• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Zen5 only 16 core.

16 core is fine, cost a lot so if we got a 20core then the price would be way to much for most.
i sure would benefit from 20 cores but cost is too high and only need that performance in rare occasions.
 
Here I made a screenshot while running GTA V, as you can see, even the 4 e-cores are helping.

Also depends on the GPU of course.

In my case it didn't run fine previously, I had a i7 6700K @ 4.5Ghz and it was bottlenecking my 2070 Super with GTA V.

Since I've been running a i7 12700K @ 5Ghz no more bottlenecking.

@1440p high-refresh that is.
if you read the article I linked and Steve's previous ones on the subject hes not stating six cores is enough for gaming or eight cores is enough but rather judge the CPU or GPU on performance not one aspect of it. So six cores may be enough, it may not be or it may be overkill because its a meaningless number without knowing the exact CPU and its performance.

If you look at Toms review of GTA 5 and the 5800X3D (can't link it so have to copy and paste) you have the eight core 5800X3D slightly ahead of the 12 core 5900X because the 5800X3D is the better gaming CPU based on performance. Clearly the 12900k is utilized by the game as well so not surprised your 12700k has its E cores utilized. The case is not for less core or more cores, the case is for performance.

Extra 1080p Games - fps
Tom's Hardware - 1080p ExtrasRyzen 7 5800X3DCore i9-12900KRyzen 9 5900XRyzen 7 5800X
Grand Theft Auto V184.9187.2179176.4
Project Cars 3273.7257.2217.5215.5
Shadow of the Tomb Raider235.7197.9174163.9
Far Cry 5182.8156.2123118.8
Borderlands 3158.4140.6118.1114.6
 
:confused: :confused:

Everyone went rabid and ripped my comment apart.
Nobody bothered to read your entire post apparently, just went straight to uber indignant "must protect my purchase" mode lol.

I've been waiting for another couple to mis-quote you, so as to award you the most misunderstood gold medal of the week.
 
I graciously accept :)
 
Low quality post by evernessince
Nobody bothered to read your entire post apparently, just went straight to uber indignant "must protect my purchase" mode lol.

I've been waiting for another couple to mis-quote you, so as to award you the most misunderstood gold medal of the week.

This isn't adding anything to the conversation, it's merely a character assassination attempt on anyone you disagree with. One could use the same logic to imply that you are just trying to validate your purchase being an Intel owner. Neither statements have any basis in fact, only create conclusions based on assumptions.
 
I can see why for non gaming workloads you might want more, but gaming? Hasn't it been proven multiple times that to a large extent the amount of cores don't really matter, single core speed and overall CPU performance seem to matter much more.

Like a 7600X, 6 core 12 thread, performs much the same as a 5800X3D, 8 core 16 thread with a slab of vcache on top. I'd imagine that Zen 5 could be similar, a 9600X could perform much like a 7800X3D in games.

Mixed workloads make some sense, but even then a 7950X3d or say hypothetical 9950X3D walks the line between both more than adequately.
 
Here I made a screenshot while running GTA V, as you can see, even the 4 e-cores are helping.

View attachment 350191



GTA V PC was released in April 2015.

GTA 6 PC will probably be released somewhere in 2026.


Years ago, @CAPSLOCKSTUCK already proved that GTA V was using all his xeon cores.

So , there you have it, once again...
What is the FPS difference then versus say, my 8700K which runs the game just fine? Utilization does not necessarily translate to performance. Its not rhetorical - but its important to separate your perception from reality. You lacked core count (or better said CPU grunt) for the game on a 4c8t CPU. You didnt anymore on the 12700K.

:confused: :confused:

Everyone went rabid and ripped my comment apart.
No not at all? Im just curious how an X3D in your mind is any different than say, an Intel CPU with and without E cores in the same line up. Because thats the Intel analogy here. And it happens too.
 
What is the FPS difference then versus say, my 8700K which runs the game just fine? Utilization does not necessarily translate to performance.
His 12700k is probably ~2x your 8700k in something like TLOU / Cyberpunk / Spiderman. GTA V is kinda old but still I expect a 2x even there
 
Wow, 4 pages in less than 24 hours. Hot topic?

Seeing usage on more than X cores doesn't mean the game benefits or even uses more than X cores for it.
Apologies to Vayra86 - I replaced your posted value with a variable to illustrate my point. And that is, just because a game is able to "utilize" x cores, that does NOT mean it "needs" those x cores to function, or more importantly, to provide good, "enjoyable" game play.

