Even 8 FP64 units per SM would be overkill for a '4'-series chip. GM204 had a 1/32 ratio - that would equate to a mere 2 FP64 units per SM if the modules still have 64 standard CUDA cores as they do on GP100.
Even 8 FP64 units per SM would be overkill for a '4'-series chip. GM204 had a 1/32 ratio - that would equate to a mere 2 FP64 units per SM if the modules still have 64 standard CUDA cores as they do on GP100.
RussianSensation
Elite Member
- Sep 5, 2003
- 19,458
- 765
- 126
GP104 cards are 560Ti/670/680//970/980 successors. For 99.9% of consumers, it would be a waste of power usage and money if these cards were 12GB. Besides, any of these should be a HUGE upgrade for GTX580/780Ti owners (and of course 680/980 and lower). If you want a card with 16-32GB of VRAM, Pascal Titan will fit the bill.
It's not exactly an apples-to-apples comparison to compare a flagship Fermi GTX580 to upper mid-range Pascal cards and crap on them. The lineage successor to GTX580 is the GP100 (rumoured GP102) line. Either way, with 8GB of GDDR5X, 1080 GP104 will already be a tangible improvement over 980Ti/Fury X for anyone who skipped those cards. It'll also use less power, have updated video codec engines, most likely have DP1.3/HDMI 2.0a/HDR monitor support, and be better prepared for DX12 games with Async Compute. Finally, you'd also get NV's more recent driver support which is crucial, as we've seen in the last 4 years.
Last edited:
I've been reading both... that it will match those and that it will exceed them. I can't really picture anything but the latter, because then NV would be holding back like crazy given the process advantage. But like many others I have the worry in the back of my mind that even with competition, performance this time will be held back to the annual average increase on 28nm despite a huge advantage with the 16/14nm jump.
They will both do as much as is commercially feasible. If one of them could drop a 600²mm chip next month on the other, they would. I just don't think the process or the RAM tech is ready.
Will they? They will have to deliver steady sales numbers to their stockholders on 14/16nm for 4+ years probably. They have plenty of motivation to hold back and not make it more difficult to themselves to come up with products in 2017/18/19/20.
ShintaiDK
Lifer
- Apr 22, 2012
- 20,378
- 146
- 106
AMD won't hold back with Vega, it'll be the last big GPU they make. Whether that's 450mm2 or 550mm2 is another question though.
I've always believed with leading with your trump card and draw out whatever the comp has. I know there are other strategies. Leading with your strength always worked for me though.
ShintaiDK
Lifer
- Apr 22, 2012
- 20,378
- 146
- 106
If they didn't hold back it would be 600mm2. And Polaris certainly wouldn't be the size it is.
Holding back and being sensible are two different things. 600mm2 is effectively unmanufacturable right now for the desktop and will remain so for a couple of years yet at least.
Assuming 232mm2, Polaris is the size it is for some unknown reason, but i doubt it's due to holding back. I don't think AMD can afford to hold back.
AMD does not have the HPC marketshare to shove 600mm2 14nm FF dies for $10K or more to offset the yield issues of a new node.
I don't expect Vega 10 to be bigger than 500mm2.
If we compare to Polaris, the next logical jump will be Vega 11 as Polaris x1.5 and Vega 10 will be Polaris x2.
I don't expect Vega 10 to be bigger than 500mm2.
If we compare to Polaris, the next logical jump will be Vega 11 as Polaris x1.5 and Vega 10 will be Polaris x2.
Nvidia ceases deliveries of Geforce GTX 970, GTX 980 and GTX 980 Ti VGAs
www.hwbattle.com/bbs/board.php?bo_table=news&wr_id=18732
www.hwbattle.com/bbs/board.php?bo_table=news&wr_id=18732
Last edited:
Krteq
Golden Member
- May 22, 2015
- 1,009
- 729
- 136
- Jun 10, 2004
- 14,600
- 6,084
- 136
You are thinking based on past tech. Interposers allow exceeding the previous 600mm^2 single die limit. Navi being scalable hints at this being the next step.
Stuka87
Diamond Member
- Dec 10, 2010
- 6,240
- 2,559
- 136
RussianSensation
Elite Member
- Sep 5, 2003
- 19,458
- 765
- 126
Why?
- SLI/CF support requires even more developer work and driver work from AMD/NV. Seems in the last 6 months a lot more AAA games have poor support
- NV isn't going to stop making large die flagships
- Mid-range die strategy failed with HD4000-6000 series
- Since 14nm node is here to stay for a long time, we should expect die sizes to grow as the node matures (7970->290X -> Fury X)
- As the node matures, it will become more affordable to produce 550-600mm2 die
All of these reasons suggest AMD should go through a similar transition of increasing die sizes (as outlined above from Tahiti to Hawaii to Fiji). Then with 7/10nm node, it starts all over again.
Even if Vega is only a 450-480mm2 die, it doesn't mean it will be AMD's last large die flagship, unless you have inside sources?
