Fansub group designed for fansubbers

[getfresh]

In regards to the request that the diff versions of the scripts be all released I have decided it is worth doing and worked out how I will go about doing it. Since I have set up a bt tracker they will get hosted there as well but they will be tagged to not display on asuki since they are not actual anime epis. Also, since I find releasing 1 pack at a time rather wasteful time wise I will instead organize a group of epis(number determined by the series and how it is split), pack those scripts into a rar and torrent. The same will be done on irc from now on.

If anyone has any good ideas to add on how to go about this I look forward to them.

peace

[pichu]

Originally Posted by getfresh

In regards to the request that the diff versions of the scripts be all released I have decided it is worth doing and worked out how I will go about doing it. Since I have set up a bt tracker they will get hosted there as well but they will be tagged to not display on asuki since they are not actual anime epis. Also, since I find releasing 1 pack at a time rather wasteful time wise I will instead organize a group of epis(number determined by the series and how it is split), pack those scripts into a rar and torrent. The same will be done on irc from now on.

If anyone has any good ideas to add on how to go about this I look forward to them.

peace

Try Rome!

(just kidding). it's actually not a bad idea to do it via website for people to learn - but that system is a bit overly complicated.

If you're just sharing scripts and no videos... it will awfully be lame to put some 45kB or less as torrents... Torrent, in this case, is great for advertisements, but there are other and better options than torrents. Website hosting (basic website -- not even forum-based... WordPress is probably a decent option to start with) is the way to go. I would have gone for IRC bot/hosting, since I knew RFC1459 quite well.

---
sorry about my comments earlier... I also like fansubbing... I've been fansubbing for five years, and I always said to myself: I should retire now... But I never did! ;_;

[getfresh]

aye, thats why I was discussing with aya about setting up an SVN system on fansubbers.org for hosting scripts that groups are willing to share. Since SVN supports a system of have numerous version of a file and has a compare function I think it would be great for showing how scripts evolve over the course of the process. I'll be discussing and working out the system with the fansubber.org people more tonight. What may have to happen though is that someone with a server hosts the svn script archive which is linked to the fansubbers.org site because it is likely that the fansubbers.org server does not have apache running on it.

[comatose]

Originally Posted by Quarkboy

The problem with releasing avs' is that in general they wouldn't teach anyone much.

Being an effective encoder is about crafting a filter chain that works for a given raw and a given show: there's no one-size-fits-all solution to a really good encode.

Also, you aren't going to learn how to encode by analyzing avisynth files, e.g. if you take the output from yatta after fixing combing problems, you learn nothing about the skills needed to actually identify and fix the problems.

When it comes to the philosophy of the group and openness, obviously all materials should be available, but I just don't think it's really possible to learn much about encoding by looking at script files.

Forget YATTA... for somebody who's just starting, seeing which filters (I'm talking "simple" stuff, like denoising, etc) other encoders use can give them a jump start.

Learning from examples is a lot easier than learning from reading.
Yeah, you have to make your own script for every source, but how will you make it when you have no idea which filters you should use, or even as far as which filters are out there?


I think scripts should be offered. Just have the encoder comment them, explain why he chose that filter and those settings - and maybe even provide before/after screenshots.

[getfresh]

Originally Posted by comatose

Forget YATTA... for somebody who's just starting, seeing which filters (I'm talking "simple" stuff, like denoising, etc) other encoders use can give them a jump start.

Learning from examples is a lot easier than learning from reading.
Yeah, you have to make your own script for every source, but how will you make it when you have no idea which filters you should use, or even as far as which filters are out there?


I think scripts should be offered. Just have the encoder comment them, explain why he chose that filter and those settings - and maybe even provide before/after screenshots.

Like I've said again and again, unless I can get the encoder to share them there is nothing I can do. All of my staff know that scripts used directly for the subs will be released and there is no arguing this point one way or another. I don't force my encoders to share their scripts though. They have their reasons and I have agreed to them as meaningful. In the future if they want to that is another thing.

