Re: There are some substantial improvements.
Eh, kinda sorta - remember, at the stock power profiles it's evening out or better than it's predecessor - which was running in a higher power class (7700X = 105W TDP as I recall vs 9700X = 65w TDP)
Phoronix benchmarks show solid improvements across the board at noticably lower power loads for real world tasks that the likes of us might do (Transcode, compile, etc) and DerBaur did a very quick and dirty PBO, and ended up using 114% of the power of a 7700X, but with 120% performance in Cinebench - which isn't really representative of anything, but is a good indicator of the efficiency/IPC gains from that fancy pants branch predictor and all that.
It's also similar to the gains Phoronix shows on the 9600X which is less power constrained by nature of having to feed fewer cores, and so isn't using 'more' power than the 7600X - less, in fact.
The PBO Doesn't make much difference in games or owt (mostly a couple of heavy threads, and a few idly doing NPC/physics/etc stuff rather than spreading the load out at full chat), but then I'm not too fussed about that. People who want that will likely be waiting for the 9800X3D I expect.
I do wonder if they'd have been better releasing this as a 9700 at 65w, and then having a separate 9700X at 105w with the leash loosened a bit.
Oh well, the 12 and 16 core chips should be less impacted by power restraints.
People have been pissing and moaning about the Intel chips chewing through huge amounts of power (And that's coming back to bite Intel, rather badly it seems) and praised the Zen4 parts for their efficiency when power reduced, so you'd think the reception for these chips would be a bit warmer, but I guess you can't please everyone....
Steven R
Biting the hand that feeds IT
