Tuesday, January 27th 2015

GTX 970 Memory Drama: Plot Thickens, NVIDIA has to Revise Specs

It looks like NVIDIA's first response to the GeForce GTX 970 memory allocation controversy clearly came from engineers who were pulled out of their weekend plans, and hence was too ambiguously technical (even for us). It's only on Monday that NVIDIA PR swung into action, offering a more user-friendly explanation on what the GTX 970 issue is, and how exactly did they carve the GM204 up, when creating the card.

According to an Anandtech report, which cites that easy explanation from NVIDIA, the company was not truthful about specs of GTX 970, at launch. For example, the non-public document NVIDIA gave out to reviewers (which gives them detailed tech-specs), had clearly mentioned ROP count of the GTX 970 to be 64. Reviewers used that count in their reviews. TechPowerUp GPU-Z shows ROP count as reported by the driver, but it has no way of telling just how many of those "enabled" ROPs are "active." The media reviewing the card were hence led to believe that the GTX 970 was carved out by simply disabling three out of sixteen streaming multiprocessors (SMMs), the basic indivisible subunits of the GM204 chip, with no mention of other components like the ROP count, and L2 cache amount being changed from the GTX 980 (a full-fledged implementation of this silicon).
NVIDIA explained to Anandtech that there was a communication-gap between the engineers (the people who designed the GTX 970 ASIC), and the technical marketing team (the people who write the Reviewer's Guide document, and draw the block-diagram). This team was unaware that with "Maxwell," you could segment components previously thought indivisible, or that you could "partial disable" components.

It turns out that in addition to three SMX units being disabled (resulting in 1,664 CUDA cores), NVIDIA reduced the L2 cache (last-level cache) on this chip to 1.75 MB, down from 2 MB, and also disabled a few ROPs. The ROP count is effectively 56, and not 64. The last 8 ROPs aren't "disabled." They're active, but not used, because their connection to the crossbar is too slow (we'll get to that in a bit). The L2 cache is a key component of the "crossbar." Think of the crossbar as a town-square for the GPU, where the various components of the GPU talk to each other by leaving and picking-up data labeled with "from" and "to" addresses. The crossbar routes data between the four Graphics Processing Clusters (GPCs), and the eight memory controllers of 64-bit bus width each (which together make up its 256-bit wide memory interface), and is cushioned by the L2 cache.

The L2 cache itself is segmented, and isn't a monolithic slab of SRAM. Each of the eight memory controllers on the GM204 is ideally tied to its segment of the L2 cache. Also tied to these segments are segments of ROPs. With NVIDIA reducing the L2 cache amount by disabling one such segment. Its component memory controller is instead rerouted to the cache segment of a neighbouring memory controller. Access to the crossbar for that memory controller is hence slower. To make sure there are no issues caused to the interleaving of these memory controllers, adding up to the big memory amount figure that the driver can address, NVIDIA partitioned the 4 GB of memory to two segments. The first is 3.5 GB large, and is made up of memory controllers with access to their own segments of the L2; the second segment is 512 MB in size, and is tied to that memory controller which is rerouted.

The way this partitioning works, is that the 3.5 GB partition can't be read while the 512 MB one is being read. Only to an app that's actively using the entire 4 GB of memory, there will be a drop in performance, because the two segments aren't being read at the same time. The GPU is either addressing the 3.5 GB segment, or the 512 MB one. Hence, there's a drop in performance to be expected, again, for apps that use up the entire 4 GB of memory.

While it's technically correct that the GTX 970 has a 256-bit wide memory interface, and given its 7.00 GHz (GDDR5-effective) memory clock, that translates to 224 GB/s of bandwidth on paper, not all of that memory is uniformly fast. You have 3.5 GB of it having normal access to the crossbar (the town-square of the GPU), and 512 MB of it having slower access. Therefore, the 3.5 GB segment really just has 196 GB/s of memory bandwidth (7.00 GHz x 7 ways to reach the crossbar x 32-bit width per chip), which can be said with certainty. Nor can we say how this segment affects the performance of the memory controller whose crossbar port it's using, if the card is using its full 4 GB. We can't tell how fast the 512 MB second segment really is. But it's impossible for the second segment to make up 28 GB/s (of the 224 GB/s), since NVIDIA itself claims this segment is running slower. Therefore NVIDIA's claims of GTX 970 memory bandwidth being 224 GB/s at reference clocks is inaccurate.

Why NVIDIA chose to reduce cache size and ROP count will remain a mystery. We can't imagine that the people designing the chip will not have sufficiently communicated this to the driver and technical marketing teams. To claim that technical marketing didn't get this the first time around, seems like a hard-sell. We're pretty sure that NVIDIA engineers read reviews, and if they saw "64 ROPs" on a first-page table, they would have reported it up the food-chain at NVIDIA. An explanation about this hardware change should have taken up an entire page in the technical documents the first time around, and NVIDIA could have saved itself a lot of explanation, much of it through the press.
Source: Anandtech
Add your own comment

138 Comments on GTX 970 Memory Drama: Plot Thickens, NVIDIA has to Revise Specs

#51
the54thvoid
Super Moderator
v12dockOn the flip side AMD is up 4% :laugh: One of the few companies that is not taking a massive hit today.
Unfortunately the AIB's for AMD are reducing orders. They need to start peddling the new tech coming 2nd Half 2015
Posted on Reply
#52
Uplink10
newtekie1They didn't lie about ROP count. The card has 64 Active ROPs. That is not a lie. It only uses 56 of them because using the others would actually slow the card down. But there are 64 active ROPs.
They could activate all disabled parts of the GPU and advertise it as they do 980 but activated parts are not important unless they can be used. You have a car with V8 (eight cylinders) engine but you can only use six cylinders (V6).
Posted on Reply
#53
No Nrg
Everyone who bought a 970 because of the ROP count raise their hands. *crickets*

The fact that they got the spec wrong does suck but the benchmarks are what most everyone based their purchase on and those numbers won't change.

Hopefully there is no underlying issue like this with the 980 I bought.....
Posted on Reply
#54
newtekie1
Semi-Retired Folder
Uplink10They could activate all disabled parts of the GPU and advertise it as they do 980 but activated parts are not important unless they can be used. You have a car with V8 (eight cylinders) engine but you can only use six cylinders (V6).
My car has 8 Cylinders, and only uses 4 most of the time because it gets better gas mileage. It is still advertised as an 8 Cylinder, but not using all 8 has benefits.

In theory the ROPs can actually be used, but as said they would actually slow the card down.
Posted on Reply
#55
Casecutter
Here's the thing if gelding 3 of the SMX's in some way caused the memory controller to be unstable and found it best to fuse off one of the L2 to get the performance; great then just explain that’s part of the differences between the two and show how and why it was done as... they have now. We have to ask, do the GTX980M (12 SMX's) have any L2 fused off? Perhap a as complete block doesn't affect the control in the same way IDK. Perhaps the explanation is simpler...

Nvidia found they had a large volume of chips with at least one damage L2, and to fulfill the volume of 970's they needed to put to market they fused off one L2 on all cards... Which is not the issue "in and of itself", but the more I see it appears the scenario of defective/damaged L2, and Nvidia figure out a way to "weasel" around it because nobody ever checks or questions L2 in reviews and would take them at their word (spec's)... why bring it up!

Either way Nvidia misrepresented the product in more ways than one. It wasn't just one obscure specification, they went as far as touting the GTX 970 ships with THE SAME MEMORY SUBSYSTEM AS OUR FLAGSHIP GEFORCE GTX 980. So Nvidia should feel obligated to make it right for any of those who have purchase 970’s and consider Nvidia misrepresent product. They should be eligible for some amount of reimbursement, or complete refund if they no long want the product.
Posted on Reply
#56
Toothless
Tech, Games, and TPU!
Aaanndd there goes my wishlist.
Posted on Reply
#57
BiggieShady
No NrgEveryone who bought a 970 because of the ROP count raise their hands.
You don't buy car because it has four wheels, but you don't buy a car if it doesn't .... see?
Posted on Reply
#58
Casecutter
BiggieShadyYou don't buy car because it has four wheels, but you don't buy a car if it doesn't .... see?
Actully just 3-1/2 and then the 1/2 spare that cause you to run slower.
Posted on Reply
#59
hyp36rmax
SeleneThis is no different than the dual GPU cards, they physically have double the memory but only half is usable. This changes nothing, the card does have 4gb, and as you said all the benchmarks are still the same. I dont agree with them doing this and not telling people but if you got the card based on reviews and benchmarks you got what you paid for.

The truth is once the card gets a to a rez where 4gb would even be worth having the GPU cant handle it and it would make maybe 1-2fps difference at best, its been show time and time again, 256bit bus really can only handle 2gb.
I agree with most of your statement however this is nothing like a Dual GPU card as the memory allocation is split and mirrored between each of the two GPU Cores to operate accordingly compared to the Single GPU GTX 970 with 4gb of available GDDR5 Memory with 3.5gb useable. As a Dual GPU card is akin to a Crossfire or SLI on a single PCB. You don't claim 8GB of total useable GDDR5 RAM on a Crossfire R9 290X 4gb cards (Unless of course you have two 8gb models)
Posted on Reply
#60
FreedomEclipse
~Technological Technocrat~
For those that care - theres a petition up to make a class action lawsuit against Nvidia. Though, the objective of it is to get refunded. They havent mentioned a percentage so im guessing a full refund rather than a partial one.

- Nvidia could just save themselves so much hassle if they just gave away some game keys to people who purchased a 970, Sure it wont 'fix' the 970 but for most of us the card performs flawlessly despite the misadvertised specs.

Heres the link to the Nvidia forums where the petition is posted - It will be interesting to see what steps Nvidia will take to fix the issue.

Already some suggestions from some of the users saying that Nvidia should accept their 970s back and step them up to 980s for a little cash on top - I wouldn't mind this option :p though i'll be happy with a partial refund.
Posted on Reply
#61
john_
I like how some people see things.

Hawaii GPU throttling:
Who cares about performance? Throw AMD into the fire.

NVidia lying about GTX 970 specs:
Oh, come on, it's only -3%!
Posted on Reply
#62
64K
FreedomEclipseAlready some suggestions from some of the users saying that Nvidia should accept their 970s back and step them up to 980s for a little cash on top - I wouldn't mind this option :p though i'll be happy with a partial refund.
I wouldn't mind putting $100 with my 970 return to step up to a GTX 980. That would make it $450 US for me which is what the GTX 980 should have been anyway imo considering the performance increase over the 970.
Posted on Reply
#63
Xzibit
FreedomEclipseFor those that care - theres a petition up to make a class action lawsuit against Nvidia. Though, the objective of it is to get refunded. They havent mentioned a percentage so im guessing a full refund rather than a partial one.

- Nvidia could just save themselves so much hassle if they just gave away some game keys to people who purchased a 970, Sure it wont 'fix' the 970 but for most of us the card performs flawlessly despite the misadvertised specs.

Heres the link to the Nvidia forums where the petition is posted - It will be interesting to see what steps Nvidia will take to fix the issue.

Already some suggestions from some of the users saying that Nvidia should accept their 970s back and step them up to 980s for a little cash on top - I wouldn't mind this option :p though i'll be happy with a partial refund.
I'd go for the refund or the setup.

What if the game key they offer is for a game that allocates over 3.5GB of memory. It would be an insult to injury.
Posted on Reply
#64
REAYTH
newtekie1My car has 8 Cylinders, and only uses 4 most of the time because it gets better gas mileage. It is still advertised as an 8 Cylinder, but not using all 8 has benefits.

In theory the ROPs can actually be used, but as said they would actually slow the card down.
Most people buy a card off of performance reviews. Not how many ROPSs it has. As a matter of fact I would be willing to bet 99.999999999% of people don't even know WTF a ROP is. But they do know they get 20 FPS more in Battlefield 4.

