- Mar 3, 2017
- 1,684
- 6,224
- 136
Considering it has 33% more CUs than Phoenix, other improvements and is probably incredibly mem-bandwidth-bound anyway, that seems like a decent clockspeed to me.
That won't necessarily do much for cases like games, where L2/L3 size is seemingly more important, considering how much V-Cache improves perf/clk.L1/L2 b/w is actually wayyy up so there's that.
thats probably because of full width AVX512 and only in that use caseView attachment 100280
Base clocks are down 200-700 MHz for 9xxx vs 7xxx. Biggest loss is for the 9600X and 9700X. nT performance might be the same as Zen 4 for these SKUs (or at least workload dependent)....
I really, really don't think it's good. It's not terrible, but again, it's not good either.AMD just continues to produce solid across-the-board gains with great efficiency.
While I have a 14900K (underclocked), the reality is that I suspect an efficient cool running 9950X will put up 45,000 CB R23 points. A 14900KS has to be volted and clocked to within an inch of it's quickly degrading life to approach such a score and needs a custom loop and chiller as well. Yes I know Zen 5 will compete against ARL but Zen 5 will stand alone for quite some time I think even if Intel manages to push ARL out the door in quantity by the end of the year.
In addition, the enormous rumored IPC gains of Skymont are going to have to be a reality for ARL to compete with Zen 5 considering the former's lack of HT.
Looks like AMD just "called" on Intel. We'll see if Intel is bluffing or not soon enough.
+5% IPC for Zen 5 would have been a stumble. +10% would have been "meh." +16% is good in my book. 20% or more would have been great.
PPA-wise it's fine, the CCD is barely bigger.I really, really don't think it's good. It's not terrible, but again, it's not good either.
View attachment 100280
Base clocks are down 200-700 MHz for 9xxx vs 7xxx. Biggest loss is for the 9600X and 9700X. nT performance might be the same as Zen 4 for these SKUs (or at least workload dependent)....
Skymont is the only core that exceeded expectations lolArrow Lake: Flop
Zen5: Flop
Apple M4: Flop
Cortex-X5: ? (Flop, obviously, we know how this story ends)
Compared to the dingus MTL E-c cluster yeah.Skymont is the only core that exceeded expectations lol
Well how close was IUnhappy with Cinebench speculation?
Then let me show you some happy data?
9950X
GeekBench 6
single 3628 more than 23868
(Only test data, does not represent the final result)
if those SKM IPC slides (emphasis on IPC slides, not perf/power slides) were in comparison to crestmont, and not crestmont LP, then that would put SKM at ~GLC levels of IPC.Compared to the dingus MTL E-c cluster yeah.
If we go by year. Its each 8% IPC each year since Zen 4. Thats not impressive but not bad. its average. ARM brought bigger improvements YoY.+5% IPC for Zen 5 would have been a stumble. +10% would have been "meh." +16% is good in my book. 20% or more would have been great.
I don’t lol. I hope he’s right. This is hilarious and AMD is just not it. I’m tired of getting slop mobile parts from big team blue/red when even MediaTek could build us better PC parts for anything that isn’t a desktop chip or Strix Halo.I hope you're wrong 3 times in a row.
Excluding L3, which uses more transistors though Zen 5 or Skymont?Zen 5 is gonna have ~16% higher IPC, so like 1 generations worth of IPC, than an E-core lol
Iirc they're comparing compute in the LP cluster.were in comparison to crestmont, and not crestmont LP,
Well... not everyone was expecting it lolIf we go by year. Its each 8% IPC each year since Zen 4. Thats not impressive but not bad. its average. ARM brought bigger improvements YoY.
Lets not sugar coat it. People expected this to be the EVERYTHING killer, including myself. Next time I will be cautious
That would be so cheap by Intel.... smhIirc they're comparing compute in the LP cluster.
No it's right, LNC has no LLC for the Atoms.That would be so cheap by Intel.... smh
Yep.No it's right, LNC has no LLC for the Atoms.
Just 8 megs of SLC and there you go into the far beyond.