Please don't focus on the topic of encoding anymore. I don't want this whole concept turned into another encoding discussion. There have been enough already.

[Quarkboy]

Hey, here's a constructive thing you might consider...

This might get a bit too "tutorial" for your liking, but I see a whole lot of videos on niconico douga and even youtube about things like "how to encode" or "how to use mikumikudance", etc, that are quite popular.

One way to get around some of the issues I and others have pointed out about the resources by themselves not being as useful as they could be, perhaps you could actually record yourself as you are working (using hypercam, with commentary perhaps). Not the whole time, of course, but a representative 10 minute segment.

So far the only comparable things I've seen on youtube have been from complete know-nothing newbies that do little to help anyone learn the actual methods people use.

I'm not saying do this for every episode or anything, but just once for each major "step" as sort of a companion to the files you plan to make public. They're actually pretty easy to make, just hook up a microphone and use a program like hypercam or any of the freeware alternatives.

[getfresh]

Originally Posted by Quarkboy

Hey, here's a constructive thing you might consider...

This might get a bit too "tutorial" for your liking, but I see a whole lot of videos on niconico douga and even youtube about things like "how to encode" or "how to use mikumikudance", etc, that are quite popular.

One way to get around some of the issues I and others have pointed out about the resources by themselves not being as useful as they could be, perhaps you could actually record yourself as you are working (using hypercam, with commentary perhaps). Not the whole time, of course, but a representative 10 minute segment.

So far the only comparable things I've seen on youtube have been from complete know-nothing newbies that do little to help anyone learn the actual methods people use.

I'm not saying do this for every episode or anything, but just once for each major "step" as sort of a companion to the files you plan to make public. They're actually pretty easy to make, just hook up a microphone and use a program like hypercam or any of the freeware alternatives.

I tried this once with fraps, and it is a good idea in theory. I have just found that when I sub I am very focused on subbing and tend to move extremely fast, use keyboard shortcuts, and fly through diff windows. Perhaps if I set out to do nothing but a video would be better than an actual episode I am working on?

[Quarkboy]

Originally Posted by getfresh

I tried this once with fraps, and it is a good idea in theory. I have just found that when I sub I am very focused on subbing and tend to move extremely fast, use keyboard shortcuts, and fly through diff windows. Perhaps if I set out to do nothing but a video would be better than an actual episode I am working on?

Yeah, it would probably be better to have it be slightly artificial. But teaching keyboard shortcuts is good, too! I bet a lot of timers out don't realize they can cut off 15 minute or so just by using keyboard shortcuts.

[getfresh]

Originally Posted by Quarkboy

Yeah, it would probably be better to have it be slightly artificial. But teaching keyboard shortcuts is good, too! I bet a lot of timers out don't realize they can cut off 15 minute or so just by using keyboard shortcuts.

lol, in the end I'd have to subtitle my demo on subtitling ;p

[D404]

Originally Posted by getfresh

lol, in the end I'd have to subtitle my demo on subtitling ;p

/me is reminded of aegisub's tutorial videos.

[getfresh]

lol, the last fansub tutorial video I made was in like 2003 I think. And it was on using {\r} to do by syllable transformation while that was still like alchemy to ppl or something. So I'll need to majorly think of what to show ppl since my ways of doing things are most like very out of date.

[getfresh]

Sorry for the double post but this is a totally different subject that I believe requires to be singular.

I have had the series World Destruction proposed to my group under the pretense that while there are subs for it there are not "Quality" subs. I am not bashing what is subbed by whoever already. I have not watched any of it. By quality they refer to the extra time taken to add all the bells and whistles to the series. As I have made it my utmost point with this group I do not wish to "start a fight", sub something that is being subbed in a satisfactory manor(mostly refers to translation and editing). Or bash other groups based on bias.

I am asking the general mass of fansubbers their opinion on this matter for this reason. I'll say again, I have not checked any of the subs myself. I want to be a blank slate that draws upon the communities over all opinion on this subject. I also apologize to any groups subbing this if you feel I have insulted you, that is not my intention at all. I am doing this for the opposite reason, to keep from absolutely one sided, self profiting, slander.

That being said, if you replay, do not "trash", "flame", or use phrasing that can be taken as picking on any group. You can flame me all you want, since I put myself out there on this, but please don't take out any grudges on any of these groups if any of you have them, by trashing them and say "yeah, sub it" when there is no real need.

Thank you,

getfresh.

[comatose]

All of the groups subbing WD could use an improvement in one area at least, and it seems like a really interesting anime. (Lagann minus HUEG mechas)
Also, it has very detailed backgrounds for the most part, so if you use Share/PD raws, it could prove as a challenge to your encoder to denoise flat areas without hurting the background.
Do it >:O


Also, I feel it is really important for the encoding scripts and raw to be available to ANYBODY.
It's difficult getting help on doom9 with anime, since they auto assume you didn't get it legally, and once that happens you're doomed unless your source is a TS or DVD.

Encoding was the part of fansubbing that took me the longest time to get a "quality" hold of, because of doom9 and other encoders not really willing to help.
Some will only help you if they've heard your name before... god knows why they do that... which is really stupid, since the time you need the most help is when you're getting started.

The only one who let me bug him is martino, and there aren't many others like him ;_;
So, here's hoping your encoders will release and comment their scripts.

If encoders don't want to see eyecancer encodes, then they need to start helping beginning encoders or else these encoders will be left to learn through [horrible] mistakes.

tl;dr almost nobody wants to help a beginning encoder, and there aren't many up to date resources on filtering anime, so getting started with encoding is very hard. Publish your scripts ;-;

[Zerox20]

Originally Posted by comatose

All of the groups subbing WD could use an improvement in one area at least, and it seems like a really interesting anime. (Lagann minus HUEG mechas)
Also, it has very detailed backgrounds for the most part, so if you use Share/PD raws, it could prove as a challenge to your encoder to denoise flat areas without hurting the background.
Do it >:O


Also, I feel it is really important for the encoding scripts and raw to be available to ANYBODY.
It's difficult getting help on doom9 with anime, since they auto assume you didn't get it legally, and once that happens you're doomed unless your source is a TS or DVD.

Encoding was the part of fansubbing that took me the longest time to get a "quality" hold of, because of doom9 and other encoders not really willing to help.
Some will only help you if they've heard your name before... god knows why they do that... which is really stupid, since the time you need the most help is when you're getting started.

The only one who let me bug him is martino, and there aren't many others like him ;_;
So, here's hoping your encoders will release and comment their scripts.

If encoders don't want to see eyecancer encodes, then they need to start helping beginning encoders or else these encoders will be left to learn through [horrible] mistakes.

tl;dr almost nobody wants to help a beginning encoder, and there aren't many up to date resources on filtering anime, so getting started with encoding is very hard. Publish your scripts ;-;

I 100% agree with this. I take lots of heat sometimes from people going "this encode is horrible!" Then it turns into why didn't you use this or run this, or try this Not too much people teaching encoding anymore, just more everybody complaining about it.

I offered getfresh my encoding services, and if FLF wants scripts published, I don't see why some screenshots can't be posted, and just some filters used to help steer newer encoders in the right direction for anime.
If a mod sees this, I want a Fansubber tag toooo!

[martino]

Originally Posted by comatose

If encoders don't want to see eyecancer encodes, then they need to start helping beginning encoders or else these encoders will be left to learn through [horrible] mistakes.

Isn't that how most learning is done? Through our own mistakes? I myself can go back, and look at some of my first encodes and shake while looking at them thinking "Did I really encode that?" and "What the hell was I thinking?" I think that it's the most effective way to learn. Of course it's nice to have someone there to tell you what you did wrong, and how you could improve, but that's not always going to be there. As you said, and has been mentioned, you'll often just get a "sux". But those come mostly from leechers anyway, who have little to no clue. Most fansubbers seem to stay away from these sites (AniDB for one, Anime-Planet sure has some representation there, but it's not a very thriving place).

Originally Posted by comatose

tl;dr almost nobody wants to help a beginning encoder, and there aren't many up to date resources on filtering anime, so getting started with encoding is very hard. Publish your scripts ;-;

Nobody wants to help? Problem is, that a newbie encoder will hardly get into a good group (makes me think how I managed to get into Ayu to start off with, I guess it was just luck) where you may have encoders that are willing to help you (I for example was given a crash course on IVTC in Ayu, although using IT as advised was not really the best option heh). Or it may have ones that are not. If you do get into one, it sure is an advantage, but chances are you will go for newer groups, which well... don't have those "resources" to start off with.

It's a hard one to come over; encoding. You'd really just need a view-only remote desktop to see what's going on during the process. Filtering, scripts... yes, there you go. Some encoders may feel that it is their intellectual property and would not want to share them, I think it's understandable, plus everyone has to choose for themselves.

[TheFluff]

Originally Posted by comatose

If encoders don't want to see eyecancer encodes, then they need to start helping beginning encoders or else these encoders will be left to learn through [horrible] mistakes.

tl;dr almost nobody wants to help a beginning encoder, and there aren't many up to date resources on filtering anime, so getting started with encoding is very hard. Publish your scripts ;-;

Just FYI I've never minded people PM'ing me with questions on IRC, at least as long as they are good questions. There's also #darkhold.

Originally Posted by Zerox20

I take lots of heat sometimes from people going "this encode is horrible!"

To be insultingly honest, in your case it's most likely because it IS horrible.

[Schneizel]

Originally Posted by Zerox20

I offered getfresh my encoding services, and if FLF wants scripts published, I don't see why some screenshots can't be posted, and just some filters used to help steer newer encoders in the right direction for anime.
If a mod sees this, I want a Fansubber tag toooo!

You can set the fansubber tag by yourself in your user control panel. (Group memberships.)

[getfresh]

I'll leave the subject of WD open for 2 more days for comments. Unless I receive clear reasons, not "they could improve in one area" I can't really justify it. And if the reason is because they have a ugly ts or encode I will ignore it under the basis of those being purely objective for the most part.

comatose

I am fine with scripts being offered, but I am not going to start giving out full releases with every file used to create it plus the raw those files are directly based off of mainly for the reason that I know people will begin to misuse the gesture. If they are interested in the video from a stand point of why those filters were used, direct slips are enough.

I do not want to make this group into a way for ppl to release fansubs without actually fansubbing. Of course some people do this anyway, but even if they are "borrowing" the translations and editing and timing they at least had to do some work to "borrow" it.

[Zerox20]

Originally Posted by TheFluff

Just FYI I've never minded people PM'ing me with questions on IRC, at least as long as they are good questions. There's also #darkhold.



To be insultingly honest, in your case it's most likely because it IS horrible.

Yeah I have done some bad encodes in the past, but I feel like I've changed quite a lot with what I've encoded recently.

I got taught Yatta, I updated all my filters, learned some new avs stuff, got rid of all that old school crap. So It is possible for encoders to change. and I agree with what they said earlier.... Can't believe released that stuff encoded before. I have plenty of examples in my head and recent encodes that I realize the mistakes I made.

An encoder saying something is horrible is different then saying here is what you did wrong, which is what I would prefer. I am tired of RTFM on everything, I've went all over doom9, read readme's, etc. Its almost useless. So If you want to bash my recent encodes feel free to let me know, because I know the older encoders had major issues because of REALLY bad filtering.

[getfresh]

Zerox, I'm sorry but I'm gunna have to say no to WD for now. I didn't get enough feedback for or against the idea so I cannot hop on a series that has I believe it is 4 groups subbing it. That is more than enough versions for at least one to be good enough, right?