given how they've hidden the oxidation via "minor incident", I bet many will prefer them to manufacture in other foundaries, at least they surely will instantly recall and claim TSMC if that happens, and not hiding it for a year and tell you "O, that's history"
-
You heard it here first: Black Friday 2024 has arrived!
MSFS 2020 is actually more main thread limited than other "sub threads", some workloads can't be divided effectively for all those environmental calculations.
And given the main stream lower end i5 and Ryzen have like 12-16 threads, especially for 10th gen and 11th gen, getting crazy amount of threading coding for the 1% of ppl didn't really make sense.
For the P and E core architecture is that a lot of games don't know how to differentiate and prioritize to use P cores, so ppl disabling e cores gets higher frame rates as it won't let the background apps use up the P cores while the game uses all those E cores for 8-16 threads. that's where the Intel APO came out for, enforcing them to use the P cores for games tested by Intel to try minimize the issue.
Cyberpunk isnt even close to being the most multithreaded game though, and 16 threads means 8p Cores + 8ecores since ecores take priority over HT.
In ratchet and clank for example 8+8 is around 25% faster on average and almost double the minimums compared to 8+0. I have videos if you are interested.
Disabling ecores drops performance by a truckload in heavy games like tlou spiderman ratchet etc. It is not reccomeneded
Last edited:
Pierce2623
Prominent
- Dec 3, 2023
- 485
- 368
- 560
Disabling e cores hurts in SOME games and speeds up others. Acting like e cores are always a benefit is totally disingenuous. If that style of e cores had been a success then their new e cores wouldn’t be wider than Redwood Cove bud.
Besides that space game that is vaporware for 10 years I don't know of any game that loses performance with ecores. Can you name one or two so I can test?
If zen 4 was good zen 5 wouldn't have released.... Huh?
thestryker
Splendid
- Apr 19, 2016
- 4,326
- 2,584
- 30,690
I mean these must be fakes, but here:
116: https://www.notebookcheck.net/Asus-...ets-RTX-4070-laptop-and-4K-OLED.871739.0.html
111: https://www.anandtech.com/show/21485/the-amd-ryzen-ai-hx-370-review/7
122.5: https://www.notebookcheck.net/Dell-...rcutting-the-Intel-Core-Ultra-7.869106.0.html
107/123 (rec/max win11 profile): https://www.pcworld.com/article/2374439/surface-pro-2024-review.html
Looks like it only "smokes" the Surface on recommended which runs them lower power profile than laptops do.
- Jan 20, 2010
- 16,847
- 8,257
- 79,540
systemBuilder_49
Distinguished
- Dec 9, 2010
- 101
- 35
- 18,620
4 cores 8 threads in 2024. (Note: an e-core is about as fast as an extra thread). I thought we moved beyond that in 2011?
When the stock market got the news, Intel tanked today ...
When the stock market got the news, Intel tanked today ...
- Jan 20, 2010
- 16,847
- 8,257
- 79,540
Apple is doing 4P + 4E in its entry-level laptop SoCs. That seems to be the market Lunar Lake is targeting.
The new Skymont E-cores will be much faster than the performance lost by removing HT from the P-cores. Even Gracemont was faster than doubling-up threads on Golden Cove, but perhaps not by much.
LOL, the share price probably tumbled on news of Intel cancelling its 20A node, with unspecified delays presumably on Arrow Lake (which is being ported to TSMC).
Pierce2623
Prominent
- Dec 3, 2023
- 485
- 368
- 560
Any of the multitude of games that spawn 4 threads or less can run faster with e cores disabled because it allows you to raise voltage and clocks further in the big cores than you can with the e cores enabled.
My point with the e cores is they went from being 25% of the area of a p core to 80% on the new design. It’s a completely different design direction and the only similarity is calling it an “e core”. It NOT a small core anymore. Zen5 was just a continuation of the same design direction just a little wider. Your implication that having newer evolution in the same style is the same thing as completely changing your design direction makes no sense.
This guy did some good testing:
And 3 of the negative ones are on the Intel APO list and almost all of the ones on the APO list weren't tested in this review. I'm guessing that many of the ones adjusted in APO have bad performance with either E-cores and/or HT and would be good candidates to test.
In case you don't have APO the list is:
Open the Intel APO app and you will be amazed about the no. Of games where limiting the threads on p cores do make a difference
If raptor lake was good… wait what is Intel releasing for this thread topic?
thestryker
Splendid
- Apr 19, 2016
- 4,326
- 2,584
- 30,690
To be fair that's not exactly what APO does from the testing that analyzed threads upon the initial release at least. It optimized where the threads were doing so fewer clusters of E-cores lit up and those that were happened to be more consistently used. If Intel could figure out how to do that without having a software layer hybrid architectures would basically have zero negatives beyond pure capability.
They've optimized thread director again with LNL so I wonder if APO will still be a thing for ARL and beyond (it's only desktop CPU die right now).
I think some test back then that is prioritizing the use of P cores and arranging the less imporant tasks to E cores instead to make it not run slower than previous gens. That actually was a defect from early P+E core architecture together with the fact that a lot if not most games don't go on bumping up the codes just to make use of the 1% users with 16+ threads. And ironically that's also kind of true for AMD, that's why the 8 Core 7800X3D just tops the gaming performance charts while the like of 7950X3D have essentially a "Pcore" CCD with the X3D and the "E core" CCD without X3D for games, the games parking the wrong CCD.
thread director issue will be interesting to observe.
APO doesn't limit the threads to PCores. It actually runs worse with ecores off.
Just a quick testing
There is "only" a 30-40 fps difference on the averages but look at the lows.....don't turn off your ecores lads.
I said prioiritze, not limit, that's how they move it. that's just how it works. of course turning all e cores off will have 1% low lower, since you don't have the e cores to run all your background works, like the FPS monitoring
In the video i just posted there is 50-60% utilization on all 8 ecores. You thinkthat's for fps monitoring and background tasks? Clearly the game is heavily using them.
So you are talking about turning off the ecores and then overclocking? Sure, but for those games that spawn 4 threads or less the same applies to P cores and HT. You can turn off 4 P cores on top of the ecores and overclock even further. But why would you do that? Realistically youll get 5% extra clocks by overclocking most chips. Too much hassle for too little gains.
Nice video, there are basically 5 games that perform worse with Ecores on (2% or more difference) and 16 that perform better with it. And he hasn't even tested some heavy hitters like TLOU, cyberpunk, once human or in general games that use 16+ cores. And it's more impressive later on in the video when he tests 1% lows. I've always been saying that turning off ecores destroys your lows and that video shows it quite clearly.This guy did some good testing:
View: https://youtu.be/LcQUUmi3rWI?t=681
And 3 of the negative ones are on the Intel APO list and almost all of the ones on the APO list weren't tested in this review. I'm guessing that many of the ones adjusted in APO have bad performance with either E-cores and/or HT and would be good candidates to test.
In case you don't have APO the list is:
Not in THIS game, but you saw the APO list of games? those obviously have a significant impact on the e cores. I was on a phone and not even realize you again choose a random game that just spawn more than 16 threads.
AMD Ryzen 7 7800X3D vs. Intel Core i9-14900K: The Definitive Test
In this comprehensive shootout comparing AMD's Ryzen 7 7800X3D and Intel's Core i9-14900K, we explore Intel's Extreme and Performance profiles, stability issues, and their impact on gaming...
www.techspot.com
This was one game where it do benefit from more threads in general and trade blows but a ton others benefit the 8cores X3D more and take the lead. good try Chief
APO does proper scheduling. It doesn't disable the ecores or what have you. The majority of the games in that list get incredibly high fps to begin with so I haven't ever properly tested them but ill give it a go. I'll start with F1 22 and compare ecores on vs off.Not in THIS game, but you saw the APO list of games? those obviously have a significant impact on the e cores. I was on a phone and not even realize you again choose a random game that just spawn more than 16 threads.
AMD Ryzen 7 7800X3D vs. Intel Core i9-14900K: The Definitive Test
In this comprehensive shootout comparing AMD's Ryzen 7 7800X3D and Intel's Core i9-14900K, we explore Intel's Extreme and Performance profiles, stability issues, and their impact on gaming...
This was one game where it do benefit from more threads in general and trade blows but a ton others benefit the 8cores X3D more and take the lead. good try Chief
Just tested RDR2, it's on the APO list.
Ecores on = 170 fps average, 141 1% lows, 130 0.2% lows
Ecores off= 169 fps average, 140 1% lows, 104 0.2% lows
I have videos of both runs if you want me to upload them to youtube. Both were done at 1080p with DLSS ultra performance to remove gpu bottleneck (im using a 4090). So pretty similar results beside the 0.2% that are much higher with ecores on.
Ecores on = 170 fps average, 141 1% lows, 130 0.2% lows
Ecores off= 169 fps average, 140 1% lows, 104 0.2% lows
I have videos of both runs if you want me to upload them to youtube. Both were done at 1080p with DLSS ultra performance to remove gpu bottleneck (im using a 4090). So pretty similar results beside the 0.2% that are much higher with ecores on.
I didn't even say disable ecores for faster games, what I say was that a lot if not most games don't get that high thread count, so giving up a few more P cores for more threads in the form of E cores have been troubling for games performance, and hence, APO being prioritize the gaming threads on P cores improve the FPS, and that's "Why some ppl disable E cores (Especially in early days of ADL) to gain more FPS for the games they play"
That is exactly a show of e cores don't do much for the game, 0.2% lows are other background tasks occupying and interrupting the p core for the game, and 99.8% of the time they get essentially the same FPS with or without E cores. And if one don't open the game instantly or have some other tasks pre-occupying the P cores due to thread schedulling, especially in early ADL era, the Threads put to E cores get lower average FPS as a result of a few threads pushed to E cores
E cores don't do much for games that don't use a lot of cores. P cores don't do much either for those games. That's not an ecore problem, that's a "game doesn't scale with cores" problem. Still, if you don't get any performance regression from those games, while you get a huge boost on the games that use those extra cores, what are we even talking about?
Im not running any background tasks btw, i'm running ghost ultra light with every windows fluff removed (store, xbox, windows defender, cloud, what have you).
TRENDING THREADS
-
-
Review AMD Ryzen 7 9800X3D Review: Devastating Gaming Performance- Started by Admin
- Replies: 202
-
Question The usual black screen and blue horizontal lines of death.
- Started by VæX
- Replies: 8
-
Discussion What's your favourite video game you've been playing?- Started by amdfangirl
- Replies: 4K
Facebook
Twitter
Instagram