Tuesday, June 26th 2018

NVIDIA GV102 Prototype Board With GDDR6 Spotted, Up to 525 W Power Delivery. GTX 1180 Ti?

Reddit user 'dustinbrooks' has posted a photo of a prototype graphics card design that is clearly made by NVIDIA and "tested by a buddy of his that works for a company that tests NVIDIA boards". Dustin asked the community what he was looking at, which of course got tech enthusiasts interested.

The card is clearly made by NVIDIA as indicated by the markings near the PCI-Express x16 slot connector. What's also visible is three PCI-Express 8-pin power inputs and a huge VRM setup with four fans. Unfortunately the GPU in the center of the board is missing, but it should be GV102, the successor to GP102, since GDDR6 support is needed. The twelve GDDR6 memory chips located around the GPU's solder balls are marked as D9WCW, which decodes to MT61K256M32JE-14:A. These chips are Micron-made 8 Gbit GDDR6, specified for 14 Gb/s data rate, operating at 1.35 V. With twelve chips, this board has a 384-bit memory bus and 12 GB VRAM. The memory bandwidth at 14 Gbps data rate is a staggering 672 GB/s, which conclusively beats the 484 GB/s that Vega 64 and GTX 1080 Ti offer.
Looking to the top edge of the PCB we see a connector similar to NVIDIA's NVLink connector, but it's missing half of its pins, which means daisy chaining more than two cards won't be possible. Maybe NVIDIA plans to segment NVLink to "up to two" and "more than two", with the latter of course being much more pricey, similar to how server processors are segmented by their multi-processor support. It could also be a new kind of SLI connector, which I'm not sure about, as GPU vendors want to get rid of this multi-GPU approach.

My take on this whole board, mostly due to the overkill power supply (up to 525 W) and the amount of test points and jumpers is that this board is used to test and qualify performance and power consumption in an unconstrained way, so that engineers and marketing can later decide on acceptable power and performance targets for release. The NVLink connector and functionality can also be tested at this stage, and the final PCB for mass production will be designed based on the outcome of these tests. On the bottom left of the PCB we find a mini-DP connector, which should be perfectly sufficient for this kind of testing, but not for a retail board.

Near the far right of the photo, rotated by 90 degrees, we see some mechanical drawings that to me, look like a new retention plate for the cooler. You can clearly see some space inside, which seems to be for the graphics processor itself. Around that are some mounting holes, which look like they are for a cooling solution.

Update:
I tried to estimate die size from the photo. We do know from the datasheet that the GDDR6 memory chips are 14 mm x 12 mm. Based on that information I rescaled, warped and straightened the image, so that each GDDR6 memory chip is 140 x 120 pixels. With all memory chips around the GPU now being at the correct size, we can use the GPU's silkscreen print to estimate the actual size of the chip package, which I measured at 48.5 x 48.5 mm. Assuming that the inner silk screen with the solder balls represents the surface of the GPU die, we get a length of 26 mm for each side of the die, which brings die size to 676 mm². This makes it a relatively large die considering NVIDIA's existing lineup: GV100 (815 mm², Titan V), GP100 (610 mm², Quadro GP100), GP102 (471 mm², GTX 1080 Ti), GP104 (314 mm², GTX 1080), GP106 (200 mm², GTX 1060). So my initial assessment that this could be the GP102 successor seems accurate, especially since GV100 die size is quite a bit bigger than GP100 die size, by roughly 33%. Our calculated GV102 die size is 40% bigger than GP102, which falls in that range.
Source: Reddit
Add your own comment

