Welcome to Doom9's Forum, THE in-place to be for everyone interested in DVD conversion. Before you start posting please read the forum rules. By posting to this forum you agree to abide by the rules. Domains: forum.doom9.org / forum.doom9.net / forum.doom9.se |
|
|
#345 | Link |
|
Registered User
Join Date: Nov 2009
Location: Northeast Ohio
Posts: 447
|
2015 re-upload of my F-Zero GX test video is now up:
https://www.youtube.com/watch?v=DFy6yKkV-Jg Note however that, as of this post, there are no VP9 encodes just yet. For reference here's the 2014 version: https://www.youtube.com/watch?v=BLxHjixnf3c EDIT: Well here's a surprise, the 720p DASH h.264 encoding for the 2015 version looks quite a bit better than the very same format in the 2014 version! And yet they both practically use the same bitrate. EDIT 2: Uhhh... is it just me or does the 2015 version of the non-DASH fmt22 format look worse compared to the same format on the 2014 version? O_o EDIT 3: Now I know it's not the best idea to compare video quality via screenshots, but I just know I'm going to get asked for them anyway (like I was previously). For reference, screenshots A and B are the same exact frames that I used for the screenshots back on page 2 of this very thread but this time I've added in a third screenshot as well. [Source] MP4 AVC (6169kbps): Screenshot A ~ Screenshot B ~ Screenshot C 2014 DASH AVC (2112kbps): Screenshot A ~ Screenshot B ~ Screenshot C fmt22 AVC (2886kbps): Screenshot A ~ Screenshot B ~ Screenshot C WebM VP9 (2182kbps): Screenshot A ~ Screenshot B ~ Screenshot C 2015 DASH AVC (2125kbps): Screenshot A ~ Screenshot B ~ Screenshot C fmt22 AVC (2894kbps): Screenshot A ~ Screenshot B ~ Screenshot C WebM VP9 (1686kbps): Screenshot A ~ Screenshot B ~ Screenshot C Last edited by Nintendo Maniac 64; 3rd May 2015 at 22:26. |
|
|
|
|
|
#346 | Link | |
|
Registered User
Join Date: Oct 2009
Posts: 938
|
Quote:
|
|
|
|
|
|
|
#347 | Link | |
|
Registered User
Join Date: Nov 2009
Location: Northeast Ohio
Posts: 447
|
Quote:
I'll probably make a better source encode at some point in the future seeing how the original recording was 1080p (and in lossless too!), but I felt that I needed to at least upload the original encode for comparison reasons. Also, even this mere 4-minute not-so-great 720p 30fps encode (which is 200MB) took 50 minutes to upload...yay for my DSL only having ~75KB/s upload. Nevertheless, in the YouTube video descroption(s), you can find a link to the original uploaded MP4. For reference, said MP4 does still look considerably better than the YouTube encodes, but remember - this is a video of a 12 year old GameCube game, so any textures or the like won't be that high of a resolution. Last edited by Nintendo Maniac 64; 3rd May 2015 at 05:26. |
|
|
|
|
|
|
#348 | Link |
|
Registered User
Join Date: Jul 2011
Posts: 1,121
|
I looked and compared the old and new vp9 versions.
If i am not doing something wrong, the new was is (for most part) quite a lot better in quality. It's really noticable when there is small detail, old one is a blurfest compared to the new one in most cases. EDIT: Question: Was the video originally a lossless upload? Last edited by zerowalker; 3rd May 2015 at 10:49. |
|
|
|
|
|
#351 | Link | |
|
Registered User
Join Date: Nov 2009
Location: Northeast Ohio
Posts: 447
|
The VP9 encodes have come in now (they weren't there 12 hours ago) so I'll get right on that. And since you guys are quite aware now that the original upload was not lossless, I'll probably stick in some screenshots from that and maybe even add in some actual screenshots from the lossless recording as well.