I see this all going back to @Assimilator's insightful post here where he said,
God forbid they just play the game and enjoy it, instead of anger-wanking about things that have zero impact on their experience with said game. It's the same kind of brainrot as those who claim they will literally die if they don't play a game on a high-refresh-rate display.

I also fear the drive for more and more cores has the [not-so] hidden agenda by CPU marketing weenies to convince the naïve they MUST have the latest and greatest with the most cores. And without, there will be no way they could actually enjoy playing the game. :(
 
Wow, 4 pages in less than 24 hours. Hot topic?


Apologies to Vayra86 - I replaced your posted value with a variable to illustrate my point. And that is, just because a game is able to "utilize" x cores, that does NOT mean it "needs" those x cores to function, or more importantly, to provide good, "enjoyable" game play.

I see this all going back to @Assimilator's insightful post here where he said,


I also fear the drive for more and more cores has the [not-so] hidden agenda by CPU marketing weenies to convince the naïve they MUST have the latest and greatest with the most cores. And without, there will be no way they could actually enjoy playing the game. :(
Thank you Bill. Well said

His 12700k is probably ~2x your 8700k in something like TLOU / Cyberpunk / Spiderman. GTA V is kinda old but still I expect a 2x even there
Yes but what part of that is attributable to core count rather than just, say, DDR5; more IPC, more frequency?

We know by now core count is barely relevant for gaming tasks. Its really the overall CPU performance on offer here that makes the difference - and (!!!) whether the game is even remotely CPU bound to begin with.

So far nobody has proven you need or really benefit from >8 cores for gaming. And even true 8 core scaling is hard to find. The X3Ds furthermore underline that games are NOT struggling for core count beyond 8 either.
 
Last edited:
Yes but what part of that is attributable to core count rather than just, say, DDR5; more IPC, more frequency?
Well I just tested 6c/12t vs the full chip in TLOU and the difference is massive. Just looking at 0.2% lows, the full chip gets 3 times the lows. Now granted, TLOU is a particularly heavy threaded game - but nowadays lots of games scale like that. I can test it for you if you want, just pick the game and ill compare 8+8 vs 6c/12
 
Well I just tested 6c/12t vs the full chip in TLOU and the difference is massive. Just looking at 0.2% lows, the full chip gets 3 times the lows. Now granted, TLOU is a particularly heavy threaded game - but nowadays lots of games scale like that. I can test it for you if you want, just pick the game and ill compare 8+8 vs 6c/12
Massive theoretical if you push a game into being CPU bound, ergo, unrealistic use cases. You can keep your testing where the sun dont shine.

Take special note of the words 'need' and 'really benefit'.
 
Wow, 4 pages in less than 24 hours. Hot topic?

Topic has been the same since the very first PC tech forum, it's only the part names that change. My PC part is better than your PC part, here are the cherry picked benchmarks to prove it.
 
What is the FPS difference

In GTA V at some areas of the game I would get around 60fps with my i7 6700K @ 4.5Ghz, now with my i7 12700K, P-cores @5Ghz I'm getting around 130fps in those areas...
And overall higher fps throughout the game. GPU stays around 97~100% load, the i7 6700K couldn't do that.
 
Massive theoretical if you push a game into being CPU bound, ergo, unrealistic use cases. You can keep your testing where the sun dont shine.

Take special note of the words 'need' and 'really benefit'.
Even at 1440p the difference in lows remains massive though. If you are not trying to make it CPU bound then you are not talking about cores anymore, you might as well say "games don't benefit from cpus when gpu bound". Well yeah, obviously.
 
Even at 1440p the difference in lows remains massive though. If you are not trying to make it CPU bound then you are not talking about cores anymore, you might as well say "games don't benefit from cpus when gpu bound". Well yeah, obviously.
Bingo, you understood what a CPU needs to do for proper gaming. Well done!

Topic has been the same since the very first PC tech forum, it's only the part names that change. My PC part is better than your PC part, here are the cherry picked benchmarks to prove it.
Yeah it was pretty obvious. CPU perf is a non issue now in gaming for the better half of a decade.
 
Bingo, you understood what a CPU needs to do for proper gaming. Well done!
The topic is about the core count specifically though, it's not about how fast of a CPU you need. 6c is a limit right now and it impacts performance in pretty obvious ways regardless of your GPU. Dropping to below 60 at lows due to 6 cores in TLOU - even at 1440p - in fact im pretty confident that's the case even at 4k.
 
