5

On my phone, for example, HDR is a feature. Whereas in music production, a good compressor is seen as essential.

Why the apparent discrepancy?

10
  • Have you not seen Instagram filters? The kids with all their pop culture like low dynamic range photos and music. Old folks are more likely to want hi res HDR on both areas Jul 7 '21 at 21:12
  • 1
    @piiperiReinstateMonica - tbh, this is a very confused explanation. It really misses the point by a long way, without quite going so far as to be 'completely wrong'.
    – Tetsujin
    Jul 8 '21 at 6:12
  • 2
    HDR as found in cameras is a compressor. Paradoxically, it produces a low dynamic range image by crushing a high-dynamic range input, so the name is a bit backwards. The sRGB colour space doesn't have enough room for an actual high dynamic range image, so "HDR" takes the high dynamic range information and compresses it to fit in a low dynamic range, limited colour space like sRGB (so you can share it on Facebook where everyone will see it on a poor display).
    – J...
    Jul 8 '21 at 14:11
  • You appear to be confusing lossless with lossy compression, and further you are confusing commercial product with personal product. Cable/satellite tv, for example, often compresses the heck out of picture resolution and color resolution. Jul 8 '21 at 17:40
  • @Tetsujin By HDR images I mean this en.wikipedia.org/wiki/High-dynamic-range_imaging i.e. various techniques for producing images with more information per pixel than can be stored in the 8 bits per color component available in conventional image formats. Do you mean something different? I removed the comment anyway. Jul 8 '21 at 18:35
15

Sound compression is used to respect the signal limits of microphones and/or speakers. Proper compression prevents sound distortion and also helps ensure across tracks that the dynamic ranges are consistent. The more range the mics or speakers have, the less compression is needed.

HDR is actually quite similar. It takes several pictures at different light ranges and then uses post-processing software to merge them into a consistent image. (See https://www.digitaltrends.com/photography/what-is-hdr-photography/.) This is, in a sense it's own form of compression.

Consider reading this Audacity manual page on dynamic compression.

The Wikipedia page on high-resolution audio may also be of interest.

13
  • 6
    It's a good answer, but I think the link on lossless compression concerns data compression, not dynamic compression – these are two different things. Jul 7 '21 at 18:20
  • 2
    @Aaron, I think the Audacity page about compression gives a pretty good overview manual.audacityteam.org/man/compressor.html Jul 7 '21 at 18:27
  • 2
    HDR is actually technically a misnomer. It's really Low Dynamic Range ;) As you say picture-level or audio-level [as opposed to data] compression are really very similar.
    – Tetsujin
    Jul 7 '21 at 18:52
  • 6
    @Yorik. No, it's a technique to compress the ratio of highest to lowest luminance. It takes several shots at different EV values then composites so that the darks get more exposure & the lights get less… exactly the same as compressing audio, make the quiets louder & the louds quieter. Just the same as audio it only has so many numbers it can handle, no matter what those numbers are or bits used to express them, it is finite. [I'm ignoring floating point audio here, for simplification]
    – Tetsujin
    Jul 7 '21 at 19:01
  • 2
    @Yorik both HDR and audio dynamic compression reduce the dynamic range of the input in order to fit desired (and usable) range of the storage medium and/or reproducing system and/or consumer perception. The dynamic range of the output is lower than that of the input – which can be also phrased as: the output represents input of higher dynamic range, that would be possible to present without processing. HDR represents quite specific technique to achieve it, while audio dynamic compression is a bit wider term, but the general idea is the same. Jul 7 '21 at 19:55
8

There's compression in both mediums, it's just that in photography it's there by default, and it's usually done by the camera for you. When you go into Lightroom and bring down the hightlights or pull up the shadows, that's compression. You're doing it so that you're able to comfortably see both of them on a piece of printed paper, or a consumer grade monitor.

