AMD Ryzen 7 1800X vs. Intel Core i7 7700K Linux Gaming Performance

[mitch074]

It's not the first time a brand new chip needs tinkering with its firmware - personally I'll wait for later revisions of the core (my Haswell CPU is still working well), but my next rig will probably be a 6- or 8-core AMD chip, except if Intel decide to drop the price on their 8-core i7's - something I doubt will happen. Intel are much more likely to enable HT on i5s and drop the price of that generation - it only requires a firmware update and they can spin it as a new chip for that reason alone. After all, they just did that with the Pentium G4600 which is basically an i3 with both AVX and 1Mb of cache disabled.

[mibo]

Michaels test sounds too negative to me.
One could also look at the min-fps (where provided) and hail Ryzen for significantly improving that *important* number.
Showing that Ryzen delivers 200fps instead of 300fps in cpu bound situations...

There was a reddit Q&A where even AMD Lisa commented on the 1080p gaming performance.

[2ezhi]

My conclusions from watching and reading reviews an benchmarks - for pure gaming i7-7700k is still king of the hill. But for users who needs multithreading and dream processor was i7-6900k Ryzen 7 lineup is very tempting and appealing. Especially because it costs half of 6900k and has lower TDP.

[andrebrait]

Before you guys judge acpi-cpufreq's quality, please notice that Ryzen also has problems with Windows, where the Performance power preset doesn't make a difference for the Intel CPUs but results in a boost of uo to 10% for the Ryzen CPUs (compared to Balanced).

There's some bugs AMD has to fix, it seems.

[dungeon]

tomshardware is always thorough !

OK i readed that too and what i found there that 1800X exceed 95W TDP with some workloads... but that is probably safe and expected as Wraith Max cooler is also rated at 140 W

But somewhat that also means to me how even better cooling solution than Max is needed if someone wanna OC it more than 3.8 and to be total safe +200Mhz made +30W in Luxrender Time there what is already at WMax rating... it is not surprising overclocking failing for some people

But OK OC is not for everybody, one need to know and test what he is doing

[FireBurn]

The Vulkan results were very surprising, you'd have thought that having the load spread across all the cores would have boosted speeds. What I'm guessing has happened is the load was so little over each core, the CPU has been down clocked

Would be great if these tests could be repeated using the Performance governor

[Filiprino]

Once BIOSes have the possibility dto disable cores, I'd like to see tests with 4 cores disabled.

TDP == Thermal Design Power. That's the heat the processor can emit. But the actual power consumption is higher.

[dungeon]

Once BIOSes have the possibility dto disable cores, I'd like to see tests with 4 cores disabled.

Why that? Just wait for something like Ryzen 5 1400X - that would be 1800X cut in half for $199 probably

Single stock would still have same perf (it would be even easier to be clocked more), only you would just lose half threaded perf.

[Michael]

Once BIOSes have the possibility dto disable cores, I'd like to see tests with 4 cores disabled.

TDP == Thermal Design Power. That's the heat the processor can emit. But the actual power consumption is higher.

I can control cores, I will have tests up this weekend.

[zboson]

Apparently AMD has suggested to disable SMT for gaming benchmarks - seems to have made a marked difference for some games on the Windows benchmarks I have seen. It would also be cool to see how it compares to the FX-8370 like the previous article

Good point. I wonder if the performance difference being seen is related to the thread topology. Last time I looked into thread topology of AMD bulldozer based processor the default topology I noticed on Linux and Windows was compact. In other words if you asked for two threads it would pack two threads to one module. That's normally going to much less efficient than putting two threads on two different modules (especially as bulldozer based processors only have one FPU per module). However with Intel hyper threading on each Linux system I have tested the topology is scatter. Each thread goes to a separate core until all cores have a thread and only after can two threads go to the same core.

So what could be happing with Ryzen is that an application/game asks for two threads and those go to one core or four threads which only go to two core, eight threads to four cores. That could account for a factor of 2 difference with Intel.

OpenMP by default (at least on all the systems I have tested it on) will use all logical processors. So on Ryzen it would use 16 threads. That could explain why OpenMP based rendering applications perform so well. If you use the maximum threads (16) then there is no problem with topology but if you use less than 16 threads the topology can have a big effect on performance.

Edit: I just checked cpuinfo and I think it's compact for Ryzen but scattered for Intel

On my Skylake 4C/8HT system with Ubuntu 16.10

grep "core id" /proc/cpuinfo returns

core id : 0
core id : 1
core id : 2
core id : 3
core id : 0
core id : 1
core id : 2
core id : 3

But the same command with the 1800X returns
http://openbenchmarking.org/system/1...01800X/cpuinfo

core id : 0
core id : 0
core id : 1
core id : 1
core id : 2
core id : 2
core id : 3
core id : 3
core id : 4
core id : 4
core id : 5
core id : 5
core id : 6
core id : 6
core id : 7
core id : 7