[johncarm]

The sound of that trumpet is different if you are playing it than if you are standing directly in front of it. It sounds quite different if you are 15 rows back in the orchestra section, and different still if you are on the third balcony in the back. All the other instruments in the orchestra sound different from different perspectives too. A good classical mix will make compromises to balance all of the different perspectives to create an optimal perspective that might not even reflect any real perspective. But it sounds real if everything is balanced properly and playing towards strengths. An orchestra produces a very complex sound. The venue adds a whole different level of complexity to that. There may not be one single optimal location, especially with concertos featuring weaker instruments like flutes or violins. A little bit of a closer perspective on the soloist with a little more distant on the tympani and brass will sound better than recording everything from the same distance. I'm not sure if I'm stating it clearly, but there is physical real and your eyes help you sort out distances and allow your brain to compensate. When you take away the visual aspect, it's better to create an optimal sound that goes beyond physical reality. Something clearer than the live sound. When you're on a mixing stage and you're working with the elements, all of this is clearer than explaining it in words.

I think I get what you are saying. Balancing these factors is important. I've created sampled versions of some of my compositions, so I'm working with mixes of software synths and samplers. I'm also aware as a composer that the presentation of, say, a symphony by Mahler is a presentation of many elements which need to be in the proper perspective and balance in order to communicate a beautiful and clear vision of that symphony.

By the way, I'm not talking about the trumpeter himself evaluating the recording, but rather a musician who is very familiar with that trumpeter's playing.

What I'm saying is that you don't need to bring in multiple instruments or mixes in order to talk about the "balance" that a recording presents. A single-point recording of a single instrument still presents us with a need to balance all the elements in that performance.

What I'm doing is simplifying the situation so that we can discuss the role of a musician's perception.

 

[bigshot]

What if the optimal distance from the mike for a flute is different than a trombone, but they're sitting right next to each other? And it isn't just the volume level. The timbre of an instrument changes as the mike is placed closer or further away. Some instruments may have more character close and others may sound better with more of the room reflection wrapped around them. This is particularly important with concertos. A solo violin sounds better if it is a little closer miked than the violin section is.

Single point stereo miking can sound good, but it requires everything to be working perfectly all the time, and you lose some of the control to finesse the sound of individual instruments in the mix. Multi-miking offers more options and control than simply capturing sound from one position. Optimized sound can be better than accurate sound. That is what sound engineering is all about... creating a sound that is better than real.

 

[johncarm]

What if the optimal distance from the mike for a flute is different than a trombone, but they're sitting right next to each other? And it isn't just the volume level. The timbre of an instrument changes as the mike is placed closer or further away. Some instruments may have more character close and others may sound better with more of the room reflection wrapped around them. This is particularly important with concertos. A solo violin sounds better if it is a little closer miked than the violin section is.

Single point stereo miking can sound good, but it requires everything to be working perfectly all the time, and you lose some of the control to finesse the sound of individual instruments in the mix. Multi-miking offers more options and control than simply capturing sound from one position. Optimized sound can be better than accurate sound. That is what sound engineering is all about... creating a sound that is better than real.

I'm not arguing for single-point microphone recordings of orchestras. I'm creating a simplified situation so we can talk about it and build from there.

I hope you would agree that the ability to produce a good recording of a single instrument is a prerequisite to recording an entire orchestra.

Yes, the timbre changes.

As I said, the engineer makes two recordings, R1 and R2. Maybe the mic distance is different and the timbre is different.

We then can listen to both of them and make a few judgement calls.

One judgement would be how much each one sounds like the live sound. A similar idea is asking how each one presents the details that the musician practiced and performed in a specific way.

Changing anything may alter the balance. Maybe the attacks have different intensities. In R1, the attacks have greater intensity, and perhaps that doesn't work musically. Perhaps our listener L says that it distorts the live experience, in which the attacks had the proper intensity. In that case, R2 would be the more accurate recording.

 

[bigshot]

The point of recording isn't recreating the live sound of the performance accurately. The only time you do that is when you don't have any choice because there is a live audience and the recording priorities have to follow the priorities of the show. If you're recording in a studio, you want as many options as possible so you can use the power of recording technology to make it sound even better than front row center.

