Navi Dicussion Thread

[SteelCrysis]

https://www.tomshardware.com/news/amd-gp...let-patent

AMD jumped over to using a chiplet-based CPU design with the introduction of Zen 2 in the Ryzen 3000 series CPUs, enabling the chipmaker to cram more cores into a single CPU. Now, a new patent appears to reveal that AMD wants to do the same thing with GPUs (via ComputerBase).
...
As great as it all sounds, we don't want to get your hopes up for such a product to come out anytime soon. SLI and Crossfire died because getting multiple GPUs to work together across different cards is a pain, and even with AMD's proposed solution of bringing the GPU chiplets closer together with a high-bandwidth interconnect, there is still a lot of work to be done.

Chances are that if this manifests into a real-world product, it will first happen on a research-level scale, being aimed at supercomputers or scientific-purpose GPUs for users with high GPU power needs in single workstations. Such a ludicrous amount of GPU horsepower will likely need to be coupled to HBM memory just to keep up, so it's likely that you can rule out consumer products for some time to come.

All that being said, it's also very possible that this will never become a thing. Tech companies file a lot of patents, and most of them never end up being used.

https://www.techpowerup.com/276746/amd-p...adeon-gpus

AMD reports that the use of multiple GPU configuration is inefficient due to limited software support, so that is the reason why GPUs were kept monolithic for years. However, it seems like the company has found a way to go past the limitations and implement a sufficient solution. AMD believes that by using its new high bandwidth passive crosslinks, it can achieve ideal chiplet-to-chiplet communication, where each GPU in the chiplet array would be coupled to the first GPU in the array. All the communication would go through an active interposer which would contain many layers of wires that are high bandwidth passive crosslinks. The company envisions that the first GPU in the array would communicably be coupled to the CPU, meaning that it will have to use the CPU possibly as a communication bridge for the GPU arrays. Such a thing would have big latency hit so it is questionable what it means really.

The patent also suggests that each GPU chiplet uses its own Last Level Cache (LLC), instead of separate LLCs for each GPU, so each of the LLCs is communicably coupled and the cache remains coherent across all chiplets. Rumors suggest that we are going to see the first chiplet-based architecture from AMD as successor to the RDNA3 generation, so it will happen in the coming years. AMD already has experience with chiplets from its processors, with Ryzen processors being the prime example. We just need to wait and see how it will look once it arrives for GPUs.

https://www.tomshardware.com/news/amd-us...shirapoint

AMD might have brought its RX 6800, RX 6800 XT, and RX 6900 XT graphics cards to market by now, but that doesn't mean the chef is done cooking up new recipes. In that light, three new entries (1, 2, 3) have surfaced in the USB-IF, as spotted by Komachi, pointing to a handful of new GPUs. There isn't much information to go on right now, so take the news with a pinch of salt until more details emerge.
...
This would leave the XTXH and XLE GPUs in the open, assuming these entries are for the Navi 21 family. If they're for the Navi 22 family, then it's very possibly a pointer to the RX 6700 and RX 6700 XT, among others.

The same principle applies to the Nashira Point codename -- at this time we simply don't have enough details to say which product (family) it refers to.

As a result, all we can do with this information is tell you that new products are on the horizon. The RX 6700 XT's BIOS has already been leaked late last year, and it pointed to some serious overclocking potential. With AMD's historic tendency to announce products at CES, it wouldn't come as a big surprise to see some official news next week.

https://www.techpowerup.com/276747/amd-r...d-incoming

Currently, it is unknown what the additional "H" means. It could indicate an upgraded version with more CUs, or perhaps a bit cut down configuration. It is unclear where such a GPU would fit in the lineup or is it just an engineering sample that is never making it to the market. It could represent a potential response from AMD to NVIDIA's upcoming GeForce RTX 3080 Ti graphics card, however, that is just speculation. Other options suggest that such a GPU would be a part of mainstream notebook lineup, just like Renoir comes in the "H" variant. We have to wait and see what AMD does to find out more.