tdiguy 96 Posted December 2, 2019 Author Share Posted December 2, 2019 what quality did you select? So luke, curious, why is it when i use hardware acceleration the file size increases but when I dont the file size decreases as expected? Link to comment Share on other sites More sharing options...
Luke 37007 Posted December 3, 2019 Share Posted December 3, 2019 But it's still under 1.5 mbps, correct? Link to comment Share on other sites More sharing options...
tdiguy 96 Posted December 3, 2019 Author Share Posted December 3, 2019 But it's still under 1.5 mbps, correct?Settings wise they were both identical except the larger file was from having hardware acceleration on. Sent from my moto g(6) play using Tapatalk Link to comment Share on other sites More sharing options...
Luke 37007 Posted December 3, 2019 Share Posted December 3, 2019 I mean the bitrate of the converted file once it was completed. Link to comment Share on other sites More sharing options...
tdiguy 96 Posted December 4, 2019 Author Share Posted December 4, 2019 (edited) I mean the bitrate of the converted file once it was completed. I did not check that and i already deleted it. I dont suppose that is mentioned in the log? Edited December 4, 2019 by tdiguy Link to comment Share on other sites More sharing options...
Luke 37007 Posted December 4, 2019 Share Posted December 4, 2019 Yes you'll see the bitrate throughout the log. Link to comment Share on other sites More sharing options...
tdiguy 96 Posted December 4, 2019 Author Share Posted December 4, 2019 (edited) Yes you'll see the bitrate throughout the log. Ok, so in the log i found this: BitRate":518260 I take it that means that with hardware acceleration on it is not honoring the 1.5mbps setting? The bitrate for the one with hardware acceleration off is: BitRate":1305856, Which is significantly higher but the file is significantly smaller.. Edited December 4, 2019 by tdiguy Link to comment Share on other sites More sharing options...
Luke 37007 Posted December 4, 2019 Share Posted December 4, 2019 Sounds like that value came from the source and not the output. Link to comment Share on other sites More sharing options...
Luke 37007 Posted May 10, 2020 Share Posted May 10, 2020 Are you still having an issue with this? Link to comment Share on other sites More sharing options...
tdiguy 96 Posted May 10, 2020 Author Share Posted May 10, 2020 Are you still having an issue with this?I have not been using hardware acceleration. I will give it a try and see what happens. Sent from my moto g(6) play using Tapatalk Link to comment Share on other sites More sharing options...
tdiguy 96 Posted May 14, 2020 Author Share Posted May 14, 2020 I tested this out on a somewhat small file for testing purpose. The original in .ts format was 30 megs. Without hardware acceleration converting to .mp4 hevc it reduced it to under 6 megs. With hardware acceleration on it was close to 7 megs in size. Settings were left the same so that the only thing tested was hardware acceleration. I am happy with with outcome of this test file the size is at least similar between acceleration and no acceleration. Thank you guys very much. Its hard to believe i can get so much bang for my buck using your software. Now that i wont be pegging the cpu on my server i can set the conversions to just run 24x7 Are you still having an issue with this? Link to comment Share on other sites More sharing options...
Luke 37007 Posted May 14, 2020 Share Posted May 14, 2020 Thanks for the feedback ! Link to comment Share on other sites More sharing options...
hawk223 8 Posted June 17, 2021 Share Posted June 17, 2021 Hi has there been any work done on this I'm currently having this issue too. If I use my 980ti its way faster and uses way less power. Using my CPU my server sounds like a jet taking off and consumes around 300-400W more power. I was consistently lowering file sizes using the software converter then I put my 980ti in the server. Worked ok on my first try but then on another show all of a sudden I had a significant file size increase. Below are some tests I ran on a file. You can see a significant difference between hardware and software encoding. I'm not sure exactly what original quality means and how it selects the settings. Also not sure why such a difference. Would be great to fix this up. Most peoples hardware can play back HEVC now so this saves me a ton of disk space and bandwidth. But I care mostly about space. Thanks. Size, Codec, rate setting, method 365 MB H265 original quality hardware 980ti 239 MB H265 1.5M hardware 980ti 121 MB H265 1.5M software 212 MB h265 original quality software 268 MB h264 original file Link to comment Share on other sites More sharing options...
Luke 37007 Posted June 17, 2021 Share Posted June 17, 2021 16 minutes ago, hawk223 said: Hi has there been any work done on this I'm currently having this issue too. If I use my 980ti its way faster and uses way less power. Using my CPU my server sounds like a jet taking off and consumes around 300-400W more power. I was consistently lowering file sizes using the software converter then I put my 980ti in the server. Worked ok on my first try but then on another show all of a sudden I had a significant file size increase. Below are some tests I ran on a file. You can see a significant difference between hardware and software encoding. I'm not sure exactly what original quality means and how it selects the settings. Also not sure why such a difference. Would be great to fix this up. Most peoples hardware can play back HEVC now so this saves me a ton of disk space and bandwidth. But I care mostly about space. Thanks. Size, Codec, rate setting, method 365 MB H265 original quality hardware 980ti 239 MB H265 1.5M hardware 980ti 121 MB H265 1.5M software 212 MB h265 original quality software 268 MB h264 original file Hi there, yes there has. They're never going to be exactly the same though given the nuances of the different encoders. Link to comment Share on other sites More sharing options...
hawk223 8 Posted June 17, 2021 Share Posted June 17, 2021 2 hours ago, Luke said: Hi there, yes there has. They're never going to be exactly the same though given the nuances of the different encoders. What's changed? Seems something isn't makin sense to me. If I choose 1.5 Mbps shouldn't that dictate the file size? Differences aside I can't understand how the hardware encoded file is basically twice the size of the software encoded one. Link to comment Share on other sites More sharing options...
