Recording tracks for each instrument

[Dan Bank US]

I have wondered if the industry has considered creating multitrack recordings, allowing each instrument (or group of instruments) to have its own track, fed into the electronics separately and then sent to its own speaker. There are speakers that are better at reproducing some instruments, such as ribbons for strings and horn speakers for winds. This would allow them to be specialized speakers and the signal itself would be cleaner or simpler. Even the digitization would be more precise as it is not trying to replicate a conglomerate of sounds. This would have been virtually impossible with analog, but seems like digital recordings could do this and then the electronics market could adapt the multichannel capabilities. And the speaker manufacturers could hone their designs to focus on specific instrument reproduction. Seems silly actually that we use different instruments to create the music but then try to reproduce all of their sounds accurately through our speakers.

 

[lovinthehd]

Be a helluva system to handle a full orchestra....

 

[Gmoney]

Dan, you gotta point,

 

[lovinthehd]

Possibly cheaper just to have the orchestra play at your place the more I think of it. All those different speakers needed for all the instruments out there boggles my mind Let alone the stack of amps! Altho I suppose they could all be active speakers....but the wiring mess! I'm just not coming up with any good reasons overall....

 

[davidscott]

Good idea but we will need what maybe 20 pairs of speakers or more? Separate speakers for male and female voices?
Sounds like an idea Kramer from Seinfeld would have thought of.
Seriously in a perfect world this is what we would own..

 

[sterling shoote]

I have wondered if the industry has considered creating multitrack recordings, allowing each instrument (or group of instruments) to have its own track, fed into the electronics separately and then sent to its own speaker. There are speakers that are better at reproducing some instruments, such as ribbons for strings and horn speakers for winds. This would allow them to be specialized speakers and the signal itself would be cleaner or simpler. Even the digitization would be more precise as it is not trying to replicate a conglomerate of sounds. This would have been virtually impossible with analog, but seems like digital recordings could do this and then the electronics market could adapt the multichannel capabilities. And the speaker manufacturers could hone their designs to focus on specific instrument reproduction. Seems silly actually that we use different instruments to create the music but then try to reproduce all of their sounds accurately through our speakers.

JBL has demonstrated surround sound amplification and speaker arrays which essentially recreates an orchestra. One of their amps has 64 channels.

 

[mtrycrafts]

Be a helluva system to handle a full orchestra....

Or a half orchestra.
Then, how do you arrange your speakers. Do we know where all the instruments were located?
Or, if they were even all present for the recording?

 

[mtrycrafts]

Good idea but we will need what maybe 20 pairs of speakers or more? Separate speakers for male and female voices?
Sounds like an idea Kramer from Seinfeld would have thought of.
Seriously in a perfect world this is what we would own..

Even just 20 speakers, 20 instruments, where are they located? What instruments? And the next recording has 8 instruments, rearrange once again?
Right speaker for the instrument?

Just a real silly idea.

 

[mtrycrafts]

By the way, in the past musical recordings were recorded for each instrument then mixed down.

 

[Verdinut]

Even just 20 speakers, 20 instruments, where are they located? What instruments? And the next recording has 8 instruments, rearrange once again?
Right speaker for the instrument?

Just a real silly idea.

And if you want to screw up an original acoustic performance, then it's one of the ways to do it.

 

[sterling shoote]

It's all about what method realizes the best illusion. Seems multi-track/multi-channel could be viable with technology to stamp out relatively small scale speaker arrays inexpensively.

 

[highfigh]

By the way, in the past musical recordings were recorded for each instrument then mixed down.

Most recordings are still made that way, although some artists like to record 'live', with the whole band or most of the people who will be on the track(s).

 

[Dan Bank US]

Be a helluva system to handle a full orchestra....

Do not need a speaker for every instrument. Could have one for strings / acoustics another for horns and another for winds. Center channels seem to replicate voices very well. Have 5 Maggies set up for my surround, and when I listen to acoustic guitar like Estas Tonne on YouTube it is quite amazing. You would think he was here. But when you mix in other types of instruments the realism factor drops.

 

[Dan Bank US]

Most recordings are still made that way, although some artists like to record 'live', with the whole band or most of the people who will be on the track(s).