 

[johncarm]

The point of recording isn't recreating the live sound of the performance accurately. The only time you do that is when you don't have any choice because there is a live audience and the recording priorities have to follow the priorities of the show. If you're recording in a studio, you want as many options as possible so you can use the power of recording technology to make it sound even better than front row center.

Now we're getting somewhere! Hopefully we won't be talking past each other as much if we can get this sorted out.

Did you see that I'm talking about how the musician practiced and performed in a certain way, choosing dynamics of individual notes, brilliance of attack, shaping notes, tempos, and even the trumpet model itself in order to convey a musical message?

I'm not talking so much about recreating the "live sound" accurately, as there is no simple way to recreate and sound field -- and no need to recreate the sound field -- in order to do what I'm talking about -- presenting the musical message accurately.

Do you see the distinction?

 

[bigshot]

The way the musician performs is recorded. That is ultimately what you hear. But there are other people who have input as well... the accompanying musicians, the conductor and the producer (aka "Tone Master"). An orchestral recording is a team effort and each member of the team does his part and hands it off to the next person to make it even better. In general, the essence of the performance is retained. If the conductor isn't happy with the phrasing the musician has practiced, the player is going to follow the conductor's lead, not his own. Once the conductor has completed the recording session he turns it over to the producer with his notes and its up to the producer to add his magic to the mix. The engineering team is there throughout planning and optimizing. A good example of this is the Decca Ring. You had the Vienna Philharmonic, and fantastic Wagnerian singers, following Solti's direction on dynamics and tempo, and everything following Culshaw's lead on placement of the singers and the "sound stage" that was created by the mix. Culshaw took the performance to the next level. He was as much a part of the proceedings as the musicians. The same is true of George Martin and the Beatles. The performers get to be on the album cover, but there are stars who just get their names in little letters in the credits who are a big part of it too.

When you're recording a blend of sounds that is complex, you record it in pieces and put it all back together, balancing one element against the other. This is a bigger task than one would expect. I spent three 10 hour days on a minute and a half of music once.

 

[Glmoneydawg]

The way the musician performs is recorded. That is ultimately what you hear. But there are other people who have input as well... the accompanying musicians, the conductor and the producer (aka "Tone Master"). An orchestral recording is a team effort and each member of the team does his part and hands it off to the next person to make it even better. In general, the essence of the performance is retained. If the conductor isn't happy with the phrasing the musician has practiced, the player is going to follow the conductor's lead, not his own. Once the conductor has completed the recording session he turns it over to the producer with his notes and its up to the producer to add his magic to the mix. The engineering team is there throughout planning and optimizing. A good example of this is the Decca Ring. You had the Vienna Philharmonic, and fantastic Wagnerian singers, following Solti's direction on dynamics and tempo, and everything following Culshaw's lead on placement of the singers and the "sound stage" that was created by the mix. Culshaw took the performance to the next level. He was as much a part of the proceedings as the musicians. The same is true of George Martin and the Beatles. The performers get to be on the album cover, but there are stars who just get their names in little letters in the credits who are a big part of it too.

When you're recording a blend of sounds that is complex, you record it in pieces and put it all back together, balancing one element against the other. This is a bigger task than one would expect. I spent three 10 hour days on a minute and a half of music once.

30 hours on 1 1/2 minutes sounds like a lot to an outsider....it was worth it right?

 

[bigshot]

It was a very complex track. Believe it or not, it was for Alvin and the Chipmunks. We had a minute and a half to get across three decades of musical progress... all using the sped up Chipmunk voices. We did a medley that included the Chipmunks doing the Witch Doctor song in the style of a bunch of different musical stars. I can't remember them all, but it included Chuck Berry, Little Richard, Michael Jackson, Jimi Hendrix, Elton John, Bob Dylan, Bruce Springsteen... We had a verse to get across the arrangement and performance style of each one. No visual cue, no announcing who it was they were doing. The listener had to be able to just "know" from the sound.

