AnimeSuki Forums

Register Forum Rules FAQ Members List Social Groups Search Today's Posts Mark Forums Read

Go Back   AnimeSuki Forum > Anime Related Topics > General Anime > Fansub Groups

Notices

View Poll Results: What do you use for your loseless pass?
Lagarith 11 20.00%
Huffyuv 23 41.82%
MSU 1 1.82%
I use the option in MeGui... 1 1.82%
Other 4 7.27%
Loseless pass? I don't make one 5 9.09%
I don't encode... Just passing by... 10 18.18%
Voters: 55. You may not vote on this poll

Reply
 
Thread Tools
Old 2007-06-20, 16:13   Link #41
pichu
Senior Member
*Fansubber
 
Join Date: Jul 2004
FFV1... I started to use that codec to import clips into AfterEffects, though I'm not sure if I should use this codec to create RGBA overlays.
pichu is offline   Reply With Quote
Old 2007-06-20, 18:48   Link #42
TheFluff
Excessively jovial fellow
 
 
Join Date: Dec 2005
Location: ISDB-T
Age: 37
You shouldn't, because AFAIK it doesn't support RGB32.
__________________
| ffmpegsource
17:43:13 <~deculture> Also, TheFluff, you are so fucking slowpoke.jpg that people think we dropped the DVD's.
17:43:16 <~deculture> nice job, fag!

01:04:41 < Plorkyeran> it was annoying to typeset so it should be annoying to read
TheFluff is offline   Reply With Quote
Old 2007-06-23, 17:56   Link #43
pichu
Senior Member
*Fansubber
 
Join Date: Jul 2004
Quote:
Originally Posted by TheFluff View Post
You shouldn't, because AFAIK it doesn't support RGB32.
That shouldn't matter... I can create two videos (one flat RGB and one Alpha) for overlaying... You just have to load the alpha channel from a separate video. I did it before for an AE karaoke, using XviD (since it's a karaoke, Q-loss won't apply here).
pichu is offline   Reply With Quote
Old 2007-06-23, 18:31   Link #44
dj_tjerk
Ana-chan~
 
 
Join Date: May 2006
Location: Netherlands
I did it once too, when i didn't understand anything about RGBA. I wondered what to do with my white (so it looked in media player) karaoke >_> Tried DivX and whatever (that said it had RGB + Alpha support in AE iirc), but didn't work :P I failed once too with using Premultiplied RGBA instead of Straight, you don't really notice anything until it start fading or opacity isn't 100% of whatever effect it is.

[offtopic]
what about a new topic about worst mistakes (your own.. or ones you know of) you made when fansubbing, or did i miss something somewhere >_>
dj_tjerk is offline   Reply With Quote
Old 2007-06-24, 09:48   Link #45
pichu
Senior Member
*Fansubber
 
Join Date: Jul 2004
No, it wasn't a mistake on my behalf. I had the karaoke created as two separate XviD videos, one being RGB and the other being Alpha (Grayscale XviD encode), which would have been the proper way to do so if your codec doesn't support Alpha channel.

Edit: The biggest mistake for me in fansubbing is to start fansubbing in 2003. I can't seem to get out anymore.
pichu is offline   Reply With Quote
Old 2007-06-24, 11:13   Link #46
TheFluff
Excessively jovial fellow
 
 
Join Date: Dec 2005
Location: ISDB-T
Age: 37
Why the christ would you ever want to do that? Granted, I had no idea you even could do it in AE, but why two files instead of one? And why with a LOSSY codec?

Quote:
Originally Posted by pichus View Post
(since it's a karaoke, Q-loss won't apply here).
What does this mean? I've been trying to make sense of it for 10 minutes now and I still don't understand what the heck you mean.
__________________
| ffmpegsource
17:43:13 <~deculture> Also, TheFluff, you are so fucking slowpoke.jpg that people think we dropped the DVD's.
17:43:16 <~deculture> nice job, fag!

