three models of GeForce GT 730
Good reality check. In any case, two (or three) wrongs don't make a right. I'm sure the 2nd party folks aren't happy about the confusion either... marketing has to work overtime on this one.
Posted on Reply
#11
Vya Domus
RejZoR said:
NVIDIA always didn't do nuffin, but as soon as AIB's (not even AMD themselves) screws users over, whole internet implodes in rage over AMD. So typical in last years...
Don't try to bring logic to this matter. The anti-AMD squad is on the way most likely and will stomp every bit of common sense.

Dj-ElectriC said:
Let's all ignore the fact that in many, possibly most cases, Nvidia's cards actually get stronger with updated architectures.
Let's ignore the fact that they are doing that simply because it's cheaper to manufacture them on the newer nodes associated with those architectures.

But let's ignore the already ignored ancient practice of having the same name for different products and spin it in such a way to make it look like a good thing. :laugh:
Posted on Reply
#12
neatfeatguy
john_ said:
AMD did what Nvidia is doing for ages. Let me be clear from the beginning.
THEY SHOULD HAVE NOT DONE SO.





Here is a list with articles about Nvidia's misleading branding: Nvidia GT 730 articles
And here is Nvidia's apology: Nvidia explains the three models of GeForce GT 730
I don't see my particular GT 730 listed on that pic you posted.
Mine has:
1GB DDR3
384 CUDA
64bit bus
12.8GB/s bandwidth
800MHz memory



I'm not caring one way or the other about it's overall performance since I have it in my work computer, but I just thought I'd point it out that she's different over the three listed in your pic.

After all the looking around it looks like the version I have is a mix between a GT 720 and a GT 730 of some sorts....I can't really find any legit posting of this particular model on wikipedia or nvidia.....oh well, it was cheap and did what I needed from it....plus, it's better over the crappy HD 2500 onboard my work computer was using.

Heck, I just went and benched my super powered work computer on 3DMark SkyDiver and compared it to another i3-3220 running just the onboard HD 2500, noticeable difference: https://www.3dmark.com/compare/sd/4914831/sd/2980398#
Posted on Reply
#13
Fouquin
punani said:
oh AMD's so sorry... as who could have known identical names on different products could cause confusion :shadedshu:
Yeah this was a shitty thing to do. Alternatively, nVidia differentiated on the two 1060 names and it hasn't seemed to reduce the number of, "What's the difference between 1060s?" threads that pop up on tech forums everywhere.
Posted on Reply
#15
john_
Dj-ElectriC said:
Let's all ignore the fact that in many, possibly most cases, Nvidia's cards actually get stronger with updated architectures.
GT 730 is supposed to be a 700 series card and one version of that card is using a 400 series GPU(cut down GT 430). Maybe I misunderstood your post. If not, then how is a 400 series GPU, disguised as a 700 series GPU an updated architecture?
And even if we ignore the Fermi version you can't sell under the same name a card with 64bit DDR3 and a card with 64bit GDDR5. The bandwidth difference is huge. In the past that kind of bandwidth difference was enough to justify two DIFFERENT models. GT 640 and GTX 650.

Sasqui said:
Good reality check. In any case, two (or three) wrongs don't make a right. I'm sure the 2nd party folks aren't happy about the confusion either... marketing has to work overtime on this one.
No it doesn't. As I think I make it clear, it was a bad idea from AMD. By now, they should have known to been much more careful and not think like this

Fluffmeister said:
"yeah but Nvidia"
neatfeatguy said:
I don't see my particular GT 730 listed on that pic you posted.
Mine has:
1GB DDR3
384 CUDA
64bit bus
12.8GB/s bandwidth
800MHz memory



I'm not caring one way or the other about it's overall performance since I have it in my work computer, but I just thought I'd point it out that she's different over the three listed in your pic.

After all the looking around it looks like the version I have is a mix between a GT 720 and a GT 730 of some sorts....I can't really find any legit posting of this particular model on wikipedia or nvidia.....oh well, it was cheap and did what I needed from it....plus, it's better over the crappy HD 2500 onboard my work computer was using.

Heck, I just went and benched my super powered work computer on 3DMark SkyDiver and compared it to another i3-3220 running just the onboard HD 2500, noticeable difference: https://www.3dmark.com/compare/sd/4914831/sd/2980398#
Your card is the worst of these three models available in 3D performance. Looking at features is probably better than the Fermi model. It's that "LOL BAD" version in the screenshot I posted. It does have a Kepler GPU, it does have 384 CUDA cores, but it also comes with only - in your case - 12.8GB/sec memory bandwidth. It's lower than those 14.4GB/sec on Nvidia's specs page, because when GT 730 came out the specs where talking about DDR3 memory running at 1800MHz.

In typical day usage scenarios - don't know about CUDA, you might not notice any difference, but a model with GDDR5 memory would be killing your card in 3DMark.
Posted on Reply
#16
jmcslob
So lemme get this straight... About the same time there was a slight price hike the product was also being cheapened.....
This is fucking Gross AMD...
Very very gross.
Posted on Reply
#17
GoldenX
Guys it's not like Intel never did this right? Oh wait, third time using 9 gen graphics up until now...
VIA master race.
Posted on Reply
#18
HD64G
Big mistake intended or not. WHoever is responsible needs to be sacked or get a payment cut off. Much unneeded press negativity in a period that AMD tries hard to gain market share. AIBs might have a share of responsibility in this but AMD should prevent this from happening from the get go.
Posted on Reply
#19
Vayra86
john_ said:
AMD did what Nvidia is doing for ages. Let me be clear from the beginning.
THEY SHOULD HAVE NOT DONE SO.





Here is a list with articles about Nvidia's misleading branding: Nvidia GT 730 articles
And here is Nvidia's apology: Nvidia explains the three models of GeForce GT 730
But... we're reading an AMD topic here. What's next, a full list of all weird releases of every company ever?

Also - you can see in your little screenshot a full spec list, including shader counts, memory bus etc. and these details are 100% correct. Let's fast forward to Intel's Skylake versus its Kaby Lake parts, now, or give their new i3 quadcore parts the middle finger because they're actually last years' i5 stack.... This is really quite normal. AMD has also been re-releasing Pitcairn until infinity, re-rebranding a whole product stack from HD- to R9, hell I could give you a dozen more examples of any hardware company.

Bottom line, your example makes no sense at all and its not the place.

And this one's for you @RejZoR
http://www.tomshardware.com/news/MSI-GTX-660-670-overvolting-PowerEdition,18013.html


Nobody is clean
Posted on Reply
#21
Dave65
Atta boy AMD, you really taking moves out of Nvidias playbook.. BAD MOVE!
Posted on Reply
#22
newtekie1
Semi-Retired Folder
john_ said:
AMD did what Nvidia is doing for ages. Let me be clear from the beginning.
THEY SHOULD HAVE NOT DONE SO.





Here is a list with articles about Nvidia's misleading branding: Nvidia GT 730 articles
And here is Nvidia's apology: Nvidia explains the three models of GeForce GT 730
I'm not a huge fan of different variations of a card all being given the same model number, I definitely don't think that should be done. But comparing it to what AMD did with the RX 560 isn't a fair comparison. What AMD did is notably worse.

With the GT730, that was literally a bottom of the barrel card. AND they released all the versions of the card at the same exact time. So everyone knew they existed from day one, and you knew what to look for. If reviewers reviewed the card, they knew to look at the version and let people know which one they were reviewing. (Did any major site ever even review the GT730?) Plus, the target demographic wasn't gaming, if you tried to game on a GT730, I feel bad for you, every version of that card was shit.

But AMD released a model on the market, then later on released another weaker card under the same name. This is very different from what nVidia did, and worse. It also isn't the first time they've released a card, let it be reviewed, then reduced the performance. They've been pulling that shit since at least 2008...
Posted on Reply
#23
Gmr_Chick
newtekie1 said:
I'm not a huge fan of different variations of a card all being given the same model number, I definitely don't think that should be done.

But AMD released a model on the market, then later on released another weaker card under the same name. This is very different from what nVidia did, and worse.
GTX 1060. Nuff said.
Posted on Reply
#24
EarthDog
medi01 said:
nVidia doing shit is justified by us knowing Huang goes as low as it gets.
AMD doing shit is not justified by that, clear let down, boo, AMD...
Gmr_Chick said:
GTX 1060. Nuff said.
Christ... how many threads can we spill this into...

The fundamental problem is the fact that amd changed the specs on the website so the same name card has two different specs. 560d was asia... in the us its jusst a 560. The 1060 on nvidias website, clearly shows that the specs are different.

This will look familiar to vya... lol....Incoming car analogy!!!

NVIDIA - You can buy a base Mustang (1060 3gb) and a Mustang GT (1060 6gb)
Your argument - they are both called a Mustang and we cant differentiate easily enough. However if i look it up, clearly we can tell there is a difference..see specs.
https://www.nvidia.com/en-us/geforce/products/10series/geforce-gtx-1060/

AMD - You can buy a Mustang GT only...however you arent sure what motor is in it, the slower 6 cyl turbo or the 5.0L, and cant tell until you get it home and check. The specs from the factory show both motors for the same car and trim level, no help there.
Argument - Those dealerships are being shady even though the ford factory itself has it listed incredibly ambiguous. Ok.. its not ambiguous, it is clear there are two specs for the same card...no "d"
http://www.amd.com/en/products/graphics/radeon-rx-560


While nvidias naming scheme could most certainly be made more clear, it is a different situation as nvidia makes it clear there is a difference starting from their website on down to the card partners.

Edit: and please stop the random nvidia injection into these threads. We get it... :)
Posted on Reply
Add your own comment