Unless you believe AMD's new strategy is to put a 450-500mm2 die against NV's 300mm2 GP104, how do you expect them to match or beat a 600+mm2 Big Pascal? Don't tell me dual-chip flagship Vega card priced at $699? Radeon Pro Duo is $1499, R9 295X2 before it was $1499. So this strategy will not work.
----
970/980/980Ti going EOL was reported a long time ago. Not sure why this is suddenly news again. Also, cards launching for June is a consistent theme. Again, not new information.
Even if NV low-balls early GP104 with a crappy 40% increase in perf/watt, 1080 will already beat 980Ti. I think after-market 1080 GDDR5X will deliver 50-80% performance increase over 980 at 1440/4K. The time is running out to sell those 980Ti cards for close to market prices....
They'll go multi die acting as one larger gpu. Not impossible if designed properly from the getgo.
It seems you missed out on all the talk of multi-dies on an interposer communicating as one-large die. To software/drivers, it will be indistinguishable because the inter-chip communication will appear the same as intra-chip communications from various blocks, CU groups and such..
Remember that GPUs are already quite modular, there's no reason they cannot exists apart but linked via an interposer.
Raja stated this twice in public, one in PCPER's interview with him, he talks about the future where node yields will be much lower than today and one way to solve that is exactly with multi-smaller dies on the same chip. The second time, some tech journalist asked him on twitter whether interposers can be used to connect multiple GPUs together into one and his response was a solid "Yes".
This will be the norm at 10 and 7nm.
Btw, the usual folks who are anti-AMD having been very dismissive of this approach as usual, saying it can't be done etc... so it will be true.
yeah in an interview with pcper Raja Koduri said the economics of the smaller die is much better at the bleeding edge nodes. My guess is AMD saw the writing on the wall with yields and design complexity at 10nm/7nm. Navi highlighted scalability as a key feature. My guess is Navi might be the first to go that route of forming large GPUs from a collection of dies on a very large interposer.
http://www.anandtech.com/show/10145/amd-unveils-gpu-architecture-roadmap-after-polaris-comes-vega
It seems you missed out on all the talk of multi-dies on an interposer communicating as one-large die. To software/drivers, it will be indistinguishable because the inter-chip communication will appear the same as intra-chip communications from various blocks, CU groups and such..
Remember that GPUs are already quite modular, there's no reason they cannot exists apart but linked via an interposer.
Raja stated this twice in public, one in PCPER's interview with him, he talks about the future where node yields will be much lower than today and one way to solve that is exactly with multi-smaller dies on the same chip. The second time, some tech journalist asked him on twitter whether interposers can be used to connect multiple GPUs together into one and his response was a solid "Yes".
This will be the norm at 10 and 7nm.
Btw, the usual folks who are anti-AMD having been very dismissive of this approach as usual, saying it can't be done etc... so it will be true.
yeah I am sure we will see AMD adopt it earlier and Nvidia will also eventually after realizing the reality of manufacturing complexity and yields at 10nm/7nm/5nm.
Last edited:
It's way too early for a multi-die on interposer acting as one but obviously that will be the end goal. What happens in-between could be more interesting.
It seems you missed out on all the talk of multi-dies on an interposer communicating as one-large die. To software/drivers, it will be indistinguishable because the inter-chip communication will appear the same as intra-chip communications from various blocks, CU groups and such..
Remember that GPUs are already quite modular, there's no reason they cannot exists apart but linked via an interposer.
Raja stated this twice in public, one in PCPER's interview with him, he talks about the future where node yields will be much lower than today and one way to solve that is exactly with multi-smaller dies on the same chip. The second time, some tech journalist asked him on twitter whether interposers can be used to connect multiple GPUs together into one and his response was a solid "Yes".
This will be the norm at 10 and 7nm.
Btw, the usual folks who are anti-AMD having been very dismissive of this approach as usual, saying it can't be done etc... so it will be true.
I'm quite skeptical about the feasibility of this myself. Someday? Sure. Foreseeable future? I don't think so.
Why? You can get 4096bit HBM2 to communicate with a GPU.
All that's important are lanes of communications.
When you look at GPU uarchs, they have blocks communicating with cache & each other. Within each block, there's TMUs, ROPS, and cores. Thus they often cut a few blocks out to make a lower end SKU.
A multi-chip approach wouldn't work without the interposer giving it the lanes of communications that it needs.
We'll see it with Navi, late 2018 or whenever 10/7nm comes around early on.
Why? You can get 4096bit HBM2 to communicate with a GPU.
All that's important are lanes of communications.
When you look at GPU uarchs, they have blocks communicating with cache & each other. Within each block, there's TMUs, ROPS, and cores. Thus they often cut a few blocks out to make a lower end SKU.
A multi-chip approach wouldn't work without the interposer giving it the lanes of communications that it needs.
We'll see it with Navi, late 2018 or whenever 10/7nm comes around early on.
I know.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 22K
-
News NVIDIA and Intel to Develop AI Infrastructure and Personal Computing Products- Started by poke01
- Replies: 384
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
Facebook
Twitter