Silly thought - HT v no HT

[FatRakoon]

This is just a silly thought, so nothing serious.
I have noticed when converting Videos that my CP has only really gone to the 40-49% mark. I sometimes run 2 apps to do 2 of them, and then the PC can often sit at close to 100% and indeed I can still use the PC just fine without any noticeable drop in speed ( unless Im gaming etc of course )

Now, just for a giggle, I did a few mess abouts on an I5 ( 4 Cores no HT ) and that was plopping along at 80-90%

I have also played with this PC with HT off and sure 80-90%

This made me think ( perhaps ignorantly ) that the program or the system, is simply ignoring HyperThreading? when coding but using it to find out the CPU useage?

Dont get me wrong, I dont particularly care if it is or not. I am simply converting a DVD Collection of War documentaries so I can add them to my Media Player ( Yes, its all legal stuff ), but its got me curious.

But it just so happens, that iots not really any quicker when using hyperthreading, unless I actually force 2 versions of the app

Just sillyness as I said.

 

[Cyber-Mav]

Depends on complexity of video you encode and codec used. If doing sd video on x265 you may not see full core loading. On 1080p output of animation i get all threads active and cpu load at 99% or 100 on my 5600x. But did notice a lot lower usage on low rez so bumped up preset from slow to slowest. I dont use placebo mode.

You may find that its better to run 2 instances of the encoder like i did for low rez dvd stuff to get more utilisation

 

[FatRakoon]

Yeah, I ripped the DVDs with DVDShrink 3.2
Then used Bigasoft Total Video Converter to MPG using HI on Video and AUDIO, keeping the same codec as DVD, so its pretty much MPG to MPG and as close to the DVD Standard as I can get it, although it has shrunk each one down from about 2.5GB to below 1GB ( 750-950MB ) so that means that its definitely NOT the same codec.

But in truth, I have seen this before and wile, I cannot be 100% sure, I am fairly sure that the codec does make the difference!

I do find it ironic really, but The time it takes does not really bother me and there is not really all that much difference in most things no matter what mode I convert to. I have however found, a level of coding that is just fine and it does lose quality, but not enough to annoy me. This one I wanted the best quality and so I have kept the quality as much as I can and the resolution, and the ourput files are NOT at all noticably less but are physically smaller.
Yeah, what I did, was, since its an 8 DVD collection, I set 8 copies of teh program up, each to do each DVD, and when it finished the convertion, it quit, and so I just left them all on the screen and every few minutes, it would beep and then I shrunk the windows I was on, and then clicked on GO and then that would do the next DVD.
The whole convertion for each DVD was only about 2-3 minutes, so it was no biggie this time.
I am going to have a mess about however with these same DVD files, but I am going to copy them to my Server for a giggle and then convert all 8 at the same time, just to see how long it takes? The Server is a 64Core ( 4x16 Core CPUs ) with 512GB RAM ( It has a 32GB RAM DISK so I will put them all into that and convert purely from/to the RAM DISK.
I love doing uneccessary stupid things.

 

[Tetras]

This made me think ( perhaps ignorantly ) that the program or the system, is simply ignoring HyperThreading? when coding but using it to find out the CPU useage?

Yeah, I think it is usually counted, because when HT is on they're seen as cores (even though they're not). But, I'm not sure if that is true of everything that measures CPU usage.