I was surprised to find out how many people play at 2560x1080, on 1080/1080Ti's - guessing they want UW screen with 144HZ on the cheap, and care more about FPS/Refresh/Immersion than IQ.
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Why Benchmarking CPUs with the GTX 1080 Ti is NOT STUPID!
- Thread starter gavinh87
- Start date
I was surprised to find out how many people play at 2560x1080, on 1080/1080Ti's - guessing they want UW screen with 144HZ on the cheap, and care more about FPS/Refresh/Immersion than IQ.
I'm shocked that people except such dumbass comments on this forum that my comment was taken seriously.
Yeah, why do you think the video was posted in the first place?
So absorbed by perceived fanboyism that they cannot even see the sarcasm in your post.
says humbug.......
Notice i didn't react to it, i maintain my sense of
Sure, It simulates higher performance GPU's relative to the CPU.
The GPU is pushing out perhaps 30 to 40% higher Frame Rates than it is a 1080P, If the CPU cannot keep up with that then it cannot keep up with future GPU's that are perhaps 30 to 40% faster.
So if you have a 1080TI now and in a years time might want to up grade it to an 1180TI you will know if the CPU is fast enough to keep up with the new faster GPU.
Nobody would expect you to get it, a potato could push that 290
TBH it's more a test of the graphics driver, API and game engine performance.
But it doesn't at all. This is the problem it just makes current trends larger. Test a 4core Vs 8 core on a 2013 game the 4 core will fare fare better than a 2016 game will show. It shows nothing of real value unless you are planning maybe on running the same game at 200 FPS next year and only 120fps now.
As coding habits and game engines change we always have seen better utilisation of hardware so it's realistic to expect better utilisation and makes these tests moot.
I mean run these 720p tests or why not even 480p on a 7700k at 5ghz compared to a 1800x @ 4.0. the 7700k demolished it historically when we did them last round. So would you say a 7700k is better option second hand than a 1800x for future use in gaming?
Most likely you would recommend the 8 core over the 4 core despite less frames at 720p
How much does Intel pay you?
The 7700k is a good (maybe the best) choice if you want to pair it with a 1080Ti and run at resolutions from 10 years ago with DX11. If you want to build a modern PC then look elsewhere.
Right now the 7700K is as good as if not even better than the 8700K.
I would recommend the 8700K all day everyday as the best gaming CPU, by the same token i would recommend the Ryzen 2600X over the 7700K because the 7700K being a 4 core is already on its way out, even with SMT.
That is not a contradiction, this, believe it or not is nothing unusual with the 7700K and others like it, look at the thread utilisation, this thing is right on its limits.
The fact that you know plenty of people with clearly too much disposable income doesn't mean it's the norm.
"Most gamers" do not have a GTX 1080 (Ti), stop spouting this nonsense.
Associate
- Joined
- 31 Oct 2012
- Posts
- 2,243
- Location
- Edinburgh
As others have covered in many previous discussions videos and threads, historic data proves using high end GPUs on lower resolutions incorrectly suggested some CPUs would be more future proof.
The best option is realistic combos - the trouble is, what are they? Some claim pro gamers all use 1080tis on low res (seems unlikely as many pro gamers share details of their machines and it's seldom a 1080ti, and many of them stream and it seldom is low res, low settings as claimed) but most people are not pro gamers anyway.
It is fair to say "realistic combinations" vary widely so a range is probably a good idea e.g. test cheap CPUs with a cheat GPU and a mid-range, test mid range CPUs with a mid range GPU and an expensive one etc. Even this is just going to cause arguments - what is mid range? Which GPU? Etc etc. As different GPUs and drivers can have wildly different performance profiles including CPU loading, it'll always be an approximation. Unfortunately we are not in a place yet where we can get good reviews that cover lots of games and do all the permutations with the hardware. I suspect we won't get there any time soon but who knows?
The best option is realistic combos - the trouble is, what are they? Some claim pro gamers all use 1080tis on low res (seems unlikely as many pro gamers share details of their machines and it's seldom a 1080ti, and many of them stream and it seldom is low res, low settings as claimed) but most people are not pro gamers anyway.
It is fair to say "realistic combinations" vary widely so a range is probably a good idea e.g. test cheap CPUs with a cheat GPU and a mid-range, test mid range CPUs with a mid range GPU and an expensive one etc. Even this is just going to cause arguments - what is mid range? Which GPU? Etc etc. As different GPUs and drivers can have wildly different performance profiles including CPU loading, it'll always be an approximation. Unfortunately we are not in a place yet where we can get good reviews that cover lots of games and do all the permutations with the hardware. I suspect we won't get there any time soon but who knows?
thats my exact point, in the future the 1700 will gain while the 7700 can only stay the same in terms of overall use. And these game benchmarks are usually run on a test bench with only the game and skeleton windows running, the fps on the 7700k might be lower with 8 chrome tabs, anti virus and spotify in the background.Right now the 7700K is as good as if not even better than the 8700K.
I would recommend the 8700K all day everyday as the best gaming CPU, by the same token i would recommend the Ryzen 2600X over the 7700K because the 7700K being a 4 core is already on its way out, even with SMT.
That is not a contradiction, this, believe it or not is nothing unusual with the 7700K and others like it, look at the thread utilisation, this thing is right on its limits.
People seem to think im on intel's side, im just on the lower utilization is typically better than current 10% lead.
My next rig will be a ryzen or intel 8 core minimum so I dont get caught as I dont update my cpu more than once every 4 years or so. At the mo looking ryzen...
my whole point is testing a scenario that would only happen in 2006 isnt appropriate way of discerning future performance gaps it only really shows what would happen if these chips were thrown in a time machine to the mid 2000s
the point is while you probably have the pc which is faster ? generally the i7 for eg will remain faster for the whole period most of us here own it. the amd option may catch up at a later date a bit or even over take a older cpu but what does that matter if you used your cpu been faster all through that period you had the pc and then upgrade.being faster at the end of the life cycle is utterly pointless.
So why buy CPUs based on GTX 1080 Ti benchmarks done at 1080p then?
So if it's not about future performance trends or pure gaming why not base your choice off cinebench or passmark
edit: this is not me genuinely saying use synthetics for day to day performance I'm trying to make people realise the further you go from what you actiactu do and the more contrived the test the less useful it is as a whole.
The whole point of a test is representing a use case the only thing gained by this type of testing is max fps which is only really applicable to hi hz gamers. But it is always used as something different even you seem to think bieng able to push 180fps instead of 173 is important. Well it's not to most people it's about minimums rather than synthetic maximums when they are using the machine...
Last edited: