digitalFAQ.com Forum

digitalFAQ.com Forum (https://www.digitalfaq.com/forum/)
-   Encode, Convert for discs (https://www.digitalfaq.com/forum/video-conversion/)
-   -   HuffYUV multithreaded + hyperthreading faster? (https://www.digitalfaq.com/forum/video-conversion/7076-huffyuv-multithreaded-hyperthreading.html)

Towncivilian 03-05-2016 09:47 AM

HuffYUV multithreaded + hyperthreading faster?
 
Will using the multithreaded version of HuffYUV on a Pentium 4 Extreme Edition 3.2ghz with Hyper-Threading provide any benefit?

sanlyn 03-05-2016 10:16 AM

No. Huffyuv_MT is pretty ancient these days. CPU's aren't the main bottleneck with this type of video compression. Later huff versions work just as well. As it is, I seldom use huff these days especially for post-processing, as huffyuv doesn't work with YV12. I've replaced it with Lagarith, which is newer and uses multi-threading.

Multi-threading and hyper-threading aren't the same things.

Towncivilian 03-05-2016 10:19 AM

Quote:

Originally Posted by sanlyn (Post 42625)
No. Huffyuv_MT is pretty ancient these days. CPU's aren't the main bottleneck with this type of video compression. Later huff versions work just as well. As it is, I seldom use huff these days especially for post-processing, as huffyuv doesn't work with YV12. I've replaced it with Lagarith, which is newer and uses multi-threading.

Multi-threading and hyper-threading don't mean the same thing.

Thanks for your prompt reply. I will try Lagarith with my next VHS capture since it is a more modern compression algorithm and more compatible. I've been using HuffYUV 2.2.1 single-threaded 32-bit without issue so far, but I haven't done any significant post processing thus far either.

I know that MT and HT are not the same - I was just wondering whether use of a multithreaded encoder still may realize some small benefit of my HT processor. But if it's easier and safer to just use the single-threaded version, I'll continue to use that.

sanlyn 03-05-2016 10:30 AM

huffyuv _MT, plain ole huffyuv, and Lagarith display so little difference in operating speed, given the same setup, that you'll need a timer app to measure a difference. There must be a ton of threads in internet forums where someone measures the performance of a handful of lossless compressors. On average, tests in one thread usually negate tests in some other threads, then somebody will use a different PC and everything changes again. Conclusion: it doesn't make enough of a difference to worry about.
:)

lordsmurf 04-26-2016 02:32 AM

Just to confirm: no.
There is no difference to speed. In fact, the non-original MT/64/etc versions have less support.

Faster = SSD, no CPU, more RAM. That helps. Depends on the task as to which resource is needed.

Towncivilian 04-26-2016 07:29 AM

Thanks for your input. I've got plenty of RAM and a fast-enough HDD (though I use the 500GB for captures):

HP Compaq d530 Convertible Mini Tower
Pentium 4 Extreme Edition 3.2ghz with HyperThreading
4GB DDR-400 RAM
ATI All-In-Wonder 9600XT AGP video card
Creative Sound Blaster X-Fi XtremeMusic sound card
80GB Western Digital Velociraptor WD800HLFS hard drive
500GB Western Digital Caviar Blue WD5000AAKS hard drive
Microsoft Windows XP Service Pack 3


All times are GMT -5. The time now is 10:40 PM

Site design, images and content © 2002-2024 The Digital FAQ, www.digitalFAQ.com
Forum Software by vBulletin · Copyright © 2024 Jelsoft Enterprises Ltd.