cleeve
Illustrious
cryoburner :
What exactly do some of these software optimizations for Ryzen entail? The performance optimizations for Ashes of the Singularity, for example, were pretty significant, and at least in Tom's benchmarks brought performance of the Ryzen 1800x slightly above that of the 7700k in their tests. Certainly part of it is the game's heavy use of multithreading at play, as evidenced by the 6900k's performance over the 7700k there, but that particular round of optimizations improved performance on Ryzen without doing the same for Intel's chips to a simlar degree, and in a way that didn't seem to simply show a preference for more cores.
I've heard that communication between cores on the same CCX might be significantly faster than it is across different CCXes, so is this partly a matter of keeping threads that communicate often on the same CCX, to avoid that increased latency? If it is, could that also be something that affects communications with graphics driver threads, and might there be ways for graphics driver developers, such as the Radeon team, to make sure their driver is operating on the same CCX as the thread that's communicating with it the most? It seems like something like that could potentially improve performance even for games that don't get specifically optimized for Ryzen.
Its all over the map, there's no silver bullet, even though that's what people want to hear. The CCX latency is there, but it's not that bad and it's not responsible for the outliers.
I'll give you an example of the kids of things that are holding Ryzen back: a developer found that their game code automatically assumed that AMD CPUs had all-physical cores, because we didn't have SMT before now. Once the game was guided to behave as it does on Intel hyperthreaded CPUs, we saw a notable boost in performance.
it sounds simple, but this is what happens when a new architecture is introduced. It sounds trivial, once you know what's happening it can be easy to attack, but finding it takes work.