Support NeoGAF

[Insane Metal]

https://translate.google.com/transl...m-chimera-squad-benchmark-test/2/&prev=search

[/URL]

There has never been such a difference in a game afaik.

CPU benchmarks show interesting results

It's hard to believe, but the editors reviewed it several times - always with the same result. By far the fastest processors in XCOM: Chimera Squad are all offshoots of AMD's current Zen 2 generation. That in itself would not be so surprising, since Zen 2 is sometimes ahead of the Intel derivatives even in games. So far, however, there has never been such a huge lead as in the new turn strategy game.

The Ryzen 9 3900X in Full HD on a GeForce RTX 2080 Ti FE delivers 28 percent more FPS and 29 percent better frame times than a Core i9-9900K. Since the Ryzen 5 3600 is almost as fast, AMD's mid-range model also places itself ahead of Intel's consumer flagship.

 

[diffusionx]

RyzenChads assemble

 

[DoctaThompson]

RyzenChads assemble

Lmao! I'm an Intel guy personally. Started off Intel, went AND for a while, back on Intel, and might go back to AMD this time around. Only for cpu of course. Their gpu's have fallen to the wayside, and my last one from them was a 390x. But interesting benchmark nonetheless.

 

[Dane]

Perhaps developed with Zen+ and 2 in mind?

 

[Unknown Soldier]

There's some speculation that the X-Com engine really likes large cache for some reason. There are other games which are also faster on Ryzen but really a lot of times random things like this will never be known.

One of the great mysteries of a bygone era was how Quake games always ran faster on Pentium 4 than on Athlon 64 even though the Athlon 64 was trashing the Pentium 4 in all other games. Carmack even commented on it at one point and admitted he didn't know why this was the case either, and he was the guy who wrote the engine!

 

[Insane Metal]

There's some speculation that the X-Com engine really likes large cache for some reason. There are other games which are also faster on Ryzen but really a lot of times random things like this will never be known.

One of the great mysteries of a bygone era was how Quake games always ran faster on Pentium 4 than on Athlon 64 even though the Athlon 64 was trashing the Pentium 4 in all other games. Carmack even commented on it at one point and admitted he didn't know why this was the case either, and he was the guy who wrote the engine!

Interesting take on the cache thing.

Now this story about Carmack I have never heard! That's quite a mistery hahah

 

[Solarstrike]

Any comparison tests running it at 4k (3840 x 2160p) and higher, resolutions?

 

[diffusionx]

Any comparison tests running it at 4k (3840 x 2160p) and higher, resolutions?

CPU tests are normally conducted at low resolutions because you don’t want the GPU getting in the way.

 

[GHG]

Any comparison tests running it at 4k (3840 x 2160p) and higher, resolutions?

No point, if you have a modern CPU everything tends to be GPU bound 1440p and above in the vast majority of games.

 

[D.Final]

https://translate.google.com/transl...m-chimera-squad-benchmark-test/2/&prev=search

[/URL]

There has never been such a difference in a game afaik.

The hight specs

 

[Mista]

God bless AMD. Real talk though, it all depends on the game at the end. I saw lots of tests of games and some of them performed better on AMD CPUs and some of them performed better on Intel CPUs

 

[Insane Metal]

God bless AMD. Real talk though, it all depends on the game at the end. I saw lots of tests of games and some of them performed better on AMD CPUs and some of them performed better on Intel CPUs

It's usually close in the end, but this result is quite interesting. The 3600 is almost on par with the 3900X, which is quite weird.

 

[Mista]

It's usually close in the end, but this result is quite interesting. The 3600 is almost on par with the 3900X, which is quite weird.

Yeah thats an interesting benchmark and kinda weird at the same time. The gap isn’t justifiable

 

[MightySquirrel]

It's usually close in the end, but this result is quite interesting. The 3600 is almost on par with the 3900X, which is quite weird.

It's a safe guess 195 fps avg is 2080 ti ceiling in that game. [3900x would be even further ahead probably]

 

[RedVIper]

It's a safe guess 195 fps avg is 2080 ti ceiling in that game. [3900x would be even further ahead probably]

Yeah, they really should run the game at a lower resolution so they can actually test the cpus.

 

[Urprobablyright]

I remember when i was a kid, i really wanted a Radeon 9[something] graphics card. I can't remember what got me on to it so hard, but i thought it was the coolest thing ever.

#myinput

 

[Armorian]

God bless AMD. Real talk though, it all depends on the game at the end. I saw lots of tests of games and some of them performed better on AMD CPUs and some of them performed better on Intel CPUs

Usually heavy multithreaded games performed great on AMD cpus, this one doesn't use more than 8 threads (7700k and 9900K results are the same).

 

[h3ad0rZ]

I'm not really an expert, but didn't Intel had some security issues with their CPUs which required a patch that would lower their performance? Could that be related in any way?

 

[Kagey K]

Dev tools are starting to favor AMD because of consoles and Nivida can’t brute force their way out?

Seems to be the only real answer.

Nividia really screwed themselves by not playing nice with MS after the OG Xbox deal.

 

[Dane]

Dev tools are starting to favor AMD because of consoles and Nivida can’t brute force their way out?

Seems to be the only real answer.

Nividia really screwed themselves by not playing nice with MS after the OG Xbox deal.

Its CPU side. AMD has been pushing foward what Intel didn't because they held the monopoly on mid to high end for +10 years. To think that back in 2016, intel high end mainstream offer was a 4 core 8 thread CPU and it was going with that since 2009.

 

[Rentahamster]

I'd like to hear what the devs have to say about this. Intriguing stuff.

 

[PSlayer]

Expect more and more of this as developers stop using only 4 cores/8 thread in their games.

 

[Rentahamster]

Expect more and more of this as developers stop using only 4 cores/8 thread in their games.

The 9900K is 8 core though.

 

[Leonidas]

I'm not sure if I'd believe the results from the OP without more places testing the game. And if it is a cache related thing hopefully some place will test the game with the 5675C/5775C as those CPUs have a lot of cache. Maybe HEDT would perform well too as those have more cache than 9900K.

This result of the same game shows Intel in the lead for minimums but with Ryzen slightly ahead on averages.

 

[Rentahamster]

I have a 5960x HEDT chip. Does that have a lot/same of cache, relatively speaking, compared to the new Ryzens? I'm not able to look up specs the moment.

 

[Leonidas]

I have a 5960x HEDT chip. Does that have a lot/same of cache, relatively speaking, compared to the new Ryzens? I'm not able to look up specs the moment.

Compared to the new Zens no, but it does have 25% more than the 9900K.

EDIT: 5960x is on the chart I posted, didn't seem to help...

 

[michaelius]

Expect more and more of this as developers stop using only 4 cores/8 thread in their games.

There's 4 fps difference between Intel 4/8 7700K and 8/18 9900K - there's clearly something else going here but I have no idea what is it.

 

[Rentahamster]

EDIT: 5960x is on the chart I posted, didn't seem to help...

Ah, so it is. Neat.

 

[alabtrosMyster]

I'm not sure if I'd believe the results from the OP without more places testing the game. And if it is a cache related thing hopefully some place will test the game with the 5675C/5775C as those CPUs have a lot of cache. Maybe HEDT would perform well too as those have more cache than 9900K.

This result of the same game shows Intel in the lead for minimums but with Ryzen slightly ahead on averages.

The funny thing is that according to this benchmark if you have a 60hz monitor ALL CPUs on the chart are functionally the same (all have their % low above 60fps).