[◯PMan] Idol Time PriPara Creditless Openings & Endings (BD 1080p FLAC) :: Nyaa ISS

[◯PMan] Idol Time PriPara Creditless Openings & Endings (BD 1080p FLAC)

Category:
Date:
2021-12-02 23:06 UTC
Submitter:
Anonymous
Seeders:
2
Information:
No information.
Leechers:
0
File size:
1.6 GiB
Completed:
53
Info hash:
b65b6a9ce6b987b13dda87f001eea338e34d3525
![](https://cdn-eu.anidb.net/images/main/197078.jpg) Encoded upon request. These BDs are incredibly blurry, which is pretty ironic considering it was mostly a sharp bicubic upscale. Filtering is pretty standard with a couple of fun caveats. The show uses a mix of Spline16 and Sharp Bicubic for the upscaling. That is, some scenes are upscaled using one, others with the other. To deal with this I wrote an easy function that compares a Spline16 descale and a Sharp Bicubic descale with the original pre-descaled frame, and have it pick whichever it deemed to be closer to the original frame. This worked for the most part, except for very blurry scenes (where the difference is not really perceivable anyway, so it's not really a problem). This is not the end of the descaling issues, though. A certain scene in OP3 seemed to have been upscaled per field. This means every field (or odd and even sets of rows of a frame in layman terms) were upscaled independently, causing a very bad-looking combing-like effect on the lineart. I worked around this by simply descaling the fields independently, merging them together, and rescaling the combined frame. This had to be scenefiltered, but I didn't notice it anywhere else. The rest of the filtering is fairly standard. Reupscaled using nnedi3, denoised using dfttest to get rid of compression gunk, weak CCD to denoise the chroma, decsized, AA'd using based_aa and sraa clamped together (with scenefiltered transposed sraa for specific scenes), very slight line darkening to offset the line brightening from descaling sharp bicubic, averaged debanding (f3kdb @ low settings + placebo @ low settings to deal with certain types of banding f3kdb couldn't handle), and then regrained. Requests are welcome, but don't count on me to follow through with yours from the get-go. A sheet with links to all my releases, as well as OP/EDs I plan on encoding, can be found [here](https://docs.google.com/spreadsheets/d/1VYJymlJMgQG36upLIT_P_evI5mUkEYnmAcg06Z0P28U/edit?usp=sharing) (not guaranteed to be up-to-date). Encoding scripts and additional stuff can be found [here](https://github.com/LightArrowsEXE/Encoding-Projects/tree/master/%5BOPMan%5D).

File list

  • [◯PMan] Idol Time PriPara Creditless Openings & Endings (BD 1080p FLAC)
    • [◯PMan] Idol Time PriPara Creditless ED1v1 (BD 1080p FLAC) [4F95AF10].mkv (83.7 MiB)
    • [◯PMan] Idol Time PriPara Creditless ED1v2 (BD 1080p FLAC) [E114C8E6].mkv (92.2 MiB)
    • [◯PMan] Idol Time PriPara Creditless ED1v3 (BD 1080p FLAC) [47A8D634].mkv (99.0 MiB)
    • [◯PMan] Idol Time PriPara Creditless ED1v4 (BD 1080p FLAC) [716F6921].mkv (94.2 MiB)
    • [◯PMan] Idol Time PriPara Creditless ED1v5 (BD 1080p FLAC) [6AECC514].mkv (87.3 MiB)
    • [◯PMan] Idol Time PriPara Creditless ED2 (BD 1080p FLAC) [22885169].mkv (151.6 MiB)
    • [◯PMan] Idol Time PriPara Creditless ED3 (BD 1080p FLAC) [FCEF35DD].mkv (85.5 MiB)
    • [◯PMan] Idol Time PriPara Creditless Ep50 ED (BD 1080p FLAC) [564383F9].mkv (38.0 MiB)
    • [◯PMan] Idol Time PriPara Creditless Ep51 ED (BD 1080p FLAC) [DF61D726].mkv (81.8 MiB)
    • [◯PMan] Idol Time PriPara Creditless OP1v1 (BD 1080p FLAC) [45EA0B0A].mkv (139.2 MiB)
    • [◯PMan] Idol Time PriPara Creditless OP1v2 (BD 1080p FLAC) [9ABBC003].mkv (138.2 MiB)
    • [◯PMan] Idol Time PriPara Creditless OP2 (BD 1080p FLAC) [1B053D06].mkv (172.2 MiB)
    • [◯PMan] Idol Time PriPara Creditless OP3v1 (BD 1080p FLAC) [8512C1CF].mkv (180.7 MiB)
    • [◯PMan] Idol Time PriPara Creditless OP3v2 (BD 1080p FLAC) [7EA2E79E].mkv (183.8 MiB)
>A certain scene in OP3 seemed to have upscaled per field. This means every field (or odd and even row of a frame in layman terms) were upscaled independently, causing a very bad-looking combing-like effect on the lineart. so it's cross-converted? if so, there's some simple filters to correct it (though i'm not entirely sure if any have been ported to vapoursynth).
oh yeah apparently there is https://github.com/YomikoR/VapourSynth-Destripe
Yeah, I just forgot about the term/module name and was too lazy to look it up at the time ([but not lazy enough to just write my own function for it lol](https://github.com/LightArrowsEXE/Encoding-Projects/blob/master/%5BOPMan%5D/Seasonals%20%5BTV%5D/2017-2%20-%20Spring/%5Ba12737%5D%20Idol%20Time%20PriPara/project_module/filter.py#L155-L176)). Fortunately it's not a particularly hard concept to write once you get over the fact that People Actually Do This:tm:. Basically the same thing happened with the auto-kernel-picking descale, where I totally forgot [kageru had written a function](https://github.com/Irrational-Encoding-Wizardry/kagefunc/blob/master/kagefunc.py#L135-L179) that does basically the same thing forever ago and I just basically [rewrote the function](https://github.com/LightArrowsEXE/Encoding-Projects/blob/master/%5BOPMan%5D/Seasonals%20%5BTV%5D/2017-2%20-%20Spring/%5Ba12737%5D%20Idol%20Time%20PriPara/project_module/filter.py#L116-L152). Good thing I actually prefer my code in the end, so it wasn't for naught!
yeah well i have no idea how to even do basic programming so 100% of my vapoursynth scripts are copied and pasted from shit i found online
Programming is for nerds.
I'm not a programmer
Oh, neat, thanks <3
super hacker light
only programmers use exe in their handles