In GTA V at some areas of the game I would get around 60fps with my i7 6700K @ 4.5Ghz, now with my i7 12700K, P-cores @5Ghz I'm getting around 130fps in those areas...
And overall higher fps throughout the game. GPU stays around 97~100% load, the i7 6700K couldn't do that.
Im almost triggered to test it on the 8700K but we havent got same gpu.. curious now :)

The topic is about the core count specifically though, it's not about how fast of a CPU you need. 6c is a limit right now and it impacts performance in pretty obvious ways regardless of your GPU. Dropping to below 60 at lows due to 6 cores in TLOU - even at 1440p - in fact im pretty confident that's the case even at 4k.
TLOU. Okay so we have one game. Maybe 3. Or 4. And then there is the other couple thousand released every year. Still tho I agree with you 6c12t is yesteryears core count at this point. But the overwhelming majority I play still runs over 60 FPS today. More often than not I see 90-100. So maybe TLOU runs at 40 Fps on a recent 4c8t. So what? Even going that low still amounts to playable frames.

Since dx12 we are way past ye olde quadcore paradigm , is my point really. And it doesnt quite apply to the 6 core CPUs either just yet.
 
The topic is about the core count specifically though, it's not about how fast of a CPU you need.
Yes, but "core count" is frequently being used to claim that processor is better simply because it has more cores. And that is simply false - especially as a blanket statement.

It is similar to cylinders in an car engine. Some believe their big V8 must be better because it has 8 cylinders. Yet there are plenty of 4 cylinder cars out there that will blow your socks off!
 
But the overwhelming majority I play still runs over 60 FPS today. More often than not I see 90-100. So maybe TLOU runs at 40 Fps on a recent 4c8t. So what?
I'm sure it does, but when you got eg. 120hz monitor you don't want to play at 60 or 40 fps. Also 1% lows will be substantially lower than that. Is the game playable? Of course, here I am gaming on my switch at 22 fps. That wasn't the question though, the thread is more about how many cores can games actually use, and by use I mean to give you a big difference in performance. Like 10% or greater.
 
I'm sure it does, but when you got eg. 120hz monitor you don't want to play at 60 or 40 fps. Also 1% lows will be substantially lower than that. Is the game playable? Of course, here I am gaming on my switch at 22 fps. That wasn't the question though, the thread is more about how many cores can games actually use, and by use I mean to give you a big difference in performance. Like 10% or greater.
Yeah that is true just the same, but its crossing the line into scientific value if you manipulate the game to make it cpu bound. After all, people cant, dont, nor NEED to buy a super cpu to game proper. Getting every game to run above or at 120hz is a hard thing to crack anyway. Its high expense gaming for questionable returns. The majority of games limit you in some other way. Youll waste shitloads of money just to play the worst optimized games at your desired FPS target.

In my world, that is called stupidity. Others call it a hobby. Where it goes wrong is when the others elevate insane targets to 'the norm'.
 
Well I just tested 6c/12t vs the full chip in TLOU and the difference is massive. Just looking at 0.2% lows, the full chip gets 3 times the lows. Now granted, TLOU is a particularly heavy threaded game - but nowadays lots of games scale like that. I can test it for you if you want, just pick the game and ill compare 8+8 vs 6c/12
This directly contradicts other sources I’ve seen though. Like here:
The delta between the 13600K and 13700K is within expected parameters considering the cache and frequency differential. Other CPUs also fall where expected. Nothing here screams out at me as core-count bound.
 
This directly contradicts other sources I’ve seen though. Like here:
The delta between the 13600K and 13700K is within expected parameters considering the cache and frequency differential. Other CPUs also fall where expected. Nothing here screams out at me as core-count bound.
Double thread count and 7 fps more on the 13900K vs the x3d. Nuff said about cores imho ;) its ONE way to approach more cpu performance, but for gaming clearly the inferior way.
 
This directly contradicts other sources I’ve seen though. Like here:
The delta between the 13600K and 13700K is within expected parameters considering the cache and frequency differential. Other CPUs also fall where expected. Nothing here screams out at me as core-count bound.
Cause the 13600k has 14 cores, not 6. Compare the 7700x to the 7600, or the 10700k to the 10600k, there are differnces of around 16-18% in lows.

8700k vs 9900k, there is a 29% difference in lows.
 
Back
Top