It's very unlikely there'll be more than 40 - 44 CUs on next generation consoles even if the die size is around 400 mm2 and they take I/O out of it.
60 - 80 mm2 will be the Zen CPU which we know the size already. The ray-tracing cores occupy around 180 mm2 in the high-end Turing GPUs. Shrink it to 7 nm and you're talking about 100 - 110 mm2 at the very least. That leaves 210 - 240 mm2 for the GPU - smaller than the 250 mm2 Radeon 5700 chips and their 40 CUs.
Of course, you might opt to severely gimp the ray-tracing cores or take L3 cache out of the Zen CPU, but all of that will come with severe performance costs (such as effectively disabling any meaningful ray-tracing or reducing CPU IPC by 20-30%). Is the trade-off worth it? They've probably studied it far better than we can surmise. But even then, the die space saved wouldn't be enough for the 11 - 14 TFLOPS some people thoughtlessly dream of.
These Radeon cards spit blood above 1600 - 1700 MHz, and we've discussed elsewhere how power consumption influences everything in design, including up to consumer-class product rules and licencing, which are a very significant dealbreaker. Not to mention the costs...
(edit - corrected number of CUs)
That's a lot of random numbers you got in your post. First I believe both Xbox and PS5 will have 44CU's and if AMD made a 48CU's discrete Gpu I expect that to have a die size of 270-280mm2.
We know from Flute leaked that Sony very likely has 8mb GameCache instead of 32MB found in 3700x, this reduces the Zen2 from 70mm2 to 40mm2.
Your numbers for RT+tensors cores are completely off, the TU104 (geforce 2070) adds 35mm2 for a 445mm2 die size gpu. I'm assuming the RT stuff will add between 10-20mm2 to our discrete gpu so 280-300mm2 now.
We can look at PS4, PS4 pro and Xbox one X Soc's die sizes to make an estimate.
PS4, 212mm2 discrete gpu + 50mm2 cpu = 348mm2 (86mm2 wasted die space)
PS4 pro, 220-230mm2 gpu + 25mm2 cpu = 320mm2 (65-75mm2 wasted die space)
Xbox one X, 270-290mm2 gpu + 25mm2 cpu = 360mm2 (45-65mm2 wasted die space)
Xbox one X gpu we can't do perfect estimate, radeon 480 have a die size of 232mm2 and 36CU's, Xbox one X 44CU's but 4 disable, a 384-bit bus and fabbed at TSMC 16nm+ instead of 14nm Glofo/samsung. Tsmc 16nm+ is less dense and will add 8-9% die area.
Let's do next-gen Socs 40mm2 cpu + 280-290mm2 gpu (hardware RT included) = 370-380mm2, I'm assuming 50mm2 wasted die because of another shrink. I think both Sony/Microsoft will use tsmc 7nm+ this shrinks the SOC 20% but I don't expect perfect scaling.
370-380mm2 x 0.85 = 314.5-323mm2
Edit: Forgot you mentioned clock speed, my post is long enough already, I don't agree with you. I expect the clock speed on the gpu to be from 1790-1899mhz.
Last edited by Trumpstyle - on 29 October 2019
6x master league achiever in starcraft2
Beaten Sigrun on God of war mode
Beaten DOOM ultra-nightmare with NO endless ammo-rune, 2x super shotgun and no decoys on ps4 pro.
1-0 against Grubby in Wc3 frozen throne ladder!!