"MSharpen(strength=45) #if you really want to do this..." Normally I use Hysteria() but with the ConvertToYV24() I couldn't use it. But now that you showed me the real way to encode Hi444p, I can switch back to Hysteria (because I don't need the convert one). Which filter do you use for sharping?
lsfmod with strength 50-75 and defaults slow, though I don't use it on every show
Oh hey, I tried sending you a friends request again and didn't crash, how weird. Anyway I wanted to ask you, are there plans for Gakkou Gurashi's BDs or that's up to fnord?
dunno, probably not but if other people want to encode and qc them i won't stand in their way
when you're working in avisynth you need to be mindful of the colour space you're in. avisynth doesn't have much support for it and won't stop you from doing stupid/destructive things, so you need to keep track of it yourself and only perform appropriate actions for that colour space._to444 returns a stacked 16-bit image, but the other filters you're using are designed to run on a normal, 8-bit image. so you need to use them first (which is always best practice anyway - when possible, filter before downscaling). so move your antialias and sharpen above _to444 and use gradfun3(thr=0.5, lsb_in=true, lsb=true). that tells it to expect 16-bit stacked output and also to output 16-bit stacked output, so it will correctly operate on the output of _to444 without munging your video....and cross out that last sentence because i just looked at my scripts again and remembered that _to444 is designed to run on a stacked image in the first place. so just do something like this:FFVideoSource("C:\Anime\ep1.mkv") maa2(mask=0,chroma=true) MSharpen(strength=45) #if you really want to do this... gradfun3(thr=0.5, lsb=true) #after this, the video is now in stacked 16-bit _to444(1280,720) #still in stacked 16-bit dither_out() #now in interleaved 16-bit ready for x264 to work on it as raw video with --input-depth 16that should output what you want, provided you set up x264 right. the raw video should be 2560x720 and it will look like a mess in avsp but if you tell x264 it's 1280x720 with a depth of 16 it should work
http://ask.fm/Xythar/answer/132605113720 fair enough. Those terms are there mainly to differentiate some stuff, like Magician/Sorcerer (魔法使い - mahoutsukai) from Magus (魔術師 - majutsushi), or Magic (魔法 - mahou) from Magecraft (魔術 - majutsu). btw only magicians use magic and magi use magecraft.
I'm aware of that. We translated majutsushi as mage and the one use of mahoutsukai (when Kiritsugu visits Shirou in hospital) as magician, so you can probably see why I don't want to put "magus" for the latter (plus, "magus" isn't something you'd use to describe yourself to a young child - it's a much more obscure term than the simple Japanese)
Magician because the root of the term is magic and magus sounds too similar to mage Fortification I don't remember why but the TLC chose specific terms to be used in reference to Shirou's ability so I'm not going to change them for no reason As far as I know we are doing the BDs but the only thing I plan to change is the previews and songs for consistency.
I don't know if you'll work on UBW cour2 BD but if you will, I was wondering if you can consider doing some changes to some terms, like "magician" to "magus", "fortification" to "strengthening" and so on.
nah, the terms we used were used for a reason
>phone crash every time I try to add you in SS but can send you messages just fine. W-Well it's not like I wanted to have you in my friendslist anyway.
no, using _to444 on its own is fine. it outputs stacked 16-bit, similar to using gradfun3(lsb=true) so you can run dither_out on that and pass it straight to x264 10bit as raw video with --input-depth=16or if you're doing an 8bit encode and don't mind dithering it back down in avisynth space, just add ditherpost(mode=7) or something
http://ask.fm/Xythar/answer/132568844920 my source is 1920x1080 and I downscale it with "nnedi3_resize16(1280,720)" - Okay, I add the "input". Can't hurt, I guess.
you can't do it that way either because you destroy the extra chroma info when you downscale in 4:2:0 land. just doing a convert after the fact won't change that.here are the filters i use for converting to 4:4:4 - http://pastebin.com/VULUCMJJ i ended up with a few different versions for if you want to use debilinear/debicubic etc on the luma as well, though maybe try the basic _to444 to begin with unless you know how to mask high-detail stuff like credits
I want to mamoru Mamoru Miyano's voice acting.
yes
I don't mind group specific ones, but are they for randoms / leechers as well?
not really no
Do you know whether there's a teamspeak for fansubbers and leechers? I mean like a massive public one.
dunno. i think the most you'll find are group-specific ones. commie used to have one (not sure if that still holds) and i remember kaylith having one too when they were alive
You know you want to buy some jewels too, don't let SSR Miku go.
.t totally not Scamco
yes, i look forward to paying for my third useless pull