Even a good DSLR sensor will struggle to capture the entire dynamic range of human visual perception. We have to resort to all sorts of tricks, like taking multiple pictures at various exposure levels to get acceptable results. At the same time, even the brightest screen with the latest HDR technology is not going to give you anything close compared to, say, looking at the sun in real life. Photography focuses a lot more in getting as much range out of the medium as possible, because the medium is already so compressed to begin with, and we haven't even gotten close to fully representing reality.

On the sound side, our playback devices generally have less dynamic range than our microphones. We easily record a dynamic range that matches the limits of the human ear, and a good sound system will be able to play that back that with high fidelity, but cheap earbuds will distort when the sound is too loud, and you won't be able to hear the quiet sounds if you listen in a noisy environment. So for the sake of accessibility, we compress sound down to a range that's comfortable for general use.

Both mediums have limitations, and in both cases the dynamic range of reality is much greater than we would often want to reproduce. I don't want my computer screen to be able to blind me if I take a picture of the sun, nor do I want my headphones to blow out my ear drums if I record a jackhammer. But the latter is a lot easier to acomplish than the former.

Finally, on a subjective note, if we're talking about music and photography as art forms, then photography is much more focused on capturing reality, while music is much more about evoking feelings through hyper-real experiences. You also have genres of recorded music that focus on recreating the original experience, like classical, and in that case you generally don't want to compress a lot. In photography, you might purposefully want to get a low dynamic range to invoke nostalgic associations, but generally speaking the goal is to represent reality in some way, and that means working against the limitations of the medium to get as much detail as possible.

6
  • 2
    I actually have some albums, mainly classical, that I had to 'remaster' myself so I can listen to them in the car without having to keep one hand constantly on the volume control ;))
    – Tetsujin
    Jul 8 '21 at 16:06
  • 2
    @Tetsujin I used to do choir recordings, and the director always told me that he wanted to keep as much dynamic range as possible. He would then listen to the recordings in his car, and invariably turned up the volume as much as it went, because the music would usually start softly and he wanted to hear the details. This would then cause the speakers to distort whenever the sopranos started shrieking in the loud bits. And that's how I learned mastering.
    – Alex J
    Jul 8 '21 at 18:10
  • What seems back to front to me is having to compress the dynamic range in the recording, when a more sensible solution would be to provide that possibility as in option in the reproduction equipment. Listening to classical music in the car is an extreme case of that, but even at home the ambient noise can vary a lot with the time of day and what else is going on in the home. It makes no more sense than saying the volume should be set in the recording, not by a volume control!
    – PJTraill
    Jul 8 '21 at 21:24
  • @PJTraill - I don't normally do that at home for music [though I have for parties], but I do for movies. I have a whole bunch of Waves multi band comp/lims strapped across the 5.1 output - lift the whispers whilst damping the explosions. I do comp it hard, so I can hear the dialog clearly without driving the neighbours crazy.
    – Tetsujin
    Jul 9 '21 at 10:13
  • @Tetsujin: Good to hear that, just a shame it is not a standard function on everyday equipment.
    – PJTraill
    Jul 9 '21 at 21:09
2

The name HDR, High Dynamic Range in photography is a bit misleading. It refers to increasing the camera input range, but typically involves processing of subsequent reducing that range, similarly to audio compression. However the reasons and conditions, and techniques are a bit different.

Typical human hearing dynamic range is 120 dB. Not the whole range is however practically usable for music. Too loud sound might be unpleasant and damage hearing. Too quiet sounds won't be heard over ambient noise. A large orchestra can reach perhaps 60 dB difference between the loudest and the most quiet notes. For most other music sources this difference will be less. Moreover listening to such dynamic program requires both good acoustic conditions, as well as high focus of the audience.

Since many decades the capabilities of audio equipment quite decently match the human hearing range. Today a 120 dB audio interface can be bought cheaply, 16 bit storage standard offers over 96 dB of dynamics, and it doesn't cost much to buy hifi system for you and your neighbors to enjoy the music.

The reasons to limit audio dynamic range are therefore mainly artistic and practical, including the infamous loudness war. Some of these reasons are addressed in other answers.

Often quoted human vision range is 20 stops, sometimes more, that is 20 increases of brightness by factor 2, corresponding to brightness contrast of 1:1000'000. Best cameras offer range of 13–15 stops. Best displays can reproduce 10–11.5 stops. Contrast of printed images is just 6–7 stops. At the same time it's easy to darken a room to get rid of any unwanted light when e.g. viewing a movie.

Therefore, unlike for audio, there are still significant technical challenges in capturing the full vision range. HDR is therefore primarily a technique to extend the dynamic range of a camera, but combining several photos with camera sensitivity set to various brightness ranges. Most typically however the dynamic range of the resulting image is then reduced to match the dynamic range of reproduction medium, and this process is also called HDR.

Summarizing: Music dynamics is lower than the range of human hearing and audio equipment, in many styles of music it's much lower. There are practical reasons not to utilize full available dynamic range in many occasions.

In contrary, details relevant for an image might be present at any point of the dynamic spectrum, and capturing them is a challenge. HDR increases the dynamic range of the camera, but HDR typically means also subsequent limiting this range to match reproduction technical capabilities.

It should be noted, that both audio compression, and HDR result emphasize small (in terms of dynamics) details of the input. In both cases artistically it might be desired or not.

1

A wide dynamic range is valued in music too. But in some modern styles of recorded music, ‘loudness’ is valued even more! The compression, limiting etc. that produces maximum loudness destroys dynamic range.

4
  • 3
    While this Answer could be more detailed and use more citations, it's not wrong. There is something called "the Loudness Wars" and a trend in certain types of music to reduce the dynamic range. And there is other music that attempts to represent, as well as possible, the dynamic range in the original instruments being played. It's why you have to so often crank up the volume on classical music.
    – trlkly
    Jul 8 '21 at 11:05
  • 1
    You like some waffle? :-) My philosophy is to say what needs saying, then stop. Jul 8 '21 at 14:06
  • The loudness war is over… fortunately. Spotify, Youtube & Apple Music, amongst others, made any track mastered like that heavily punished… so people stopped doing it [not a moment too soon]
    – Tetsujin
    Jul 8 '21 at 16:07
  • 7
    @LaurencePayne: What waffling are you talking about? I simply added more information to the same thing you said. I saw your Answer had been downvoted below zero (until I upvoted it because it's not technically wrong), and decided to try and help so it wouldn't be further downvoted. A good Answer may be short, but it will have sufficient information, and will be well cited. Yours has neither. I even tried to help by giving you the keyword "loudness wars." If you don't care enough to improve a downvoted Answer, then I feel I wasted my time and shouldn't have upvoted.
    – trlkly
    Jul 8 '21 at 18:46
0

In both music and photography, the optimal dynamic range will depend upon the environment in which a medium is experienced. If one will be listening to music in a noisy room, or will be viewing a picture projected on a wall in a brightly lit room, an overly high dynamic range will cause parts of the content to be indistinguishable from the background. If one will be listening to music or watching a movie in a dark theater that is free from extraneous noise, however, a higher dynamic range will enhance the experience.

The purpose of a compressor in music is to allow a sound source which has a high dynamic range to be usable in contexts where the optimal dynamic range would be much lower. If a sound or picture is assembled from multiple overlaid elements, attempting to adjust the dynamic range of the composite result will often cause the elements to interact with each other in undesirable ways. For pictures, this isn't usually a problem because most scenes would contain material which should have comparable dynamic range. In music, however, this is much more likely to be a problem.

If one has a guitar whose transients are at -1dB but whose notes sustain at -20dB, along with a keyboard player which is sitting at a volume of -15dB, trying to flatten the overall volume profile to fit within 6dB would make it necessary to have the keyboard part increase and decrease in volume by 8dB. If the guitar part had been compressed so the peak transients were at -10dB, then the keyboard volume could sit steady at -5dB.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.