This is a question everyone is curious about, how will the X3D SKU with 2 CCDs be priced compared to non-3D SKUs with 2 CCDs. After all, now you will get the situation that 7900X3D/7950X3D gets higher frequency CCD and CCD with larger cache, but there is a frequency limit, while 7900X/7950X not only has the advantage of larger cache, but also has no CCD disadvantage Frequency limited.

So what you’ll probably see is that the X3D models expand exponentially on workloads that really benefit from the added 64MB cache, but workloads that don’t benefit from more cache, but it’s frequency based, and it shows no improvement or Maybe even regression. After all, I’ve seen this with the 5800X3D compared to the 5800X.

Another unknown (until the reviews come) is how well the windows (or Linux) scheduler handles this. Until now, AMD only had SKUs where each CCD was equipped with a 3D V-Cache when it was in the SKU, so that each core always had the same amount of cache available.

Now for the first time we’ll see SKUs where this isn’t the case, so not every core has the same amount of cache and so the scheduler has to choose, do I run this process on the slower core with a large cache, or do I run the process On fast kernels with less cache. If AMD and Microsoft can’t do that, it may be the case that, just as you saw with Intel’s E and P, some workloads will show very strange performance results.