Vega dieshot out, though it's stylized, you can clearly see the individual CUs, memory controllers and the frontend here.
Most interesting thing about it is the HBM2 memory, which seems to Samsung and Hynix.
Vega dieshot out, though it's stylized, you can clearly see the individual CUs, memory controllers and the frontend here.
Most interesting thing about it is the HBM2 memory, which seems to Samsung and Hynix.
Other urls found in this thread:
videocardz.com
youtube.com
twitter.com
Hawt.
>which seems to be Samsung and NOT Hynix.
Fixed
Dang, that HBM is about 1/3rd the whole freaking die, hope it's worth it.
>2 stacks
>not 4
DOA. What a joke.
The block between the two memory controllers is likely the command processor, below that is likely the entire frontend, which might or might not include the ACE, which if I'm reading this correctly are now part of the CUs.
The interesting thing is the memory PHY like thing on the far right, which could be a IF link.
Fiji for comparison.
And the same amount of memory as a 4 stack GV100
Really makes you think.
Really weird Nvidia didn't wait for 8GB stacks for a product that clearly needs all the memory it can get.
How strange.
The new Mac is using a cutdown Vega with 56CU instead of 64