platform-configs/novacustom-*: Bump benchmark results for 155h platforms - #1243
platform-configs/novacustom-*: Bump benchmark results for 155h platforms#1243mkopec wants to merge 2 commits into
Conversation
Signed-off-by: Michał Kopeć <michal.kopec@3mdeb.com>
297a148 to
3508a3f
Compare
| ${CRAY_4_K_RENDER}= 326.895 # sec | ||
| ${CRAY_1080_P_RENDER}= 80.547 # sec | ||
| ${ZIP_MULTI_COMPRESSION}= 69245 # MIPS | ||
| ${ZIP_MULTI_DECOMPRESSION}= 47858 # MIPS |
There was a problem hiding this comment.
So the expected CPU performance went down for V56 variants despite configuration being adjusted to increase the performance? I've noticed that COREMARK_SINGLE wasn't updated and thought that's because it's higher here than on OpenBenchmarking.
In general, do we expect 155H behave the same for all devices? If so, maybe all this should be defined once in a file like include/cpu-perf-155h.robot to make updates simpler.
There was a problem hiding this comment.
Done: f30fa95
Updated coremark results to match OpenBenchmarking and moved to separate include.
So the expected CPU performance went down for V56 variants despite configuration being adjusted to increase the performance?
I'm not sure how the original results were obtained. I think we should expect to be within some margin from the average 155H results in any case.
There was a problem hiding this comment.
Forgot to update platform-configs/novacustom-v560tne.robot in the new commit?
Signed-off-by: Michał Kopeć <michal.kopec@3mdeb.com>
2 participants
Update benchmark values with ones from https://openbenchmarking.org/vs/Processor/Intel+Core+Ultra+7+155H .
After recent performance fixes, the performance exceeds the previous values. Set the expected values to average 155h performance.
ref: ncm-2109