Trigun (new Funimation release)

[Castor Troy]

YATTA works fine on it, also stop using YATTA as you are not trained and mirko is not qualified to teach you. I don't even know why you are trying this. YATTA can do everything it needs to. If you honestly think the program cannot do what you want, then YOU are just incapable of using it.

@Castor Troy: um no, stop being dumb please. The R2j disk absolutely shits on this. Do you even have eyesight? Like a tiny bit? At all?

I think you need to take the beer googles off, my friend.

[Cannonaire]

Field from the Japanese R2 release:


Oh hell. It seems to be like this for the whole episode.

[mirkosp]

OTOH the R2J seems to be hella more detailed than the funi release. Something tells me it was given a much more proper bitrate...

[Cannonaire]

The scenes are from different episodes. The first scene posted (from the Funi release) is from episode 2, which is really blurry in the R2J release as well. There is no consistency of quality in either the R1 or the R2 release. The scene shown from the R2J release looks pretty much the same in the R1, except they removed all the wonderful grain that was in the R2 and also the blending doesn't looks nearly as bad in motion somehow on the R1s. Maybe later I'll post the same field but from the Funi release to show the difference.

Regardless, the R2s are not magically and perfectly superior to the new R1 release. I would actually say that, aside from the grain, the new Funi release is a lot easier to work with. It basically seems that all of the issues in the (new) R1 release were present in the R2, just covered by a healthy grain. This includes the haloing as well, which seems to result from aggressive over-sharpening in the R2 release.

[Cannonaire]

Here is a comparison of the two releases. I think this is the same field, but I could be wrong:

Spoiler :

Region 1 Funi:


Region 2:

It seems the R2J release got a much better bitrate and they removed a lot of the nice grain for the R1 . Fundamentally though, they are both crap. It seems this episode doesn't have the haloing problem, but most episodes have significant haloing due to most likely over/improper sharpening in both the R1 and R2 release, as are most of the problems. Like I said before, regardless of the release there is no consistency in quality between different episodes, as some look really good but others look like... well whatever you would call the screens I posted.

Here are a couple bonus shots from the R1 funimation release in which I tried to clean up the haloing. It's not perfect, and it kills detail, but I think it looks better without the halos.

Spoiler :

Original:


DeHalo_alpha(rx=1.8, ry=1.8, darkstr=1.3, brightstr=1.7, lowsens=10, highsens=10, ss=2.0)
(and some LSFMod):

So sad. I hope when/if they finally release a blu-ray version they can fix some of the fundamental problems we're seeing.

[Mister Hatt]

The main reason R1's look worse is the fact that funi insist on cramming far too many episodes onto much smaller disks.

I don't like bluray anime most of the time. Everyone has a huge hardon for 1080p, so content that ISN'T 1080p always gets hurfed on. Trigun is ofc a cell animated series, so it can be recaptured at 1080p, but most recapturing companies in Japan at least seem to such fairly bad. There are a few exceptions though. America however just pointresize().warpsharp()'s everything from 480i to 1080p and thinks it looks :AWSUM:

Re your clean fields comment, it might just be a strange pattern, but if you cycled every possible match and didn't find anything then I guess it's just shitty mastering. The Japanese DVD isn't covered by extra grain to hide issues, it is more that the American DVD smoothed the shit out of it. Grain doesn't get in the way of matching, but other factors might be confusing TFM. If this is happening, you didn't do your metric collections properly, if you collected metrics at all. If you skipped on metrics, then you either don't know how to use YATTA properly, or you know far more about manual IVTC than anyone else on this forum. I think even I have only done a fully manual IVTC a dozen or so times.

[Cannonaire]

Thanks for the input, Hatt. Yeah, definitely looks like Funi decided for no good reason that grain was a defect.

When i first tried YATTA, I just went straight in without collecting metrics which obviously didn't work. I've since collected metrics with YMC, but I haven't had time to go any further with learning the correct way to use YATTA yet. Maybe tonight or tomorrow.

[Mister Hatt]

Actually for the best result you don't WANT metrics. However that is some high level shit and I don't know too many people who do that unless they know for a fact that hey are dealing with single-pattern material (hi TokyoMX)

I don't think funi wanted to remove the grain, rather I think they just over-quantized the shit out of it to fit so many episodes on a disc.

[Scintilla]

I don't think funi wanted to remove the grain, rather I think they just over-quantized the shit out of it to fit so many episodes on a disc.

Anyone know if the original Pioneer release has the same problem? I remember that being something like six to eight disks.

[Cannonaire]

I don't have any working experience with the Pioneer/Geneon release, but I have looked back at a bunch of old AMVs. Due to the high compression, I can't say for certain whether the grain was destroyed, but there appears to be a lot of random blocking/noise that looks an awful lot like what happens when you compress very grainy footage. I would guess the Pioneer release keeps most of the grain, however the transfer is pretty bad. There is severe rainbowing all over, and though I haven't found a specific example yet, dot crawl often goes hand-in-hand with rainbowing. Looking at footage from different episodes I have worked with in the context of the old AMVs, they also have the same problems plaguing the R2J release, episode 2 for example being very blurry with blended fields but no halos, and most of the rest of the series having bad haloing.

Bottom line:
Until someone decides to make a new R2 release without these problems (if that's even possible at this point), all sources you can get will have problems. Right now the best option is still the R2J DVD release, and choosing between the new and old R1 releases leaves you with a second-hand mess.