Question Alder Lake - Official Thread

Page 56 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

zir_blazer

Golden Member
Jun 6, 2013
1,184
459
136
AVX512 seems like a real flaw of the Alder Lake design. A lot of silicon that is just being turned off.

I wonder if they are going to squeeze AVX512 into to future E-Cores.
Wasn't there another Alder Lake die in the works that was 6P + 0E? Those could come with AVX-512 enabled out of the box since there is less of an excuse to not have it enabled, except market segmentation.
 
Jul 27, 2020
17,876
11,659
116
All this E-core brouhaha could have been easily avoided with right-click context menu options "Run with P-cores only" and "Run with E-cores only". Would have been piece of cake for Microsoft to add those options.
 
Reactions: ryan20fun

Hulk

Diamond Member
Oct 9, 1999
4,372
2,247
136
All this E-core brouhaha could have been easily avoided with right-click context menu options "Run with P-cores only" and "Run with E-cores only". Would have been piece of cake for Microsoft to add those options.

In the BIOS you can set the scroll lock to enable/disable the E cores. Problem solved.
 

TheELF

Diamond Member
Dec 22, 2012
3,991
744
126
All this E-core brouhaha could have been easily avoided with right-click context menu options "Run with P-cores only" and "Run with E-cores only". Would have been piece of cake for Microsoft to add those options.
Leaving that kind of performance changing decisions to the clueless end user is terribad, anybody smart enough to understand which group is going to give him better performance or efficiency or whatever can easily do this with task manager' affinity settings, or by making a shortcut including these settings.
 

Heartbreaker

Diamond Member
Apr 3, 2006
4,262
5,259
136
All this E-core brouhaha could have been easily avoided with right-click context menu options "Run with P-cores only" and "Run with E-cores only". Would have been piece of cake for Microsoft to add those options.

The "brouhaha" only really exists for people looking to find fault. In which case they would find something else to find fault over. Like the 12900K using too much power... Like Alderlake is only the 12900K.
 
Jul 27, 2020
17,876
11,659
116
Leaving that kind of performance changing decisions to the clueless end user is terribad, anybody smart enough to understand which group is going to give him better performance or efficiency or whatever can easily do this with task manager' affinity settings, or by making a shortcut including these settings.
Still, Windows has plenty of power user options available in the UI, like show "Show hidden and system files" or *shudder* "Format" right there in the Drive right-click context menu which doesn't really provide any helpful information about what it really does. I bet there are quite a few people out there who found out the hard way what it does. I think my dad was one of them because I remember when I asked him what the Format option does in Excel back when we got our first PC, he immediately forbid me to click it, saying it would destroy all data! I clicked it anyway when he wasn't around and found, much to my delight, that I could make colorful cells. Also, I would love to know if anyone other than him clears the gridlines in Excel and then writes a frickin' letter in it
 

coercitiv

Diamond Member
Jan 24, 2014
6,391
12,817
136
All this E-core brouhaha could have been easily avoided with right-click context menu options "Run with P-cores only" and "Run with E-cores only". Would have been piece of cake for Microsoft to add those options.
The entire point of Thread Director is offering a seamless experience to the user. Microsoft adding these options would make it clear that for the time being the scheduler cannot offer optimal performance.

I'd rather see continuous development effort from Intel/MS than a crude UI switch that tells the user the most advanced chips from Intel could use human scheduling help. The upcoming desktop & mobile launch in 2022 could be a great occasion for a software update.
 

Hulk

Diamond Member
Oct 9, 1999
4,372
2,247
136
The entire point of Thread Director is offering a seamless experience to the user. Microsoft adding these options would make it clear that for the time being the scheduler cannot offer optimal performance.

I'd rather see continuous development effort from Intel/MS than a crude UI switch that tells the user the most advanced chips from Intel could use human scheduling help. The upcoming desktop & mobile launch in 2022 could be a great occasion for a software update.

Yes I agree. I also think the Thread Director is getting it right the vast majority of times.
 
Jul 27, 2020
17,876
11,659
116
The entire point of Thread Director is offering a seamless experience to the user. Microsoft adding these options would make it clear that for the time being the scheduler cannot offer optimal performance.
Even Ian complained that his particular use case wasn't handled gracefully by Thread Director. The whole Alder Lake affair seems kinda rushed, like they were tweaking stuff till the last minute. I understand that Microsoft can't provide an Alder Lake specific option without go ahead from Intel and Intel desperately wanted to make the hybrid cores seem as seamless as possible in general use. They succeeded for the most part. Let's hope AMD shows them how it's really done.
 