Luke 37007 Posted June 17, 2021 Share Posted June 17, 2021 7 minutes ago, hawk223 said: What's changed? Seems something isn't makin sense to me. If I choose 1.5 Mbps shouldn't that dictate the file size? Differences aside I can't understand how the hardware encoded file is basically twice the size of the software encoded one. That dictates the bitrate, not the size. When we request a specific upper limit from an encoder, some will come closer to that limit than others, and that is the reason for the difference. Link to comment Share on other sites More sharing options...
hawk223 8 Posted June 17, 2021 Share Posted June 17, 2021 Do you have any suggestions to get similar results? My understanding is that the file should be half the size for equal quality when going from h264 to h265..That's what I was hoping choosing original quality would do. Link to comment Share on other sites More sharing options...
hawk223 8 Posted June 17, 2021 Share Posted June 17, 2021 Did some more testing and I can get the same file size as the software encoder with the original quality setting using the hardware encoder and setting a custom bitrate of 0.8. Interestingly setting it to 1.0 gives the same file size. Perhaps emby needs to be updated to fix this so that the hardware encoder produces a file size that makes sense because it doesn't seem to right now. Link to comment Share on other sites More sharing options...
Carlo 4330 Posted June 17, 2021 Share Posted June 17, 2021 It that 0.8 as in 800KB bitrate? If so, why so low? Link to comment Share on other sites More sharing options...
hawk223 8 Posted June 18, 2021 Share Posted June 18, 2021 23 hours ago, cayars said: It that 0.8 as in 800KB bitrate? If so, why so low? Just testing to see what it took to make the file size the same as h264. Since H265 is supposed to be around half the size for the same quality it would make sense the bitrate should be half correct? Link to comment Share on other sites More sharing options...
Carlo 4330 Posted June 18, 2021 Share Posted June 18, 2021 It's not 1/2 the size always. Sometimes a conversion can get bigger. Sometimes you only save 10%, some times 80%. On average you likely save 30% to 50% but it depends on the media you start with. I wouldn't base anything on a small sample size of just a couple files. Try 25 files of various types to get an idea. Personally when I convert my media to HEVC I do not set any upper limit as that just decreases quality. Instead I do the conversion and only replace the file if it's 10% smaller. I keep HEVC in MKV containers and AVC in MP4 containers for easy knowing what is what and for being able to have them side by side for comparison. Link to comment Share on other sites More sharing options...
tdiguy 96 Posted June 19, 2021 Author Share Posted June 19, 2021 16 hours ago, hawk223 said: Just testing to see what it took to make the file size the same as h264. Since H265 is supposed to be around half the size for the same quality it would make sense the bitrate should be half correct? I am not sure how well the bitrate adjustment works when using hardware acceleration. Link to comment Share on other sites More sharing options...
Carlo 4330 Posted June 19, 2021 Share Posted June 19, 2021 It works fine but it's an upper limit of what bitrate can be used. Link to comment Share on other sites More sharing options...
hawk223 8 Posted June 21, 2021 Share Posted June 21, 2021 (edited) On 6/18/2021 at 3:03 PM, cayars said: It's not 1/2 the size always. Sometimes a conversion can get bigger. Sometimes you only save 10%, some times 80%. On average you likely save 30% to 50% but it depends on the media you start with. I wouldn't base anything on a small sample size of just a couple files. Try 25 files of various types to get an idea. Personally when I convert my media to HEVC I do not set any upper limit as that just decreases quality. Instead I do the conversion and only replace the file if it's 10% smaller. I keep HEVC in MKV containers and AVC in MP4 containers for easy knowing what is what and for being able to have them side by side for comparison. Well it costs a lot of power to convert using the software converter but it does work as expected with a significant size savings. My power bill went up significantly. Also my server sounds like a jet taking off. If I'm not getting a significant saving it's not worth doing. The hardware encoder uses hardly any power and my server is relatively quiet but sometimes the files get significantly larger. Something just isn't working right in my opinion. I don't want to take the time to understand exactly what's happening and what's different either. I just want it to work as expected. A decent size saving and original quality as selected. Edited June 21, 2021 by hawk223 Clarification of software encoder saving space. Link to comment Share on other sites More sharing options...
Luke 37007 Posted July 28, 2021 Share Posted July 28, 2021 @hawk223 have you tried just lowering the bitrate on the conversion job? Link to comment Share on other sites More sharing options...
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now