Quote:
EDIT: Screenshots! [2015] WebM VP9: Screenshot A ~ Screenshot B ~ Screenshot C EDIT 2: Well from the screenshots alone the quality really doesn't seem all that different from the 2014 encode...only Screenshot A really looks distinctly better on the 2015 version, and even then I wouldn't really say it's by that much. EDIT 3: mzso, I would greatly appreciate it if you didn't quote my post in post #346 that you made - that way I can edit and update my post and the like without there being an non-updated version quoted directly below it... EDIT 4: Holy crap, no wonder the 2015 VP9 encode doesn't look much different - the bitrate is only 1686kbps! Wow, there's definitely been some quality improvements if they can reduce the bitrate by 25% and still have about the same quality. Man, if only I could have tested 1080p...well maybe I could, but it would have to be a really short video - like only 30 seconds at most. EDIT 5: Remember, it wasn't the greatest encoding... [Source] MP4 AVC (6169kbps): Screenshot A ~ Screenshot B ~ Screenshot C Last edited by Nintendo Maniac 64; 3rd May 2015 at 22:26. |
|
|
|
|
|
|
#352 | Link |
|
Registered User
Join Date: Apr 2012
Location: Germany
Posts: 153
|
I hate that. Why dont they just treat (I hope its the right word for "gönnen" xD ) us the better quality? ._.
Thats the general problem .. if there is more efficiency they react with reducing the bitrate instead of treating us the better quality.. And this worries me - because this is then a deadend. A black hole. We will never see better quality if they react like this to new codecs. Gladly 1440p @ 1920x1200 looks okay at least. Especially it would with VP9. At least there it is that 30fps gets about 10000 kbit @ h.264 and at HFR it gets 10000 kbit VP9 - so even better quality then. Damn I cant wait anymore for 1440 HFR to be activated for all.. |
|
|
|
|
|
#353 | Link | ||
|
Registered User
Join Date: Nov 2009
Location: Northeast Ohio
Posts: 447
|
Quote:
http://youtube-eng.blogspot.com/2015...e-youtube.html tl;dr - more bitrate isn't useful if most people's connections can't handle it. Quote:
Now to clarify, nothing actually needs even 360 vertical pixels - my point is that there are situations where the content in question was designed with the expectation of an HD display to the point that even 480p at lossless quality is worse than 720p at 1.5Mbps. Last edited by Nintendo Maniac 64; 3rd May 2015 at 23:34. |
||
|
|
|
|
|
#354 | Link |
|
Registered User
Join Date: Apr 2012
Location: Germany
Posts: 153
|
no bandwidth would be increased if using the same bitrates. But reducing it further is just annoying. Sure this helps lower end internet connections - but if you select 1080p or higher - it should be decent quality then. These are not quality levels with intention for low end internet anyway. And 1080p with only 4000 kbit is already low enough anyway. Why reducing it further with vp9? : /
4000 kbit can watch a DSL6000 user - thats already fair enough imo if a DSL 6k user can watch even 1080p. Why reducing 1080p further? Just pointless .. |
|
|
|
|
|
#357 | Link |
|
Registered Developer
Join Date: Mar 2010
Location: Hamburg/Germany
Posts: 10,406
|
webm is VP8/VP9+Vorbis/Opus.
A quick check in the ffmpeg sources also suggests that it allows all these formats. Make sure your version is up to date.
__________________
LAV Filters - open source ffmpeg based media splitter and decoders Last edited by nevcairiel; 8th May 2015 at 15:28. |
|
|
|
|
|
#358 | Link |
|
Registered User
Join Date: Jul 2011
Posts: 1,121
|
Hmm, i got this guys builds: http://ffmpeg.zeranoe.com/builds/
But do you mean is limited to VP9+Opus or something? Or can it be any supported video with any supported Audio (like MKV, as i am pretty sure Webm is like a branch of it or something). I can't combine libx264 with libopus. |
|
|
|
|
|
#359 | Link |
|
Registered Developer
Join Date: Mar 2010
Location: Hamburg/Germany
Posts: 10,406
|
webm does not allow H.264, so using that will not work. It needs to be VP8 or VP9 for Video, and Vorbis or Opus for audio. Nothing else is valid in webm.
If you want a MKV, then name it MKV. That'll allow all codecs.
__________________
LAV Filters - open source ffmpeg based media splitter and decoders |
|
|
|
![]() |
| Thread Tools | Search this Thread |
| Display Modes | |
|
|