affect of denoising on encode time

HandBrake for Windows support
Forum rules
An Activity Log is required for support requests. Please read How-to get an activity log? for details on how and why this should be provided.
Post Reply
theotherwhitematt
Posts: 3
Joined: Wed Jul 01, 2015 1:53 pm

affect of denoising on encode time

Post by theotherwhitematt »

I'm working on encoding an mkv of Back to the Future into mp4, and the resulting file is around 14 GB while Cars is about 8.5 GB using the same settings. It looks like I can reduce the BTTF mp4 file size be using the denoise filter, but I'm concerned about how much time that may add to the encode time.

On the older hardware I'm doing the encoding on, it already takes about 8 hours to encode the BTTF mkv to an mp4. Will I see any increase in encode time by using the "light" preset for the hqdn3d denoise?

For that matter is the hqdn3d denoise the better setting to use compared to NLMeans? It looks like the less complicated one since there's only one setting to choose instead of two.

Thanks in advance for any help.
Deleted User 13735

Re: affect of denoising on encode time

Post by Deleted User 13735 »

Sure it will take longer.
No way anyone could tell you how much longer.
Please post back with your test results.
kpflugshaupt
Novice
Posts: 63
Joined: Sat Feb 06, 2010 8:00 pm

Re: affect of denoising on encode time

Post by kpflugshaupt »

musicvid wrote:Sure it will take longer.
No way anyone could tell you how much longer.
Please post back with your test results.
You can easily do a test with some part of your source (e.g., just one chapter), and extrapolate.
moneymatt4life
Veteran User
Posts: 440
Joined: Fri Mar 09, 2012 5:26 am

Re: affect of denoising on encode time

Post by moneymatt4life »

nlmeans is slower, but it also yields better results ;) never tried using hqdn3d though :? just use a slower x264 preset and you won't notice the speed hit much :P
theotherwhitematt
Posts: 3
Joined: Wed Jul 01, 2015 1:53 pm

Re: affect of denoising on encode time

Post by theotherwhitematt »

I kicked off the encode last night, and it looks like the denoising changed the encode time from 8 hours to 13 hours. When I get home I'll check what the file size difference ends up being.
moneymatt4life wrote:nlmeans is slower, but it also yields better results ;) never tried using hqdn3d though :? just use a slower x264 preset and you won't notice the speed hit much :P
What do you mean by "better results" and "slower x264 preset"?
moneymatt4life
Veteran User
Posts: 440
Joined: Fri Mar 09, 2012 5:26 am

Re: affect of denoising on encode time

Post by moneymatt4life »

it'll look nicer :P and it's more customizable...
on the Video Tab, there is a slider under Optimise Video... the more left the slider is, the faster the preset, and likewise to the right, the slower the preset (don't use ultrafast/placebo though)... the more right the slider, the better the compressibility/quality/whatever-i-dunno
it's been a while since i used nlmeans, so i can't really remember if i noticed a speedhit since i use veryslow all the time :? so i could be wrong, and i can't test it now unfortunately :(
theotherwhitematt
Posts: 3
Joined: Wed Jul 01, 2015 1:53 pm

Re: affect of denoising on encode time

Post by theotherwhitematt »

So after 14 hours of encoding time the file size for the mp4 was reduced from 14.7 GB to 9.25 GB, and I couldn't see any loss of picture quality for the smaller file size. Since I will still need to encode The Godfather as I go through my collection, what settings would you recommend as a starting point for the nlmeans denoising?
Deleted User 13735

Re: affect of denoising on encode time

Post by Deleted User 13735 »

what settings would you recommend as a starting point for the nlmeans denoising?
Glad you are making progress.
A lot of people are dropping hints that the information you are requesting ("starting point", "the better setting", "will I see", "how much time"), does not exist!

Others have suggested a "weak" nlmeans starting point. I suggest and use the "strong" setting. There 'ya go. Take your pick.

And DO let us know how your tests turn out with your source, your system, your eyes and ears, your expectations, and your viewing/listening environment.
Post Reply