i7-8750h HyperThreading on or off? proofs inside.

[chumacher]

Hi,

Using HyperThreading in most games is useless. (if you have 6 cores or more)
Disabling gives you same FPS or even better in popular eSports FPS title CSGO.

You can achieve the same amount of FPS in most games with 6 cores without HyperThreading and the best thing is, less power, means less heat.

Video for proof, sorry for quality, honor 6 camera is not the best.




ps. there is info @ video description.
I don't have many newer games in my steam library. Doom 2016 is newest FPS that I do own. Maybe I Will test that next.

[xeromist]

FULLMETALJACKET7 wrote:
Oh well, where do I begin?

First of all, this isn't the correct subforum to post something like this. This is the GL series subforum.

Second, you proved nothing other than CSGO runs better without HT,which is no news. I said that on the other thread. It runs on an ancient engine that takes no benefit from it.

Third, that's a sample test size of 1. You can't make any conclusions out of it.


Do some proper testing.

I don't have a problem with him posting it here. It's pointless to do this on a desktop so the notebook area seems as good as any.

As long as everyone keeps things civil this forum is open to discussion. It may not make sense for most people to play with these settings but if he wants to test and show situations where it could make a difference (good and bad) then I'll let him.

[xeromist]

This is a start. You've demonstrated a difference under heavy load but I'd argue that's when you need the extra oomph, not to kneecap your processor. I'm not sure it's worth modding the BIOS and having to restart to switch either. As your CB runs demonstrate you get worse performance with CPU intensive tasks so it's a trade off for questionable gain. For older games and GPU bound games you might not see much difference in playability but there are games that really hammer the CPU and you'd have to switch back if you want to play those with the best performance.

The machine isn't overheating in either scenario so the temps don't really matter. Power might be a different story but even that is mostly relevant when on battery. You may want to run a few tests on battery and see what performance is like. Unfortunately battery drain is a pain to test because you have to have a consistent load. People used to use Battery Eater but I'm not sure what the tool of choice is these days. Maybe just loop Youtube? Not sure.