| To: Vattila who wrote (42279) | 7/30/2021 11:11:53 AM | | From: Joe NYC | | | | Intel DG2 graphics card now delayed until CES 2022 launch.
wccftech.com
It's kind of a mystery that Intel can't make a graphics card that can run Grand Theft Auto, yet is going to make a graphics card for and exascale supercomputer... |
| | AMD, ARMH, INTC, NVDA | Stock Discussion ForumsShare | KeepReplyMark as Last ReadRead Replies (3) |
|
| To: Joe NYC who wrote (42280) | 7/30/2021 12:55:28 PM | | From: neolib | | | | Raja's doing his best...
It's kind of a mystery that Intel can't make a graphics card that can run Grand Theft Auto |
| | AMD, ARMH, INTC, NVDA | Stock Discussion ForumsShare | KeepReplyMark as Last Read |
|
| To: Joe NYC who wrote (42280) | 7/30/2021 12:56:26 PM | | From: Vattila | | | | Intel is stepping into a highly competitive arena shaped by the long and intense competition between AMD and Nvidia. No wonder they have a hard time pushing out frames at high rates and reasonable power. I guess they have more design leeway in the HPC space.
By the way, one of the discussion points about RX 6600 XT, is that the underlying Navi 23 chip has only 32 MiB of Infinity Cache (cut down from 96 MiB on the larger Navi 22 chip in the RX 6700 XT), which is part of the reason the 6600 XT product is aimed at 1080p gaming only, AnandTech claims. If there is such large effect on performance from the amount of Infinity Cache, you would think GPUs are also ripe for a V-Cache solution. It will be interesting to see how GPU design evolves in the 3D packaging era.
 |
| | AMD, ARMH, INTC, NVDA | Stock Discussion ForumsShare | KeepReplyMark as Last ReadRead Replies (1) |
|
| To: Vattila who wrote (42282) | 7/30/2021 2:52:27 PM | | From: Joe NYC | | | | Yes!
There was some discussion on Twitter, and I also posted a patent that shows a long and narrow silicon bridge connecting / stacked on top of multiple graphics dies, loaded with L3 cache.
Which would bring together both of the AMD architectural home runs - partitioning large chips to chiplets and stacking.
One of the reason why partitioning never quite worked right in the past was because all the units need to work on the same image, and need super fast communication abilities between them. The active silicon bridge accomplishes it.
So the L3 would be stacked and 256 mb and 512 mb were mentioned as possibilities for RDNA3.
We don;t know how much of this will turn into reality. But there is a real chance that AMD will go as far ahead of NVidia with RDNA3 as it did ahead of Intel with Rome / Milan. |
| | AMD, ARMH, INTC, NVDA | Stock Discussion ForumsShare | KeepReplyMark as Last Read |
|
| From: Joe NYC | 7/30/2021 3:14:02 PM | | | | | | No graphics cards from Intel, but they have the merch ready:
 |
| | AMD, ARMH, INTC, NVDA | Stock Discussion ForumsShare | KeepReplyMark as Last Read |
|
| |