I know. Which is part of why it came to me, why mix together only to create a soundwave that is already compromised.

 

[TLS Guy]

I have wondered if the industry has considered creating multitrack recordings, allowing each instrument (or group of instruments) to have its own track, fed into the electronics separately and then sent to its own speaker. There are speakers that are better at reproducing some instruments, such as ribbons for strings and horn speakers for winds. This would allow them to be specialized speakers and the signal itself would be cleaner or simpler. Even the digitization would be more precise as it is not trying to replicate a conglomerate of sounds. This would have been virtually impossible with analog, but seems like digital recordings could do this and then the electronics market could adapt the multichannel capabilities. And the speaker manufacturers could hone their designs to focus on specific instrument reproduction. Seems silly actually that we use different instruments to create the music but then try to reproduce all of their sounds accurately through our speakers.

That would be totally the wrong approach. Good speakers do not favor particular instruments and are impartial to all. I know favoring certain instruments is a trait of many speakers, but that is a significant defect. Think about it. If you record an oboe for instance and play it back through a speaker that sounds like and oboe, you have double oboe!

Now all sounds are made up of sine waves, which combine to make a complex wave. There is only ONE wave however, so a speaker only has one vector of travel at any instant. This can all be shown by Fourier analysis. You have lots of company with many harboring this miss conception.

The first time I encountered this question was when I was ten years old. I had a school master who knew I was very interested in the reproduction of sound. He asked me time and time again how a loudspeaker could possibly reproduce all the instruments of the orchestra at once. I attempted to go over the principals of Fourier analysis over and over again. I could never get him to understand, and unfortunately he remained convinced it was sort of alchemy! Anyhow that was one of a number of sentinel events from my childhood, that convinced me, and gave me the confidence, that I would make out OK.

 

[highfigh]

I know. Which is part of why it came to me, why mix together only to create a soundwave that is already compromised.

Why mix the individual tracks together? Because they're making a product to sell, not an item of perfection.

 

[slipperybidness]

That would be totally the wrong approach. Good speakers do not favor particular instruments and are impartial to all. I know favoring certain instruments is a trait of many speakers, but that is a significant defect. Think about it. If you record an oboe for instance and play it back through a speaker that sounds like and oboe, you have double oboe!

Now all sounds are made up of sine waves, which combine to make a complex wave. There is only ONE wave however, so a speaker only has one vector of travel at any instant. This can all be shown by Fourier analysis. You have lots of company with many harboring this miss conception.

The first time I encountered this question was when I was ten years old. I had a school master who knew I was very interested in the reproduction of sound. He asked me time and time again how a loudspeaker could possibly reproduce all the instruments of the orchestra at once. I attempted to go over the principals of Fourier analysis over and over again. I could never get him to understand, and unfortunately he remained convinced it was sort of alchemy! Anyhow that was one of a number of sentinel events from my childhood, that convinced me, and gave me the confidence, that I would make out OK.

The incredibly useful, but somewhat difficult to calculate Fourier Transform! I always hated running through these types of calcs, while I completely understand the underlying principles.

Extremely useful in the scientific field! The most notable examples of course is FTIR analysis, where your raw data is a complex EM wave form that must be processed by FT to obtain a useful analytical signal.

 

[mtrycrafts]

Do not need a speaker for every instrument. Could have one for strings / acoustics another for horns and another for winds. Center channels seem to replicate voices very well. Have 5 Maggies set up for my surround, and when I listen to acoustic guitar like Estas Tonne on YouTube it is quite amazing. You would think he was here. But when you mix in other types of instruments the realism factor drops.

Then you compromise the signal again per your goal.

 

[sterling shoote]

Do not need a speaker for every instrument. Could have one for strings / acoustics another for horns and another for winds. Center channels seem to replicate voices very well. Have 5 Maggies set up for my surround, and when I listen to acoustic guitar like Estas Tonne on YouTube it is quite amazing. You would think he was here. But when you mix in other types of instruments the realism factor drops.

Well then maybe you do need a speaker for every instrument, since when "you mix in other types of instruments the realism factor drops".