They were merely estimating. Like everyone else so far.Have I posted this before? Jim Keller was saying around 30% uplift
Ex-AMD & Legendary Chip Architect, Jim Keller, Reveals Next-Gen Zen 5 CPU Performance, Frequency & Power Estimates
The first AMD Zen 5 CPU architecture performance, frequency and power estimates have been revealed by Ex-Chip architect, Jim Keller.wccftech.com
Or is that publishing wrong?
Well it seems this time there's a general consensus which didn't happen before. Besides, the leak, this time, is SPECint which is pretty solid, for me at least anyway.I think everyone is wrong and just like they've always been the past 4-5 years, AMD is tight lipped and none of the "leaks" are more than 75% accurate.
I even foresee a near future where all of these rumor mongers all begin to hinge on one or two specific benchmarks to suddenly validate their wild predictions.
I've been actually deep in tech since the days of sandy bridge, and these leaker names like MLID and AdoredTV didn't come out of the woodwork until the time of 2018/2019 when techtubers became a gold mine. I think its obvious a lot of people stand to gain by grasping at straws to build up clout and thus generate streams of revenue/help bolster their existing income streams.
I've been re-reading the RDNA2 speculation thread lately to try and see which leakers were and weren't correct and its very interesting how much the forum has changed as of late. People seemed to get way more hyped about little tidbits these days. Zen 5 is going to be good, if not then it will be AMDs worst blunder since bulldozer. But I think people should be more proactive to ignore hype and wild claims, in my opinion.
Agreed. For some reason I don't think @dr1337 has been reading the thread regularly to know where the 40+% value comes from.Well it seems this time there's a general consensus which didn't happen before. Besides, the leak, this time, is SPECint which is pretty solid, for me at least anyway.
It’s a good pick because the 86% increase brings up the average quite a lot.WPrime 86%? Wow. Surprised anyone still benches that though. What's next, SuperPi 1.5 mod XS?
Monthly corrections + plus "I told you so" incoming. There's also a recent video about the number 37 so I'll take 37% ipc uplift
However, heat is a major for me and for many others as the 7800x3d is just that good/cool.
There is leaks about 7-Zip numbers from Allthewatts, 28% faster score is 8 percent faster. I am not surprised cause I believe IPC numbers would be lower. What do you think?
They should use a random number generator instead:Yes, people are generally bad random number pickers, but for stupid reasons. People tend to avoid numbers that end in 0 or even numbers in general.
That is such a bad take because Fornite was included. Either choose a wide range of applications or don't.It’s a good pick because the 86% increase brings up the average quite a lot.
WPrime 86%? Wow. Surprised anyone still benches that though. What's next, SuperPi 1.5 mod XS?
Veritasium!Monthly corrections + plus "I told you so" incoming. There's also a recent video about the number 37 so I'll take 37% ipc uplift
Idk what's behind your 15-20% numbers, but I'd generally expect the same considering AMD may have found potential for improvement from 10-15% (for the A0 tape-out) with the respin, on top of possibly increased Fmax as a bonus, so we'll seeBut IPC of Zen5 would be much lower, my estimate IPC would be around 15-20% faster...Definitely not 40% uplift as adroc claimed.
Fortnite 7%, Passmark 11%, R23 1T 20%, 7-Zip 28%, V-Ray CPU 33%, Metro Exodus 38%, Dolphin Bench 71%, WPrime 86%, Avg 3
The numbers Mason, what do they mean?
Mind telling me your estimate IPC if higher than my 15-20%?Idk what's behind your 15-20% numbers, but I'd generally expect the same considering AMD may have found potential for improvement from 10-15% for A0 with the respin, on top of possibly increased Fmax as a bonus, so we'll see
They should use a random number generator instead:
Having a lot fun of reading this. Yesterday we had to exclude games from the sample because they brought Zen 3'a average up, today we need to exclude WPrime because it does the same.
I severely doubt it has anything to do with AVX512.Possibly that it has nothing to do with AVX512 but to an improvement in the integer divider and root square extractor, these are attached to one, or perhaps two, ALU and they are not displayed in the diagrams published by AMD, they didnt even stipulate which radix was used for those computations.
Your expectations aren’t going to match reality. Zen 5 is a beast and the 40% number isn’t even the whole story.Idk what's behind your 15-20% numbers, but I'd generally expect the same considering AMD may have found potential for improvement from 10-15% (for the A0 tape-out) with the respin, on top of possibly increased Fmax as a bonus, so we'll see