How to become an encoder?
 

Alright, so I've come to the conclusion that I want to start contributing to the scene in some way, and the thing that has undoubtedly interested me the most is to practice the art some folks might consider a religion on this board. It's the art of encoding. So I was wondering if you guys could be so kind to perhaps provide any links / give pointers on the basic stuff on the get abouts which would help me and perhaps other desirous people who wants to know, but do not know how. Also, I want to make it clear that I'm somewhat new(b) (Completely so to be perfectly honest, but hey, who hasn't?), so please explain in the most articulate manner as possible, thanks.

As a side-note, I just want to say that I've indeed read some basic info, info which is located on doom9 (the guide) as well as lurking on here -> (http://forum.doom9.org/forumdisplay.php?f=17)


:)


Mods: If this is the wrong sub forum, I ask you to please move this thread to the correct sub forum instead of locking it. If this were to be the case, I'm very sorry for the inconvenience I've caused.

Search/Read lots of threads on animesuki forum and doom9 forum, experiment, practice, get it down to key, get some experience, and have fun.

:D

thread on some filters on here if you haven't read before.

Don't forget to stock up on children, and get the Zelman900 Altar Table. It is quarlity.

Turn back before it's too late.

Doom9 isn't exactly the best place to dive into if you want to have a pleasant first experience with encoding... A lot of the people there are elitist.

And what's wrong with that? Reading the stuff on Doom9 certainly doesn't hurt, and asking neither (just make sure that you have read all the guides/faqs before asking)...

And the rules. For example, fansubs are NOT allowed on Doom9 so don't post questions that specifically relate to a fansub encode or post clips of a fansub encode.

That said, just reading and searching the threads there can get you started on specific tasks. For example, you could search for: 'denoise', 'blocky', 'sharpen', 'interlaced' or other such keywords to find threads that might be similar to whatever your particular objective is.

Not many shows have digital and/or HD sources, and even if they do you need to know filters really well. My best advice is to start off with simple 24fps analog broadcast raws and work on your filter skills, everything else you can learn with time. If you master filtering and codec settings you are more than well equipped to work for most fansub groups. Once you've established yourself in a group you can work with existing encoders on expanding your skills. When I started there wasn't a lot to learn, but nowadays trying to "master" encoding is basically biting off more than you can chew ;)!

Haha.

Unlike text related jobs, encoding isn't really rewardable. It's necessary in fansubbing, but it also waste your time. What I recommended is try not to over involved into encoding. Avoid those groups which do hardsub QC again and again.

Get the popular programs, and experiement, then when you can encode - like, simple "re-encode" - set yourself a goal, like:

"I want to learn simple IVTC" or "I want to learn how to remove dotcrawl" or "I want to learn how to remove X artifact..."

You know? Take it easy - but most importantly, encode and encode and encode... And compare! And use programs like AVSP with like 2-3 tabs of the same video open with the Orignal(no filters) and then (filter chain A) and (filter chain B).

Oh, and buy a strong computer :).

That's how I got into encodig. ^_^

*cough*anime-keep*cough* :rolleyes:

Hey guys,

New to the forums and learned the basics of encoding and muxing. Only thing I don't get at the moment is splitting the raw's video and audio.

I downloaded a gundam raw with D-TBS WMV9 1280x720). How do I go about splitting the audio into vobis or mp3, and the raw without sound later for encoding with subs?

Usually when I encode it, it's always in PCM format :/. I'm also having trouble trying to load the file in vDub, as I'm getting an error saying "Couldn't locate decompressor for format 'WMV3' (unknown)", which means the wmv9 codec correct?, Which I've installed multiple times to ensure.

any help around would be great, thanks.

For the FourCC error download WMV9VCM from M$'s website.

If you want to encode just the video, I don't think that there's any need to separate them before doing so, because after all you can just do an audio=false for both DirectShowSource and AVISource.

However if you really do want to separate them then mkvextract for MKVs, AVI-Mux(GUI) for AVIs (or VDubMod) and mp4box for MP4s would be the simplest solutions... (This might be helpful for audio encoding, depending on how you do it - since there are many ways.)