Is this an issue in advertisement? Yes. Should Nvidia address it? Yes. Should people be able to cry foul and return the card......No. They bought it for the performance. Not the damn ROP count.
Posted on Reply
#65
L337One91
All I want to know is whether or not I will experience stuttering playing at 1440/1600p.
Posted on Reply
#66
64K


Well, it's a little early to tell how this will all work out. Nvidia is just putting fluff responses out there to stall until they can figure out what is the smartest long term decision that they can make.
Posted on Reply
#67
Xzibit
64K

Well, it's a little early to tell how this will all work out. Nvidia is just putting fluff responses out there to stall until they can figure out what is the smartest long term decision that they can make.
Well they could be putting this out with the intention of minimizing penalties down the road in a False Advertising Lawsuit case. One of the severe penalties that can be levied by the courts is if the advertiser had intent. The first address Nvidia made was "Miscommunication" to try and minimize public perception and an early sign of Cover Your own A**.
Posted on Reply
#68
Uplink10
64KI wouldn't mind putting $100 with my 970 return to step up to a GTX 980.
Nvidia would not mind this since they would get another 100 USD and we all know that if GTX 980 would cost 450 USD this is still too much, I mean they deserve this since they intentionally disable parts of GPU on GTX 970.
REAYTHMost people buy a card off of performance reviews.
And because of specifications, with more resources it should be faster. This is technicality (in contrast to performance) and when you violate them it is going to stab you in the back.
Posted on Reply
#69
newtekie1
Semi-Retired Folder
REAYTHMost people buy a card off of performance reviews. Not how many ROPSs it has. As a matter of fact I would be willing to bet 99.999999999% of people don't even know WTF a ROP is. But they do know they get 20 FPS more in Battlefield 4.

Is this an issue in advertisement? Yes. Should Nvidia address it? Yes. Should people be able to cry foul and return the card......No. They bought it for the performance. Not the damn ROP count.
That is exactly my point. When I read the reviews for the 970 I didn't really even care about the specs. In fact I jumped straight to the performance section. In the end, that is all that matters.
L337One91All I want to know is whether or not I will experience stuttering playing at 1440/1600p.
I haven't yet @1440p.
Uplink10And because of specifications, with more resources it should be faster. This is technicality (in contrast to performance) and when you violate them it is going to stab you in the back.
I've never bought a card based on specs. If it used 64 ROPs and was slower, I'd be more disappointed than only using 56 of the 64 active and being faster.
Posted on Reply
#70
HumanSmoke
IkarugaI have to disagree. I can't imagine a single reason why would an engineer lie 64 ROPs when there is only 56. These monster companies always have dedicated teams for communications with the outside world (press, developers, retailers, etc). The only thing I can imagine is that somebody in that department failed big time, (regardless if it was deliberate or just a stupid a mistake from that person).
I'm inclined to agree. Hardware vendors know their products go under the microscope of the community they sell to. I do have to admit that If I have a choice between believing a labyrinthine conspiracy theory, or the communication between engineering and marketing screwed up, I'm inclined to go with the latter. As I pointed out in another thread, wasn't Bulldozers missing 800 million transistors just such a case?The alternative conspiracy theory (which some oddballs gave credence to) would be that AMD boosted the trans count to make it seem like a more complex chip.
v12dockClass action lawsuit?
Certainly a shitty situation, but is it actionable in court? Might be good leverage from a bad PR standpoint to get some action but it has to be an all-or-nothing scenario - refund or trade-up. The heart of the matter would seem to be the memory and bus width discrepancy, but even the testing shows that albeit slower it is active. The only revised specs I've seen are the ROP count and L2 cache, neither of which appear in the official product specifications (this a cached copy of the original listed spec sheet) of Nvidia or their partners. It might be a shitty situation, but I'd be a little dubious if that it alone constitutes a case. The only evidence I've seen that might be indictable is the claim in the reviewers guide stating that the GTX 970 shares the same memory subsystem with the 980 - but that isn't part of the official product specification. If reviewers guides with their cherry-picked best case scenario benchmarks are litigation fodder then I don't think many products are safe from civil suit.
Posted on Reply
#71
the54thvoid
Super Moderator
Class action lawsuit might not get very far. Even if it was going somewhere, it would be settled out of court to the small number of complainants and then be signed off to a confidentiality clause. Nvidia would ride that out and take the hit to it's image. It'd bounce back by releasing it's next chip and charge lower than expected, bringing back the love.