01:04:41 < Plorkyeran> it was annoying to typeset so it should be annoying to read
TheFluff is offline   Reply With Quote
Old 2007-06-24, 13:19   Link #47
dj_tjerk
Ana-chan~
 
 
Join Date: May 2006
Location: Netherlands
I think he means, if it'll get shitty anyway, why bother doing a lossless output. The final encode will always (most of the time anyway) be more vague/blurred so you won't notice the quality loss from using a lossy codec to output from AE. Something like that anyway..
dj_tjerk is offline   Reply With Quote
Old 2007-06-24, 14:59   Link #48
pichu
Senior Member
*Fansubber
 
Join Date: Jul 2004
Hence, you spent ten minutes trying to understand what i'm saying there, as you failed to see the point of using a LOSSY codec in that scenario. The idea is quite simple: Karaoke is the effect for the eyes, not the effects for the scene, which is why it won't matter whether I'm using a lossy codec or a lossless codec. As, Q, in our terms, stands for "Quality," the actual karaoke isn't meant to match up with anything but as an eyes appealing subject.

As for two files, you shouldn't be against it, as long as it's doable.

In aftereffects, you can render RGB and Alpha separately, allowing you to make RGBA into two separate files. (To speed up the process, use video proxy in AE) A simple script could automate this part.
pichu is offline   Reply With Quote
Old 2007-06-24, 18:25   Link #49
dj_tjerk
Ana-chan~
 
 
Join Date: May 2006
Location: Netherlands
What is video proxy? Two output modules of the same renderqueue item is a lot easier (or is that what you meant?) I tried it btw, and it doesnt even look that awful (xvid with w/e settings). Proof of principle.. but i still like it big
dj_tjerk is offline   Reply With Quote
Old 2007-06-24, 20:09   Link #50
TheFluff
Excessively jovial fellow
 
 
Join Date: Dec 2005
Location: ISDB-T
Age: 37
Quote:
Originally Posted by pichus View Post
Hence, you spent ten minutes trying to understand what i'm saying there, as you failed to see the point of using a LOSSY codec in that scenario. The idea is quite simple: Karaoke is the effect for the eyes, not the effects for the scene, which is why it won't matter whether I'm using a lossy codec or a lossless codec. As, Q, in our terms, stands for "Quality," the actual karaoke isn't meant to match up with anything but as an eyes appealing subject.

As for two files, you shouldn't be against it, as long as it's doable.

In aftereffects, you can render RGB and Alpha separately, allowing you to make RGBA into two separate files. (To speed up the process, use video proxy in AE) A simple script could automate this part.
Uh...
I think you mean that "since the karaoke is so distracting anyway, noone will notice the artifacts", which is a pretty funky statement in itself (disregarding your, uh, not very straightforward way of saying it), but hey, I've seen weirder things on these forums before...

I still don't see what the point of using a lossy codec in this scenario is (especially when the codec in question forces you to use two files instead of one) though. Saving a few MB on uploading the stuff? Intentionally making things look worse, "ethical" subbing style? It may or may not be more work for you as the AFX'er, but you can bet your ass it'll be more work for the encoder.
__________________
| ffmpegsource
17:43:13 <~deculture> Also, TheFluff, you are so fucking slowpoke.jpg that people think we dropped the DVD's.
17:43:16 <~deculture> nice job, fag!

01:04:41 < Plorkyeran> it was annoying to typeset so it should be annoying to read
TheFluff is offline   Reply With Quote
Old 2007-06-25, 11:01   Link #51
pichu
Senior Member
*Fansubber
 
Join Date: Jul 2004
1. Being a Smartass [again]? And no, that's not what I'm saying. I think it's better to drop it here, as you're drawing the wrong conclusion here based on your [inept] experiences.

2. A few MB's saved = 300MB from 400MB - 100MB (Lagarith - Q1 XviD) per 720p overlay, sure. (mathematically speaking)

3. With proper functions/scripts/templates, your job as an encoder won't be as difficult as you're claiming so. Also, I'm providing avs scripts if I were going to do this... So you only need to do 0 work, except to adapt the function.

