Here is the complete zeppelin die. Outside of the 2 CCXs, there are SerDes and mem controller, each of which is about the size of a Zen core. Assuming the failure is completely random and catastropic, the chances of getting the "dead" dies could be just as common as getting 6 functional cores...
My wild guess would be that some of the dies have broken memory/PCIe controller which could very well be considered catastrophic regardless of how many cores would otherwise be non-defective. There are bound to be those kind of dies would it?
If we pay attention to the cache size in that table, we will see that the 7920X has 16.5 MB cache. By taking the highest-clocked variant with 16.5MB cache, we only get the Xeon Gold 5118 with 2.3/3.2 GHz in the link below...
The $999 one is for the 1950x the top-end 16c/32t so this does not invalidate the BnC's claim yet. BnC was only speculating $850 for the bottom-tier 16c/32t, which could very well be the vanilla 1950 (analogous to the vanilla 1700).
At least, gluing 4 "mainstream" dies to form a server CPU is still better than gluing 1 mainstream die to a bigger socket and call it a HEDT.
Sorry, I couldn't resist posting this.
In this screenshot it's not 8 CCX of course, but rather 8 dies in total for this setup since this system is 2S Epyc (pay attention to "2 processors" in the system information). Now looking back to the Ryzen with only one die, that's equals to just 1 NUMA node.
Now the question is, would Windows...
The minimum may not be necessarily true though, but nonetheless very likely. Remember that the 16c EPYC still have twice the PCIE lanes and DRAM channels while still having the same core count. That said, the 16c TR is made of the full-fat ccx and will have higher clock speed. So, the pricing...
Here is the "source" of the Platinum 8180 score.
http://www.coolaler.com/threads/xeon-platinum-8180-56112.345288/
Which can be roughly translated, with my rusty knowledge in Chinese to:
Judging by the poster's tone, he seems to be disappointed with the performance and thermals he get...
That 8-core part... That one is likely be 1 core active per CCX. I wonder how much yield of those dies with that bad of a defects.
I expect that AMD will not make this too appealing price-wise to the mass, or they will have to cripple the better dies for it.
Hmmm, that could mean up to 25-30% advantage in the Multi-threaded performance for the 18c i9. More than what I once estimated.
Oh boi...
Btw, I am curious to see if those Xeons are soldered or TIM'd. That would also somewhat influcence how much clock speed those i9s would have without...
That's all depends on the clock speed wouldn't it? It's worth noting that 18c is "only" 12.5% more than 16c, multiply that with the IPC advantage of 10% for the total of about 24%. This 24% is the (hypothetical) maximum all-core clock speed advantage TR allowed to have over the SKL-X's 18c parts...
On the same clip, what is the 2442 cb score doing here with the 7900X on 2:27? Would that suggest this is the pure 7900X performance when the thermals has yet to kick in during the first run?
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.