- Joined
- Mar 3, 2017
- Messages
- 2,383
Sub 60 fps…that’s unplayable
![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
-
Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.
Intel Readies Two 22-Core "Nova Lake-S" SKUs with 108 MB bLLC Cache for Gaming
- Thread starter erek
- Start date
littlegreenmenfromhome
Limp Gawd
- Joined
- Aug 27, 2025
- Messages
- 385
It is almost like you ignored the entire post to do a fallacy which tells me you are a waste of time
- Joined
- Mar 3, 2017
- Messages
- 2,383
I forgot the /s, sorryIt is almost like you ignored the entire post to do a fallacy which tells me you are a waste of time
1080p/720p tells CPU performance.
4K tells GPU performance.
How people don't understand this in 2026 is astonishing.
The point of a 1080p (or lower) test is to tell if how CPUs compare when they aren't GPU constrained. A 4k test only tells you if a CPU gets so slow that it isn't GPU constrained. Now I agree that is the relevant thing to think about for a given game, like if you are playing Cyberpunk on a 5090, then your concern is if your CPU is fast enough to handle that. However that is of limited total usefulness. There are two major things it doesn't tell you:
1) How does your CPU compare for games that are more CPU intensive? The games you are playing right now might not be, but perhaps the next ones are more CPU intensive. Testing at a lower rez helps show what you can expect in that case.
2) How will your CPU do with the NEXT GPU you get? Ya maybe your 9900K is fast enough to max out your 3090, which is what they were testing with in the graphs you linked. However what if you go to a 5090? Is it still enough? How about when the 6090 comes out? While we can't look in to the future to an unlimited degree, this helps. It helps tell you that when you get a certain FPS level on your new GPU, well now you are hitting CPU limits. Like on those graphs it would imply that if in Cybepunk you are hitting 72fps with your 9900K, you are now CPU bound. Well guess what? That is now something that could very much be a reality with a 5090. If you turn on Path Tracing you are under that, but with regular RT and using DLSS upscaling (no framegen), you can go above that on a 5090 at 4k... or I should say you can if your CPU can support it. A 9900k wouldn't be able to.
3) Related to that, most people use upscaling these days. If you don't, you should, it looks great and you really are just leaving FPS on the table by not doing it. Lower rez testing helps expose what kind of performance blocks you might have with a CPU that is too slow if you do. You could have a game where you turn on DLSS, FPS goes up by only a couple percent, you say "Man this game sucks with DLSS," when in reality your CPU is nearly maxed at 4k so using DLSS doesn't get you any headroom.
That's why there's 1080p testing: You want to see how CPUs actually compare with each other if the GPU isn't holding them back. Gives you an idea of where you sit in terms of needing a CPU upgrade as new GPUs come out and so on. If benchmarks show that at 1080p your CPU can get lots more FPS than you get gaming at 4k, then cool, you know you aren't leaving GPU performance on the table and you know that you'll probably be good with the same CPU for the next generation of GPU. However if the benchmarks how that 1080p performance near, or the same as, the 4k performance, then you have an expensive GPU being held back by a CPU.
The inverse of this is why they always test GPUs with the biggest CPUs out there, even when testing lower end CPUs. No, you shouldn't pair a 9850X3D with a 5060Ti, you are wasting money on CPU you don't need. However they'll test it that way to give the GPU the best chance to flex its muscles and make sure the results are as GPU limited as possible.
1) How does your CPU compare for games that are more CPU intensive? The games you are playing right now might not be, but perhaps the next ones are more CPU intensive. Testing at a lower rez helps show what you can expect in that case.
2) How will your CPU do with the NEXT GPU you get? Ya maybe your 9900K is fast enough to max out your 3090, which is what they were testing with in the graphs you linked. However what if you go to a 5090? Is it still enough? How about when the 6090 comes out? While we can't look in to the future to an unlimited degree, this helps. It helps tell you that when you get a certain FPS level on your new GPU, well now you are hitting CPU limits. Like on those graphs it would imply that if in Cybepunk you are hitting 72fps with your 9900K, you are now CPU bound. Well guess what? That is now something that could very much be a reality with a 5090. If you turn on Path Tracing you are under that, but with regular RT and using DLSS upscaling (no framegen), you can go above that on a 5090 at 4k... or I should say you can if your CPU can support it. A 9900k wouldn't be able to.
