Question Significant gains with extra threads/cores (1080p): myth or reality?

jana519

Senior member
Jul 12, 2014
771
100
91
There's a popular myth that extra threads/cores will result in significant performance gains for gaming at 1080p.

I present to you exhibit A:








Here we see from none other than our own Ian Cutress that in 3 games benchmarked at 1080p, there is less than 2% performance gain realized from 4 additional threads.

Based on this data, it's ridiculous to recommend extra threads/cores for the average PC gamer. We, as a community, need to stop propagating this myth.

I look forward to your vigorous and data based rebuttals.
 

rbk123

Senior member
Aug 22, 2006
694
254
136
FPS for FPS's sake is just bragging rights in a hall of fame toplist. Not for your actual gaming experience... If you want bragging rights for having 10-20 higher FPS on a screenshot?
That's all this thread is. Jana's tired of taking the beating from other gamer's with superior CPU's over his, so he had to lash out with this old and tired argument. It's all he's got since he doesn't want to cough up the dough to upgrade to save his pride.

Gamer's are very insecure; it's why the 9900KS sells so well - regardless of how many times Intel has stuck it to them.
 
  • Like
Reactions: Makaveli and lobz

eek2121

Senior member
Aug 2, 2005
437
315
136
I just want to throw in a word of caution to anyone who takes AnandTech's gaming numbers seriously. Don't. They are using a GTX 1080 (non-Ti). They are GPU bottlenecked, and in addition, NVIDIA's drivers make assumptions about certain CPUs they should not. In order for a benchmark to be valid, things like 'multi-threaded optimizations' need to be disabled in the NVIDIA control panel.

I have the utmost respect for AnandTech's reviewers, but until things improve I cannot trust their benchmarks.
 
  • Like
Reactions: CHADBOGA and lobz

Gideon

Senior member
Nov 27, 2007
809
1,204
136
My hypothesis stands if you're using a 1080p 60Hz monitor. Which most people do. But yes, you're right that I was wrong about gaming on a 144Hz monitor. :D
Even on a 60Hz monitor it's not quite that simple. Tech Report has a classic article that provides a very thorough explanation of why FPS isn't a good metric and frame-times are much better. I think someone already posted it, but here's the link again. Here is another, more recent one, explaining why even using min-FPS is not good enough.

The thing is that your average FPS might be above 60, but if some frames take much longer than others you'll still notice very distracting "stutters" and "janks".

With a constant FPS rate pegged at 60 FPS each frame would take exactly 16.7 ns, with 30 FPS it would be 33 ns. But framerate is not constant. You can be easily getting over 60FPS but have tens or hundreds of frames taking longer than 33ns or even 50ns. It will be noticable and it will be distracting.

For instance look at Far Cry 5 results of Tech report 3700x review

All the CPUs are comfortably above 60 FPS:


yet they are not running perfectly smoothly over > 60 FPS all the time:



this is the amount of time in the benchmark spend below 30 FPS per-cpu. Bear in mind this is in milliseconds, meaning that even 9900K had one spike during the benchmark (oddly 8700K did not). The spike look like this:



and on slower CPUs there are multiple spikes that are usually longer:


Each of them would definitely be noticeable on-screen. Bear in mind that this is despite the fact that even 1800x ran over 100FPS on average.

Overall you are more-or-less correct in saying that at 1080p and 60Hz with 4 really fast cores (e.g. i3-8350K and above) is still usually enough, but even then in some games 6+ cores are smoother (linked to Civ 6 results wher even 8350K cant hold 60FPS all the time, yet i5-8400 can). For 144hz, there is no question that a lot of cores are needed
 

Arkaign

Lifer
Oct 27, 2006
20,481
937
126
One important thing to note is that globally, most PC gaming is not in the AAA big budget sphere, but rather stuff like LoL, WoW, SC2, CS, and indie/basic stuff. Indeed I believe I read that more PC gamers play with integrated graphics than with dGPUs as well (though that info was as of like 2016, not sure if it still holds, but it probably does).

There are damned lies and statistics, or something to that effect. As always, 'it depends'.

It's like the use case for the 9900KS (and yes, it's more than 1080p, lol). I'm a big proponent of it because as of yet, it's the closest I can bring my monitor to the higher refresh rates in the games that I play.

As always, I really REALLY wish people would remember that sites nearly always use 1080p Ultra/Max for testing. Which is ok, but you have to look at it for what it is. For example, I usually get better than 1080 ultra performance even at 3440x1440 Ultrawide, because there are many settings I can live without. I value high refresh, which translates as motion clarity, more than dialing everything to the max on my 2080ti. Hah, besides Q2 RTX, I don't even leave RT enabled. But that logic goes both ways. I live in my own edge case, and what is right for me is not for everyone, and vice versa.

Can 4C/4T work fine for lots of stuff? Yup. Can 4C/4T also be terrible for some? Yup.
 
  • Like
Reactions: scannall

rbk123

Senior member
Aug 22, 2006
694
254
136
I never said I wanted 'bragging rights', nor called the 9900KS the best CPU ever. That's just kind of bizarre.
He's just speaking about the majority; there will always be exceptions which you appear to be. The majority are far more like the OP, than you.
But that behavior isn't just limited to Intel gamers. AMD gamers are no different when they end up in the same situation.
 

lobz

Golden Member
Feb 10, 2017
1,202
1,166
106
I'm not sure why the hostility. I noticed the upgrade, both from the 8086k, as well as the 3700X. Now the 3700X is overall superior, and still my recommendation outside of purely gaming setups for numerous reasons, but I'm also not the guy running DDR4 2666 and stock clocks. The difference I see is often bigger than 15-20%, which is definitely something you can feel A/B if you are used to it. Due to regular site benchmarks being maxed out ultra settings, you do see more GPU bottlenecking masking some of the gaps.

