karlrupp / microprocessor-trend-data

Data repository for my blog series on microprocessor trend data.

Microprocessor Trend Data

This repository contains the raw data used in my blog series on microprocessor trend data:

• 50 Years of Microprocessor Trend Data
• 48 Years of Microprocessor Trend Data
• 42 Years of Microprocessor Trend Data
• 40 Years of Microprocessor Trend Data

License

All raw data and my own plots are available under a Creative Commons Attribution 4.0 International Public License, see file LICENSE.txt for details.

microprocessor-trend-data's Issues

48 year data link wrong

The link on the main readme.md page says "48 Years of Microprocessor Trend Data" but it links to the 42 year page "https://www.karlrupp.net/2018/02/42-years-of-microprocessor-trend-data/". The embedded diagram is correct, just the link is wrong.

How to cite

Suggestion to add a comment on the README to instruct people on how to cite this work

Which Processor?

This is not an issue but a feature request. It would be great to associate to each data entry (row of the "data" table) the corresponding microprocessor with possibly a hyperlink to its Wikipedia page... Then we could implement interactive plots with the data and interrogate the scatter plot with a datatip.

External memory bus bandwidth, external memory latency, cache size

The widening CPU/memory speed disparity is an important piece of the picture.

Can this information be extracted programmatically from your collection of datasheets or does it require manual labor?

Will there be a 44 (or 48) year version?

Hello. Thanks for sharing this figure (and the corresponding data) with us.
I would like to ask whether you plan a new/updated 2020 version of the figure?

In case you need help, just give me a hint which data sources you used in the past.
(I have already seen some URLs in the newdata.txt file.)

42 years ?

Hello. Thanks for compiling these data.
Is there any reason why the dataset and article mention "42 years" while they actually cover late 1971 to late 2017, i.e. 46 years instead of 42 ?

Wrong number of cores

In newData, the number of cores in the Xeon Phi 7290 is stated to be 272 cores. I think this is a typo. According to the specs. it should be 72 cores only.

This should be an easy fix!

Dataset

Hi Karl,

Hope all is well. I am doing a research project and I need your dataset. I tried to download it but it give me your pdf paper only. Could you please help me with this?

Thanks,
Kim

Recent single-core SPECint data misleading

I know that it's just the way things have progressed in the benchmarking game, but most of the "single-core" speed results use auto-parallelization and do not truly represent single-core performance. The last non autopar'd x86 results were submitted in 2009 and it topped out at 23.1. I would guess that without autopar current systems would be in the ~50 range. So not a 4x increase in performance over 10 years, but closer to 2.