Doctor M

I recall a lot of discussion awhile back about some interlace artifacts remaining in the GOUT transfer.  A lot of filters were tossed around as being useful, but what is available has improved.

I don't see a specific filter being used for cleaning that, and they are probably removed as a by-product of something else (probably the anti-aliasing), but I'd like to suggest one more thing:

TFM(pp=2,clip2=TempGaussMC_beta2(edimode="nnedi2").SelectEven)

TempGaussMC with nnedi2 will bring most computers to their knees (I think I can render 1 frame every 30 seconds with it), but TempGaussMC turns on ONLY when TFM decides a frame has interlace artifacts.  The rest of the time it does nothing.

Image quality is pretty much a perfect extrapolation for these frames.

Hey, just a quick OT thank you.

I was working on another project today and the source had some unpleasant aliasing.  I remembered a line in this script about it:

########## anti-alias
NNEDI2(dh=true, field=0)
NNEDI2(dh=false,field=1)

I've never seen that done before, I'm not sure HOW it works, but it does a beautiful job.  Thanks.

I suppose someone could just rip the GOUT discs, use vobdnorm and do a listening test.

Not completely true.  Dialog normalization sets the level that is assumed to be normal listening level for the dialog.

Now maybe I misunderstand how it works, but I believe by assigning a dialnorm value you crush the softer audio.

You might turn it louder, but I don't think the range is the same.

I think it would be correct to say that all films in the franchise are "black sheep".

The first film, in and of it self worked, and it is good.

The second film was garbage, and pretty much everyone pretends it doesn't exist.

You've trashed the third film, while being questionable in its continuity, is quite enjoyable and really doesn't suck too badly.

The fourth film was a horror I've tried to forget.  Combining Connor and Duncan?  Ugh.  The unbreakable move or whatever?  WT?

Highlander 5?  Pleeease.  It was recut so many times, you'd think it was a Matrix fanedit.  It might as well have been made in Canada by The Asylum direct for the SyFy channel.

H3 is the closest thing we have to a watchable movie outside of H1.  Wouldn't it just be a better to keep the movies and TV series separate entities and make #3 continuitous with #1 (or for that matter #1 continuitous with #3?) and just smooth things out a bit.

Is continuitous even a word?

Curious to see your change list though.

Doctor M said:

TheDigitalBits.com reported today that there will be a new take on Pastoral's censoring on BD.  "More seamless" using newer digital tools.

What got me, is they commented that bootleg uncensored cuts are available if you know where to look.

I wasn't aware of this.  If we don't know where to look, who does?

hairy_hen said:

Combined with the non-defeatable 4 db reduction in total volume that comes from the Dialnorm parameter in the AC3 stream, the result sounds very anemic, much less powerful than it should.

Non-defeatable?  Why didn't you ask: http://www.coises.com/software/vobdnorm.htm

The only problems are that it's command-line, only works with VOB files, and you will need to do each VOB separately (it doesn't automatically move to the next file).

Still, it does just what it says on the tin: it lets you rewrite the dialog norm to any value without re-encoding.

I had previously made corrections to the second disc with the extras.

Rechecking it now I'm seeing a weird behavior.

Can someone that has disc 2 tell me if every extra after playing exits to the root menu instead of the submenu that you just made your selection from?

One more suggestion: fft3dfilter includes a dering filter... and fft3dgpu is mod of it that offloads the work to video hardware.

Could improve performance by quite a bit since that would offload 2 steps to the gpu.

Probably a stupid question, but did you try dgdecod with the cpu > 4?

I've had pretty good success and it's fairly fast.

cpu: 0 to 6 (default: 0)
Post-Processing Quickset Options.
(Y=luma, C=chroma, H=horizontal, V=vertical)
- 0: DISABLE POST-PROCESSING
- 1: DEBLOCK_Y_H
- 2: DEBLOCK_Y_H, DEBLOCK_Y_V
- 3: DEBLOCK_Y_H, DEBLOCK_Y_V, DEBLOCK_C_H
- 4: DEBLOCK_Y_H, DEBLOCK_Y_V, DEBLOCK_C_H, DEBLOCK_C_V
- 5: DEBLOCK_Y_H, DEBLOCK_Y_V, DEBLOCK_C_H, DEBLOCK_C_V, DERING_Y
- 6: DEBLOCK_Y_H, DEBLOCK_Y_V, DEBLOCK_C_H, DEBLOCK_C_V, DERING_Y, DERING_C

You can use cpu2 if you want to disable/enable specific options (such as the strong deblocking).

I wonder what condition those deleted scenes will be in.  Hopefully they'll get some finishing so some diligent editor can give us a 6 hour cut or whatever.

Btw, I finally finished the most perfect edit of Matrix that fixes everything wrong: I got rid of Matrix 2 & 3 from my collection.  Problem solved. ;)

Having compared more what is available, I will make this completely uninformed statement:

The PAL GOUT discs of ANH and ESB were cropped and zoomed when being transferred from the master, removing any advantage PAL might have had in preserving detail.

ROTJ is cropped more at the top but LESS at the bottom allowing it to retain a few more lines of resolution than the NTSC disc.  Probably still the same master though.

And one piece of wild speculation: The master used in both versions probably contains MORE detail than is apparent on the DVDs. 

I'm starting to wonder if it was intentionally reformatted to letterboxed to reduce detail (and in hopes of a resell in anamorphic later should the inital release go well enough).

The PAL ROTJ would suggest that (since if you used the same NTSC master, there would be no more detail (or lines when formatting) to gain.)

Edit: Although a less conspiracy minded answer might be that some nitwit just overcropped the NTSC ROTJ during tansfer.

http://aptirrelevance.com/otscreenshots/index.php has screenshots for comparison which include PAL & NTSC GOUT, but that is ANH only.

The gist of the other thread is that ROTJ is probably slightly improved in PAL vs NTSC, but not ANH and ESB.  The reason is unclear.

Also be aware, the PAL DVDs seem to be cropped more (in ANH and ESB, just differently in ROTJ).

http://originaltrilogy.com/forum/topic.cfm/Are-the-PAL-GOUT-DVDs-upscaled-from-the-NTSC-masters/topic/6531/

http://originaltrilogy.com/forum/topic.cfm/GOUT-improved-versions/topic/11377/

Leaked BD Aliens screenies: http://hcc.techradar.com/playback/coming-soon/blu-ray-screenshot-gallery-alien-anthology-aliens-07-09-10

The clean up was definitely NOT a problem (although I may have gotten some drool on the screenshots).

This is sort of a cross-post, but only because I scrapped the original project (which was to be a fan edit using the graphic novel) and transitioned to a preservation.

The original thread is here, which the admins can feel free to lock or delete.

This is a custom 5th disc to go with The Middleman Complete Series DVD set.

The final episode of the series was never filmed, but later turned into a graphic novel.  In 2009 at Comic-Con most of the cast was reassembled and they performed a complete table read of the episode.

It ties up all of the plot threads left over from season 1, and is riddled with great call-back gags.

This DVD contains that table read with subtitles (that were necessary because of audio quality and the rapid line delivery).

The menus are custom and closely match the rest of the discs in the set.

It also includes:
The Middleman Exclusive Interviews by ComicsOnline (following the table read).
“Dinner” faux Public Service Announcement (missing from disc 4)
TV Premiere Trailer
Wendy Watson Trailer

All video is culled from the net so expect YouTube-ish quality or less.
The table read is cut together from 2 sources to fix some video issues.

DVD_ROM extras include a backup .SRT of the English subtitles and .PDF files of The DDA Graphic Novel Script and The Pilot Episode Sanction Script.

I’ll post screenshots, but I get the impression there aren’t a lot of MM fans here… so this is more of an FYI.

Okay, CompMovieGuy, it sounded like you were assuming DGIndex could physically inverse telecine.

Mind you, if I had notice you also asked g-spot a question, I probably wouldn't have said anything. :-)

Btw, I HAVE seen over 95% cause problems.  I usually like to know WHY a video isn't reporting 100%.

CompMovieGuy said:

g-force said:

Chewtobacca said:

Stupid question: you guys do force film the video in DGIndex before using the script, don't you? 

Yes, force film is a must.

I think I would rather IVTC in AVS then use DGIndex's Force Film since its very hit n miss. Although I have never tried it on the GOUT so it could work well in that instance.

Any thoughts on that g-spot?

CompMovieGuy: What?  I know you've done a lot of great conversions, but this doesn't make any sense.  I'm assuming I'm misunderstand you, but for others who might not know...

DGIndex's force film doesn't inverse telecine video!

DGIndex INDEXES mpeg video so a frame based program like avisynth can work with it frame accurately and non-linearly as it would AVI content.

Force film or honor pulldown flags just determines if avisynth (or whatever program) will soft interlace progressive content whenever a pulldown flag is present.

I.e., add pulldown & interlacing where it isn't present in progressive material.

The only reason to use "honor pulldown flags" instead of "force film" is if your source is not 100% progressive (less than 95% is the rule of thumb, DGIndex will give you this number).  I have seen films/tv shows that have patches of hard telecining within progressive content.  (Anime is notorious for variable frame rate).

This is a problem because mpeg can have variable frame rate, AVI content must have a constant frame rate.

Using honor pulldown flags will make avisynth (or whatever) generate 29.97 interlaced video adding soft telecining in progressive areas and leaving hard telecined areas alone.

From 100% telecined you can then use an IVTC filter like telecide/decimate which will restore progressive frames from material that is soft telecined during playback.

Mind you, that still assumes that the interlaced portions are film sourced.  They can just as well be true video (59.94 unique fields) in which case you will need to deinterlace to 29.97p or use something like AnimeIVTC().

Or to rip off Doom9:

FORCE FILM WHEN YOU CAN, IVTC WHEN YOU CAN'T FORCE FILM, AND DEINTERLACE WHEN YOU CAN'T DO EITHER.

http://www.doom9.org/ivtc-tut.htm has more info for those looking to read up.  It refers to DVD2AVI but that's pretty much the same as DGIndex (which is build from it).

Really?  Do you have a link that says the S.E. will be 40 minutes longer and not the same as the extended cut currently in theaters?

I'd heard him say there were 40 minutes he could add, but never that he was intending to do it.

Thanks for the refresher Moth3r.  I haven't used IfoEdit in years.

Muxman isn't turning your subtitles red.  .SUP files do not contain color information, just the subpicture images, so Muxman's default color palette is being used (which does not match up with your original colors).

You can use a program like DVDSubEdit (free) (http://download.videohelp.com/DVDSubEdit/) or DVDRemake Pro (pay but awesome) to correct the colors manually after authoring.  (IfoEdit and PGCEdit might also work, but I've never tried.)

Now if you are muxing with Muxman and then using a program like VobBlanker to substitute your newly remuxed movie into the original full DVD, the original DVD's color scheme will take over and self correct your problem.

Updated a couple sentences in the subtitle conversion section: I now recommend Bilinear resizing instead of Mitchell or Lanczos3.

Having now played with BDSup2Sub more, I find that Lanczos3 is the sharpest, but it crushes some of the detail and can make text look a smooshed or lopsided.

TheDigitalBits.com reported today that there will be a new take on Pastoral's censoring on BD.  "More seamless" using newer digital tools.

What got me, is they commented that bootleg uncensored cuts are available if you know where to look.

I wasn't aware of this.  If we don't know where to look, who does?

For 35 pounds?  That's a pretty amazing deal.

Chewtobacca said:

Thanks, Doctor M. 

I am really interested in whether this latest version of the script takes NAB into account, as I am almost ready to encode with it.  One option might be to make a non-anamorphic MKV or AVI with the script that would automatically be resized on playback.

Feeling stupid, but what is NAB?

I think you'd want an anamorphic MKV if you are resizing it anyway during playback.

I used to only use Lanczos4 for upscaling as well, but it can produce a ringing effect.  AviSynth added spline because it produces similar quality without the artifacts.