Because it's pretty good, maybe the best way to quickly bench the CPU overall.
Linus did not have any major complaints with v4. They are at v6 now.
https://www.realworldtech.com/forum/?threadid=159853&curpostid=159860
Its godlike good for benching your own CPU and catching any regressions.
7600 doesn't have 3D cache so i'm not entirely certain why the comparison. It's not about the quality of the silicon if the 3D cache is the limiting factor, won't matter how "S-Tier" the die is.
but time will tell, i'd be glad to be wrong in the expectation that it won't be much higher then 5Ghz if at all, but given the TDPs and etc. it only makes sense.
the vcache hurts thermals, thermals don't effect zen4 that much, cores happily sit pegged >5.5ghz at 95c. in terms of clock speed. 10c only yields about ~50 mhz +/- , so even a 40 c regression isn't going to result in much more than a ~200mhz drop on its own. so there has to be something else to it.
We can use the non-cache skus as a reference point for a "what if" for frequency where there are no thermal or power constraints (few cores using little power) and subtract from it based on a variable thermal penalty, the 7600 is a good candidate because it is the lowest bin of silicon currently used. therefore if we assume the worst, which is that the vcache die is low quality, the highest clock it should do is likely similar. if there is no other as of yet unknown factor, maximum observable boost would be 5.20-5.40ghz with a variable thermal penalty added based on the information I have at least.
we could probably do better by more accurately estimating the thermal penalty of the 5800x3d, and extrapolating, but I feel that that a maximum 40c penalty estimate is good enough for our purposes.
Anybody still gaming at 1080p needs to take a good hard look at themselves in a mirror, and conclude there is life beyond living in a basement, eating Cheetos, drinking Mountain Dew and playing Apex Legends.
I made an account just to point out those numbers.
Geekbench probably doesn't know how to distinguish its not 96x2.
That, or this 7950x3D was an experiment that had the 3D cache on both dies, which isnt what processor we are getting
208 is not correct either which way
I am leaning towards an early model that may have ending up in third party hardware developers hands or possibly Microsoft for scheduler development. I also considered the possibility for the benchmark not being able to accurately being able to to identify the the chip, or possibly completely false information generated by the use submitting the results. In any of the following situations the results do not provide us with an accurate representation of the end product.
idk why folks are wigging out over the 7950X3D. while it is a halo product the primary market for this CPU isn't (just) gaming, it's productivity and gaming. the only people who will be buying this are gamers who work hard (beyond playing games).
the ultimate gaming numbers should be less than the 7900X3D or the 7800X3D
idk why folks are wigging out over the 7950X3D. while it is a halo product the primary market for this CPU isn't (just) gaming, it's productivity and gaming. the only people who will be buying this are gamers who work hard (beyond playing games).
the ultimate gaming numbers should be less than the 7900X3D or the 7800X3D
And those games that dont really take advantage of the 3D Cache
its funny looking back at the 5900x3d amd demoed initially, the sample they showed only has 1 vcache die , didnt register at the time.
they've been planning to do this from the start it seems. potentially put some minds at ease that this isnt a last minute decision.
https://www.amd3d.com/wp-content/uploads/2021/06/AMD-VCACHE-1536x1092.jpg
I made an account just to point out those numbers.