I have absolutely no clue about all when it comes to older shows when there's half a dozen or more releases. I just look for pretty subs because I think it implies an extra level of care. For modern stuff... well, fansubs are mostly dead but I get those edits when I can, although I try to see what people are saying about them on nyaa and such first.
Hmm....

Like you, I'm watching on 1080p screens. I'm perfectly fine with EMBER releases and even the CR rips from HorribleSubs/SubsPlease.
I'm not going to notice a real difference/improvement from one release to another unless there's some substantial filtering done such as de-noising or some sharpening filter thrown on it. One thing I can't stand is black bars, though. I'll go out of my way to find a release that crops those out. Overall, though, if I look at it and think "good enough, works for me" then I'll grab it.
Personally, I don't really see why a particular release needs to have episodes at 5+ GB apiece, when there's other encodes that are virtually the same video quality for a greater space savings.
Whenever I see Nyaa comments complaining about file sizes being so small, I wonder why they even bother since they were never going to obtain that release anyway. It sometimes feels like the audiophile community where there's snobs that'll call you names for opting for an "inferior" release that is virtually the same but they used up 50+ GB of HDD space instead.

I do not understand the space argument when you can just delete after watching.
If you want to archive something, that means it's not just some "average" show that you don't really enjoy. Otherwise, the space argument goes out of the window. But if you do decide to archive a show that you really enjoy, you should be going for highest quality anyway.
So either way, as far as I can see, the space argument collapses under its own weight.

I don't need the highest possible quality for watching. For seasonal stuff, I usually go with automated rips, because these satisfy a criterion that is actually important to me: reliable releases.
I sometimes make an exception for uncensored versions or releases from groups that I trust.

For anything not-seasonal, there is no reason ever to not pic the highest quality (as far as your senses can tell). That said, I have sometimes downloaded shows in segments because I could not be bothered to put the entirety of the dozens (or hundreds) of gigs onto my drive at once.

>>2980
>If you want to archive something, that means it's not just some "average" show that you don't really enjoy.
Not really. I archive kuso shows I didn't like both for the purposes of seeding alongside just wanting to archive them. I enjoy having a personal collection, even if it's not entirely made up of stuff I personally enjoy.

>>2980
Some people are on metered connections/capped and the ISP throttles them for using too much bandwidth. Other times, it's horribly slow connections for those living in the hinterlands. Nobody wants to sit there forever waiting on huge filesize downloads.
As for saving stuff, I keep what I watch. Nothing gets deleted since I share the collection with family and it turns into a local Netflix-type thing with Jellyfin.
If I delete everything and watch to re-watch the same release, only for it to disappear due to either no seeders or it's been taken off the XDCC bots then it's not so easy to find again for that favored release.

not as i really download stuff
they closed allthe streaming platforms in my area and all those global simply dont serve it
id prefer watch streams and even still miss daisuki.net

For newer stuff, I don't really care. I just stick with SubsPlease. It's all pretty much the same anyways. Older stuff is really what the differences lie because there can be genuine differences between 480i NTSC and 576i PAL releases, DVD vs Bluray upscale, LD vs VHS, etc. Those warrant more investigation, and really are a case-by-case type thing that I don't think you can give a simple answer to.

>>2984
Agreed, new stuff is pretty much the same. SubsPlease and call it a day.
When trawling older releases, that's where the actual investigation and comparison matters.
Such as seeing BD rips where the disc itself is just a upscale and there's no point to grabbing that both quality- and size-wise, or where the encoder was a little too trigger happy with video filters on an old DVD release, etc.

>>2979
Nyaa comments tend to be really abrasive I've noticed, especially over what seems (to me) like pretty minor stuff. I can understand being passionate about it, but a lot of the users seem to fail to understand different releases exist for different people, 十人十色 and all.
But yeah, outside of the more obvious stuff I usually can't notice a difference unless I get close to the monitor. Aliasing and banding I do notice, but all the decent releases usually fix those to an extent.

>>2986
Yep, there's unfortunately some real elitists in Nyaa comments sometimes. Like man, I'm just watching cute anime girls doing cute things, calm down lol

I've been doing EMBER and other smaller stuff lately for airing things and I can't say I notice much of a difference. My monitor is 1440p if that means anything.
Most edited/fansub releases go with big filesizes so I end up with those, which makes sense because they feel strongly enough about it to not sacrifice on quality even if it's a placebo. The only time I'd notice is with a direct comparison of the same static frame, and that's similar to other things like which gaming screen to buy for the Vita and so on. For old shows I love, well, I never get the gigantic BD stuff that's like 60GB for 12 episodes unless it's the only one available, which is quite rare thankfully.
Also, for airing shows the TV bitrate stuff means that you can't really make it beautiful in particularly colorful animated scenes anyway. This is most reliably seen in Precure transformations, but the recent Jellyfish show is also this way. This show looks beautiful at times, but when you take a screenshot like I did here you really notice the bitrate problem. It's a 1.5gb file but the loss is already there in the source.

I always go for a high bitrate since I have a fairly large sized screen and a really bit-starved release sticks out a LOT when I fullscreen it. Also I tend to look for good audio quality if not lossless since bad audio is a cancer on the ears. Then, normally on older stuff, I check to see whether the BD release or DVD release is the best. Since sometimes you'll get a better quality on the DVD since the BD is retarded and just a smeared upscale (lots of sailor moon ones)

Aside from that I try to avoid the dolby vision whenever possible.

>>2988
Sadly the Jellyfish show got licensed by HIDIVE which means it has awful video and audio since they compress the hell out of it.. noticeably worse than CR even. Their version of the Jellyfish show actually has holes in the spectrogram too, which I'd never even seen before. Thankfully Okay-Subs is doing an improved version.

>>2981
If your purpose is ensuring a show is available to other people, that's all the more reason to pick a quality release. It's retarded to have 50 webrip seeders when there's a BD fansub with only 5.

I'll use SP for reliability unless they do something stupid with the translation like dropping honorifics, but I never keep those files past the season they air in.

>>2986
There is an aspect of the cartel hazing people, but there's also a lot of shit getting thrown around that doesn't need to be there. Outside of filtering out reencodes, there isn't really a way to sort through things meant for you and things meant for some theoretical audience with an bizarre usecase. There's nothing more frustrating than putting effort into making something of high quality only to have a bunch of blind retards go after an objectively worse release because it's 100mbs smaller or has a green highlight.

>>2991
I do seed quality releases, just not always the best possible for reasons already given. I don't think I'm seeding a single TV or webrip actually, outside of series that haven't had a decent BDrip or whatever posted yet. From my experience most of the best releases have plenty of seeds already anyhow, like Okay-Sub's 100GB Fate/Zero release or hchcsen's Gundam remuxes. I personally don't have a need for those releases, but I do still seed quality alternatives for the series I do archive so that those with desires similar to mine can still reliably get them.

And yeah, there's a lot of useful feedback given on Nyaa too and it's helped me avoid awful releases several times, though I was mainly referring to stuff like https://nyaa.si/view/1800560 where the first commenter seems incapable of understanding that not everybody wants a large release. Notice how he didn't provide any meaningful feedback other than whining about the size?

>>2990
Oh, that makes sense then. Yeah, I saw okay-subs and got them for episode 2. I'm not sure if it's better since I'd need an equivalent scene, but it probably is.
As someone mentioned earlier I'll be deleting it after the season ends, anyway. I tend to take lots of screenshots and if I feel nostalgic I can just look at them... or something like that.

There is zero noticeable difference in quality between the ASW release and SubsPlease. People need to stop normalizing bloat. It didn't used to be like this. Everyone used to shit on group like Coalgirls for releasing bloated garbage.

>>2994
Most people dont care about bloat or kludges if it works

>>2995
>Most people dont care
Exactly. And this in turns normalizes everyone to release their stuff in a bloated state. It takes longer to download, kills your SSD/HDD with more writes and has negligible benefits. The only reason to download the SubsPlease release over ASW is that SubsPlease comes out first.

I noticed that Fran's show had a guy named Isekai that also had 'Isekai-mini' releases that were much smaller. It would be nice if people did that more often, especially with an easy to notice label.

>>2994
I generally agree with you, but it depends on the scene and your eyes. For example:
ASW: https://i.slow.pics/KB0ANuqn.png
https://i.slow.pics/Ziwob76G.png
SubsPlease: https://i.slow.pics/6LA9ZPcp.png
https://i.slow.pics/0rEqfml3.png
Granted these are cherrypicked, but during busy scenes noise is far more common in ASW releases and if you're watching on a display where that needs to be upscaled it's going to be even more noticeable. Though with non-mini encode stuff like Coalgirls vs. the next best, it's significantly less noticeable for most.

>>2998
I have to get very close to the screen to see anything though. And if you're that obsessed with quality you should wait for the BD release anyway. Just doesn't sit well with me downloading 1.4gb per ep.

>>2994
>>2996
You are way behind the times, man.
Normalization does not matter. The product that you are discussing is an automated release, a direct rip from a source that is for a completely different audience (paying customers).
Around 2010, fansubbers still pretended to be relevant by putting their own spins on the official releases. Nowadays, everybody knows that they are a marginal existence next to the official versions.
Fansubbers today exist largely to address whatever gaps open up, and that's a thankless job.
That's why whatever technical skills or standards were common back then are now mostly forgotten.

CR streams in large file sizes because that's the simplest way and is good for the greatest variety of toasters, even if it costs them bandwidth. You can boycott subsplease or complain about it on an anonymous image board. But that's not going to impress the CR leadership.

Smallest size with watchable quality
Usually just torrent Judas releases

>>3000
I know it's ripped directly from CR (that's why it always comes first).
>The only reason to download the SubsPlease release over ASW is that SubsPlease comes out first
I sometimes download the SubsPlease release based on this if I want to watch something as soon as it comes out.

CR however, was made to be streamed, not downloaded. Otherwise you're killing your poor SSD/HDD with those huge ass file size (specially SSD, don't complain if it stops working after a few years of doing that).

>>3002
The point being that it's pointless to complain about people normalizing it.

>>3003
I feel like this will cause everyone to abandon torrents and just watch streams instead (already happened to some extent, nyaa used to be more popular).

Random comments I found there about this.

>>3004
>abandon torrents and just watch streams instead (already happened to some extent
I believe that future to be inevitable to some degree. Anime being one of the few holdouts with a lot of people still downloading properly and using real players, contrasting western television that is almost exclusively watched on stream. I don't think that newcomers to anime will ever see a single reason to acquaint themselves with the tools that we use.

echoing what was already said releases literally do not matter anymore since they're all just streaming site rips and fansubs have been dead for ages

the only thing i watch out is for hidive rips, as already mentioned in this thread they're bit starved garbage and they're immediately noticeable especially if you take screenshots, and what i usually try to get then is amazon rips, i dunno how licensing works but for some reason things tend to be available there when they're not on CR
>>3002
>Otherwise you're killing your poor SSD/HDD
modern ssds have vastly better endurance than when the first drives started hitting the market
i don't really think that was even an issue for those, but for the average 1 TB ssd drive nowadays you'd have to overwrite the entire drive every day for more than a year... on that account it's really a non issue

For seasonals I just get whatever has the most seeds.

For old shows I look at the AB comment section and see if anyone has posted a comparison, otherwise I check seadex, nyaa comments, etc.

I delete everything after watching except for a few favorites that I find particularly rewatchable, like Azumanga Daioh, Galaxy Angel, PPD, Patlabor 2 and Gunbuster.

>>2977
i do my best to avoid h265 because i think the grain looks better with a high bitrate h264, or even better h264 with 10 bit color profile aka 'Hi10p'. but generally speaking i go for filesize. i only minmax subs for anime i really care about, for example sora no woto i prefer chihiro subs.

>>2977
I completely agree though.
I'll get space-saving encodes at 720p with openings as seperate files for super long series and stuff I just kinda like and large HD stuff for the shows and movies I love the most.
If I happen to love a super long series, I'll either make space for the entire thing or just get my favourite episodes in HD.
I prefer when subtitles are in separate files so I can edit them or easily put them in a translator so people who don't speak English can watch with me.
I like hi10 for space-saving and erai raws for HD with multilanguage subtitles.
I like animencodes for dual audio stuff, but I only get it for certain shows.
...I don't actually know anything about this stuff myself, I just know that I can't torrent.
These days, I don't really download shows anymore anyway and just watch streams instead anyway...
...unless I want to do video editing or make gifs ofc

>>2980
if you're on private trackers, archiving suddenly becomes really important if you want to maintain your ratio

>>3153
Not the case at all on AB which is the main anime private tracker.

I posted many walls of text about this months/years ago that no one bothered to read. So I'm not going to do it again. But a summary;

1) File size is not an indication of quality
2) Fansubbing isn't dead but most people claiming to be 'fansubbers' now are just ripping groups that have no idea what they're doing
3) The vast majority of the above groups are just automating the process (badly) and begging for money. I don't understand why people give it to them.
4) Codecs are not an indication of quality but newer codecs (x265 vs x264 etc) have many features that can improve quality at a given bitrate provided the person encoding them knows what they're doing
5) Again; Most people don't know what they're doing and parrot myths about encoding. The vast majority of people claiming to be experts on places like nyaa have no idea how codecs work. This INCLUDES the 'respected uploaders'
6) Private trackers are the most retarded thing in the world. Not only are they not really 'private' they do nothing to protect you as a user. It's just a circle jerk where people make you beg them for access and gloat about having access. They do not help preserve anything. Many things important to the history of fansubbing will be lost in the coming years because people insist on using them.
7) The so-called 'cabal' is not only real and annoying. It actively screws over viewers and translates thing wrong on purpose to 'troll' them. The same people running the private trackers are the same people that control nyaa and all the 'best fansubs lists/wikis' and a lot of them also work for CR and other streaming websites. They have normalized rips over actual fansub translations because they do not want competition. They spread rumors and FUD about people outside of the circle. They conspire together to get other group's releases taken down. They make thousands of dollars a month by providing horrible releases to clueless idiots. They have ruined the industry and are the reason why AI translations and horrible human translations have become normalized. They are the reason for the drop in quality for things like typesetting. They are the reason most things are automated with scripts now. They get off on controlling/trolling 'leechers' because they have nothing else happening in their lives and no real power outside of the 'community'.
8) Video quality doesn't even matter honestly. Most people can not tell the difference between a 'bad' encode and a 'good' encode no matter what hardware they're using. It's mostly placebo for 99% of the population.
9) The _most important thing_ about a proper fansub release is the TRANSLATION and presentation of said translation. If there is a release with an original translation that's good it would be preferable to download that as compared to the kuso CR translation that translated things wrongly on purpose to 'troll' the viewers or otherwise make it more acceptable to a small insane portion of the population's idea of 'proper politics' and 'non-offensive language'. So many releases since about 2010 have this problem and it's always excused with excuses like: "Well at least we're getting it the same day as Japan" and "Look at this trollsub from the early 2000s this is so much better". The same people putting out those 'trollsubs' back then are the same people that work for places like CR now. They're still trolling you. They're just being more subtle about it these days.
10) These people have somehow convinced the Japanese side of the industry that this is 'what the west wants' and the only way to make the shows acceptable to a mostly American audience. But they're not pulling the same stuff elsewhere (SEA, Spanish, non-English speaking Europe) because they can not push their bullshit there due to the language barrier.
11) Most people 'translating' anime do not have a very good grasp on the Japanese language. Some CR employees working on translation openly admit that they can't pass the N4
12) I already went on for too long didn't I?

If you'd like me to sit down and type out the nitty gritty for video/audio quality again I suppose I could do it sometime in the next few days. But it's going to be a lot of math involved and most of it is not going to agree on what the so-called expert's blogs/wikis are going to say. When I've attempted to do it on other public places in the past typically such threads are instantly linked in some discord server somewhere. The thread gets flooded with shitposting and lies. Then I eat a ban or the post is otherwise buried and/or hidden.

>>3156
(cont.)

I for one am not a fan of discord. But I'll admit I made a burner account two or three years ago when I noticed my own releases being taken down or shouted down in certain places like nyaa. I also needed it to access certain software because these people refuse to use public places like git repos or the old forums like sane people. So I had an actual reason to lurk for a bit beyond my own curiosity about who these people are.

What I discovered was very disheartening and horrible. If you thought the IRC autism and in-fighting back in the day was bad you haven't seen anything. They not only openly conspire to prevent people from making releases outside of their control. They openly conspire to get new torrent trackers and nyaa mirrors taken down. They were the ones that got nyaa.pantsu taken down. They are the reason why you can no longer make nyaa accounts (they gatekeep uploading access and make a lot of money selling accounts). They have bots set-up to monitor for mentions of their names on large boards like /a/ so they can get alerted as a group and flood any thread to spread FUD or censor them. They have moderation access most everywhere relevant and use it to censor and ban people. They follow people around and harass them. They spread rumors about other people. They use influence to ensure only people in their in-group get hired at the major streaming/media companies. They do a lot of bad things.

The worse part though is they speak as if they're the authority on what makes a good 'fansub' and/or rip when I didn't find one among them that could read the source code of x264/x265/other codecs. They parrot a lot of audiophile/videophile myths and used them as justification to shit on other people's work and talk up their own. They have armies of bots and accounts to flood discussion and mass report other people's accounts on social media, reddit and a bunch of other places I don't use (but where such things are discussed anyway). They've even weaseled their way into certain software communities (the various auto-downloaders for things like Plex/Jellyfin etc) to ensure their own releases and accounts are always listed at the top and always the first to be auto-downloaded by casual viewers. They also control all the various lists of 'best releases' and wikis. Pretty much anything involved with modern 'fansubbing'.

Here is the truth: Most of us from back in the day moved on. A few of us are still around but we're excluded from participation for the most part. Aside from one or two groups that are mostly contributed to anonymously and more focused on things like preserving/archiving rare LaserDisc/VHS content before it's lost forever. It's very rare for an actual group to work on an airing show. Mostly because to do it invites so much trouble from the people mentioned above. But also because it's rare for a currently airing show to be interesting enough to take precedence over archiving rarer stuff.

Plus anyone that is actually good at translation can make far more money translating other stuff. Which is why all the people working for outfits like CR can barely pass the N5.

My personal background is with A/V work. I've been working with audio/video since the early-90s. I've been working with digital audio/video since about 1998. I have been around since we distro'ed mostly on VHS and used Amigas. I was one of the first people that was capturing VHS fansubs. First in RealMedia then later DivX 3.11 ;-) when it became an option. I helped develop things like xvid, virtuadub, nandub and various other A/V software. I still contribute to stuff like ffmpeg and x265 (well patches). I've been around.

I'll try to remember to come back later and dispel some of the common myths again. This has gotten longer than I expected and I have other things to attend to at the moment.

>>3156
>>3157
You are awesome. I suspected all of this for a long time.

Remember DameDame!? That was fun.

>>3156
>I posted many walls of text about this months/years ago that no one bothered to read. So I'm not going to do it again.
The threads are probably still up. I recall e.g. >>>/qa/120408, >>>/qa/122423, >>>/secret/28874, >>>/secret/30414

>>3156
>>3157
This is some very interesting stuff. I'd like to hear more, if there's anything else.

>>3156
>8) Video quality doesn't even matter honestly. Most people can not tell the difference between a 'bad' encode and a 'good' encode no matter what hardware they're using. It's mostly placebo for 99% of the population.
This is the only thing I might take contention with, but it really isn't that much of a problem on the fan encoder side (most of the time) and usually is a larger issue with the source itself being crap. Like BDs for Sailor Moon being crap while Utena had an absolutely gorgeous re-release "recently". Sorta makes me want to join up with some of those hobbyist groups that go around trying to find the original cels, or other potentially superior sources to a current release, for anime and then getting proper sources out there. Those people are the real heroes of classic anime.

https://www.youtube.com/watch?v=9YUNBvX6d3E
https://www.youtube.com/watch?v=sEnY1kdqbFU

>>3163
>but it really isn't that much of a problem on the fan encoder side
No most of them are actually really bad.
>usually is a larger issue with the source itself being crap.
and they typically make it worse

>>3166
How bad are we talking? I thought the source mattered a lot more for major differences in video quality.

For airing stuff I usually get the smaller encodes like ASW. For finished stuff I usually look around various encodes for opinions but unless there's a large difference in opinions I'll go with the most seeded one. For especially early 00s digital anime, bluray releases can be just upscales and sometimes with bad filtering, so getting the smaller DVD release might be better. I've made my own encodes too by combining aspects of different ones. Typically if there's a combination of video, audio, and subs that don't exist. I need to spend more time removing dub audio from some of my files.

>>3167
Do you have time to hear about our lord and savior Moozzi2?

>>3167
I don't have time to explain right now. I'll come back later today when I'm done cooking for my imouto's birthday party. I'll just leave it at this for the moment: Most of the 'filtering' they're doing is really bad and they parrot a lot of myths about what is and isn't 'quality'.

There are very very few people that actually know what they're doing and almost none of them are working in the industry (authoring BDs for both domestic and export markets), the streaming companies and least of all the 'experts' in the modern 'fansub community'. It's actually a huge insult to real fansubbers to call the modern community by that name. I prefer to call them ripping groups.

There is a real good reason why most of them aren't willing to share their *synth scripts.

>>3172
I don't know the specifics of him, but he seems to have a bunch of great releases including Utawarerumono and includes the BD menus and scans, which is really nice.
It lacks subs, which I need, but the effort to do all the conversion and scanning is very much appreciated.

>>3174
I was joking. His encodes for old (pre-digital era) anime are comically terrible.

>>3175
Oh. Well, he's the only option I saw for a bunch of stuff so I can't complain.

Filesizes have gotten too big for me to go for pure quality outside of BDs and fansubs are rarely fast or reliable enough to go for unless the official is especially atrocious.

>>2980
I usually don't clean up until the end of the season in case I need to go back and grab a screenshot for a post or something. The issue isn't that I don't have storage space for it, it's that I download a lot of stuff and have limited bandwidth.

>>3156
>a lot of them also work for CR and other streaming websites
You know, I never understood this. How are fags like h___z able to find work in the industry when they don't make even the slightest effort to hide their piracy-related activities?

>>3157
>They are the reason why you can no longer make nyaa accounts (they gatekeep uploading access and make a lot of money selling accounts).
I bought a game for a trilogy that only had game one and three on sukebei and my response when asking for an exception to the rule about not asking for accounts so I could actually upload the game bought directly from dlsite in its original zip was and this is a direct quote: "We also do not upload content by request nor do we have a preference for what is or isn't there". They literally do not care about preservation. They are actively harmful the scene.

For watching, I download from whoever's releasing the show, mostly prioritizing ASW and Ohys-Raws.
For archival stuff, I wait for BD encode of the shows I like, usually ReinForce, VCB-Studio, Moozzi2 (people say its shit, I don't see it), and the other well known ones. For shows that weren't really the best, I either get the 720p Dmon/Ohys BD release, or Judas and the like's BD release.
Haven't gotten around to downloading BDs for shows from the past two years or so though, been low on space.

>>3225
>Moozzi2 (people say its shit, I don't see it)
He puts ridiculous filters on it that sometimes completely change the color range. I think it's more visible on some releases than others, but you should avoid him if at all possible. VCB are probably the best of the main releasers.

>>3226
I see. Yeah, I go for moozzi2 as a last resort, despite not being able to tell the difference. Just in case.
>VCB are probably the best
That's nice to hear. Most of what I have is VCB. I can understand nihongo pretty fluently so I've even downloaded their non-English releases over other English releases.

>>2977
I just look for it to look nice, I have a 4k monitor and as long as it look good in fullscreen and there are no compression artifacts or stuff like that it's all good for me, I don't care about miniscule differences in color saturation that elitists hyperfocus on, I also don't care for upscaling or artificially increasing framerate of anime

I've seen many picture of this release vs picture of some other release comparisons, and they are trying to convince me that one is superior to the other while it literally looks like the same picture (just taken at slightly different timestamp to create an illusion of being different), it's some emperor's new clothes shit, I laugh upon the buffon

>>2998
>it depends on your eyes

So you're saying if the person can't see any difference something is wrong with their eyes? Perhaps also the brain?
You are part of the problem, anon.
As for your example, I see slightly different pixelation in the bushes in the background, none of this matters because if the scene is in motion no one will be paying attention to it. If you need to strain your eyes to see a miniscule difference on a static screenshot, you won't see any difference during the miliseconds of animated movement.

>>2977
For me it depends on which computer i'm using at the moment but most of the time I usually go with the smallest filesize since I don't have a big fancy monitor, I will go for a larger size release if it looks noticeably better and I'm planning to share it with friends.

>>4644
>>2993
I forgot to mention I also take a lot of screenshots, as was the intention of the attached image in the post.

How do niggers nowadays just upload a 24 minute chapter that weights almost 2 gb?

>>4648
AI upscale to 4k 60fps

>>4648
They have no idea what they're doing but the echo chambers they sit in (discord) encourage them and provide a lot of bad information. They've run out everyone that actually knew how to encode long ago. Same goes for anyone that can translate or typeset.

>>4648
>>4650
Another thing is they've harassed anyone that isn't in their circle to the point where they've fucked off to do something more fun. Imagine putting in a lot of effort on a show you really like only to have it banned from all the trackers while being harassed by a bunch of people that simply ripped a DVD/BD (and didn't even do that correctly). At most they might have taken an old fansub script and edited it to suit whatever their mental illness is. It becomes so frustrating dealing with them that you find something more fun to do. Since you know there is no chance more than 10 people will see your work anyway and no company is going to hire you for a living wage because these same types of people work for them and control staffing.

Same thing that happened to most hobbies. Attention whores flooded in and took it over. Viewers don't care about quality so they jump on the first thing available every time. They aren't willing to wait a few hours or a day for something of quality with love put into it. Even when you beat these people to the punch they just make up stories or have your release banned from all distro channels. They've even paid people off to ensure their own stuff is always listed above anything else in the automated torrenting applications lots of people use now to pull down content for their home media servers. Really nothing you can do about it either since they devote all of their time to this for some reason. I guess they're all rich NEETs or living on Government checks or something. If you bother to do the minimal effort to read the backlogs of their chats none of them speak English or Japanese as a first language either.

Modern releases are mostly all done by ESLs either using AI or editing old scripts (which aren't usually that great to start with). I guess they make enough off donations in those third world countries that they can devote time to doing it full time. They're all using automated scripts other people wrote for audio/video as well. They really have no idea about the underlying technology beyond spamming random commands at ffmpeg.

The smartest thing you can do is for your own enjoyment of this content is to learn Japanese then seek out untouched rips. BDs are usually garbage quality from the studio but at least it's untouched garbage. No one is really doing proper encodes anymore. If you don't want to store large BDs you can do much better than any ripper by using a simple degraining filter and using the default settings for x264/x265 along with encoding the audio to AAC or any lossy codec using the defaults. That's how bad things have gotten. The defaults in any GUI application like FFBatch produce better results than someone claiming to be encoding GURU using a bunch of filters.

The worst thing these people have done is push all the old good fansub releases from the 90s-mid 2000s off the public trackers. They flood all the public trackers with re-releases of that content and it's becoming really hard to find untouched scripts/subtitles these days. They also lock up untouched BD rips and scripts behind log-in walls like private trackers. You'd think in 2025 that there would be a nice public repo of scripts like we had in the 80s-2000s but it doesn't really exist anymore. If you didn't take the time to archive that stuff back then a lot of it is now lost media.

I learned enough Japanese so for seasonals I just download the first garbage quality download I can find and delete it when I'm done watching. Subs don't matter so if the quality is watchable I don't see why I shouldn't grab the first streaming rip available.
For older shows I try to find the old fansub releases just because it's a convenience to others if I seed them, and I'm likely to rewatch them.

Fansubs died with official subtitles. They're not coming back no matter what except for like a few shows here and there by bored neets (kissusubs).
The best thing that could happen would be to push official subs releases like Crunchyroll to start with writing an as faithful as possible subtitle script, and then creating a localized version out of that and having both available. Something like [en/honorifics/tlnotes] for those who want to understand as much as possible of the culture, and [en/localized] for those who don't like anime in the first place and want to replace everything that could possibly remind them that anime isn't made in their country.

>>4652
Any streaming release is going to be able equal quality these days (e.g. not great). There is no point in waiting for another streaming source unless the first is censored or something like that.

Be aware that a lot of the modern honorific scripts are simply the same localized garbage with -san/-chan/-senpai tacked on. So there is barely an advantage to getting those because nothing else has been modified. Like for example, food names or other cultural references. Modern alt scripts pretending to be less localized are mostly an excuse to shit out the horrible main script. Can't expect people that don't understand the language to do a good job in the first place.

Used to be you could use CR's script as an okay starting point but these days the only thing that might be of value is the timing. But even that isn't good most of the time. Last show we did we didn't even bother looking at CR's script because the translator could write his own from scratch faster. Then we'd time it in about an hour tops. It took longer to modify the existing CR script than to do our own.

Also a lot of the time for older shows it's best to avoid the BD re-release and seek out the old VHS, LD or DVD rips. Which are becoming hard to come by sadly. The BDs coming out for such shows are usually bad upscales of the DVD/LD where they didn't bother doing a proper detelecine before retelecining it. I don't understand why they continue to use telecine on modern discs considering all DVD and BD players could handle true progressive 24fps for decades now (player does the 24->29.xxfps convert on the fly). But that's how things are. The tools they're using in the studios are worse than the FOSS stuff. It's all done really lazy. But the majority of people can't tell the difference so they get away with it. The only redeeming thing about a modern BD for such shows is the higher bitrate by default. So in a few scenes you might be able to recover better frames that were broken on older discs due to compression and mpeg2 codec shitting itself at scene changes. But it's a huge pain in the ass to extract them and deal with the broken telecine.

LD sources are usually the best going for 70s-90s shows because video is stored as analog. With a domesday you can by-pass the capture card+composite video feed and get really good video from them. Much better than DVD sources for a lot of stuff because the video wasn't run through mpeg2 compression.

Also for some shows the broadcast versions can be better than the streamed ones. But it's getting harder to find those and the people producing them haven't updated their methods for capture+encoding in decades. So you have to get lucky and run across one done correctly with modern tools instead of it being 120fps .avi stuff from circa-2002.

Hmm, maybe this is a good thread to ask since people seem to pay attention to encoding and stuff.
What, if any benefit could there be to using a slow speed when encoding something? I've compared file sizes and it's completely negligible, so is it a quality thing? I just do it to burn subs for streams so maybe my usage is different from where this would matter.

>>4677
For burning in subs using a slower setting probably isn't a great idea unless you have time+energy to burn. The size differences between the profiles like slow vs. slower vs. placebo has more to do with quality than file size. You should be getting slightly improved quality at the same file size the further down the list you go. But the profiles don't enable every quality setting. Some things you have to enable manually. There are a ton of CLI options you can enable/disable manually, see: https://x265.readthedocs.io/en/master/cli.html

For x265's. A similar page for x264 exists too of course. A lot of these may or may not be placebo then there are others that aren't well supported by certain media players.

I'm not on my main PC I use for encoding right now. I can come back later and post some of my own settings if you want. But in general for what you're doing using medium (or the slow profile/preset) is probably good enough. When I'm doing a lossy -> lossy encode to burn in subtitles I don't really worry about the quality that much. As long as it's watchable it's good. Most people won't be able to tell anyway.

Also note that following some of the guides/blogs/posts on the web isn't the greatest idea. People think enabling a setting that takes 4+ more hours per episode much mean it's better quality. Even if the gains are minimal (or don't exist at all). You have to test them yourself and more importantly test the combination of them to see if enabling something in worth it.

For most people the defaults are fine and anything below the "slow" profile is a waste of time and CPU cycles.

>>4678
Thanks for the answer!

>I'm not on my main PC I use for encoding right now. I can come back later and post some of my own settings if you want.

Oh, you really don't need to bother. It was more of an offhanded question. These are videos compressed at 720p for a stream and then deleted afterwards. I'm not even sure how many people watch it at full 720p size.

>As long as it's watchable it's good.
Exactly.

>Even if the gains are minimal (or don't exist at all)
Yeah I did this a few years ago and took screenshots of multiple settings and didn't really notice much.

>>4680
>Yeah I did this a few years ago and took screenshots of multiple settings and didn't really notice much.
It's hard to notice from individual screenshots. You have to judge improvement in motion when we're talking about those little gains from going from slowest -> slower presets. Compressing individual frames is easy and achievable in any modern codec from the default fast/fastest setting. What the slower settings do is take those few real frames and make what's in between them (partial frames) look better. Slower settings are mainly built around looking further ahead during encoding and optimizing for when there are drastic changes between frames. It's smarter about where it puts I-frames when it can look further ahead so the bitrate is spread more evenly and optimal throughout the entire video. Where you see degrading quality first is always at scene cuts, pans and anything else with a lot of motion and/or stuff going on.

The main issue you'll have with what you're doing is the mild artifacts around the subtitle lines. If that is the case instead of going down an entire speed tweak the command it's giving to the codec that sets lookahead. For x265 this is called "--rc-lookahead". There are a few others related to it but that's the main one. I always set it to 120+. This only consumes more RAM and doesn't increase over all encoding time/CPU load. But it gives the codec a much larger batch of frames it can see at the same time to make decisions about where to place I-frames and how to place P/B-frames. You can also mess with b-frame bias and how many references they can use. But going really high on those can break some media players and you get very diminishing returns beyond about 6-8.

Lookahead is basically free boost in quality with no increased encoding time on any profile as long as you have the RAM. For me 120 frames of 1080p video takes about 14GB of RAM.