quote: Charles Greenlee

---

And the main limiting factor on analog's quality

---

...is the dynamic range in itself - which would peak out at 50 to 60dB.on a good day..whereas digital can push the envelope over 100dB without trying..

Like the analog recordings of vinyl, open reel, and cassette (we won't touch 8tracks..they didn't even come close..) having the same, basic dynamic output range within these above factors..

One can take an analog source and convert it to a 128k bitrate Mp3, but the sound would slightly suffer due to the 'lossy format" that the Mp3 format uses during the conversion from WAV to the Mp3.

A good analog soundtrack when read and reproduced correctly, if you want to compare such, can sound as good (or even better) as a 320k plus variable bitrate Mp3. For using the constant bitrate isn't the real good way to compress since so much info is lost in keeping the filesize at a budget level and try to keep quality intact at the same time.

Yes, definitely an Apple to Oranges comparision - analog against digital..how you slice and eat the fruit.
 

---

Posted by Louis Bornwasser (Member # 3063) on 06-25-2006, 09:42 AM:
 
A few rules, first? Assume 2 track, Dolby SR optical. Probably 70 db effective s/n and 20-16000 Hz. (Dolby tracks are filtered above that and you can't project much above that in a medium cinema-sized room) Louis
 

---

Posted by Marin Zorica (Member # 1716) on 06-25-2006, 11:02 AM:
 
Hmmm sticky question!

But in mine opinion, I think is closest to 192kbps mp3, considering You have SR and well system!
 

---

Posted by Charles Greenlee (Member # 3856) on 06-25-2006, 11:59 AM:
 
Yes, I was considering 2channel Dolby SR. I was throwing around a few ideas in my head, and was looking at a current mono print. Stereo has 2 tracks that take up that small space. Mono, has one track, that is twice the size as a single track from the stereo. So it should, in theory, be able to have a better dynamic range, and a lower media noise level.

A mono analog print, that is 2nd gen., meaning the soundtrack is straight from the master, and not copy of a copy of a copy, could actually have a higher quality than a digital recording? But just on a single track. That is if you had decent analog sound equipment, of course.

True we were comparing apples and oranges, but were relating their quality, which they both have, just doing it using digital terms. Thx guys, I got something to play around with.
 

---

Posted by John Hawkinson (Member # 1135) on 06-25-2006, 03:45 PM:
 
The apple/orange thing is a big deal here. If you have a reasoanbly perfect Dolby SR soundtrack, you won't be able to hear any defects. Even if you a 192kbps MP3 stream, you'll be able to hear compression artifacts (well, it depends on your ear).

Trying to equate compression artifacts to SNR isn't so straightforward.

--jhawk
 

---

Posted by Charles Greenlee (Member # 3856) on 06-25-2006, 06:36 PM:
 
Well, I don't buy the 192kbps so much because of lack of clairty in the analog, and lack of dynamic range. Mainyl what I was looking for is a comparitive benchmark on quality. Like DTS would be about 256k, etc. I have CD's, which are 160k, that come out clearer and w/ negligable artifacting, than the onboard analog in just about every theatre I've worked at. Of course part of that could be poor tone adustment, or equalization, at playback. Too much low end will make it sound murky, etc.

I'm just goofing off with some sound stuff, and wasy trying to figure out where the breaking point is on the optical track, what can be adjusted to make it keep up if not suprass the current digital formats. The mono format, being double sized compared to a stereo track, seems like I'm barking up the right tree. Only problem is it's just 1 channel. A remarkably clear channel, but only 1. It's more just for kicks than anything, but I am learning from it.
 

---

Posted by Patrick de Groot (Member # 858) on 06-30-2006, 07:04 AM:
 
Your question doesn't make much sense cause Dolby Digital is compressed audio (just like mp3). With compressed audio the following things are important for the resulting perceived quality:

-audio source is uncompressed audio with a high sample rate
-high enough resulting compressed bitrate
-used compressing algorithm (codec) mp3, aac, ogg, ...

The bitstream of a normal audio cd is 1411.2 kbps (kilo-bit-per-second) which is uncompressed audio (16 bit 44,1 KHz)(16 * 44100 * 2).

Compression artefacts sound a lot different than analog recording/playback "problems".

A better question would be what equivalent an analog soundtrack has in terms of digital recording quality (sample rate, resolution).
 

---

Posted by Robert Minichino (Member # 3495) on 06-30-2006, 07:43 AM:
 
You're talking about a compressed bitrate, which isn't comparable as stated above because of different codecs, etc., which all throw away different parts of the original signal (hopefully ones we can't hear ).

But you can compare an analog source to an uncompressed digital audio bitrate.

For the sampling rate, multiply the analog -10dB point by 2.2 (Nyquist frequency is 2x, but you need some room for the transition band), so that gets you about a 35kHz sampling rate for a 16kHz -10dB point.

For 70dB dynamic range, we can see how many bits gives us a quantization noise floor -70dB or more down from 0dBFS. 12 bits gives us a 72dB dynamic range (log2(10**(70.0/20)) = 11.6 bits, so round up to 12, 20*log10(2**12) = 72dB).

So, stereo means we've got a pair of 12 bit samples occurring at 35kHz, so 2*12*35 = 840 kbit/sec, compared to CD at 1,400 kbit/sec as stated above.
 

---

Posted by Charles Greenlee (Member # 3856) on 06-30-2006, 09:11 PM:
 
I a round about way, then optical, in it's current form, even with a new print, clean heads and good processor, is no where near CD quality, relatively speaking.
CD quality should be no where near digital cinema formats quality, that is unless the compression used by the digital formats degrades the signal below CD quality, which I wouldn't think it would. Thanks for the formulae. I can use it. So I'll go write it down in my noteboock when I get to it tomorrow. Left it at work.
Ok, on to my other questions. What is the relative quality then of Dolby Digital, DTS, and SDDS?
I've been expiramenting with the only optical cell I could scrounge before they scrapped our old white light readers. I found out that I can emulate a mono cell by ganging the two indiviual cells, up together. This seemed to have a greater dynamic range. Though this was done in the lab at school, with an led powered by a function generator, And with the cell hooked to our o-scope. Unfortunatly, I don't have a real soundhead I can experiment with, yet.
 

---

Posted by Louis Bornwasser (Member # 3063) on 07-01-2006, 10:31 AM:
 
Charles; there is even a difference in what analog can do. For example, an expert alignment of a red reader through a decent preamp can get you 16,000 hz flat. Very few soundheads are adjusted to or remain in that condition. To the extent that you fail this "maximum" you degrade the analogue. Same goes for level, crosstalk, etc.

The fact that mono is more robust is correct. Without Dolby, stereo would have been too noisy to tolerate. Hence it had to wait until suitable NR was available. 2 channel analogue optical was experimented in the 40's, but was too noisy.

Now we are comparing apples, oranges, pears & bannanas! Louis
 

---

Posted by Stephen Furley (Member # 1277) on 07-01-2006, 11:36 AM:
 
Louis, experimental two-channel optical stereo goes back further than that; at least to the mid '30s, when the Alan Blumlein recordings were made. web page I don't know if anybody else made any before this.

Some of these recordings still exist, I heard the amateur dramatic one mentioned on the web page a few years ago. The quality wasn't great by modern standards, but I wouldn't say it was much worse than contempory mono tracks.

Basically the same system was revived in the '50s, and got nowhere. There were many 'high-quality' sound systms over the years: Fantasound, Warnerphonic, Perspecta, Cinerama, Cinemascope, Todd-AO etc. All of these were either short-lived, or little-used. Most films were released in mono optical, and even where there was a stereo release, most cinemas ran them in mono. These systems were expensive of course, but they were available, if the demand had been there for them.

Both Dolby 'A' type NR, and the Sansui QS matrix encoder/decoder were available several years before Bolby stereo film tracks were introduced. Even when they were introduced, the conversion didn't happen overnight; it took many yearsfor cinemas to convert to stereo, there are still a few mono only equipped ones today.

I think the delay in conversion to stereo film sound was more economic and market led than technological.
 

---

Posted by Charles Greenlee (Member # 3856) on 07-02-2006, 02:35 AM:
 
I just came up with an idea on how to use the existing sountrack real estate, to allow for another descrete channel, at full mono fidelity and response. Damn, now I just need the resources to test it out. It's a solid idea but I don't want to broadcast it so it can't be jumped, but I need some more resources. Grrr.
 

---

Posted by Michael Schaffer (Member # 1204) on 07-02-2006, 03:20 AM:
 

quote: Louis Bornwasser

---

The fact that mono is more robust is correct. Without Dolby, stereo would have been too noisy to tolerate. Hence it had to wait until suitable NR was available. 2 channel analogue optical was experimented in the 40's, but was too noisy.

---

Also, - correct me if I am wrong - effective matrixing technologies were not available before the Schneider/Sansui matrix in the late 60s, so actual multi (more than 2, anyway) channel sound was not really possible. Maybe 3 front channels with a sum for the center, but no "effects" or surround channel.
 

---

Posted by Charles Greenlee (Member # 3856) on 07-02-2006, 03:34 AM:
 
Yeah, but as someone mentioned fantasound. Surround has been around since 1940. Fantasound had 4 optical tracks on film. 3 descreet audio, and 1 control. From that the togad, driven from the control track, was able to, ehh, kinda matrix out 3 more channels. Actually, all it did was pan the same 3 tracks to alternate speakers. But you could have a car drive by on the right, hear it approach on the center, pass to the right stage, and then via panning, pass to the right surround. I enjoyed reading the technical atricles on fantasound at the widescreen museum. Doly NR allowed the optical tracks to become small enough that 2 would fit in the same space and sound....acceptable. The matrixing is mostly for the optical, because there was 4 and later 6 channel magenetic before optical was matrixed to 5 channels.

Sum? Is that how a 3-stereo setup derives it's center channel?
 

---

Posted by Stephen Furley (Member # 1277) on 07-02-2006, 06:11 AM:
 
Michael, I'm not sure when matrix became available; your date sounds about right for a commercial product, but, very often, commercial products are based in ideas from long before. I don't know in this case. However, Dolby stereo could still have been introduced several years before it was. Even then, there was not a sudden rush by the cinemas to install it; many remaind mono for years. Quite a few installed Dolby 364/E2 mono systems, even after stereo prints were introduced. The Phoenix, East Finchley (later removed) and Screen 1 at the Odeon, Watford come to mind. Presumably this was done to avoid the cost of a stereo installation.

Dolby stereo didn't really offer anything much in technical terms which wasn't available before; what it did do was to offer it at lower cost and complexity than previous systems, which relied on either separate interlocked tracks, or mag striped prints*. Even though the prints were cheap, the cost to the cinema of a stereo installation was still considerable, as it had been for all previous systems, including Perspecta.

* The exception being the two channel Blumlein tracks, and the similar '50s Photostereo system, but the lack of a centre channel in these systems was s serious problem in a theatre auditorium.

Several of the early optical sound systems allocated a much greater film area to the sound; one used roughly half the width for each. The best-known of these systems was probably the 42mm Tri-ergon. For anybody who hasn't seen it, take a normal 35mm silent film, add an extra 7mm of film outside the perforations on one edge, print a wide VD track on this extra film area. It's a pity such a system wasn't adopted; the extra space available would have been useful a few years later when stereo experiments started.
 

---

Posted by Steve Guttag (Member # 268) on 07-02-2006, 07:59 AM:
 
Charles...at this stage in the cinema sound on film...I don't think too many people will be jumping on changes to the otpical soundtrack...then again, I didn't think they would have accepted the cyan crap either so what do I know (and I maintain it was a mistake...the effin red readers are not stable enough...their levels drift all over the place depending on the batch of LEDs....grrrr).
 

---

Posted by Charles Greenlee (Member # 3856) on 07-03-2006, 04:03 AM:
 
Hmm, I haden't notice the drift, but it makes sense. I mean, you can buy a semiconductor here, and buy supposedly the exact same thing somewhere else, and because it's simply from a different batch, it may be off. Good point. Either way, I'm writeing my ideas dow, so I can remember them, and it makes it easier when you can go back a reread. Sometimes you find stuff you didn't before. Well, I guess drop me a line if you're interested in helping, or at least interested in listening. Maybe you can find something I haven't yet, or counter something I have.
 

---

Posted by Mark Gulbrandsen (Member # 72) on 07-03-2006, 10:15 AM:
 
Semiconductors are like lamphouse mirrors... Two are never exactly the same... I supose I should add projection lenses in there too!

Mark
 

---

Posted by Charles Greenlee (Member # 3856) on 07-04-2006, 03:06 AM:
 
Ha-ha, yeah. We've had someone burn one of our lenses (remember kiddies, if the lamp is lit, the motor should be running. And limit the amount of time the lens is exposed to the unfiltered light), so we pulled a flat lens of the same stats from another projector that wasn't using the flat (was a scope feature) and put in there. Well we did it by simply unscrewing the lens from its collar, rather than fuss with pulling the collar and lens out of the turret, and having to refocus and all when we put it back. So collar and lens from the recepiant house were already at a position of being in focus and aligned and all. It should be a simple matter of unscrewing the bad lens and screwing the good one on the collar. Ths should work since, on paper, the lenses were identical. Not! The image size was slightly larger, and the focus was off more than just a smidge. Evidently, even if the lenses have the same stats, there's a difference from lens to lens, even if unintentional.
 

---

Posted by Charles Greenlee (Member # 3856) on 07-13-2006, 11:20 PM:
 
Anyone good with sound stuff? I mean on a more technical, engineering standpoint? Drop me an email.
 

---