- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I'd like to understand how the same 24 EU IGP performs differently on ultra-low power Y series processors all the way up to high power actively cooled desktop processors.
If the IGPs are functionally identical (internally) then what is being throttled?
How is throttling being expressed?
Are there any tools that can identify when throttling occurs?
Thanks!
-A
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
There may be other tools, but as far as I know the main thing to watch is the GPU clock rate. In Windows, you can use tools like Intel GPA or GPU-Z. In Linux, you can use the Media Server Studio system analyzer tool or read the clock speed directly from /sys/class/drm.
For power, the GPUs usually use so much less power than the CPUs that they can just go up to their max frequency (published for each processor at ark.intel.com). In other words, in practice I've seen the CPUs throttle but it is rarer to need to throttle GPUs.

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page