Learning to encode really is an art.

There are basically like 3 levels of encoders.

Beginners -> Intermediates -> Advanced

--------------

Beginners are like people who ask, wth is avisynth and how do I use it.

Intermediate encoders understand and can script avisynth scripts.

Advanced encoders can script custom scripts to say, apply certain filters to a certain part of the image.

I would say, I am somewhere between intermediate and advanced. I have no exactly hit that high level yet, but I am quite experienced now.

-----------------

Best way to learn is probably to find a group with an encoder willing to "teach" you a few things.
However, most common way to learn is to basically read up on doom9 and stuff...

Now that I think about it, learning this crap was hard :)

You missed the last step;

Masochists - People who use YATTA and are addicted to it

(Personally I'd put apply filters to certain portions of the script/picture into Intermediate, it's not such a hard thing to do scripting-wise, but takes many lines to achieve)

My own opinion: I wouldn't really group them into levels. I think it's mostly about experience and well, knowledge of course. You might call yourself "Intermediate" or "Advanced" and come across a badly muxed 5.1 AAC file in AVI, or shifted chroma and have no clue to do which is pretty trivial stuff imo and yet can make ugly complications and give headaches or be able to use YATTA well and don't know how to decimate a 120fps file. So I don't think that it's something that you could group into "levels"...

IMO it's a good thing to like perfection if you encode. So if you get a source with NTSC->PAL framerate conversion (every 25th frame is a copy of the 24th) you should notice and fix it back to NTSC framerate, deleting the duplicate frames. At least 95% of the anime viewers would not notice. But you, as an encoder, need to! Because this stuff is subtle. People might not know it's a duplicate frame, but they feel something's not right.

The problem with encoding is: Everyone can do it, but most people who do it don't do it right. They do things that usually work, but sometimes don't. Good encoders are not satisfied until they know they must be close to the best solution possible.

Also, it's soooo frustrating. For me, it's only damage reduction... The raws I get r usually bad in so many ways, I don't know whether to laugh or cry. :(

About doom9: Elitist it may be, but as an encoder, there's no way for you around it. Its search function will already answer most of your questions. :p

NTSC video is 29.97~ frames per second, while PAL is 25. FILM, on the other hand, is 24 frames per second. Most anime is produced in FILM and then converted to NTSC via interlacing. Most PAL anime DVDs, that I've seen, are really really really really really terrible NTSC->PAL conversions that look atrocious. As such, its usually recommended to stay as far away from PAL sources (for anime at least) as possible.

No, the FILM to NTSC conversion does is NOT "just through interlacing". The process has a very specific name, it's called TeleCine (convert into TELEvision from CINEma) and it merely exploits the field-based nature of video. While it does result in interlacing artifacts when played back on progressive scan equipment without using a proper IVTC (InVerse TeleCine) pulldown filtering it's still not true interlacing.

True interlacing means recording half frames at double the frame rate, eg. first recording the even field, then the odd field, then the next even field, and so on. Only a video recorder recording in an interlaced format produces interlaced material.
Both interlaced video and telecined film are forms of field based video, but just because a video is field based it does not mean it's interlaced.

(Disclaimer: I'm no expert on this field, but this post reflects the best of my knowledge, and is also partly personal opinion. If anyone can cite some good sources telling otherwise, or supporting this, please do so.)


Edit: Oh, and I forgot to say: NTSC->PAL conversions can never be saved fully. NTSC is, as mentioned, 29.97 (well, 30/1.001) fps field based video, while PAL is 25 fps field based video. Any conversion from NTSC to PAL will lose video frames or fields through some kind of irreversible process, unless you elect to slow down the video by about 17 percent, which will be very noticeable, and certainly not desirable. If the original material is FILM, what is usually done is that it's instead sped up from 24 fps to 25 fps, producing progressive field-based video. This speed up is not too noticeable and is standard practice on most PAL land video/DVD releases of cinema film productions.