- Mar 3, 2017
- 1,747
- 6,598
- 136
Why?
I was always under the impression Golden Cove is +19% over Willow Cove (TigerLake). This puts Zen 5 at +16% over Golden Cove. That doesn’t seem impossible to me.
Mahboi
Golden Member
- Apr 4, 2024
- 1,035
- 1,899
- 96
There are 40 days left before Computex.
What's convenient is to have a copium balloon as well furnished as yours.
I've already said it several times: there is no way AMD would've bothered redoing the entire frontend and making a large Tock just to come out with 5% IPC improvement at the exit. It is an insane degree of cope to believe that they would release a product like that after this much effort.
Which means that that convenient number you've found that barely puts it above Zen 4 is either completely false, and the other 2x TGL number is also completely false, either that your convenient number is completely false and the other one is not.
I don't have proof that the other one is true. I do have every reason to be confident that that score you're attaching yourself to is entirely false.
So I'll do as the optimists and BELIEVE. And you're free to believe whatever you want, I don't wanna argue with someone who seriously believes that AMD would come out with a 5% IPC growth as the first big leap since Zen 2. That is a ridiculous number to latch on to.
adroc_thurston
Diamond Member
- Jul 2, 2023
- 3,708
- 5,426
- 96
Neither of those samples should be valid to you, kids.
You don't know the platform envs they ran in.
Just keep the numbers in mind, they're funny.
Mahboi
Golden Member
- Apr 4, 2024
- 1,035
- 1,899
- 96
That's pretty much where I estimate Zen 5 IPC to end up. It seems pretty good to me. And if they can push clock rates a bit higher still then Zen 5 3D will be in very good shape.
@adroc_thurston Do you know about TDP of Zen5? My source only told me Zen5's TDP is higher than Zen4; also claimed by HXL.
tamz_msc
Diamond Member
- Jan 5, 2017
- 3,865
- 3,729
- 136
Still waiting for an explanation why one score in the morning completely invalidates the other score in the afternoon, on the same day.
"SF-23 will be unprecedented in terms of speed" - Ferrari CEO before the 2023 F1 season started.
Mahboi
Golden Member
- Apr 4, 2024
- 1,035
- 1,899
- 96
I'm sure you were making a point there, but all I know about cars is that mine still starts when I turn the key and hasn't killed me yet. And I don't care to know much more.
Zen 4 already had a decent IPC advantage over Golden Cove, so that would mean Zen 5 has almost no IPC uplift. Given what we know about the design, this isn't possible (edit: unless AMD completely botched the design in some way, but that also seems extremely unlikely).
Last edited:
adroc_thurston
Diamond Member
- Jul 2, 2023
- 3,708
- 5,426
- 96
So you can cope without knowing the platform variables involved.
It's real simple!
they're the exact same since it's all the exact same platforms.
well except -halo.
tamz_msc
Diamond Member
- Jan 5, 2017
- 3,865
- 3,729
- 136
I don't think there was any way to have "Power Consumption: +" compared to RPL-R.
Mahboi
Golden Member
- Apr 4, 2024
- 1,035
- 1,899
- 96
And if you knew about letting go when you've become a broken record, that would be highly appreciated. Bye.
Mahboi
Golden Member
- Apr 4, 2024
- 1,035
- 1,899
- 96
The real concerning part is that he only went "Power Consumption: -".
Frankly I expected a "- -" or even "- - -" at least...
Intel better not go from Repeller perf to Repeller - 20% but with 10% more perf. I'd be very disappointed. I'm really expecting something like 40% less power draw even if the perf uplift will be close to zilch.
tamz_msc
Diamond Member
- Jan 5, 2017
- 3,865
- 3,729
- 136
Pictures and Gif's do not give you immunity to the no profanity rules in tech.
Last edited by a moderator:
- Mar 3, 2017
- 1,747
- 6,598
- 136
I did it, lscpu shows 2.6 Base for my 7950X, but cpufreq reports 2G, I don't have GB account, so maybe someone can check the frequencies
My RAM is 6000MT/s
Bash:
cat /sys/devices/system/cpu/cpu*/cpufreq/scaling_cur_freq
400000
400000
400000
1999222
400000
400000
400000
2000130
400000
2000078
1998810
400000
400000
400000
400000
400000
400000
400000
400000
400000
400000
2000082
400000
1999932
400000
400000
400000
400000
400000
400000
1999921
1999553STX is not recognized properly, L1 and other data seems completely off.
This is a more conservative report unlike this one @1.4G here https://browser.geekbench.com/v5/cpu/compare/22423185?baseline=22425328
If these are spoofs then well... waste of time
adroc_thurston
Diamond Member
- Jul 2, 2023
- 3,708
- 5,426
- 96
Again, you don't have to cope here.
There's an Intel thread, go there.
Last edited by a moderator:
Which factors are sensitive to Geekbench performance: Processor firmware, or/and time of result upload?
To avoid misunderstandings: This question is a purely rhetorical one, as was my former question.
...and except for SP5, but that one goes along with a core count increase.
adroc_thurston
Diamond Member
- Jul 2, 2023
- 3,708
- 5,426
- 96
Obviously but they've been bumping server skt power every gen since Naples, no changes here either.
So 19.3% higher IPC than Zen 4 if that leak is real. Not bad. Especially if other rumors of still higher clock rates is also true it puts it right in line with previous Zens in generational uplift.
tamz_msc
Diamond Member
- Jan 5, 2017
- 3,865
- 3,729
- 136
adroc_thurston
Diamond Member
- Jul 2, 2023
- 3,708
- 5,426
- 96
tamz_msc
Diamond Member
- Jan 5, 2017
- 3,865
- 3,729
- 136
Attachments
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)
- Started by DisEnchantment
- Replies: 22K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 17K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter