- Joined
- Oct 24, 2013
- Messages
- 1,253
- Motherboard
- Asus ROG Z690 Formula
- CPU
- i9-13900K
- Graphics
- RX 6900 XT
- Mac
- Mobile Phone
I think we’re going to have wait and see if these rumors for Meteor Lake are true. Some of the same sources misquoted RDNA3 performance. I don’t understand why Intel would have a regression in its P-core count. And speaking of rumors, those same source claim Intel will offer 8 P-cores on arrow lake.There is a lot of good engineering in these new chips. But, if Intel could increase the P-core ratio, wouldn’t they favor doing so?
In Meteor Lake, the ratio tilts weirdly to only 6 P-cores and 16 E-cores. That’s a ratio of 1 to 2.7 in favor of E-cores. Admittedly, the IPC is expected to increase significantly so there may still be a net gain, but doesn’t it give the impression that P-cores are on their way out? As Intel improves and improves their smaller and more energy-efficient E-cores, eventually there will not be any reason for P-cores.
And that might make sense. It seems there’s no headroom left for scaling P-cores.
Intel's Next-Gen Desktop Platform Intros Socket LGA1851, "Meteor Lake-S" to Feature 6P+16E Core Counts
Keeping up with the cadence of two generations of desktop processors per socket, Intel will turn the page of the current LGA1700, with the introduction of the new Socket LGA1851. The processor package will likely have the same dimensions as LGA1700, and the two sockets may share cooler...www.techpowerup.com
I think so long as the IPC of the E-cores are increasing in new generations then that’s a win for consumers. But I don’t know how the upcoming E-core architectures (crestmont, skymont, and darkmont) will improve in ipc.
Also, Intel will sell more P-cores in its Xeon products. But we all know Intel’s Xeons have been extravagantly delayed. Sapphire rapid’s volume ramp has been anything but rapid.