We are all excited for what AMD is going to launch in the gaming PC market in a short time. Intel put Lisa Su’s men on the ropes and in one move in a masterful move that hits all the sticks, but the red team has at least one weapon in the chamber. The first real performance data on expectations Ryzen 7 5800X3D were not disclosed, but those of its superior counterpart for EPYC 7V73X servers. How much power do processors gain with V-Cache?
The preview of the new EPYC Milan-X arrived last week and we saw how its larger L3 cache size helped reduce overall system latency by having more capacity and thus exchanging less data with the RAM. The improvements in this section were slight, although surely appreciated in environments such as games, but today we already have comparative general performance data.
The increased power of AMD processors with V-Cache
The comparison pits the current EPYC 7763 with 64 cores and 128 threads against 2.45GHz of Base Clock and 3.5 GHz of Boost (256 MB of L3) with the new EPYC 7V73X with the same cores, threads and Boost, but with a base frequency of 2.2GHz and 768 MB of L3 (+512 MB). With that data in mind and being a more or less fair comparison (they differ in the base clock), how did they end up being equal?
Five tests were put to the test with the two processors to show different scenarios and thus be able to have a first contact with what the Ryzen 7 5800X3D will be on PC. What is shown reveals differences from the 7.6% in Gem5 in favor of processor with V-Cache, Y-Chuncher only one 1.5% moreover, while in OpenSSL it loses minimally.
In summary, if we assume that the differences in base frequencies are of the order of 5% and in many tests like Gem5 it is important, we could say that we have a GAP from 1% to 12.5% improvement, which isn’t bad at all, but what about desktop?
How much would a Ryzen earn in games with its V-Cache?
Can a Ryzen processor with 2 CCDs benefit from 2 x 32 MB of L3 cache?
▶️ R9 5900X, fixed 4GHz, 6 cores, 1 CCD 6-0 vs 2 CCD 3-0
SMT Enabled, DDR4-3733 C14, Win 11, Overclocked RX 6800 XT, Adrenalin 22.1.2
🧮 1 CCD 6-0 >= CCD 3-0 (except FC 6) pic.twitter.com/VzCunCplXC
— CapFrameX (@CapFrameX) January 23, 2022
The data is now theoretical, since although a Ryzen 9 5900X is used, we try to simulate the increase in L3 by modifying and balancing the CCD options. What has been done is to test the performance of a CCD with 6 active cores (2 CCX x 3 cores) and disable the neighboring CCD, while on the other hand the two CCDs with 3 cores enabled each are enabled, thus doubling the total accessible cache from 32 MB to 64 MB, since the Zen 3 topology allows this cache to be shared with all cores.
Why is this done? Well, for one simple reason. AMD is not planning to launch a supposed Ryzen 9 5900X3D or 5950X3D in principle, as it would have to place two cache stacks for each CCD and so far we don’t know how this affects performance. Well, with this demo, that effect is simulated with the simple act of selectively disabling the cores.
What are the performances obtained? Well, the expected. Doubling the amount of cache with 3D V-Cache would not have a positive effect on gaming performance for the simple fact that there is an impact on latencies between CCDs and CCXs. It can be seen perfectly like the CCD with the 6-0 configuration is clearly superior at the same frequencies. Is this the reason why AMD will only launch the Ryzen 7 5800X3D knowing that in a CCD it influences a game improvement? It could be, but it would have been masked by a supposedly higher price per CPU that would push users back in favor of Intel.
