- The "Top CPU Performance to Date" has flatlined, so even if the hypothesis that developing countries are running PassMark is true, it doesn't explain everything.
- The recent failures of high end Intel chips might be a factor.
- This data is after the initial round of developers being replaced by AI and outsourcing.
- PassMark doesn't include AI or ML benchmarks that the next generation hardware prioritizes.
- CPU top clock speeds have been going down for more than a decade now, instead prioritizing the number of cores.
- HPC labs are less likely to use PassMark because of licensing issues whereas free and open source alternatives exist.
- The average consumer doesn't know or care about the difference. Instagram is still limited to around 1000x1000 pixel images and at most 1080p video with 30 fps.
Personally, this signals that PassMark is not a good benchmark because I've seen a significant increases in my applications from recent CPU improvements like with AMD's EPYC offerings.
- The "Top CPU Performance to Date" has flatlined, so even if the hypothesis that developing countries are running PassMark is true, it doesn't explain everything.
- The recent failures of high end Intel chips might be a factor.
- This data is after the initial round of developers being replaced by AI and outsourcing.
- PassMark doesn't include AI or ML benchmarks that the next generation hardware prioritizes.
- CPU top clock speeds have been going down for more than a decade now, instead prioritizing the number of cores.
- HPC labs are less likely to use PassMark because of licensing issues whereas free and open source alternatives exist.
- The average consumer doesn't know or care about the difference. Instagram is still limited to around 1000x1000 pixel images and at most 1080p video with 30 fps.
Personally, this signals that PassMark is not a good benchmark because I've seen a significant increases in my applications from recent CPU improvements like with AMD's EPYC offerings.