Reactions: lightmanek
Jul 27, 2020
17,876
11,659
116
Reactions: lightmanek

Shivansps

Diamond Member
Sep 11, 2013
3,873
1,527
136

Possibly first gaming use of AVX512?
You need more than just disable the E cores to get avx 512 on adl.

The bios must support enabling It as well.
 

IntelUser2000

Elite Member
Oct 14, 2003
8,686
3,785
136
Just saying that a Desktop Sunny Cove/Tremont hybrid would not cut it for for desktop against Amd Current and up coming Desktop CPUs, they had to push Alderlake(the 12900K) to match a stock 5950X(When the 5950X is without limits the story changes). Do you thing a 8C/16T Sonny cove + 8C/8T Tremont Hybrid would keep up?

Did you read my post? I was saying basing it on Lakefield is terrible since the hybrid combo on that chip outperforms the standalone Tremont chip on Jasper Lake.

Alderlake completely reverses the fortune. It actually does really well. 20% difference in the big core and 30% on the e core isn't what's responsible for that - proper implementation is.
 

IntelUser2000

Elite Member
Oct 14, 2003
8,686
3,785
136
Also I believe that the superwide design of the Gracemont cores is directly related to their "tuning" for MT performance. We have 6 decoders in 2x3 clusters and if I remember correctly 17 decode. But, the OoS is much more primitive than Golden Cove, meaning MT throughput will only be high if the incoming code is already highly parallel as Gracemont doesn't the smarts to work it out on the fly.

So on the big cores they have the Complex+Simple approach and on Golden Cove it's 1+6.

Since Tremont they have their own way, which is splitting the decoders into two clusters.

It's a different way of reducing power and transistor impact of x86 decoding.

Golden Cove is very wide, since on uop cache hit it's effectively 8-wide. The higher number of pipeline stages will result in more misses and both contribute to gain when SMT is on.
 

Hulk

Diamond Member
Oct 9, 1999
4,372
2,247
136
Thread Director and Windows 11 observations with my new rig.

Last night I'm editing some video with Vegas Pro and then rendering using Render+, which is the "Happy Otter" add on for Vegas. It basically allows high quality x264/265 rendering. I notice that when Vegas is "on top" all 20 logical processors are at it but when I move to another apps the Gracemont cores take over. The thing Ian was concerned about.

I think I actually like this behavior. I was watching the UFC fight night on ESPN+ on the computer and also doing some Photoshop work so I had all of the P's at my disposal for Photoshop. The video was rendering just fine in the background and I didn't need it done immediately. Plus it can always be set to batch render.

If I was to fine tune this behavior I would have 4 P's "stay with me" on my current app while the rest stay with background processes.

But this is going to be a subjective thing, which will vary from person-to-person so I think the Thread Director is making the right decision by keeping the P's with the user. Perhaps in the future there could be a setting in Control Panel that allows you to assign P's and E's to foreground/background applications. So if you have 8 P's and 4E's like the 12700K you could select 4 P's as foreground and the rest to background as the OS deems necessary. Or if you really needed the background apps to crank you could select 2 E's for foreground for a little web browing or something?
 
Reactions: lightmanek

eek2121

Diamond Member
Aug 2, 2005
3,045
4,267
136
The problem with looking at averages is you miss issues with performance deltas during high stress scenarios, and the 1% lows on the e-cores can be worse than Sandy Bridge.

No, the problem is that people are attempting to directly bench “E” cores period. The “E” cores were added to handle 2 scenarios:

  1. Give dedicated cores to handle background tasks/services to help free up “P” cores.
  2. Enhance multicore performance in well threaded workloads.
For gaming, the demanding threads should always be run on the “P” cores. The “E” cores should be used for things like AI and other workloads.

The issue with benching “E” cores in isolation at all is that “E” cores have higher latency and were designed around the scenarios I mentioned above.
 
Jul 27, 2020
17,876
11,659
116
So if you have 8 P's and 4E's like the 12700K you could select 4 P's as foreground and the rest to background as the OS deems necessary. Or if you really needed the background apps to crank you could select 2 E's for foreground for a little web browing or something?


What do you suppose would happen if you select Background Services on your system? Would that make all except one P-core tend to background stuff?
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |