RU.08

DarthWasabbi said:

Burnt bridges on purpose because that place was toxic, and vile. There’s a reason why I deleted every damn message there, and bounced the fuck out. You’ll never know.
Good riddance.

Toxic and vile?

It is not. I won’t stand for you slagging off my server. It’s got a good, positive vibe and it’s been for the most part attracting good people. There’s a reason you don’t see me spamming and promoting it everywhere - people who are interested find their way there and I don’t want just random people joining for the wrong reasons.

I had no ill wishes against you AdmiralNoodles. That said you decided to purposefully burn your bridges and that’s on you. You should own up to your own behaviour and apologise. I actually apologised to TGR97/Wedge over your behaviour so please take some personal responsibility of your own and make your own apology.

Here is what you claimed:

This project seems dead in the water, and the project host ran off with the funds.
Nothing new, nothing lost.

1. The project lead is here. 2. One of the scanner’s representatives is here. Both HERE on OT and over on my Discord. 3. One of the Moderators of OT knows the scanner rep that I’m talking about and has many years of relationship. Same with me I know and trust this man. So claiming that there’s fraud is outrageous. TGR97 Has done nothing wrong. Scanning can take longer than expected, it’s happened MANY times before it’ll happen many times in the future. Case in point I thought I had a scanner for a print a couple of months ago, but he had retired suddenly without my knowledge and turned the business over to someone else.

TGR97’s scanner will only scan if the results are perfect, that’s why it’s taking a while it has nothing whatsoever to do with what you claim.

DarthWasabbi said:

A Discord Server. How quaint!

You weren’t banned from my Discord, you quit and burned bridges with everyone you knew who gave you a chance.

I have a print here, but I can’t use my preferred scanner for it. So just waiting until I can figure out how I want to scan it. 😉

AwesomeJ said:

Agreed, however, when you compare both the Laserdisc and the DVD, if you overlay them both over each other in a side by side comparison (something I did in iMovie as a test), the laserdisc goes out of sync with the DVD very quickly. So yes you are right, the film was not “slowed down” or “sped up”. The film just gained and lost some frames in some shots on the film out. So yeah, it was rendered frame by frame for standard 24fps projection, its just that when Pixar were processing the film for celluloid, the Avid Video Composer caused the movie to gain and lose a frame in all of the shots.

That’s not how it works, the digital format doesn’t have a frame rate. You render it out to a series of images which has no frame rate, it’s just 0000001.dpx, 0000002.dpx, 0000003.dpx and so-on, there’s no way for it to gain or lose frames in that process. The printer doesn’t know about frame rates, or for that matter audio - all it knows is that it prints frame after frame to a negative that then gets processed, and then a colour-timed interpositive is made later to strike prints from. The DVD and Laserdisc won’t be a definitive source, they are many reasons they may not have the same frame counts as they may have been edited on tape (they probably were) we’re talking about editing together a 10 reel movie, potentially adding subtitles or text over the image especially for 4x3 where credits or other text may be cropped out, potentially doing a scene-by-scene colour correction and editing it back together as well. Plus I don’t know what print they used when they made them, for all we know there were random lab splices in it that had to be removed physically, or it had other repairs made with frames removed.

There’s no “speeding up” or “slowing down” of film, it’s shot how it is. You can run the camera slow or fast and get the result that when played at 24fps you get fast or slow motion. Toy Story was rendered frame-by-frame for standard 24fps projection.

AwesomeJ said:

Considering the movie was rendered out at 23.97 fps (my source is Craig Good, who was the Supervising Layout artist for the film), the scanned film is most likely going to be the former frame rate.

I don’t know what you mean by that? Toy Story is 24fps it doesn’t matter how it was rendered in the computer before film-out. An example might be if you run a camera at 120fps for slow-motion footage, it was captured at 120fps but the film is 24fps.

AwesomeJ said:

I have some screencaps of a laserdisc copy that I color corrected a while back, so I don’t know if this is 100% accurate to the original theatrical presentation but I hope these help when the thing is finally scanned:

Laserdiscs are colorcorrected for the living room by the telecine operator, aka the colorist, like this:

So basically it has its own colortiming. The telecine prints are more expensive to make than projection prints, although they’re usually 16mm not 35mm, and they’re printed low-contrast so that they can be transferred for broadcast and/or home video (you can also transfer off the interpos or a dupe negative as those are also low-contrast film that will transfer acceptably on a telecine). For a bright film like Toy Story it may not look too different, but for films with many dark/night scenes those in particular will look nothing like how they look theatrically. Beauty and the Beast for example is really ruined by the home video colortiming. BATB is an example of a film that probably shouldn’t have been released to home video, but we are where we are and today everything now goes to home video eventually. Certain decisions are even made for both home video and broadcast at the time of filming, for example in Hollywood they shoot alternate scenes for broadcast where they know that something will have to be censored for US domestic TV.

For the theatrical prints the way it works is that the interpositive is colortimed and then when theatrical prints are struck the printer is fully calibrated for the film and the print is struck reel-by-reel. So for example if you’re printing 200 prints you print 200x reel 1 and then 200x reel 2 and so-on. Some prints will be intentionally printed 2 stops brighter and those are Drive-In prints (prints for outdoor projection). The interpos is struck off the original negative or off a dupe negative, in this case the original negative would have been a digital film-out (there may be more than one “original negative” if they made more than one film-out). Now the resolution Toy Story was outputted to wasn’t 2048x1556 so I don’t know if they upscaled it for the printer, or window-boxed it and then did an optical enlargement off the o-neg - either is possible and we should see if there was any attempt at “filmizing” it when it’s scanned.

TonyWDA said:

I never meant to cause any confusion. Until now, I had no idea about ScanStation’s software extraction option. Good to know.

The software audio extraction is pretty good, and better than AEO Light. But not all ScanStations have it, it was initially developed for the ScanStation Personal I think, for full ScanStations I think you just have to get support to enable it if they set it up for hardware-only optical audio (which requires the support contract to be paid and up-to-date). The professionals usually set up something dedicated to optical audio though.

zerocool said:
Did projectors have toslink/spdif out back in the day to allow for this?

I’m not a projectionist.

Catsyz said:

The print on the eBay listing does seem to have a Dolby Digital track in between the sprocket holes. But I agree that it would easier to use the laserdisc audio.

We don’t know if it will run or not yet though. It could quite literally be worn out now, Toy Story was a big hit and each print will have had hundreds of shows. If it still works you can record it off a projector.

TonyWDA said:

Yeah, it’s quite the hot topic in audio engineering circles. In this context, having as many audio options as possible is always best when all is said and done. It’s less convenient to get the Dolby 5.1 on the print preserved, but a lot easier to get the analog stereo track digitized using AEO Light— especially if the raw scan resolution is well past 2K. That would only be necessary if the scanner couldn’t (or simply didn’t) capture the analog audio along with the image scan or the sound on the capture was too hissy; unfortunately, LaserGraphics ScanStation units are kind of notorious for that. But all things in due time; I’m sure TristAndShout64 will cross that bridge when he gets to it.

Please don’t confuse Trist. ScanStations have two ways to transfer audio, the hissy hardware reader or software extraction that works better than AEO Light. But if you want the best optical audio transfer I can also get that at an additional expense. There’s no guarantee the SRD will play - it might, but it might not.

If you do they can be scanned in Germany and then shipped back to Spain FYI.

Yep the print is here right now sitting in my house, I should have a good option that can do the scan for Trist’s project. The film was already packaged for postage in what the seller called a “film box” and I can quite confidently say it would very likely have been quite damaged in transit. When we send it to Trist in the US it’ll be packaged more professionally and will cost a lot more.

I’ve sent you a PM, be sure you read it (speech bubble at the top next to your avatar).

When a film is professionally scanned in 16-bit color as DPX image files, every single frame weighs in at 100 MB. With upwards of 175,000 frames in each film, a complete scan requires about 21 TB of storage (42 TB if you want a backup copy! And then you need at least another 21 TB of space to work on it – over $1000 just in hard drives is therefore required for every film). Having the film cleaned prior to scanning costs another $870 (plus about $75 shipping each way) and then the scanning costs between $2000 and $15,000, depending on where you send it.

Now, I would very much like some information for the scanning process of these films and whether or not they where cleaned up. Cheers!

I don’t know where you’re pulling all those prices from and numbers from, but they’re not accurate anymore (or they at least are not representative of competitive commercial rates and storage options). The price of cleaning varies wildly depending on where you go, and your weirdly specific price of $870 would equate to .054/ft if we’re talking 178 minutes of 35mm 4-perf. I know where to go for a much better rate than that.

Yes if you’re scanning 6K DPX 16-bit it’s over 100 MB per frame (around 125MB per frame I think), but that should not be necessary unless you’re dealing with very badly faded film. Backups should go on LTO-7 or LTO-8 tape (or even LTO-9 now) not on Hard Drives.

Feel free to join my Discord friend.

Cjmp475 said:

I live in the UK so scanning would be about £600 per reel. If there’s any way to fundraise to get someone from the US to buy this it would be greatly appreciated as I think shipping to a US address would cause to many issues from here.

I have a scanner in London that could do the entire scan for less than £600, but it’s not worth sending the film from the US to the UK just for scanning. It can be done to higher quality in the US - join my discord channel and send me a message there.

This is the type of quality you can get out of the Gugusse Roller.
https://www.youtube.com/watch?v=qVFr5I7Tht4

Which is very rough on film and cannot handle 1,000ft reels let alone 2,000ft reels. The only things that DenisCR can scan on it is trailers and other similarly short reels like music videos, snipes, and theatre ads. The Gugusse Roller is a Raspberry Pi project really, not a film scanner.

You can scan it in the US if you buy it (so no import charges and postage should be more like $50). Scanning about $350 (on a real scanner).

MonkeyLizard10 said:

I compared on a D56 high CRI lightbox directly next to monitor showing the two earlier 1080P project scan .mkv files. Not sure if it is because the print they used was more faded out or maybe processed differently in the film lab or the way the scan got done or graded. (that said still all props to the original 1080P JP project!)

It’s nothing to do with the quality of the print, the scanner that was used for that scan produced duller colour due to having a low CRI light basically.

Fullmetaled said:

How much more superior is it going to look compared to the uhd?

It won’t look better, it’ll just be a more faithful to how the print looks (virtually identical to the print if you were making a 4K DCP from the scan and projecting it in a cinema).

Jurassic Park is being done as it is here and I can confirm it is absolutely professional quality.

Join my Discord server and send me a message.

They haven’t scanned JP1 yet, but it’s being scanned in a very different way to much higher quality. That said Retrofilm looks like they’ve improved the quality of their scanner based on that clip which is wonderful, it’s also a real-time scanner so it scans very fast compared with how these ones are being scanned.

felpotomous said:

Re: your last point I remember seeing Jurassic Park at an IMAX back in the 90s and it looked terrible because they were taking a regular print that was definitely not meant to be blown up that big and doing so.

That shouldn’t happen during the original run, sounds like they somehow ended up with a multiplex print.

Problem has always been the cost-cutting bean counting tight-ass distributors. If they can save a buck surrendering QC on multiplex theatrical prints they do. The optical enlargement uses Japanese made printing nikkors optics (look them up, and how much they cost - used let alone how much they were brand new back in the 70’s), so you can absolutely go to 65mm/70mm or as is the case with Terminator II Super35 to Scope while the film stays razor-sharp, it’s just a matter of how the final prints are made and the budget. The HQ 35mm ones are called show prints and are designed for special shows or theatres with very large screens.

You’re absolutely right though about the large screens. An average print can look better in a smaller auditorium, but it the places that don’t do any masking will ruin the experience with window-boxing or letter-boxing on the screen and sadly most multiplexes are like that now. They must not understand how it ruins the experience!

I’m not a fan of these films but I’ll clear up a couple of things:

https://www.youtube.com/watch?v=yn21u6j6Ywc

It’s shot on 35mm, so it’s absolutely 4K+ if you’re scanning the oneg. 2K film-out is 2048x1556 and that was the standard since Terminator 2 and Jurassic Park. Before that I believe all CGI had not been attempted to be printed full-frame, that is to say that Abyss and anything older had CGI printed that was then optically matted into the negative.

The first film from the video above does not have a DI, that means it’s photochemical and therefore yes it’s higher than 1080p. The final release prints are probably in the range of 1.5-2K and remember that 2K means 2048x1556 NOT consumer 2K at 1920x1080/2048x1152. The second and third film had DIs, that means 2048x1556. It doesn’t matter what resolution the CGI was rendered in - the CGI can be rendered at 1.5K, 1080p, 720p, whatever. The CGI is simply being added to the 35mm background which would have been scanned at 2K (2048x1556) and the DI will be 2K. Some further information is here:

https://thedigitalbits.com/item/lord-of-the-rings-trilogy-2020-4k-uhd

The Fellowship of the Ring was shot photochemically on 35mm film in Super 35 format using a variety of Arriflex, Arricam, Mitchell, and Moviecam cameras with Zeiss Ultra Prime and Angenieux Optimo lenses. Only about 70% of the film was finished as a Digital Intermediate at the time, as the process was then new and still evolving (the other 30% was finished traditionally on film). For this new Ultra HD remaster, Park Road Post (a New Zealand post facility owned by WingNut Films) went back and scanned the original camera negative in 4K, then scanned the VFX film-out elements (for VFX shots that were finished on film) in 4K, and upsampled the VFX shots that were finished digitally (in 2K resolution) to create a brand new 4K Digital Intermediate at the proper 2.39:1 aspect ratio.

That’s describing the type of CGI in Abyss and older films: VFX that is rendered and then optically matted into the film. Also it doesn’t matter how much was DI and how much wasn’t, the VFX doesn’t set the resolution the background frame does.

The resolution of the final prints will vary. Terminator 2 projected off an original 35mm which I have seen on a huge 70ft+ screen (I’d have to do some digging to find the exact size, but it’s at the size where the difference between 35mm and 70mm is very obvious even with the sharpest of 35mm) and it is clearly 1.5K or lower true resolution perhaps as low as 1K. Most 70mm makes 4K look like low resolution. So as with anything with prints, it’s a matter of YMMMV depending on where it was printed, whether the cinema had a large screen and demanded they print it to a higher standard than a multiplex cinema which was the standard at the time (and sadly still is). So for example it’s possible to print your standard quality for multiplexes and then do some at show-print quality for the Theatres that still had large screens.

Just curious, but in these situations, do scanners at least return your film? I’m worried the answer is no, which is awful.

In this case the film is owned by a friend of the scanner.

Join my Discord channel and send me a direct message I’ll be happy to tell you more about “bad” scanners and the kind of things that can happen.

SpookyDollhouse said:

Higher color gamut alone would benefit it and HDR would be icing on the cake.

Film doesn’t have a “colour gamut”, it’s basically got intensity of reds, greens, and blues which is represented by each corresponding resistive dye layer (magenta blocks green etc). Film has a bit_ more in the reds and oranges and a little bit more in cyan blue and yellow compared with Rec709, but that’s it:

The print I believe is here just needs to be scanned.