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

#76
xfia
AMD should make a commercial side by side with 970+290 that shows what happens when you go over 3.5gb.. I could see the ending now!

Posted on Reply
#77
qubit
Overclocked quantum bit
NVIDIA normally make such great graphics cards which is why I've been buying them for years. Yet, they also have a tendency to pull a fast one like this, trashing their reputation and customers' trust in their products, including mine. What's worse is that this isn't the first time we've seen underhand tactics like this result in scandals for them. A good class action win against them for this would be great at helping to keep them honest in the future. It's not gonna happen though, is it?

I never like the idea of a gimped GPU with a bit of it disabled since it reduces performance and unbalances the design and this is a graphic example of exactly why this is bad. This is the sort of reason why I always insist on getting the top GPU in the range. Yeah, it costs me a lot of money, but I also don't have to put up with garbage like this.
Posted on Reply
#78
Bad Bad Bear
I purchased 2 of these cards. Not happy.

I have lodged a formal complaint with the ACCC ( I'm in Australia ) Have forwarded my formal complaint to Nvidia and my local retailer Umart. Under Australian consumer law i'm entitled to a refund.

Let's see what eventuates from this.....
Posted on Reply
#79
damric
Bad Bad BearI purchased 2 of these cards. Not happy.

I have lodged a formal complaint with the ACCC ( I'm in Australia ) Have forwarded my formal complaint to Nvidia and my local retailer Umart.

Let's see what eventuates from this.....
Yeah I imagine that SLI would take advantage of more VRAM so I would be pissed too. I never thought my HD 7850s could use the full 2GB but they do in games like Far Cry 3 1080p even with no AA.
Posted on Reply
#80
silapakorn
damricYeah I imagine that SLI would take advantage of more VRAM so I would be pissed too. I never thought my HD 7850s could use the full 2GB but they do in games like Far Cry 3 1080p even with no AA.
But SLI won't stack VRAM of both cards.
Posted on Reply
#81
Bad Bad Bear
damricYeah I imagine that SLI would take advantage of more VRAM so I would be pissed too. I never thought my HD 7850s could use the full 2GB but they do in games like Far Cry 3 1080p even with no AA.
The Nvidia moderators just deleted my post off the offical forums.

They are in damage control for sure. Too late now !
Posted on Reply
#82
AsRock
TPU addict
I wanted a 4GB card and the 290\X was the only real option, if i had waited i probably would of got a 970 and this finding would not make happy if a performance hit was there or not.

A play a fair few games that take over 3.5GB ram and it would just annoy me just being sold some thing that is not truly what it claims.
Posted on Reply
#83
damric
silapakornBut SLI won't stack VRAM of both cards.
We know this.
Posted on Reply
#84
Fluffmeister
AsRockI wanted a 4GB card and the 290\X was the only real option, if i had waited i probably would of got a 970 and this finding would not make happy if a performance hit was there or not.

A play a fair few games that take over 3.5GB ram and it would just annoy me just being sold some thing that is not truly what it claims.
Out of interest what are those games? I'm a sucker for creating problems for myself.
Posted on Reply
#85
ensabrenoir
....ok all Nvidia has to do is release some awesomesauce that fully activate all of the 970's potential (scapping the ti version) and everyone's happy.:rolleyes: Either way I'm gonna buy a nice card really cheap due to crowd hysteria.
Posted on Reply
#86
Steevo
FluffmeisterOut of interest what are those games? I'm a sucker for creating problems for myself.
It could be modded Skyrim, modded GTA4, modded Farcry 3, Watch Dogs, Far Cry 4 without mods and high res, future games like GTA5 that want more vmem to hold textures.
Posted on Reply
#87
Xzibit
ensabrenoir....ok all Nvidia has to do is release some awesomesauce that fully activate all of the 970's potential (scapping the ti version) and everyone's happy.:rolleyes: Either way I'm gonna buy a nice card really cheap due to crowd hysteria.
Naw, They just have to rename the 970 to 970 SE (Segmented Edition). Marketing gold!!! $$$

Maybe Peter will sell you his.

Posted on Reply
#88
Justin_you
Can we bring NVDIA to court because of their false advertisement on the GTX970 spec?
Heck, can we even ask for a refund because we have bought something that not according to the spec?
Posted on Reply
#89
FordGT90Concept
"I go fast!1!11!1!"
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.
Most people don't know the difference between GB and GiB either. Seagate got sued for it and lost even though math was clearly on their side.
Posted on Reply
#90
Ralfies
Read this on the NVIDIA forum:
PeterS@NVIDIAHey,

First, I want you to know that I'm not just a mod, I work for NVIDIA in Santa Clara.

I totally get why so many people are upset. We messed up some of the stats on the reviewer kit and we didn't properly explain the memory architecture. I realize a lot of you guys rely on product reviews to make purchase decisions and we let you down.

It sucks because we're really proud of this thing. The GTX970 is an amazing card and I genuinely believe it's the best card for the money that you can buy. We're working on a driver update that will tune what's allocated where in memory to further improve performance.

Having said that, I understand that this whole experience might have turned you off to the card. If you don't want the card anymore you should return it and get a refund or exchange. If you have any problems getting that done, let me know and I'll do my best to help.

