Welcome to our community

Be a part of something great, join today!

  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

This is very closely related to film vs digital

Hm...

This could be expanded a bit upon, as i happen to have had my first wave of analogue vs digital on audio.
And in bost cases, it seems that compression is the keyword.
Harmonic distortion, is maybe an issue and a result of that.

Compression of signal levels from more (like a cannon) to less (what does not kill your earlubes) is the same as from the sun to a signal on a tv-screen. Thus
Preserving dr in a medium which has no forgiveness, is the issue often passionately discussed with little understanding both in the audio and the video world.

But i agree
The parallels are screaming!

Thanks

G
 
Thing is that compression of signal-levels, thus the capacity to represent "more" signal with "less" signal is exactly what we want and strive for. Like representing the sun on an iPhone...

Or in a cinema.

In analogue revording and playback media, this compression to a certain extent is a side effect of the media itself, while in a digital world, the mathematical representation of values and levels is very unforgiving and there happens unwanted distortion in both the lower and highest part of the signal. The answer to that in the audio buiseness is to go for a signalstandard with "enough" headroom to betyer emulate the analogue signal, than the initial 16bit revording devices could.

My guess is that is where we're heading with digital video too, and the epic pretty much represents that philosophy. But as a first attempt, not as the final stage, is my guess.
 
Film and audio do share a lot of similarities. In your article you focus on "Harmonic Distortion" but really CD sound is mainly dependent on the quality of the analog to digital converter and clocking used during transfer. Transferring analog audio to digital is an art. Although many people (mastering houses) have the ability to do it, VERY few people actually know how to do it properly.

Back to similarities.. Film celluloid overexposes nicely, digital video clips harshly. Analog audio tape clips nicely, Digital audio clips harshly.

It's all about dynamic range. Film has more DR than digital video and in the audio world tape has more range than digital audio.

Audio is not getting better unlike the fast and furious evolution of digital film sensors and dynamic range. Audio is actually getting worse with cheap MP3 compression and over the top sample rates that do absolutely nothing. 192??? And then we have the loudness war, how pathetic is it that dynamic range is being absolutely ripped apart by every single modern recordings. My advice, for your favorite records try to find CDs printed before 2005 and avoid at all costs modern "remasters". If you have the patience Vinyl is still champ but avoid the newer prints (180 grams..)

I work with a lot of audio engineers and it is very interesting how similar both worlds are. Noise reduction, clipping, dithering, dynamic range etc..
 
Thing is that compression of signal-levels, thus the capacity to represent "more" signal with "less" signal is exactly what we want and strive for. Like representing the sun on an iPhone...

Or in a cinema.

In analogue revording and playback media, this compression to a certain extent is a side effect of the media itself, while in a digital world, the mathematical representation of values and levels is very unforgiving and there happens unwanted distortion in both the lower and highest part of the signal. The answer to that in the audio buiseness is to go for a signalstandard with "enough" headroom to betyer emulate the analogue signal, than the initial 16bit revording devices could.

My guess is that is where we're heading with digital video too, and the epic pretty much represents that philosophy. But as a first attempt, not as the final stage, is my guess.


But remember, that compression was another different issue at the time when MP3 was invented by Fraunhofer in the 90s.
What I’m talking about here was way before that, and I also remember that there is no compression in digital audio CDs.
Mp3 = Digital Audio compressed 10:1 at 128Kbps.
Audio CDs = Digital Audio 16, 24 or 48 bit, 44khz Stereo. No compression.

Anyway I don’t want to get off topic here, since I’m trying to point out the fact that, we are experiencing the same thing, in the transition between film and digital, and in my personal opinion, we see many cinematographers and directors shooting on digital, but when they grade, they all try hard to make it look like film as possible, and this is happening with many blockbuster movies today, just in the same way they wanted back that "Magic Analog Tape sound” on CDs in the 80s.
And they got it, many recording studios today use very expensive reborn vacuum tube preamplifiers and other analog devices to recreate that warm sound, that we probably don't quite notice in the music we listen today.
I don’t blame cinematographers and directors, Film still has something magic that digital lacks of. (Again, personal opinion).
 
Last edited:
But it is very closely related. Thing is i am talking about dynamic compression, which in the digital domain just can happen after capture of the full range, while you're talking about data-compression, which is a totally different issue.

