- Mar 3, 2017
- 1,687
- 6,235
- 136
No need to be insulting. Go ahead and share the data.
No need to be insulting. Go ahead and share the data.
He makes a cryptic statement about "PHX2" compared to Raptor Lake U, by which I assume he means Z1. I find low power Ryzen 4 comparisons to 155h. He talks down to me. Lame way to prove a point. Adroc is good at causing commotion. Moving on, same old behavior.Good luck with that. The data part that is. Even if he did have it he wouldn't be able to share it for various reasons.
Literally any Phoenix2 against any 1355U works.He makes a cryptic statement about "PHX2" compared to Raptor Lake U, by which I assume he means Z1
I'm just telling you do compare b.L monodies with LITTLEs in the same cluster.He talks down to me
This isn’t my job and you’re not my condescending boss. Find the stuff and share, or you have nothing.Literally any Phoenix2 against any 1355U works.
It's not that hard!
I'm just telling you do compare b.L monodies with LITTLEs in the same cluster.
You made the claims, prove 'em.This isn’t my job
I’m goodYou made the claims, prove 'em.
Post the cinememe curves for Z4c and GLC and then apply the SKT data and there you go!
I am again asking the overlords about the 16% figure and the fabled 32%. I am willing to sacrifice a goat to the gods for the answer. Thank you.
Expecting anandtech do do any tests is...hopeful16% figure seems to be MT uplift and is what AMD shared as their geomean IPC claim.
32% was the SPECint 2017 1T IPC uplift claim made by adroc.
The gaming performance AMD shared puts Zen 5 about on par or a tiny bit ahead of the 7800X3D which represents around a 25% increase over the non 3d Zen 4.
Some slides have AMD showing faster 1T performance in GB6 than Snapdragon x elite but the footnotes indicate that might be an MT result so it is a bit up in the air.
Ultimately there seems to be enough wonky slides / footnotes and enough divergence between the presented geomean and the gaming results that we don't really know so it is just a case of waiting for actual benchmarks when it launches. Anandtech will do SPECint 2017 tests and then we have a definitive showing of IPC uplift and we can check it against the claim.
Expecting anandtech do do any tests is...hopeful
Agree with the thrust though, we need help to understand all of this. Still so many unknowns.
According to the slide Escher looks like hawk point refresh, just like barcelo was cezanne refesh.View attachment 100959
Strix Point : $1200+ laptops
Kraken Point : $800+ laptops
Escher : <$800 laptops??
Also if Kraken has 8CU RDNA3.5, then how many CUs does Escher have?
Exactly.+150MHz, to be exact. 7950X FMax is 5.85GHz.
There was a rumour that Zen 5 FMax is 6.1GHz. that could very well end up have been true given the clocks/voltage here, because that 5.7GHz at ~1.25v for relatively low end silicon implies that for the well binned stuff 5.7GHz at 1.2v may be possible. And in case you're wondering why 1.2v matters - that's V-Cache die territory.
Interesting that the AMD rep was asked if the 9950X would take the gaming crown but the rep answered with the 9700X being slower than the 7800X3D by a smaller amount than we would expect. FWIW, the 9700X officially boosts up to 5.5 GHz, while the 9950X goes to 5.7 GHz.AMD rep says that vanilla Zen 5 won’t top Zen 4x3d in gaming.
I asked Woligroski if the 9950X would take the crown of the fastest gaming chip on the market. "Is it the fastest in gaming? It's faster than the competition in our tests. X3D is still the king of the hill, but by a much smaller margin than typically between X3D and non-X3D," Woligroski responded. "So a 7800X3D would, yes, be faster than 9700X, but maybe not by as much as you would expect."
Asus is doing their very best to spoil the reviews in July. This appears to be an absolute apples/apples comparison with same chassis and power consumption.
View attachment 100965
Increasing cores count does the majority of the work here.Since they use the same process as Zen 4 i was expecting 27% better perf at isowatt, they are about 4% % above, exactly the perf improvement of N4X vs N4P at isowatt.
Otherwise Zen 5 would be almost 10% more efficent uarchitecturaly wise than Zen 4, wich would be surprising given that the fomer is quite bigger.
8 full Zen 4 cores running at ~4.5Ghz vs 4 full Zen 5 cores @ unknown clock (likely 4.5Ghz) and 8 clock deprived Zen 5c cores (likely ~3Ghz). Expected result.That's pretty awful for 50% more cores and higher IPC.
True however, we don't know MT clocks, and those dense cores are surely running at lower clocks than P cores, but what exact speed remains unknown.That's pretty awful for 50% more cores and higher IPC.
The Zen 5C cores appear to be running at 3.65ghz in GB6.8 full Zen 4 cores running at ~4.5Ghz vs 4 full Zen 5 cores @ unknown clock (likely 4.5Ghz) and 8 clock deprived Zen 5c cores (likely ~3Ghz). Expected result.