71 Comments on NVIDIA GV102 Prototype Board With GDDR6 Spotted, Up to 525 W Power Delivery. GTX 1180 Ti?

#1
TheLostSwede
The lack of display connectors seems to suggest that this is a prototyping solution for a compute card, rather than a graphics card though.
Posted on Reply
#2
First Strike
150×3+75=525(W) Peak Power
Though, it is not a good estimation on test samples, because there is nothing to stop you from drawing power beyond specifications.
Posted on Reply
#3
Caring1
Would be interesting if they mounted an LGA socket on the board so only the chip needs to be changed to upgrade.
Posted on Reply
#4
FordGT90Concept
"I go fast!1!11!1!"
Me thinks it is an engineering card to test what will become an NVLink product.
Posted on Reply
#6
Upgrayedd
If they made a new kickass SLI connector that fixes the downfalls I would love to buy 2
Posted on Reply
#7
bug
TheLostSwede said:
The lack of display connectors seems to suggest that this is a prototyping solution for a compute card, rather than a graphics card though.
Or a board meant to test the GPU squarely, without looking at outputs just yet. Or a rejected/failed prototype.
What we do have, is confirmation this is using GDDR6. A lot of people were riled up by the previous TPU article saying so, because the only lead was that Nvidia has signed a contract for GDDR6. This proves the article was lucky with its assumptions.
Posted on Reply
#8
silentbogo
TheLostSwede said:
The lack of display connectors seems to suggest that this is a prototyping solution for a compute card, rather than a graphics card though.
Bottom left: one mini-DP connector. There's also a wider placement pad for DP or HDMI (instead of mini-DP), but incoming traces are completely covered w/ solder mask.
Posted on Reply
#9
W1zzard
Updated with die size estimation
Posted on Reply
#10
gdallsk
TheLostSwede said:
The lack of display connectors seems to suggest that this is a prototyping solution for a compute card, rather than a graphics card though.
There looks to be a mini DP at the bottom left of the card. This could be anything tbh.
Posted on Reply
#11
bug
W1zzard said:
Updated with die size estimation
Assuming the socket must fit the largest die in the family, 600-something sq mm isn't actually that big.
Posted on Reply
#12
ppn
So tensor cores are present.
7nm will shrink 676 to 300 sq. looking forward to that.
dropping 384 bit 14gbit 12GB in favor of 256 20gbit 16GB will make this card more afordable in the future.
Posted on Reply
#13
W1zzard
bug said:
Assuming the socket must fit the largest die in the family, 600-something sq mm isn't actually that big.
Why should we assume this board has to fit various dies?
Posted on Reply
#14
las
That's a big die

Posted on Reply
#15
bug
W1zzard said:
Why should we assume this board has to fit various dies?
I don't know. Are there different sockets for GP102, GP104 and GP106?
Posted on Reply
#16
W1zzard
bug said:
I don't know. Are there different sockets for GP102, GP104 and GP106?
Technically they are not sockets, but yes. Otherwise you'd see empty PCB space around cards with smaller GPUs. Also the package of the whole chip range would have to be the same size, to match up with the solder balls, which would just drive up cost for no reason.
Posted on Reply
#17
bug
W1zzard said:
Technically they are not sockets, but yes. Otherwise you'd see empty PCB space around cards with smaller GPUs. Also the package of the whole chip range would have to be the same size, to match up with the solder balls, which would just drive up cost for no reason.
Ok, then scratch that.
Posted on Reply
#18
agent_x007
I think this is prototype board to test GDDR6 memory.
How they clock on this layout, how hot they get when operating, and what can be optimised in case of tracers. That OP power delivery is to have worst case scenario (a lot of power plane with GDDR6 data tracers near them).
I can be 100% wrong here :)
Posted on Reply
#19
jabbadap
agent_x007 said:
I think this is prototype board to test GDDR6 memory.
How they clock on this layout, how hot they get when operating, and what can be optimised in case of tracers. That OP power delivery is to have worst case scenario (a lot of power plane with GDDR6 data tracers near them).
I can be 100% wrong here :)
No video outputs on that board, so yeah it's a test board. That board takes bga packaged gpus and they are clamped down without soldering. Don't know how pin compatible nvidia gpus are, put ain't i.e. gp106 and gp104 are direct drop in replacement to same socket?
Posted on Reply
#20
iO
Yeah it's a bring-up board for testing.
VRMs look really beefy but there are no heatsinks only direct airflow on bare mosfets.
I come up with a substrate size of 52x52mm so maybe they plan a GV102 with GDDR6.
Posted on Reply
#21
W1zzard
jabbadap said:
No video outputs on that board
Bottom left: mini-DP
Posted on Reply
#22
ppn
mini-dp with 12 pin? more like USB type-c
Posted on Reply
#23
bug
ppn said:
mini-dp with 12 pin? more like USB type-c
Can still speak Thunderbolt ;)
Posted on Reply
#24
gdallsk
ppn said:
mini-dp with 12 pin? more like USB type-c
Mini-DP has stacked pins, so you only see one of the rows on the outside.
Posted on Reply
#25
erixx
Clearly a G-Sync 2 card.


:laugh:
Posted on Reply
Add your own comment