Last edited by pichu; 2007-06-25 at 11:20.
pichu is offline   Reply With Quote
Old 2007-06-26, 07:11   Link #52
shinjipierre
Computer graphist
 
 
Join Date: Dec 2005
Location: Paris, France
Age: 41
Only one thing to say : "..."
__________________
http://www.remipierre.fr - Some of my computer graphics work
shinjipierre is offline   Reply With Quote
Old 2007-07-02, 11:55   Link #53
Onniguru
 
Join Date: Feb 2006
Anyway to get huffyuv to produce < 6 Gig files for a 25 minute show, and still be be virtually lossless?

Any good alternatives for a lossless or nearly lossless pass that dont produce 6 gig files?
Onniguru is offline   Reply With Quote
Old 2007-07-02, 12:26   Link #54
martino
makes no files now
 
 
Join Date: May 2006
Quote:
Originally Posted by Onniguru View Post
Anyway to get huffyuv to produce < 6 Gig files for a 25 minute show, and still be be virtually lossless?
Huh? I get 4-5GB usually for 704x400 25min using Huffy (Planar, YV12, Adaptive Huffman Tables)... And if you use a lossless codec, it will always be lossless. Unless you or the encoder application fucks up something.

Quote:
Originally Posted by Onniguru View Post
Any good alternatives for a lossless or nearly lossless pass that dont produce 6 gig files?
http://www.compression.ru/video/code...s_2007_en.html

Take your pick. (FFV1 should be a good choice for high lossless compression; not sure about how fast it is though. Then Lags is pretty fast too while having pretty good compression.)
__________________
"Light and shadow don't battle each other, because they're two sides of the same coin"
martino is offline   Reply With Quote
Old 2007-07-02, 12:40   Link #55
Onniguru
 
Join Date: Feb 2006
Thanks for your quick response.

I didn't have "adoptive huffman tables" checked. Tried it again with this check, and its making a big difference, heh A particular "test" file went from 6.2 down to about 4.4 gigs.

As for "nearly lossless" I was thinking FFV1, which actually is supposed to be lossless, but with some limitations involving colorspace, and apparently other things. FFV1 doesn't support 32bit colorspace, according to this thread, but most sources Im working with are only 24bit anyways.

If Im working with a crappy source off of EX2 that only has 24bit color space, is there any reason NOT to use the higher compresson of FFV1, in other words?

Quote:
Originally Posted by DryFire View Post
non-linear access in ffv1 is killer.
This is not a major concern me when reprocessing RAWS, since all I do is some filtering (such as adding logo 1.5, resizing if source is a weird size, decimating frames for that one capper that caps at 119 fps for no reason at all), none of which requires me moving back and forth through the video. For a TIMER putting subtitles in, I can see where this would be a real pain.

PS: all my encoding is self taught. My professional experience is as a Network Administrator and Windows App programmer. Graphic Artist training == Zero.

Last edited by Onniguru; 2007-07-02 at 13:04.
Onniguru is offline   Reply With Quote
Old 2007-07-02, 17:33   Link #56
checkers
Part 8
*IT Support
 
 
Join Date: Jul 2006
Location: Western Australia
Age: 35
Send a message via MSN to checkers
Quote:
but most sources Im working with are only 24bit anyways.
Most sources are actually 12bit, that is, YV12. There's not much reason to use more anyway.
checkers is offline   Reply With Quote
Old 2007-07-16, 04:39   Link #57
Sakaki-
AnimeONE Typesetter
 
Join Date: Nov 2003
Dreams of truth

VBLE is what i use
Sakaki- is offline   Reply With Quote
Old 2007-07-20, 14:32   Link #58
SirCanealot
What? I am washed up!
 
 
Join Date: Feb 2003
Location: London, England
Age: 38
Send a message via MSN to SirCanealot
I usually use a YV12 version of Huffy.

I tried Lagarith recently and was pretty impressed. Might end up sticking with that for a while.
SirCanealot is offline   Reply With Quote
Old 2007-07-28, 11:20   Link #59
Darth_E_
Freelance TS'er/Encoder
 
 
Join Date: Apr 2004
Location: Canada
Age: 39
Hmm, seems like I am the only who is still using CorePNG. Decent decoding speed, and nice filesize. Then again, I am using it because of HDD limitation ( limited free space ).
Darth_E_ is offline   Reply With Quote
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -5. The time now is 00:05.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.
We use Silk.