I've been thinking of when will the RDNA4 cards come out.
As MILD mentioned RDNA 4 will come out around Q3 2024. I don't think there will be RDNA 3.5 refresh cards or RDNA 3 refresh cards out next year.
I think RDNA4 will be very similar to RDNA3 apart from very small arch improvements and update to Raytracing core.
There were rumors in 2022 that AMD had issues with TSMC 3nm node and that they will be using 4nm. Current rumors don't say anything about which node will RDNA4 use but, seeing that TSMC N3E 3nm node is just being put in production and others will be using it, makes sense that AMD has to use the 4nm node in 2024. This could make AMD release the RDNA 4 before Nvidia does its new series. So, I'm thinking RDNA4 could come out end of May and be available in June 2024.
Further, I was looking at how big the RDNA 4 flagship chip will be in mm2 and what its performance could be. Taking the N31 which is based on 5nm and 6nm nodes and combined size of 530mm2. An RDNA4 best would be around 370mm to 450mm2 chip with 90-96 CUs like Rx7900 series, but with 256Bit bus, faster memory since it will use Gddr7 and rated TDP of below 280W. I came to this conclusion that 4nm TSMC node is a very small improvement in transistor density, of just 6% for the N4, (N4x or Nvidia specific 4N might be a bit more).
Looking at the 4nm node and doing the math is no wonder that AMD can't produce a high-end GPU next year because by my math comes out that a 20-30% more performat GPU then a RX7900xtx would have to be bigger then 680mm2 and have a TDP of 410W, that's what the 4nm node does.
But here are the all the good things, the GPU, let's say it's called Rx 8800 XT is out in middle of next year has 16gb of Vram for 600$, identical performance in raster compared to Rx 7900 XTX and somewhat better performance in raytracing.
There are two AMD patents on raytracing that I’ve read few months back. The first one, released 1 year prior to first RDNA 2 GPU, talks about raytracing core. But the second one, it was released in June this year. So, the latest AMD patent describes GPU withing its raytracing core, addition of a hardware specific traversal engine and specific BVH memory cache. Not to go into details, from what I understand of the two patents first patent describes raytracing core in RDNA 2&3 and the second patent describes similar but a much improved way of doing raytracing. I’m hopeful we will see this is RDNA4(the patent did arrive this june and next june we’ll have RDNA 4 card so it matches the schedule prior to RDNA2) (https://www.freepatentsonline.com/20230206543.pdf)
Upping the CUs from 96 to 128 (and the ROPs similarly) will increase the GCD size from ~305 to ~372 mm^(2), based on the die image (and leaving some blank space at the side), and the total to 596 mm^(2). Whether performance will increase performance enough depends on the RAM bottleneck.
It's also worth nothing that RDNA 3 apparently didn't reach the expected clocks, and if AMD managed to solve this problem it would be possible to get extra performance without much or any extra die space.
In general if you're just aiming to reduce chip size by removing two MCDs, that's not really that much of a cost saving. That won't make the chip a mid-range chip as rumoured.
In my opinion, on the RDNA3 not reaching expected core clocks, its mostly down to poor performance of TSMC 5nm node to what AMD wanted to achieve. There target might been RTX4090 flagship performance but on the 5nm plus 6nm, and TDP 355W resulted in high power consumption, and to keep power down, core clock has to suffer.
Rx 7900XTX Taichi model has roughly 250mhz more in boost and a roughly 50W more power consumption then the AMD model.(https://www.techpowerup.com/review/asrock-radeon-rx-7900-xtx-taichi/38.html)
thing is, people can't seem to get much clock boost that translates into performance out of the 7900 series. 3200mhz overlocking tests doesn't do much more than 10-15% from AIBs, and consumes 500w+
if you look at timespy results, going from 2800-3000mhz game clock to power unlimited 3500mhz (which consumes over 500W), you gain a few thousand points
time spy with a random "stockish" result and the 7900xtx overclock from here. most of the gains came from the 6.0GHz oveclock on the CPU, with graphics results only being about 12% faster with this overclock
unless i've missed something, it seems like the 7900 series is limited by more than power
I don't see GPUs having more then 3Ghz anytime soon as standard clocks. High core frequencies make a small chip have much better performance, so its cost efficient for AMD to make them but high core clocks are increasing power consumption. AMD is trading and evaluating size of the chip with its frequency to get the performance.
I have Taichi on water, 3.1 GHZ clock - what more you may need? 425W~ consumption. 2.95Ghz stable game clock...
All that is before i will update the BIOS to Aqua monster...550W+~ without OC..
They could make a much larger die, if its really just 300 mm².
They could, but N5/N4 is significantly more costly than N7/N6 (was estimated to be 1.7x more costly, IIRC). That's why AMD is trying to keep the size under control by using chiplets.