In. The analogue domain, a lot of the dynamical compression is happening at capture, then the rest at playback. In the digital domain, you need capacity to record th
E full range, then compress and then playback at that compressed level.

That is why we need more bits and resolution in the digital domain, than what is resolveable from any single frame of analogue capture....

Also, when talking film, digital must replace stocks from 50 to 1600 iso, that all could theoretically have 16 stops of latitude... Just to pick a number...

That equals 21 stops of latitude to replace film.

Understand me right. I am a digital dude, and like it. But it is smart to know what we are replacing what with, and why there still will be some room for enhancement...


Eventually i think imagecapture will stabilize, like audiocapture has at 32 bit float. Maybe 24 bit float is enough for imagery. At whatever number it stops, there will still be room for improvement in the signalchain, after the math overhead is good enough to emulate he analogue domain...

Nothing wrong with digital. Love it.
But we're not done yet.
 
Have a listen to young adults, teenagers and children speaking. You may observe a trend towards alteration of speech more characteristic of subtle hearing deficits. We naturally in our learning process say what we hear. Whether a significant number of this current generation has deafened itself with ipods and loud arena music is one question. If the numbers of affected people are suffiicient, then their affected speech may become the predominant accent to be learned by the others around them. Is the reproduced speech also being affected however slightly, by the different sound being heard in heavily compressed voice tracks via cellphones and other audiovisual digital media? It would be interesting if some researchers had a bit of a look at this and published.

Words like "sleep" becoming "khhhleep". "Bruce Lee" becoming "Bruffkhlee" are extreme examples.

When CDs came upon the scene, the dynamic range was immediately evident but at the same time, I found myself reaching deep to hear something which seemed to be missing. The natural reaction was to turn the levels up. This was due as we eventually were informed that the 44Khz digitising standard eliminated audio frequencies which had been thought to be inaudible. The earlier compressed digital audio on broadcast TV was so ordinary that entire consonants could become substituted.


I think we can look safely forward to a future without becoming "fline" (blind) from watching digital imaging.

Overwork of human faculties through proportionately too much media input competing for a finite human learning ability may be another one for the researchers to continue to examine. In the meantime, despite being of the wrinkly generation, I am not in any hurry to go back to black and white picture books and distant radio stations fading and phasing with the quirks of ionospheric MW radio propagation. The current generation of kids and teens amaze me with their fluent knowledge of the technical world and ability to function in it. It makes me positively cringe when I reflect on how "dumb" I then was at the same age.
 
Mp3 = Digital Audio compressed 10:1 at 128Kbps.
Small correction: there are many available bitrates, all the way up to 320kbps. Ironically, 320K more or less corresponds to 4:1 compression, which is not far off from the lowest Redcode bitrate, and the relative bitrate of HDCam-SR.

There are some interesting comparisons you can make between analog and digital, but one key problem is, if you slam digital hard enough, you can lose a lot of its advantages over analog. This is true with sound, pictures, tape, digital, and everything else. Film is very forgiving to exposure mistakes (2 stops either way); digital is not.

I know of a major $75 million Hollywood film about six years ago where a key sequence was shot about 2.5 stops underexposed -- on a night scene -- and we were able to save it in post. No one ever knew except the colorists and the DP. Film got released, made a lot of money... everybody was happy. If it had been digital, I don't think there would have been enough signal down in the crushed part of the blacks to salvage.
 
Film has more DR than digital video and in the audio world tape has more range than digital audio. .

Well, actually that'd be incorrect. Currently DR in digital audio is actually a lot higher than the best analogue tape. Even humble CD with it's 96db DR is practically achievable and about 20db better than the best analogue tape, with 24bit analogue to digital converters reach 20+bits of precision / dynamic range (rather than the 16 of CD).

Graeme
 
But with audio you have more controll over preamps and compressors/limiters prior to recording, than you usefully have with images... (Sure... I am aware of lights and filters...)

The thing one could do analoguely, was to change to a stock with the same range, but different sensibility. That is not yet fully emulateable digitally. But we're getting there...

Or rather... You guys are, if i interpret Jim correctly...

Hahahahaha
 
Back
Top