--Peter
Posted on Reply
#91
Xzibit
RalfiesRead this on the NVIDIA forum:
It's good that they're trying to work on a fix through drivers, but the performance of the card isn't the big issue. The fact they lied about the specs is why most folk are upset.
Good catch. Here is the link to his post...

Just so people know its real and your not flamed..
Posted on Reply
#92
Prima.Vera
SteevoIt could be modded Skyrim, modded GTA4, modded Farcry 3, Watch Dogs, Far Cry 4 without mods and high res, future games like GTA5 that want more vmem to hold textures.
Let's relax a little. Those games, even ubber modded, they will only go beyound 4GB of VRAM, only if you play on 4K resolution with 4xMSAA enabled. Which even if you had 1TB of VRAM would still be unplayable due to insufficient horse power ;)

www.digitalstormonline.com/unlocked/video-memory-usage-at-4k-uhd-resolutions-idnum146/
Posted on Reply
#93
xfia
wow.. he is actually contacting manufacturers for people. they should just hash out a deal that lets you trade up for a 980.
Posted on Reply
#94
Vario
Its unfortunate but at the same time, the card does perform well for a lot of what you guys are using them for.
Posted on Reply
#95
damric
Ugh this is like finding out the doctor took a little extra off the tip then was needed.
Posted on Reply
#96
AsRock
TPU addict
FluffmeisterOut of interest what are those games? I'm a sucker for creating problems for myself.
FC4, Watch Dogs and Arma 3 to name a few.
RalfiesRead this on the NVIDIA forum:
Convenience, cannot imagine them saying that there is a chance of your game when it hits over 3.5GB usage it could take a dump.

EDIT:
Prima.VeraLet's relax a little. Those games, even ubber modded, they will only go beyound 4GB of VRAM, only if you play on 4K resolution with 4xMSAA enabled. Which even if you had 1TB of VRAM would still be unplayable due to insufficient horse power ;)

www.digitalstormonline.com/unlocked/video-memory-usage-at-4k-uhd-resolutions-idnum146/
You can hit 3.5+GB with them without mods. And i know WD can hit that under 1080P.
Posted on Reply
#97
sergionography
So to summarize what we know so far

1. each Maxwell smm can only address 4 pixels per clock meaning even with the rops present they r pretty much useless.
2 Maxwell is having this dilemma because it's the trade off nvidia made when they made Maxwell a more cache dependant design than it is on external bus width.
3. Bus width is meaningless when the gpu computer units can't handle the throughput, it's almost like having 5 lanes open at a movie theater counter but with only 2 employees working

This being said this is more reason to stop looking at theoretical specs and actually measure real time performance. Had it been that memory bandwidth was properly measured it would've been obvious that gtx 970 has lower throughout than a gtx980 despite sharing the same memory width and "rops". And last but not least I wanna see more benchmarks on amd cards measuring the effectiveness/efficiency of the memory width vs throughput of gcn and have a more lower level look at architecture in general because I can't help but think amd just goes extreme on certain aspects and slaps random hardware on their chips, idk if anyone noticed but gcn cards have insane variance in efficiency between different chips because the design has become "less modular?" Or should I say too modular? Meaning the parts on the chip are too independent of each other while nvidia has a more focused graphics module containing pretty much everything at the perfect ratio with as little bottlenecks as possible so when they scale up and down the ratio of rendering/throughput remains unchanged across the board, the downside is less flexibility to customize specs which caused this gtx970 dilemma, but the benefit is having consistency and less chance of a "hit or miss"
Posted on Reply
#98
FordGT90Concept
"I go fast!1!11!1!"
xfiawow.. he is actually contacting manufacturers for people. they should just hash out a deal that lets you trade up for a 980.
Would probably cost them more than the class action suit. Last one several years ago was settled for around $18 million if memory serves. That's chump change for NVIDIA. NVIDIA also can't force third party manufactures to do things not in their agreement. The manufacturers have to do that on their own accord and I doubt they'd jump in front of the bus for NVIDIA because they have no reason to.
Posted on Reply
#99
RCoon
The most hilarious thing about all this is that most of the people who are successfully RMA'ing their 970s are then buying a 980 as a replacement from the same retailer. It's not like NVidia is even losing anything out of this, if anything retailers and AIB's are probably going to make more profit from 980 sales now.

The world we live in.
Posted on Reply
#100
Xzibit
RCoonThe most hilarious thing about all this is that most of the people who are successfully RMA'ing their 970s are then buying a 980 as a replacement from the same retailer. It's not like NVidia is even losing anything out of this, if anything retailers and AIB's are probably going to make more profit from 980 sales now.

The world we live in.
The consumer can't sue Nvidia directly unless its a made by Nvidia since they don't put the memory, which is at issue on the board. The fury will have to be directed at the AiBs, The AiBs are extremely unlikely to sue Nvidia to recoup losses and they will take the financial hit on returns & lawsuits. Nvidia will in return compensate them and probably discount future shipment to make nice. Depending on how this plays out AIBs will probably be more cautious when ordering next go around.


We'll see how long "performance is the only thing that matters" sticks around. Next GPU rumor/release thread -->
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