Film-Tech Forum ARCHIVE: Variable Density & Red LED's

This is topic Variable Density & Red LED's in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=002764

Posted by Ray Derrick (Member # 1001) on 10-30-2001, 12:32 AM:

I don't know if this topic has been raised before ... but,

Some months ago the Astor cinema in Melbourne (Australia) had some new prints run off in LA of some old movies for a film festival. When they took delivery of the prints they tried running them on their red-LED reverse scan soundhead and the sound was low level, distorted and "harsh". In fact so much so, that they were unable to run them for the public. As it transpired the soundtracks were variable density and had been printed in high magenta. A tech was called who converted the soundhead back to the old exiter lamp setup and the prints played perfectly.

So it would seem that high magenta, a red light source and variable density do not mix. I have since wondered what would have happened if the variable density track had been printed in cyan, or if the high magenta track had been played with a high intensity green light source(green being complimentary to magenta). I suspect that the problem is due to the "grey scale" of the variable density track being rendered non-linear by the red light source.

Anyone else come across this problem and/or any technical explanations for the problems experienced by the Astor?

------------------
Ray Derrick
President
Panalogic Corporation Pty Limited
44 Carrington Road
Castle Hill NSW 2154
Australia
Phone: 61 (0)2 9894 6655
Fax: 61 (0)2 9894 6935

Posted by Steve Guttag (Member # 268) on 10-30-2001, 06:00 AM:

Some of my customers have come across this problem. RED light just isn't the best light source to begin with. The solar cell is sensitive to the infrared spectrum in the first place.

I would suspect that if you changed the LED to an infrared one that the film would have played ok as well.

I have one exhibitor that is having me do just that and we will experiment. I also believe that the motion picture academy theatre has infrared LEDs rather than the visable ones to handle all of the soundtracks best (cyan not a reality yet, and probably never).

The whole red LED thing just ticks me off every time. It is a dumb idea in this day and age that yeilds inferior results at best. Their life-span is a bit wild too and their fading is unacceptable. All of the infrared LED systems we have installed since about 1996 are still running and still at Dolby level without having to constantly tweek them (over 12000 hours later)...you can get a red LED to say constant for even a month.

Steve

------------------
"Old projectionists never die, they just changeover!"

Posted by Gordon McLeod (Member # 33) on 10-30-2001, 01:01 PM:

Most likely this is a magenta only dye track not a high magenta that still contains silver.
The variable density part should not be an issue per see
Also there were experiments using cyan tracks back in the 40's with cinecolor

Posted by John Pytlak (Member # 331) on 10-30-2001, 02:21 PM:

The movie industry Dye Track Committee is meeting at ShowEast on Thursday morning, November 1, at 9:00 am. The meeting will be held in the New York / New Orleans conference room at the Orlando Marriott World Center. I recall that committee member Jess Daily (of the UCLA film archive) did quite a bit of testing comparing old variable density tracks on both white light and red LED readers. If you still have access to the prints, I'm sure the committee would be interested in evaluating the problems encountered.

------------------
John P. Pytlak, Senior Technical Specialist
Worldwide Technical Services, Entertainment Imaging
Research Labs, Building 69, Room 7525A
Rochester, New York, 14650-1922 USA
Tel: 716-477-5325 Cell: 716-781-4036 Fax: 716-722-7243
E-Mail: john.pytlak@kodak.com
Web site: http://www.kodak.com/go/motion

Posted by Ray Derrick (Member # 1001) on 10-30-2001, 07:40 PM:

Gordon, I think you may be right about the prints being magenta, rather than high magenta. This would make more sense given the lack of silver in the former.

Steve, interesting point about infra-red LED's. The tracks may have played correctly if that option was tried. I also agree that the rate of fading of red LEDs is unacceptable. Have you run any cyan film with an infra-red system? In theory, primary red light (approx 650nm) should produce the greatest degree of "darkness" of cyan dye, so infra-red (theoretically) will produce less contrast and hence lower gain and higher film noise. But as you say, the peak sensitivity of solar cells is in the infra-red region (about 7dB better at 950nm than at 650nm to be precise) so this will tend to offset the gain loss but not the increased film noise. I look forward to the results of your tests.

John, the matter was reported to several "authorities" on the subject at the time, including Jesse at UCLA, but as far as I am aware, no one has come up with an actual technical explanation for the distortion and harshness at this time.

------------------
Ray Derrick
President
Panalogic Corporation Pty Limited
44 Carrington Road
Castle Hill NSW 2154
Australia
Phone: 61 (0)2 9894 6655
Fax: 61 (0)2 9894 6935

Posted by Gordon McLeod (Member # 33) on 10-30-2001, 07:44 PM:

Dion Hanson in the UK has a cd available of comparisons of different track types nad red/white readers. This was reported in the BSTS

Posted by Mark Gulbrandsen (Member # 72) on 10-30-2001, 09:24 PM:

I have a number of variable density films in my collection and they all play fine on my AA-2 LED reverse scanner. I agree with Gord that if it happens to be a high magenta track as some older color processes may have used then its possible it would sound horrible. I was recently told by more than one manufacturer of reverse scanners that there is now no real difference between using a red led and an infrared. This has come about due to new optics and not anything to do with the LED's themselves. Anyone else heard this news yet?
Mark @ GTS

Posted by Ray Derrick (Member # 1001) on 10-30-2001, 09:34 PM:

Steve, I quote from US patent 5,587,749 (Goldberg & Richards, Dolby Labs):

"Unfortunately, the dyes used in all color prints have a very low absoption in the infrared region."

They go on to discuss Fig 2 of the document, which shows very little absorption in the infrared region when silver re-development is not used. In other words, cyan dye (or any other colour dye) is largely transparent to infrared light in the absence of silver.

This patent makes interesting reading and can be found on the US Patent Office website (www.uspto.gov) The patent is essentially for the use of a red LED to read a cyan analogue soundtrack. Yes folks, Dolby have actually patented the colour RED!

------------------
Ray Derrick
President
Panalogic Corporation Pty Limited
44 Carrington Road
Castle Hill NSW 2154
Australia
Phone: 61 (0)2 9894 6655
Fax: 61 (0)2 9894 6935

Posted by Leo Enticknap (Member # 534) on 10-31-2001, 07:29 AM:

There was recently a thread about this on the AMIA-L listserv. The posts can be found here.

Posted by Pete Naples (Member # 761) on 10-31-2001, 07:59 AM:

Some time ago, three years or thereabouts, I heard (or rather didn't!) a non-redeveloped copy of Jiffy. The output in the theatre with the fader on 7 sounded like something was leaking somewhere, barely a whisper. This was on a Vic 8 with IR LED Reverse Scan

Posted by Michael Hunt (Member # 772) on 10-31-2001, 07:00 PM:

Strangely, I received more comments about older soundtracks when my main screen was still equipped with an exciter lamp - I converted to reverse scan LED because I was extremely impressed with the Stereo seperation and clarity in my newer screen, which had been supplied with reverse scan IR LED's, as at the time of installation, Strong were shipping reverse scan IR's as standard (c. 1995/6)

Since converting the main (smaller) screen to Red LED, I've rarely found problems with older B&W films...
I tend to get the most complaints from pensioners, who want to hear the dialogue, and none of that music... can you ask the projectionist to turn it off...

Incidentally, we play host to the local film society, and their head honcho is a pensioner... They insist we show films 'right' and have often asked for the volume to be turned up if they feel it's too quiet...

Sometime's I'm glad not to be in a multiplex chain, though I'd love to have any form of Digital sound in the main screen.....

------------------
Uncle to JEDI Hunt,
honest!

Posted by Mark Gulbrandsen (Member # 72) on 11-01-2001, 06:30 PM:

I absolutely agee with Michael. The red led reverse scanners let you hear how good some of the old tracks can really sound......
Mark @ GTS

Posted by Pete Naples (Member # 761) on 11-02-2001, 02:37 AM:

That's true guys, but unfortunately they can also pick up on some stuff that the film makers probably never even knew they had recorded.

Every christmas, someone somewhere plays 'It's a wonderful life', every christmas I will get a barrage of calls from managers and projectionists who haven't dealt with this film before. Somewhere in reel three is a noise, not unlike motor boating, we never heard it with exciter lamps, but we sure as hell are hearing it now with reverse scan LED! To be fair this is a reprint (but not a new reprint) copy that does the rounds at Christmas.

Posted by Stefan Scholz (Member # 147) on 11-03-2001, 06:38 AM:

I have a 1957 cinemascope print in my collection, that uses a magenta colored density track as optical (backup). The track is magenta colr layer only, and contains no silver.
It works, within the limitations of half track width, on a conventional lamp reader, with just a little lower output.
Using the track on a Kinotone reverse was impossible and showed the reported results.
Firstly, the reverse scan reader never sounded as clear with older tracks, as the incasdescant lamp pickups did, either foreward or reverse scan type.
Secondly I came across the distortion problem, when I converted my first setup to red Lasers in the mid 90's. Films tended to sound sybilliant, indicating cross modulation due to non linear pickup. So the processor's preamp was replaced with a current converter input, and the problem vanished.
Solar cells have a liniar current to light ratio, not voltage to light. So, if you change the type of light and or intensity, and use a voltage amp, as most processors do, the working range is severely shifted out of the quasi linear range.
I was using a 5 mW laser with line generator on my U2's, so by changing the main processor to ADSG's DCP 1k again brought up the problem. A 12 Ohm resistor paralleled to the cell input, and slightly higher preamp gain got rid of the distortions.
The Kinotone (Dolby) preamp in the retroscan reader has an input impedance of more than 400 Ohms. My final solution was to omit the internal amp, just using the cells and my external current converter preamp. It now works fine, and I could even run the LED at lowest current position, helping it staying stable.

Most preamp designers seem to think, the cell behaves like a dynamic microphone, so their preamps are basicly modified mike designs, fiddling just a little with input resistance to gain a compromise between sensitivity, noise, and IMD, and most modern appliance just go for a specificic track type.
AFAIK only Kintek featured a current preamp, probably the best feature among their processors.

Posted by Gordon McLeod (Member # 33) on 11-03-2001, 07:51 AM:

Actually PanaStereo's preamp is a true current type
Also most of the reverse scan units preamps are current as well as they feed directly into the - input of the opamp and that is usually very low impedance

Posted by Ray Derrick (Member # 1001) on 11-07-2001, 08:37 AM:

We are still trying to determine with certainty the actual cause for the problems reported by the Astor, however I am now of the opinion that it is primarily (if not completely) to do with the red colour of the light source. Red (and infrared) are about the worst colours you could chose for non-silvered magenta, because magenta coloured film is really absorbent only to green light. Therefore some green light is essential for any usable soundtrack modulation to occur in the cell.

A normal exciter lamp has it's peak output in the infrared region, however as the light output is reasonably white, there is a good deal of light emission in the green region, so a magenta dye track will play, albiet at a lower level than a silvered track. Red LEDs on the other hand are fairly monochromatic (there is not much else emitted other than pure red) so if a red LED light source is used, the magenta dye soundtrack will be effectively transparent to the light, so the modulated output from the solar cell will be very low and distorted.

What Stefan says about solar cells being a current source device is true but I don't think this is a factor in this case. As Gordon points out, all reverse scan Red-LED readers use the standard Dolby preamp circuit which is "almost" a virtual earth (or current) input. There is a small series resistor at the input but I am sure it is closer 100ohms, not the 400ohms as stated by Stephan. Why this resistor is there at all is a bit of a mystery because the solar cell would perform better without it. This is because any resistance in the solar cell output causes a voltage to develop, which then opposes the current trying to escape from the cell, causing non linearity and hence distortion. By loading the cell with a virtual earth, all of the current is extracted from the cell, resulting in excellent linearity and thus very low distortion.

As Gordon also points out, Panastereo optical preamps are true virtual earth or "true current" as he puts it, however AFAIK it is also the only preamp ever manufactured that has a "super-balanced" virtual earth input. Unfortunately the considerable benefits of this design are entirely lost when used with a reverse scan reader because the reader's inferior preamp gets in the way. We are currently (no pun intended) developing a range of super-balanced true virtual earth preamp cards as replacements for the preamps in various reverse scan readers to address this problem.

------------------
Ray Derrick
President
Panalogic Corporation Pty Limited
44 Carrington Road
Castle Hill NSW 2154 Australia
Phone: 61 (0)2 9894 6655 Fax: 61 (0)2 9894 6935

Posted by John Pytlak (Member # 331) on 11-07-2001, 12:43 PM:

Ray said: "Red (and infrared) are about the worst colours you could chose for non-silvered magenta, because magenta coloured film is really absorbent only to green light. Therefore some green light is essential for any usable soundtrack modulation to occur in the cell."

If that "high magenta" track did not have silver redeveloped, a mistake was made at the lab! Traditional analog optical tracks are printed with red and green exposure in the printer and processed with a silver redevelopment step, to produce a track having cyan dye, magenta dye, and silver. The optimum cross-modulation density is slightly different when read with tungsten "white light" than with red LED, so the track has to be optimized for one, with slight sibulant distortion on the other. "High magenta" analog tracks restrict the exposure to the magenta layer and have the silver redeveloped, producing a track that has magenta dye and silver. "High Magenta" tracks optimize at about the same density for both "white light" and red LED readers, so they are the best choice with the mix of readers now in theatres.

Cyan dye analog tracks will NOT have silver, eliminating the need for the complex and wasteful soundtrack redeveloper step in the processing. But a red LED reader is absolutely required for cyan dye tracks, as the signal level on old "white light" readers will be too low.

Posted by Ray Derrick (Member # 1001) on 11-08-2001, 03:11 AM:

So what are you saying John? That magenta dye only variable density tracks don't exist or simply that there must have been an error in the processing for these new prints? I have it on good authority that magenta dye only tracks were not uncommon 30 or so years ago. In fact I have been told there were also cyan-blue dye tracks, I think from a process called Cinecolor?

One country cinema owner here told me recently that he had reels of national anthems variously with magenta, cyan (or blue) and silver redeveloped variable density tracks. So I asked him the other day to try playing one of the magenta films on one of his red LED readers and he came back to me today to report the exact same symptoms as the Astor cinema, low level and distorted.

I am wondering if the new prints that the Astor commissioned were somehow simply copies of the orginal films complete with magenta dye tracks.

The consequence of all this of course is that cinemas that regularly play vintage material will not be able to convert to red LEDs. So what are they supposed to do when cyan tracks turn up??

------------------
Ray Derrick
President
Panalogic Corporation Pty Limited
44 Carrington Road
Castle Hill NSW 2154 Australia
Phone: 61 (0)2 9894 6655 Fax: 61 (0)2 9894 6935

Posted by Steve Guttag (Member # 268) on 11-08-2001, 06:08 AM:

>>"So what are they supposed to do when cyan tracks turn up??"<<

Be amazed! I still don't think it will happen...the conversion to red reader has mostly been a function of new installations and they are on a big decline. As such, depending on whos figures you use...the red reader is only at about 50% of the industry so a full-scale roll-out of cyan would be a disaster.

Those responsible for this red-reader insanity should just give up...it is an inferior method of reproduction...we should ride out film's last years on the conventional optical redeveloped soundtrack and those with LEDs can switch over infrared LEDs (which shouldn't take too long since the red variety have such a short life-span.

Steve

Posted by John Pytlak (Member # 331) on 11-08-2001, 08:14 AM:

Any Kodak, Fuji or Agfa color print made in the last few decades should have an analog track that is dye + silver. (Current Technicolor dye transfer prints have only silver tracks). A magenta track without silver would not have been properly applicated by the lab. A dye-only track will play poorly on any tungsten "white light" or infrared LED reader, since dyes are transparent to infrared, and most readers have significant infrared response. Silver is needed to properly modulate infrared energy. Cyan dye tracks REQUIRE a red LED reader, which has negligible infrared energy, allowing the cyan dye to be an efficient modulator of the red light.

Posted by Gordon McLeod (Member # 33) on 11-08-2001, 09:39 AM:

Cinecolor was a dye only cyan track that according to the notes I have required a red dot PE cell and a red excitor lamp
Technicolor experimented with a Diazo FerroPrussion Blue track several times to eliminate the need for a silver based blank recviever stock. There is an excellent article in the latest Cinema Technology from the BKSTS on it.

Posted by Ray Derrick (Member # 1001) on 11-08-2001, 10:24 PM:

Steve, I must say I agree with you 100%.

Gordon, yes that confirms what I have been told. And of course this is interesting considering that both Dolby and Kodak have current patents for using a red coloured LED to read a cyan dye soundtrack. How two companies can have patents that not only have such obvious prior art but also which claim essentially the same thing makes you wonder who they've got working down there at the US patent office. It sure as hell ain't Albert Einstein.

------------------
Ray Derrick
President
Panalogic Corporation Pty Limited
44 Carrington Road
Castle Hill NSW 2154 Australia
Phone: 61 (0)2 9894 6655 Fax: 61 (0)2 9894 6935

Posted by John Walsh (Member # 168) on 11-10-2001, 12:35 AM:

From what I can see, a patent issued from the US Patent Office really isn't worth that much. All they seem to do is say; "On this date, this guy says he invented this thing." If you have money and lawyers, you can wangle someone's patent away from them. The company I worked for actually did not patent some ideas. If they file a patent, that means disclosure. Since the ideas were buried so deeply in the product, they would rather take the chance.

I have to agree with Steve; this is red-reader insanity. "Stop the madness!!"

Posted by Stan Gunn (Member # 554) on 11-10-2001, 12:56 AM:

I will shortly be looking at the sound tracks that caused the problem at the ASTOR, I will report on my observations.

------------------
ALL PARTS FOR VICTOR AND KALART VICTOR 16MM PROJECTORS.
SERVICE TO 35 AND 16MM
PROJECTION EQUIPMENT.