Linus Torvalds: Too many cores = too much BS

Shehriazad

Senior member
Nov 3, 2014
555
2
46
I guess at least for gamers too many cores are BS at some point.


I mean sure you can try and optimize your game to use 6, 8, 10, 12 cores...but is that worth it?

I'd rather just see a "close to the metal" API that runs applications with 4 cores that perform so well that this CPU parallelism is just not an issue anymore.

If I had a quadcore CPU that runs at 4 GHZ but is as fast as an intel i7 8 core at 4Ghz...then my choice is clear...less cores means less trouble if they deliver the performance you need. Just look at the poor FX8 and FX9 chips from AMD. Not many games utilize 8 cores..and even when they do the performance is crap (okay, AMD is to blame for that part...but still).
More cores also always means more power draw and less space for other things. (At 14NM or so that won't matter anymore I guess)

I for one wouldn't mind a move toward capable SoCs that even have the ram on the chip already...but I don't see that happening with 35324535 CPU cores. Per core power for me will always be more important than number of course.

4 cores just seems to make the most sense for gamers...and that is what I care about the most.

P.S. 4 actual cores, please...no modular bs. Work that IPC, baby. I guess at some point in the future even 8 cores can be an established thing for gamers(FX8 can still only use its power in a minority of games)...but beyond that? I don't see 16-32-64 cores becoming a thing for us gamers anytime soon.

P.P.S. Yes, I know that some areas benefit from a ton of cores...but as a gamer..I am generally not a part of this.
 
Last edited:

VirtualLarry

No Lifer
Aug 25, 2001
56,450
10,119
126
Has he not heard about job-based queue systems, which may prove to be more scalable than traditional threaded code?
 

monstercameron

Diamond Member
Feb 12, 2013
3,818
1
0
Has he not heard about job-based queue systems, which may prove to be more scalable than traditional threaded code?


If I'm not mistaken the infamous devs used such a system for their ps4 game. Do you know what the difference is? Higher latency but better scalability?
 
Last edited:

Shivansps

Diamond Member
Sep 11, 2013
3,873
1,527
136
Has he not heard about job-based queue systems, which may prove to be more scalable than traditional threaded code?

He did said that is a waste of time for most tasks, and, it is, it mostly used for graphics and physics, going beyond that is insane.
 

monstercameron

Diamond Member
Feb 12, 2013
3,818
1
0
He did said that is a waste of time for most tasks, and, it is, it mostly used for graphics and physics, going beyond that is insane.


Maybe I am misreading your comment but are you really claiming that multicore is a waste of time outside of graphics and physics?
 

ShintaiDK

Lifer
Apr 22, 2012
20,378
145
106
Without tiles, concurrency etc that is natural parallel. Then more cores is nothing but a waste in both hardware and software manufactoring/development cost. And its quite obvious that both AMD and Intel knows this. Hence their roadmaps are filled with 2-4 cores for the consumer if we look beyond the tiny niche that buys more than 4 cores.

Mitosis was also another example in the desperation of using more cores for the consumer.
 

Shivansps

Diamond Member
Sep 11, 2013
3,873
1,527
136
Maybe I am misreading your comment but are you really claiming that multicore is a waste of time outside of graphics and physics?

multicore its not the same thing, for example in games,they generally use 1 core for the AI, another for sound/game logic, after that is graphics and physics that make sence to use several cores for those 2 tasks, even if they use just 4 cores (AI, Sound, Graphics, Physics), that is multicore, but not really parallel as each task is done on only one core.

Trying to split AI or logic intro several cores is a waste of time and human resources, you must keep in mind that in any task that you want to split and do it in parallel you are adding overhead for each thread, making it less and less efficient.

Now for other things that are not games? no, there a few more things that make sence to use multiple cores for 1 one particular task, encoders, decoders is one for example.
 
Last edited:
Aug 11, 2008
10,451
642
126
I would say he has a point, but as usual, his comments don't exactly look objectively at both sides of the issue, to put it mildly.
 

BonzaiDuck

Lifer
Jun 30, 2004
15,785
1,500
126
multicore its not the same thing, for example in games,they ganerally use 1 core for the AI, another for sound/game logic, after that is graphics and physics that make sence to use several cores for those 2 tasks.

Trying to split AI or logic intro several cores is a waste of time and human resources.

