Harmy said:
...
Here's the frame from the !!!OFFICIAL!!! retail 2004 DVD:
It's not an intentional change...
Of course it is. That's clearly another Lucasfilm Deliberate Creative Decision™.
Harmy - as I mentioned in the 70mm thread, I got a chance to watch this at the weekend and thought I'd post my comments. Sorry I didn't post earlier.
Before I start picking nits and getting Rob all distressed, I'd just reiterate how important this project is. It really does seem like, for the foreseeable future at least, it's the fans who will save Star Wars. Hairy Hen's fan made 70mm 6-track recreation is wonderful, and Harmy's video is just as impressive.
When I watched this (with my son) I didn't notice any "new" or SE elements. The 2 hours flew by; even though I knew it was a reconstruction and not authentic 1977 it felt like I was watching the film I love, but for the first time since seeing it in the cinema I could see detail that gets lost in laserdisc transfers and the GOUT DVD. Finer details in faces and the SFX modelwork really stood out for me. I didn't notice the DS battle glitch, and normally I have a good eye for spotting single frames of corruption!
(Disclaimer: the remainder of this post should be considered as positive ideas for improvement and not nit-picking.)
The crushed blacks are still an issue, but it's more than just that - the image to me feels like it has lost some of its cinematographic characteristics. An example would be the red lights in the corridor battle that cut though the smoky atmosphere as if it wasn't there. I noticed the same effect in Ady's ESB reconstruction. Mike Verta discusses the phenomenon about 3/4 of the way down this page.
There were a couple of shots that did stand out as being lower in quality - the infamous vas-o-cam shot in Mos Eisley, and the Tatooine sky colour reversion (this shot).
Unfortuately, as the above are both down to the sources available, I don't think there's much that can be done about them.
I also noticed occasional compression artefacts in the form of mosquito noise, particularly in dark backgrounds. I don't know if this is down to brightening up the image bringing out artefacts in the source, or compressing the video down for a DVD9, but I was pleased to read that you will be doing a new version using the BD video. I look forward to it!
msycamore said:
Any special reason behind why most companies do this?
I doubt it. Probably just a default in encoding hardware - the same reason why nearly all Dolby Digital soundtracks have dialnorm set to -27dB.
zombie84 said:
... even a good mono source played through the centre channel will sound poor. The centre channel not only gives you bad audio spread, but it's a very weak speaker. Ideally you should be playing it back through at least one left and one right tower/bookshelf speaker. So, it's more that your equipment setting sucks balls. I don't know why they put the default to be centre-channel though, that's probably a consequence of compromsing for 5.1 defaults or something.
Centre channel speakers should be matched to the other speakers, and not be weaker.
You can listen to mono soundtracks through the left and right speakers (effectively creating a "phantom" centre) but this is not ideal - you need to be sat dead centre between the speakers. It's better to use a "real" centre if you have one.
1.0 tracks default to the centre speaker because that is the ideal way to hear them.
dark_jedi said:
Sucks balls LOL, you are a dick.
Personal attack - you have been warned. Please read the sticky: http://originaltrilogy.com/forum/topic.cfm/Guidelines-for-Post-Content-and-General-Behavior-Or-How-Not-To-Get-Banned/forum/2/topic/12074/
I'd much rather watch films like Jaws, The Terminator or even The Jungle Book with their original mono soundtracks.
But it is super cool that DJ really likes these multi-channel remixes.
I listened to the Star Wars 70mm 6-track recreation for the first time over the weekend (on Harmy's de-specialized AVCHD). I don't have a high-end separates 5.1 system, just a "home cinema in a box" type thing.
My review in one word - flawless.
I couldn't hear any joins where the '93 mix had been patched with bits of the '77 mix, and it's nice that the added Alderaan debris and detention block glass smashing sounds have gone. I think you've got the overall dynamics and bass volume spot on.
I will always have a soft spot for the mono mix; it's the one I grew up with and in terms of content is the most complete sounding. But for the overall sonic experience this is the mix I'd listen to again and again.
I used TSMuxer to create the folder structure with split m2ts files. I also used AVCHDMe to rename non-compatible filenames to 8.3 format.
On copying the folder across to the memory stick (actually an 8GB SD card with a USB adapter) I discovered that it wouldn't fit - 8GB was actually about 7.45GiB, whereas a DVD9 is 7.95GiB. (I thought memory always used "binary" prefixes, unlike hard drives, but what the hell.)
I might have been able to make it fit if I'd taken out some audio tracks when remuxing. However, on checking the files that I had copied over last week I found a complete m2ts file, as well as two corrupted files and what appeared to be a transcode. The m2ts file on the PS3 hard drive was complete and (I think) untouched, so I watched that.
For the future, sharing the virtual drive root in PS3 Media Server was probably the wrong thing to do, and I should have just shared the stream folder or even the m2ts file itself.
It has since been discovered that the Region 2/4 PAL disc of Jedi is slightly more detailed than the Region 1 NTSC. So the best would be:
ANH - Region 1 NTSC
ESB - Region 1 NTSC
ROTJ - Region 2/4 PAL
But I must stress that the quality differences are very, very small.
As for your suggestion, resizing up, filtering, then resizing back down to target res is a common trick - some filters work better if they have more pixels to play with. Upscaling with one of the new EDI filters may give better results.
My classic space minifigure looks out of place next to the rest of you.
Seeing as Bingo can have The Master, please could I have Sylar? According to Ady, it's not far away from my looks in RL...
A compromise - I'll move the thread into the technical discussion forum, where it is problealy most appropriate.
Just run a quick test. The median function is more effective at removing drop outs - as you would expect. However, an average function is better at removing dot crawl artefacts.
For general noise the difference is negligible.
Hard to tell from those screenshots, but I don't think it's that bad. Does seem very dark, but that might just be the capture settings.
It would be interesting to compare shots with those from the smeared THX versions. I'll post some from the French LDs for comparison.
none said:
^ nice and this leads to an interesting development about all the theater performance recordings. Turns out the MeBeJedi/Starkiller is not from 77 but most likely 81. So mthr's is the true only 77 recording!
And Puggo Grande.
I can understand the GOUT having the 1977 crawl with the 1981 credits, but why would the MBJ/Starkiller version be the same?
Moth3r said:
[...]
With this [LG] recorder [...]
The main disadvantage is the weird interlaced chroma problem at the top and bottom of the colour bars. Might be due to a sub-sampling error in the MPEG-2 encoding/decoding.
Looking at this again, the issue is obvious. DVD recorders encode to MPEG-2, which uses 4:2:0 YUV (YV12) colour subsampling. The recorder has no way of knowing whether the material you are recording is interlaced or progressive; this particular model assumes it is interlaced and flags each frame as such.
When upconverting from YV12 to RGB24 for a screenshot, DGIndex looks at this flag to decide how it should upsample the chroma, the error seen on those screenshots above are due to incorrect assumptions used for the upconversion. If you force progressive type upsampling, the problem is fixed.
Another comment on pitch correction:
http://www.michaeldvd.com.au/Articles/Interviews/ShrekTech.html
AK: Is the transfer pitch corrected?
MA: We typically try not to do that only because pitch correction can add some digital stepping. So unless a director specifically asks us to do it we avoid it, as do most studios.
To be honest, I cheated slightly with this. Because it was a still frame on the laserdisc, I just used TemporalSoften(2,255,255) on the capture.
The principle is the same though, for a real multiple capture average you would use Interleave, TemporalSoften, then SelectEvery.
grisan said:
The BT878 is a pretty dated chip, do you have another capture card? Years ago I replaced my BT878 card with an SAA7134 based card and the difference in picture quality was pretty big.
Yes I have a newer card, based on the SAA7231 chip. I'm going to do some tests with my other LD player first (an analogue-only V4100) then swap the cards over and test again.
4(b). D925 - s-video - BT878 (after AviSynth processing)
This shows the result of a simple levels tweak in AviSynth with a 5-capture average (compare to the original No. 4).
The black level was originally too high, the levels tweak has brought it down to where it should be.
Multiple capture averaging is a technique for reducing noise that can, in theory, be totally non-destructive. All other denoising methods can lead to a loss of detail, either spatial (smoothed textures) or temporal (blurred motion). Even though the latest motion compensated algorithms can help, such denoisers can never be completely non-destructive.
Yes noise levels are reduced (notably in the chroma) and there is no loss of detail, but what impressed me about using this method is the virtual elimination of dot crawl artefacts at colour boundaries. You could argue that you don't actually need a 3D comb filter as this method produces results that are just as good.
Anyone still following this? ;-)
4(a). D925 - s-video - BT878 (using btwincap drivers with sharpness maxed out)
I've included this shot because the setup is similar to what I used back in 2004 when I did my first LD capture. Increasing the sharpness setting produces an image that initially looks more detailed. However, if you look closely (and compare with picture No. 4 above) you will see that the added detail is an illusion caused by the addition of false halos that are not part of the real image. It also has the effect of increasing the overall noise level; I really dislike this oversharpened look now.
Just use your ears; if you listen to the first few bars of the Star Wars theme music from each release it will be obvious if they are at different pitches.
There was no LD release in the UK. I've not checked if the French discs are pitch corrected or not.
Yes, the clipped whites are also a problem on the PAL THX laserdiscs - there's no possible way to recover this detail.
Although personally I don't think it's that big a problem, certainly nowhere near as bad as the crushed blacks on the '04 DVD.
grisan said:
AFAIK the vast majority of PAL movies are not pitch-corrected. Especially in the VHS/LD era this was really uncommon. I guess I need to gather more information about this and make a sound comparison first.
It is true that the vast majority of films released in PAL format have audio that is simply sped up. However, the 1995 UK VHS releases were pitch corrected (ie the film runs 4% faster, but the audio retains the original pitch).
These are the tapes I'm talking about:
(Wow, coincidence - second result from a Google Image Search was from a site called http://www.haku.co.uk - turns out this belongs to Citizen (see this thread) who hasn't been here since 2006...)
Thanks. I would also add the following, from my experience:
1. I stopped using Daemon Tools after it started to come bundled with adware (to be fair, it is optional). Anyway, I prefer Virtual Clone Drive these days.
2. It you're still on Windows XP, you won't be able to browse the directory until you've installed the UDF 2.5 drivers.