3) Related to that, most people use upscaling these days. If you don't, you should, it looks great and you really are just leaving FPS on the table by not doing it. Lower rez testing helps expose what kind of performance blocks you might have with a CPU that is too slow if you do. You could have a game where you turn on DLSS, FPS goes up by only a couple percent, you say "Man this game sucks with DLSS," when in reality your CPU is nearly maxed at 4k so using DLSS doesn't get you any headroom.
That's why there's 1080p testing: You want to see how CPUs actually compare with each other if the GPU isn't holding them back. Gives you an idea of where you sit in terms of needing a CPU upgrade as new GPUs come out and so on. If benchmarks show that at 1080p your CPU can get lots more FPS than you get gaming at 4k, then cool, you know you aren't leaving GPU performance on the table and you know that you'll probably be good with the same CPU for the next generation of GPU. However if the benchmarks how that 1080p performance near, or the same as, the 4k performance, then you have an expensive GPU being held back by a CPU.
The inverse of this is why they always test GPUs with the biggest CPUs out there, even when testing lower end CPUs. No, you shouldn't pair a 9850X3D with a 5060Ti, you are wasting money on CPU you don't need. However they'll test it that way to give the GPU the best chance to flex its muscles and make sure the results are as GPU limited as possible.
This whole thread is about a CPU
People said they wanted to see benchmarks on the performance of the CPU
The gameheads came in and said they wanted 4K gaming tests not 720 because 720 tests are for stupid people.
It was pointed out that 4K tests are tests of the GPU not the CPU
The gameheads said it doesn't matter because they game at 4K and all CPUs are equal and CPUs testing is stupid
People said they wanted to see benchmarks on the performance of the CPU
The gameheads came in and said they wanted 4K gaming tests not 720 because 720 tests are for stupid people.
It was pointed out that 4K tests are tests of the GPU not the CPU
The gameheads said it doesn't matter because they game at 4K and all CPUs are equal and CPUs testing is stupid
That’s exactly the point. You’re not comparing CPU performance at 4K because you’re GPU limited. The kind of person reading long form reviews in specific computer parts knows this. If you want to understand the raw differences between CPUs, you can’t test them in environments where they’re not completely stressed because the GPU doesn’t feed it info fast enough.
In any case, if you know of anyone who’s gaming at 4K and is budget constrained to the point of wanting to know specifically the level of corner-cutting they can do on a CPU, I’d love to see them.
littlegreenmenfromhome
Limp Gawd
- Joined
- Aug 27, 2025
- Messages
- 385
Yeah, I am comparing it like I play it.
At 4K, chances are you will be GPU limited before you are CPU limited. If that’s what you’re after, look at GPU performance.
that obviously fine, but it is not necessarily the norm for people that spend a lot of gaming computer to play games under 60 fps.
(It is not so much a resolution affair but a target FPS affair per title that make the CPU difference significant or not), some people that spend a lot on a GPU is not necessarily to play game at 4k with all setting set to ultra regardless of how slow it get... but the best setting they can at a minimum fps they target (can be 80-90 for slower game and 144+ for fast paced)
Armenius
Extremely [H]
- Joined
- Jan 28, 2014
- Messages
- 53,802
I thought I had read somewhere that Intel was dropping the E-cores.
it is more dropping the p-core than the e-core, Intel will made unified cores like AMD (with probably the equivalent of zen C, smaller core because less cache that clock lower but 100% exact same ISA capability to the scheduler-os-code), but they will be made by the e-core teams from the e-core base that won the battle.
The IPC per mm of die of the e-core got so incredibly good and started from a fresh architecture at the time, versus the should have been the end of its life a while ago old p-core.