Those numbers are identical to the A770, and don't match the numbers from the preview[0], so I think that's a copy paste error.
If we use the numbers from the preview:
| |Arc A770|Arc B580|RTX 4060|
|--------|--------|--------|--------|
|Process |N6 |N5 |N5 |
|Die Size|406mm^2 |272mm^2 |159mm^2 |
|Trans. |21.7B |19.6B |18.9B |
|Mem Bus |256 bit |192 bit |128 bit |
|TDP |225W |190W |115W |
|~Perf |90% |110% |100% |
In terms of performance per die area it's a big improvement over A770 but still far behind Nvidia. It's interesting that the transistor density is so much lower than the 4060 despite having the same (or at least similar) process node. Speculating about why that may be:
- Nvidia has better layout.
- Intel is using higher performance (larger) transistor libraries or layout in order to hit the higher boost frequencies (2800 vs 2460).
- Wider bus interface takes up more space.
- The B580 may have 1 render slice and 64-bits of memory bus disabled, and they're not including those transistors in the count, but they still take up area.
In the Hardware Unboxed Podcast, Intel's Tom Peterson gave some explanation (@ 24:00) for the difference in transistor density:
1. There are dummy transistors included in the design for various manufacturing reasons, and there's no standard as to whether those are included in the numbers.
2. SRAM cells in particular are highly optimized and the amount of SRAM (e.g. cache memory) on the chip versus logic will have a big influence on the transistor density.
He said he's still asking around the engineering team for a more complete explanation but that's the supposition.
Those numbers suggest that they have caught Nvidia in performance per transistor. As for the die area being larger, I suspect that the larger memory bus might be partly responsible. The transistors used for IO stopped shrinking on new nodes a while ago, so they use plenty of die area.
If we use the numbers from the preview:
In terms of performance per die area it's a big improvement over A770 but still far behind Nvidia. It's interesting that the transistor density is so much lower than the 4060 despite having the same (or at least similar) process node. Speculating about why that may be:- Nvidia has better layout. - Intel is using higher performance (larger) transistor libraries or layout in order to hit the higher boost frequencies (2800 vs 2460). - Wider bus interface takes up more space. - The B580 may have 1 render slice and 64-bits of memory bus disabled, and they're not including those transistors in the count, but they still take up area.
[0]: https://www.techpowerup.com/review/intel-arc-b580-battlemage...