For those that keep bitching on about people saying it's not making a difference - the game reviews still stand. The performance is still there. Nvidia have entered a PR nightmare and they cannot come out looking clean. They cocked up for sure - no doubt - let's not defend them.

The irony is those people that have not noticed any problems with their cards suddenly thinking - is my card bust? No, it's not - it's running as designed. The problem is not the design but the marketing of it. There were no lies as such, only technical ambiguity (that's why the lawsuit wont work). The dual gpu cards have set a precedent there. Someone said elsewhere that it's not relevant but it really is. A 12GB Titan Z or a 8GB R9 295 does not have 12 or 8 GB of functioning memory as we know it but they both have what is stated. This would be used as a defence by NV. The 970 does have all the things listed but they're just not used as we 'assumed'.

Have Nvidia been arseholes (and continuing to make themselves look even more raw?) yes, of course.

Are some forum posters being fanatically childish about it? of course.

Should you be pissed if you have a 970? Only if it's actually affecting you. FFS, my 780Ti's only have 3GB memory but they still pull 5083 in Firestrike Ultra (at stock).

The best thing about this entire debate is watching those it affects sort of going 'meh' and watching (sorry guys) AMD loyalists getting their knickers in a twist. Can't we all agree - Nvidia are dicks, but the cards still work.
Posted on Reply
#72
Casecutter
newtekie1That is exactly my point. When I read the reviews for the 970 I didn't really even care about the specs. In fact I jumped straight to the performance section. In the end, that is all that matters.
If Nvidia called it 3.5Gb Active Boost... or 4Gb Memory Compression... or something like that I think a lot of folk would've... But we weren't privy to that information.
Posted on Reply
#73
TRWOV
HumanSmokeAs I pointed out in another thread, wasn't Bulldozers missing 800 million transistors just such a case?The alternative conspiracy theory (which some oddballs gave credence to) would be that AMD boosted the trans count to make it seem like a more complex chip.
Sorry for countering so many of your posts :oops: but in that case AMD actually did what people are saying nVidia should have done. AMD saw the error on a review and corrected the mistake. Kind of a typo if you think about it: I can easily see 1.2 become 2 if at some point someone just wrote ".2" forgetting the "1" and then someone "corrected" that to 2 (a .2B CPU wouldn't make sense but a 2B wasn't much of a stretch). Marketing didn't catch that up, after reviews went up AMD engineers must have caught the error and sent the mentioned correction.



It's quite amusing how people react to mixups. Here we are sacrificing nVidia for not telling people and qubit actually berated AMD for making the correction:
qubitIf this is an attempt to make the processor look better by showing it "doing more with less", then this PR stunt has backfired spectacularly and it would have been better to have left the "error" as it was. Paradoxically, FX processors are a sales success and are flying off the shelves as we just reported, here.
Bad if you do, bad if you don't. :slap:
Posted on Reply
#74
HumanSmoke
TRWOVSorry for countering so many of your posts :oops:
Well, that is perfectly fine. Healthy debate never killed anyone. The point I was making is that screw-ups in communication (esp. involving marketing) are quite prevalent - certainly more so than vast conspiracies. Even the fallout (the delay in recognition and acknowledgement) from major screw-up's on a hardware level such as Intel's FDIV bug arose through lower-level management going missing on basic protocol. The day it made it to board level, Board Chairman Arthur Rock instituted a mea culpa.

If anyone is actually interested in the performance aspects of this issue, PCGH have quite an interesting analysis(German) comparing the GTX 970 with a GTX 980 downclocked to simulate the reduced bandwidth/lower shader count etc. but with a full 4GB of vRAM. The results tend to show significant separation in the 4K benchmark (although since neither offer playable frame rates I'm not too sure of the relevancy).
Posted on Reply
#75
damric
Maybe Nvidia will give you $10 off your next Titan purchase :)
Posted on Reply
Add your own comment
Nov 21st, 2026 12:08 PST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts