Yu-Gi-Oh! Duel Monsters - Pyramid of Light (720p, BDxHDTV, Japanese audio)

Category:
Date:
2021-06-12 18:39 UTC
Submitter:
Seeders:
3
Information:
irc://irc.hoshinet.org/fusion-rips
Leechers:
0
File size:
2.7 GiB
Completed:
159
Info hash:
425fa2c8ea16564e53a5191f2292f49231c6e1cd
![screenshot](https://i.imgur.com/ycrZQ1t.png) This is a ridiculously complex hackjob to insert as much footage as possible from the US Blu-ray release of this movie into an HDTV recording of the Japanese version. For reference, the Japanese version is 10 minutes longer and the English dub is a complete hackjob, so this was anything but easy. The final encoding script for this file ended up being *290 lines long* and took eight hours straight to write. If anybody cares to see it, [go nuts](https://pastebin.com/GCV0jLNz). The subtitles were translated long ago by Tatsu no Kiba Productions, and were edited to match the terminology used by Crunchyroll.

File list

  • Yu-Gi-Oh! Duel Monsters - Pyramid of Light (720p) (BDxHDTV) [F-R][a3051bb0].mkv (2.7 GiB)
Holy shit what is this script. Vapoursynth uses Python, so use it :notlikethis:
can we please have a filter that hides trusted torrents

LCE

User
Jesus fucking christ that's horrifying
why force people to watch mediocre encodes instead of sharing the source files you have?
I mean... I guess I can share the animax TS file if you want it? The dub BD’s on U2.
@Aergia Seriously? Like really? What a dumb fucking comment. So many choosing beggars on this site like holy shit. Imagine thinking the MPEG-2 source TS was actually better than the encode itself. Imagine someone putting effort into taking a shitty source, the only HD one available for the Japanese version, combining it with the best dub version available and complaining because the shitty source was shitty. Get a grip dude.

LCE

User
@Puto please do
> The final encoding script for this file ended up being 290 lines long and took eight hours straight to write. This isn't something to brag about. It's easily one of the most inefficient scripts I've seen in my life and could've been done in like, 40 lines, maybe less. You really need to take a python primer.
I didn’t write it to be efficient, I just went through the thing in vspreview with the videos stacked together and added an extra line every time I came across a cut. Having it written more efficiently would take a lot more work for an identical result, so why bother?
because writing this the normal way would've taken 60 minutes instead of 480
But at this point, it makes 0 difference? The source file is useless at this point.
@SeaSmoke, It really wouldn’t have, because what took 8 hours was seeking frame by frame tracking down all the edits, not writing the Trim commands. I could’ve condensed it down to less lines, but it’d take a lot more work given this specific workflow (load both videos stacked together, add a line every time they go out of sync, refresh and continue looking, rinse and repeat). @LCE I’m uploading the TS file right now, but it’s kind of enormous so it’ll take a while to get it uploaded to the seedbox.
Simply writing a couple of loops that iterate over a list of tuples/ints could've made the script vastly more readable and a lot faster to write. Efficiency isn't something that will always take *much* longer by necessity. Instead, you now have a couple hundred lines that makes things unnecessary complicated. Also some rather weird choices that make me go ??? (like, why are you doing `a+a+a+a+a+a` instead of `a*10` or whatever value you need? Why are you using `std.Trim` when simple python slicing works fine and would've been much easier to read? What is that odd `insertclip` module you're loading in, and why are you not just using `vsutil`'s `insert_clip`? Why are you running `Tweak` *3 times* with different params instead of all at once? A lot of questions keep popping up the more I read this script). Writing code you can easily read again later is fundamental, especially if you plan on also sharing it around. Right now it looks more like you want to show off how poorly you can write scripts, and it kinda baffles me. There was a good reason I said you should be using the Python aspect it offers—It makes writing code a lot easier if you do. @AwfulCaps Having the source available means more people have the ability to try their hand at encoding this. I don't get why you're so upset about him asking for the source files. It'd certainly help in making me understand some of the actions taken in the script he posted!
^ This pretty much > I could’ve condensed it down to less lines, but it’d take a lot more work given this specific workflow. Doubt. I've done similar stuff previously and never had to do something like this. I would've rewritten some of it to show you what I mean but I can barely make any sense of the code, and that's part of the problem here. Simplest example of extra code would be: ``` video = adjust.Tweak(video,bright=0.8) video = adjust.Tweak(video,cont=1.02) video = adjust.Tweak(video,sat=1.2) ``` instead of: ``` video = adjust.Tweak(video, bright=0.8, cont=1.02, sat=1.2) ```
I had it like that originally, but for some reason it wasn’t applying the cont and sat parameters in the preview, so I ended up separating them.
To be fair, I'd get it if he wanted to very specifically want the contrast and saturation to affect the brightness-adjusted clip and such, but uhh... you could certainly write at least a comment to indicate that! And looking at the Adjust code, I'm not even sure it really matters much besides maybe doing `sat` at the end after all, since it does perform that operation before `cont` and `bright` (which I'm also kinda ?? about, as I am with most ports of older AviSynth code, honestly).
> Why are you using std.Trim when simple python slicing works fine and would’ve been much easier to read? Habit, I’m used to Avisynth. Insertclip is a module I adapted from avisynth, https://kyodaimax.puto.pt/insertclip.py And honestly, the script wasn’t written with the intent to be shared, I just put it in there as a “Jesus Christ the dub was hackjobbed to ridiculous WHY” venting kind of thing.
> avisynth Explains why I can't read anything
99% of the script is plain, basic splicing. I don’t understand what there is to understand.
Fair, although I think Python slicing is easy enough to adapt to that I'd recommend using that over Trim whenever possible myself (do keep in mind that it's exclusive, unlike Trim which is inclusive, but that's easy enough to used to). For reference, this is vsutil's [insert_clip](https://github.com/Irrational-Encoding-Wizardry/vsutil/blob/513576abd4bb08605f1f91ad32780ba16c7f3fea/vsutil/clips.py#L115-L136). Doesn't have an `end` param, but for that you can simply trim the clip before passing it. ```py from adjust import Tweak from vsutil import insert_clip as ic src = core.ffms2.Source(r"PATH/TO/CLIP") adj = Tweak(src, cont=2.0) inserted_clip = ic(src, adj[:101], start_frame=100) # This replaces frames 100-199 in `src` with frames 0-100 from `adj` ``` I think we share a server or two, so if you want general VS pointers, I can help out with a simple ping if you need me to. Alternatively, some people make most of their scripts public (like [me](https://github.com/LightArrowsEXE/Encoding-Projects)!), so you could always look through their more recent scripts to see Python VS code in action.
To be honest, I only moved to Vapoursynth from AVS a few months ago, which is why my scripts tend to read like an avs translation.
Explains a lot, and it's especially hard with how awful the general packaging situation is with Vapoursynth. That said, there's still plenty of people around who are more up to speed with the generally good filters nowadays and what has greatly improved from older functions, so you could always ask around for pointers.
https://nyaa.si/view/1397429 here’s the animax TS, as requested
> I don’t understand what there is to understand. Nothing, I just get an aneurysm trying to read it.
Remember when anime used to be about the anime and not bitching about encoding?
no lol. even in the mid-late 2000s (debatably the peak of fansubbing) there was a ton of shitflinging about switching codecs. also insert "encoding isn't an important part of the fansubbing process" quote before herkz posts it
Talking about encodes is fine, being a condescending prick when someone is just trying their best is not. And I am not applying this to actual garbage stuff, like MTL releases.

sarachikorita (uploader)

Trusted
We're kind-of a magnet for choosing beggars.