GeForce GTX 1080 Ti: Core i7 7700K vs. Ryzen 7 1800X Linux Gaming Performance

[cj.wijtmans]

Possibly a consequence of games being compiled with more Intel-oriented optimisation settings?

unless compiled with ICC i dont think so. dunno about MSVC tho

[peppercats]

Look at the single thread results here: http://www.anandtech.com/show/11170/...0x-and-1700/22
Ryzen's IPC is comparable to Intel's. But given the clock speed difference, single core performance isn't. It's in the same ballpark, but lagging. Though, not by the margin Michael has found in many games...

It's expected to not perform as well as the 7700k in games, it's a competitor with the 6900k at less than half the cost. Technically, AMD really won't be releasing a direct competitor to the 7700k. The 1500x appears to be in a very weird position unless the leaked specs were false.

Interestingly from those charts, it seems to have better multithreaded scaling than the 6900k. Perhaps a better SMT implementation than Intel's hyperthreading?

[NihilMomentum]

It would be interesting to test the games with Ryzen vs i7 6900k (or a comparable CPU like the 5960k). That would tell if it's a game/port thing (just using 4 cores/threads, for example) or if it's a problem with the platform ATM (since the R7 matches the 6900k).

[indepe]

It would be interesting to test the games with Ryzen vs i7 6900k (or a comparable CPU like the 5960k). That would tell if it's a game/port thing (just using 4 cores/threads, for example) or if it's a problem with the platform ATM (since the R7 matches the 6900k).

That would indeed be interesting. However Michael doesn't have the 6900 and it is quite expensive. I haven't seen such tests anywhere else either, so far.

[NihilMomentum]

That would indeed be interesting. However Michael doesn't have the 6900 and it is quite expensive. I haven't seen such tests anywhere else either, so far.

I believe he has the 5960k... which is almost the same thing.

[Marc Driftmeyer]

I wouldn't touch MSI with a ten foot pole.

[kruger]

Why not tested with MSI B350 TOMAHAWK? It was shown in the other test that it's a bit faster currently.

[JAYL]

for the following test Tomb Raider at Resolution: 3840 x 2160 - Quality: Ultra on Ryzen, how come the average FPS is higher than the maximum?

[indepe]

I believe he has the 5960k... which is almost the same thing.

You are right, if he still has it...even did a core scaling test, back in the time, although it didn't include games... runs at 3 Ghz, so you'd need to extrapolate the results, or just make relative measurements. However, it might just confirm that the games themselves currently don't scale well. That would be good to know, but not very exciting.

[andrei_me]

Michael another interesting thing to test is the environment variable that controls Nvidia driver threaded optimizations, don't remember the name but should be something similar to the description.

This way one can test if the driver can be improved too beside the games to better use the Ryzen cores