If the arch changes are significant, and they try something new- it could easily be much worse than 10% IMO. It’s not like there isn’t precedent for large frequency drops on new Intel products.
If the arch changes are significant, and they try something new- it could easily be much worse than 10% IMO. It’s not like there isn’t precedent for large frequency drops on new Intel products.
Intel’s design teams are worse than AMD’s. That’s not just me saying it, Pat all but said it himself in a recent interview as well. They lost leadership in many key areas, not just what node they use.
Very cool
15th Gen Arrowlake, what it is so far
Bad lol
If the rumors that the -C parts are being fabbed on TSMC 3nm are true, then the idea that there will be increased margins for the cloud server variants vs Turin might not end up being true at all.
Also I’d like to know whether AMD’s approach to hybrid architectures (zen4 + zen4c) is a better or worse approach than having multiple core architectures with differing IPC and ISA capabilities.
Different trade offs, no clear answer.
And given that raptor lake consumes more power at peak power consumption than zen4, is it even worth it to have P+E cores?
Yes.
Would intel have attained higher multithreaded performance if it offered 10 or 12 p cores + Avx512 rather than using gracemont?
No.
Like what’s the point of gracemont if overall raptor lake power consumption is much higher vs older Intel CPUs ??
It’s not, at least iso perf.
overpriced
Ye, flagship CPUs are always over priced
marketing chip
How? Intel is the one with the mindshare with non-tech inclined people. The amount of mindshare they have for their poor record over the past couple years in client is really impressive.
which cripples itself under gaming
3D cache with Zen 4 is just way more impressive than the bloated, power hungry RPC core. And more performant to boot.
while running hotter at a lower power draw…
No one cares about that lmao
Also sorry you had to get a 3435x. That mesh must suck lol
better binning resulting in slightly higher frequency stock
Nah, bullying the 14900ks is valid
Kinda.
Though I will add, what Intel calls “10nm” is equivalent to TSMC’s “7nm”.
So yes, if they fail to progress past Intel 7, they will be screwed in the future. Overcoming an entire nodes difference in PPA is extremely hard, and with Intel having worse design teams than its competitors, it becomes an esentially impossible task.
Intel does have Intel 4, and they claim they are shipping it to OEMs already, with MTL (the Intel 4 product) launching in Dec. However it appears it’s mobile only, and we have no idea about the volume. If it’s just a paper launch, and yields/volumes take years to fix, (like Intel’s 10nm), then yes, they will be hurt badly.
Idk if I would go as far to say they would be screwed, but the company they would be in this scenario would be unrecognizable compared to the current Intel.
ARL doesn’t use Intel 3, Intel 3 and Intel 4 are not at all related to ARL