Is it really worth it? That's in the eye of the beholder, hell it's honestly all just fancy toys if you're talking about PC gaming anyway.

I never said I wanted 'bragging rights', nor called the 9900KS the best CPU ever. That's just kind of bizarre.
There is zero hostility towards you from me. Your posts are always reasonable even when I don't agree with some of them. Even now I understand your reasoning, I just think it's invalid. That being said, it's not up to me to decide what the 'truth' is :) I stand 100% by what I've said (please read my post again, I also never said that you wanted bragging right, nor that you thought the 9900KS was the best CPU ever) - but I actually dig your forum presence, I swear I feel no hostility at all :)
 
Last edited:

Markfw

CPU Moderator, VC&G Moderator, Elite Member
Super Moderator
May 16, 2002
20,263
7,918
136
There is zero hostility towards you from me. Your posts are always reasonable even when I don't agree with some of them. Even now I understand your reasoning, I just think it's invalid. That being said, it's not up to me to decide what the 'truth' is :) I stand 100% by what I've said (please read my post again, I also never said that you wanted bragging right, nor that you thought the 9900KS was the best CPU ever) - but I actually dig your forum presence, I swear I feel no hostility at all :)
This is a little off-topic, but I just wanted to point out to the forum that THIS is the way to disagree with someone.
While this is not moderation (this post) But I would like to say as a moderator, that this is what we need more of here, this is the way to disagree ! With respect !
 

guachi

Senior member
Nov 16, 2010
750
402
136
Anything more recent than these ancient benchmarks?

I've seen a few videos (like at Tech Deals) that show major limitations with 4c/4t CPUs.

He actually plays the games and said in many titles the stuttering and other issues was really noticable regardless of where the average fps showed m
 
  • Like
Reactions: aigomorla

Markfw

CPU Moderator, VC&G Moderator, Elite Member
Super Moderator
May 16, 2002
20,263
7,918
136
Those are old CPUs . And old benchmarks. Try something from the last year/
 
  • Like
Reactions: Makaveli

lobz

Golden Member
Feb 10, 2017
1,202
1,166
106
There's a popular myth that extra threads/cores will result in significant performance gains for gaming at 1080p.

I present to you exhibit A:








Here we see from none other than our own Ian Cutress that in 3 games benchmarked at 1080p, there is less than 2% performance gain realized from 4 additional threads.

Based on this data, it's ridiculous to recommend extra threads/cores for the average PC gamer. We, as a community, need to stop propagating this myth.

I look forward to your vigorous and data based rebuttals.
Is this a troll post or you just simply can't read?

The point is: there are a lot of games now that stutter a lot with less than 8 threads and the minimum FPS numbers are horrible, and the number of these games will only increase.

Inserting 3 graphs with 3 games where the average framerate doesn't suffer, shows only this: you either have no idea what you're talking abut, or you're just trolling. I can't decide.
 

kschendel

Member
Aug 1, 2018
64
9
41
It's easy to say "gaming" as if it's all one, and I do it myself sometimes; but this sort of thing is where the generalization breaks down.

Multi-threaded game engines benefit from more cores, up to the number of simultaneously busy threads the engine wants. More cores beyond that help relatively little, and then only by offloading background system activity.

So the question "will more cores help" can only be answered in the context of a specific set of game engines, and what versions of those engines, and how many active threads they are programmed to run. If you're running an old WoW that's essentially single threaded, then no of course more cores does nothing. And likewise for any other game engine that runs just a small handful of threads.
 
  • Like
Reactions: ryan20fun

BigDaveX

Senior member
Jun 12, 2014
440
209
116
Based on this data, it's ridiculous to recommend extra threads/cores for the average PC gamer. We, as a community, need to stop propagating this myth.
Right. I'm sure it's just a total coincidence that the end result of this would most likely be people recommending Core i7/i9 CPUs over Zen 2.
 
  • Like
Reactions: lobz

tamz_msc

Platinum Member
Jan 5, 2017
2,558
2,245
106
Post the data if you have it. The lazy argument with no facts or data presented is not a rebuttal of any standard or measure.
What's there to argue? You've posted some old benchmark charts with a GPU bottleneck that doesn't tell anything of substance whatsoever. Also, AT gaming benchmarks are useless except the 720p low and 1080p medium results because they still haven't got anything faster than a GTX 1080 to test their games.
 

teejee

Senior member
Jul 4, 2013
295
87
101
Post the data if you have it. The lazy argument with no facts or data presented is not a rebuttal of any standard or measure.
It is you that don't have valid argument for your statement. So it is up to you to give us better arguments.
 

Campy

Senior member
Jun 25, 2010
746
148
106
Yeah we should all be buying 9350K /s

There is so much good information available that you obviously either haven't looked for it, or intentionally avoided it. I don't understand which purpose this thread serves.
 

Hitman928

Platinum Member
Apr 15, 2012
2,537
1,708
136
Possibly a severe case of confirmation bias going on here?

BTW, from Iron Woode's post:

1578341376899.png

Compared to 4 cores, a 12 core CPU provided:

107% faster "average" fps
84% higher lows (taking 1% lows).

Nope, don't see much advantage there at all :rolleyes:
 

Cogman

Lifer
Sep 19, 2000
10,235
75
91

Tada! Old game that does better with more cores!

Whether or not a game will benefit from more cores is going to depend on what the game is doing. Games with a bunch of physics, pathfinding, and objects flying around are (likely) going to do better with more cores.
 
  • Like
Reactions: jana519

ASK THE COMMUNITY