Now for other things that are not games? no, there a few more things that make sence to use multiple cores for 1 one particular task, encoders, decoders is one for example.

Sure, and nobody disputes that. We'll find a range of people in the forums with varied preferences for spending money, different usage profiles like those you suggest -- different needs.

Personally, I'm at a crossroads myself. I've now got two Sandy systems. They take everything I throw at them, but what I throw at them might be slightly limp compared to the usage profiles of others.

I've been planning to build a Haswell-E system next year. Why? Like the lawyer says about the tycoon who had a snuff film made in the Nicholas Cage film "8mm" -- "because I can." The "E" system would be my first experience with water-cooling.

I don't know whether I need to build the E system to do the latter. But I'm reasonably sure if I invest the money in building an E system, I'm likely to do it with water -- even custom-water.

Why do I need six cores, or 12 with HT? I don't think I do. Or, at least -- I should admit to myself and the world at large that I "need" it for the bragging rights. But I could also spend the money on new software, new peripherals -- and maybe a new gas-grille barbecue. Or better automobile maintenance. Or maybe I should just save the money to be left to my beneficiaries in a last will and testament. I'm not sure they would be all that pleased to inherit several computers.

If you put some bottles of wine aside at the right constant temperature, they may likely improve with age. But last year's processor is just "last year's processor" slowly becoming obsolete.

That being said, disposable computing is also a budget drain . . . .
 
Last edited:

NTMBK

Lifer
Nov 14, 2011
10,269
5,134
136
Bear in mind he isn't talking about 6 or 8 cores. He is talking about Xeon Phi style manycore.
 

III-V

Senior member
Oct 12, 2014
678
1
41
He's right for lower core count configs as well.

Take a look at what else he states:
Given the choice between 16 cores and 4, I suspect most people will take 4, and prefer more cache, graphics, and integrated networking etc.

I agree, wholeheartedly.
 
Last edited:

seitur

Senior member
Jul 12, 2013
383
1
81
Linus is right.

Market reality show this.

AMD vs Intel in last couple years is basically more weak cores vs less stronger cores. Who won?

Other leading computation makers? What direction Apple went in ARM space with it's CPUs? Less cores strong IPC.


Remember that silicon will get replaced.
 

witeken

Diamond Member
Dec 25, 2013
3,899
193
106
Quote, remark, comment, sentiment etc. let's not play word games.

It does not exist. No one ever claimed such a thing, there are no facts to back it up, and there's no causal relation or reason for why should use the Moore's Law equivalent of memory scaling in a core count debate.
 

tolis626

Senior member
Aug 25, 2013
399
0
76
Saying something like this without context is stupid. If we're talking about smartphones, then I agree, more cores are stupid. I mean, 8 core CPUs in a phone? Even in big.LITTLE it's just too much. But on the desktop, and as long as we stay within reason, more cores make sense. It's not rare to see a desktop user doing multiple things at once, with some of them requiring multiple threads anyway. And I don't mean browsing the net while listening to music, but gaming while rendering or encoding/decoding or something like that. 8 cores/16 threads is as high as I would go for a consumer build with current software. It's by no means the sweet spot, but I would give the sweet spot award to 6 core CPUs, not to quad cores. Quad cores are ok now, but I suspect that that will change within the next few years. Not many things will require more than 4 cores, but if something actually uses 4 cores and 4 cores is all you have, I can't help but think that performance will take a dive, because there are also things running in the background.

But that's just my 2 cents. Also, I think sometimes Torvalds says stuff just to seem relevant. But I digress.
 

Nothingness

Platinum Member
Jul 3, 2013
2,766
1,424
136
Torvalds is mainly talking about parallelism which isn't the exact same thing as core count. For instance, I don't do any parallel programming but I'm glad I have an 8 core with HT Intel CPU (and powerful cores of course...) for building large software, and I bet Linus thinks the same.
 

SunnyD

Belgian Waffler
Jan 2, 2001
32,674
145
106
www.neftastic.com
Depending on the task, parallelism can be pointless.

However a properly designed OS can easily keep a ton of small lightweight cores busy just by handling background tasking while freeing up the heavyweight cores to do the more important heavyweight user demanded operations to keep the system responsive. This, imho, would be the holy grail of "parallelism" at the OS level for a multi-many core CPU at this point. Less places for the OS to get bottlenecked or backlogged equals a much more responsive system. It's not exactly traditional parallelism though.
 
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/    |