Jump to content

Robert Houllahan

Site Sponsor
  • Posts

    2,206
  • Joined

  • Last visited

Everything posted by Robert Houllahan

  1. Ah yes the Disastor "test" which seemed like a make work project for some academics who don't actually use or know anything about film. So this is Part II the more Disatoriniinnning... now with their own scan scan machine
  2. Vague at best. "Multi Spectral" i.e. sequential RGB? that is "multi spectral" as in red green and blue. The example on their web site I actually prefer the look of the "other" system and think it will grade to finish better than their example.
  3. The Scan Station "personal" can be upgraded to the latest software with 2-Flash HDR and that makes it a decent 12-bit scanner and the noise from the older CMOSIS camera is mostly gone. So for $20-25K used plus about $15K you can have a updated 35/16 Scan Station that works really well. Nobody complained about the noise from that camera more than Moi. GPU "pin registration" using machine vision is really excellent almost all of the time I think the LG or Xena edges out my Arriscan a bit on 16mm but is about the same on 35mm. The Arriscan definitely has superior color fidelity and color channel separation and overall detail compared to any of the bayer mask based scanners HDR or not. I am sure that is why LG makes the Director 13K ( which is a Monochrome Sony 6.5K and a pizeo stage I assume ) RGB scans for the Pic quality win and speed loss. There are a few ways to clean film I have a Lipsner XL1100 alcohol cleaner which is Ok for basic dirt / dust A Lipsner 8200 Ultrasonic which really works and for really tough stuff the option of rewashing the film in a film processor can kill all the mold and help heal emulsion problems while releasing allot of the really hard to get out dirt from the emulsion. There are new cheaper fluids for cleaning and full immersion liquid gate scanning that are environmentally ok and easily available now. I think the cameras keep getting better, as far as I can tell the 5.4K Sony in the Archivist is one gen newer than the 6.5K Sony and it looks excellent has great dynamic range and extremely low noise. The 6.5 Sony is also great but has some odd color stuff and allot of color channel cross talk when I went through allot of setup on the Xena, this is all done by LG as a turnkey scanner product. Most TV and Movies you see get scanned true RGB at Co3 or Postworks or FotoKem on the DFT Scannity or Arriscan. DFT has introduced the new Polar scanner using the G-Pixel 9.4K x 7K sensor so that can make true 8K RGB HDR scans. FotoKem and Cinelab London have the DFT Oxscan 14K 65mm / 70mm RGB HDR scanner for things which are big. I could see 4K or 8K continuous motion true RGB sequential HDR scans using X-Y GPU Machine vision registration shortly and the LG Director and DFT Polar might already be doing this, massive bandwidth from the camera and allot of GPU making it possible.
  4. I would go for it, I think any B&W will hold up very well if it is in relatively ok storage conditions. I personally got a few rolls of the 1962 Plus-X because they were just a little shrunk for the purpose someone had them for but they went through my Aaton XTRprod just fine.
  5. B&W film stocks last exceptionally well, I used a roll of Plus-X Negative for a music video a few years ago that had expired in 1962! Color turned to B&W never really matches the contrast of real B&W IMO. There are a number of options for current B&W stocks and how to process them, 7266 runs nicely as a negative and I still really like 7222 processed properly, we have dome some overexpose and pull and some underexpose and push processing for various clients that worked well for their purposes. Here is the MV with the Plus-X from 1962:
  6. The general rule is to overexpose about a stop per decade when using expired film. I think that rating it at 100iso and processing normally achives the same thing as 200iso and a push +1 and may even produce a bit nicer result.
  7. Look for a Russian Gen-1 (tube) set of Night Vision goggles with a IR lamp in the center you can turn off. We use these and ND the IR lamp and they do work with it off too. Nice green night vision analog tube light amplification goggles, work great for the darkroom. The earlier US ones (and later too) are also great and better built but allot more expensive for the oldest tech to as much as a nice car for the newer ones.
  8. I would suggest a Schneider Makkro-Symmar 80mm / 5.6 it is what we use on one of the Xena scanners for 8mm and is excellent.
  9. Here is a MV / Spot we ran which is 5222 Push +1 and some 5219:
  10. You might want to talk to the lab you are working with and have a discussion about target density and target gamma. What we do every run for process normal is shoot a test strip on a XRite Sensitometer on some fresh stock we are running which is then processed and read on a XRits 310 Densitometer. We aim for a "standard" reference D-Max and D-Min and some stocks and processes will be based around a target D-Max. For example when we make optical tracks for printing the target density is a 2.8 specified by the person who records our optical 16mm tracks. I am home now but I can get the last Sensi strip and densities for 7222 from last week tomorrow and post them, we run F76 at 75Deg F / 3min (50ft /min) and a 1:3 mix with the developed being about 100Gal with allot of turbuation so a pretty active developer, For 3378 Sound film we run at 80Deg at 20Ft/min so more than double the dev time and higher temp, that gets it to the high density. So it is a good idea to do some experiments and shoot some brackets and see how things turn out with more or less target density and then a lab can process to that density / gamma specified.
  11. The Vision3 stocks are designed to be inter-mixed so they all have similar grain and base character, I have found they can be inter-mixed in scanning without having to do a base re-calibration between stock. Similar when printing them to film a one light can be pretty good with mixed V3 stocks, scene and lighting dependent of course.
  12. I recently did a Digital to 7222 recording at Cinelab and I calibrated for one stop over and pulled 1/2 it had a nice contrastier look, it was for re-scan not a print. We also tested a number of developers a number of years back before we ran a big well known 35mm B&W feature and found F76 developer to really work best and produce much nicer looking grain. It is a negative so it will be a Log scan typically on a Scan Station or Arriscan and then as David said you can adjust contrast to your liking.
  13. I have a Nucoda setup but don;t really do any stabilization, it is rare that any film comes out of the LaserGraphics or Arri scanners that need post stabilization.
  14. IN scanner GPU stabilization works really well because it is done on the raw camera data before it is demosaiced so basically the machine looks at the monochrome info for contrast edges in a specific known general area where the perfs are supposed to be. The LaserGraphics GPU stabilization is so good the scanner just has a encoder on the capstan and then the GPU stabilization, if you turn it off the film moves around quite a bit and then is locked amazingly steady when it is on. I think Nucoda/Phoenix has better stabilization tools for film than Resolve does also.
  15. I could potentially record to 100D Ektachrome and I am working on getting around to a test and calibration for the recorder and I have recorded to Tri-X reversal if you are just looking for a one off positive to project.
  16. HI We record to a negative and then print it, choices are 50D 250D 500T / 7222 for the recorded negative and we can print that to 16mm color or B&W print stock. I am thinking about a direct to print stock recorder with sound using an Auricon but it is far down on the list due to the technical challenges of recording direct to print stock in real time like the Cinevator does. The V4 (now) of the Cinelab Digital to 16mm recorder runs at about 2FPS to record digital files to 16mm film, pin registered.
  17. Most of the film color timers learned by trial and error, Brad who was at Cinelab was an incredible color timer and did not like the hazeltine as he found it to be slow and cumbersome he timed on the bench with color cards and experience and made incredible timed prints. Unfortunately he passed away a few years ago so much of the knowledge he had was lost except for the tools which are still at the lab. For Di workflows on film you might reach out to Steve Shaw at Light Illusion as he has an extensive toolset for calibrating a Kodak digital LAD with a densitometer through the negative and print stages at a lab and building a matching 3D viewing LUT so the look of the print can be closely matched in the DI theater. For more controlled general printing you can use a Kodak LAD and measure the patches with the densitometer for ECN2 consistency and then print that LAD onto print stock and then measure the LAD patches on the print to get to a base printer light setup on the printer and process at the lab. From there color timing is primary print lights +RGB = darker in all colors -RGB = Lighter in all colors +G = more magenta -G =less magenta etc.
  18. I think it has a metal plate system that "holds down" the film and makes it run flatter through the gate. This probably really depends allot on the tolerance for being flattened by the presumably very shrunken film being scanned.
  19. Canon 11.5-138mm zoom. Canon Super16 10.6-180mm T2.7 zoom.
  20. Kodak did about 1.3B in sales last year and turned a bit of a profit, I think most of the sales volume was in advanced materials and inks etc. The MP products division and the still film division are both profitable I believe.
  21. I would go for the Arriscan if you value color accuracy and DPX is really not that hard to deal with on a modern machine, just transcode it to ProRes4444XQ if the scan house won't offer that. The Arriscan is a true RGB and true 16bit machine when run in 2-flash it is in another class of machine compared to the still excellent Scan Station. The Arriscan is definitely superior in terms of color reproduction.
×
×
  • Create New...