The writer wrote the bit into the script, but my boss didn't believe it could be done. I convinced him to give me a couple of days in the studio to try. He had block booked the musicians and studio for the special, so he just set aside days for me to work with the musicians and engineers and see if we could pull it off. I had a huge stack of notes and cassette reference of recording techniques for each bit- slap backs, phase shifts, doubled vocals, punched in woos and yelps, certain guitar sounds, etc. We were limited by budget to just a singer, a synth guy and a guitarist. Tracking it was excruciating because when you VSO up vocals, it sucks the energy and texture of the voice right out of it. You have to double the energy and double the texture to get it to come across. By the time we got to Springsteen the poor vocalist was practically spitting blood on the spit screen. The next couple of days were spent tracking in doubles and fills and roughing in the balances. The whole mix changed with each verse- over and over and over. It was like establishing the sound of 7 or 8 different songs all in one song. We were working on a 24 track master too, so we kept having to jump tracks and juggle things around when we ran out of space. The mix was like figuring out a Chinese puzzle box, but it came out reading super clear and my boss was amazed. He did a couple of quick tweaks and approved it. It ended up being the main song of both the TV special and the LP. I don't know if it's out on CD, but I imagine it is.

 

[Glmoneydawg]

It was a very complex track. Believe it or not, it was for Alvin and the Chipmunks. We had a minute and a half to get across three decades of musical progress... all using the sped up Chipmunk voices. We did a medley that included the Chipmunks doing the Witch Doctor song in the style of a bunch of different musical stars. I can't remember them all, but it included Chuck Berry, Little Richard, Michael Jackson, Jimi Hendrix, Elton John, Bob Dylan, Bruce Springsteen... We had a verse to get across the arrangement and performance style of each one. No visual cue, no announcing who it was they were doing. The listener had to be able to just "know" from the sound.

The writer wrote the bit into the script, but my boss didn't believe it could be done. I convinced him to give me a couple of days in the studio to try. He had block booked the musicians and studio for the special, so he just set aside days for me to work with the musicians and engineers and see if we could pull it off. I had a huge stack of notes and cassette reference of recording techniques for each bit- slap backs, phase shifts, doubled vocals, punched in woos and yelps, certain guitar sounds, etc. We were limited by budget to just a singer, a synth guy and a guitarist. Tracking it was excruciating because when you VSO up vocals, it sucks the energy and texture of the voice right out of it. You have to double the energy and double the texture to get it to come across. By the time we got to Springsteen the poor vocalist was practically spitting blood on the spit screen. The next couple of days were spent tracking in doubles and fills and roughing in the balances. The whole mix changed with each verse- over and over and over. It was like establishing the sound of 7 or 8 different songs all in one song. We were working on a 24 track master too, so we kept having to jump tracks and juggle things around when we ran out of space. The mix was like figuring out a Chinese puzzle box, but it came out reading super clear and my boss was amazed. He did a couple of quick tweaks and approved it. It ended up being the main song of both the TV special and the LP. I don't know if it's out on CD, but I imagine it is.

So it was worth it.....if you got it right,it was worth it...nice job bud.

 

[bigshot]

I think we paid the vocalist for the next day too because he left the studio drenched in sweat and croaking like a frog. It was an ordeal.

 

[BobG55]

It was a very complex track. Believe it or not, it was for Alvin and the Chipmunks. We had a minute and a half to get across three decades of musical progress... all using the sped up Chipmunk voices. We did a medley that included the Chipmunks doing the Witch Doctor song in the style of a bunch of different musical stars..

I feel a bit embarrassed asking this bigshot but do you think you could get me Alvin's autograph ? I've been a big fan since I was a kid

 

[bigshot]

I was a jack of all trades. I signed autographs for the Chipmunks too! I see my autographs turn up on eBay sometimes. I'll look and see if I still have some 8x10s around here. I used to have a stack of them.

 

[johncarm]

The way the musician performs is recorded. That is ultimately what you hear. But there are other people who have input as well... the accompanying musicians, the conductor and the producer (aka "Tone Master"). An orchestral recording is a team effort and each member of the team does his part and hands it off to the next person to make it even better. In general, the essence of the performance is retained. .

So you're talking about the question of how these choices are evaluated, the criteria by which one microphone placement is evaluated vs. another, or one ADC versus another, etc.

I don't know why you insist on bringing a hoard of cohorts into this. All that is needed to begin a discussion of such criteria is a single player, a single engineer, and a single "evaluator."

You say "the way the musician performs is recorded" and "the essence of the performance is retained." You're glossing over an important point here.

I'm not sure you've paid any attention to my description of the "musical message" a few posts prior.

From my own experience being present at recordings, as well as knowing the huge range of quality in classical recordings, I know there's no guarantee that "the essence of a performance is retained."

Someone is calling the shots, and that's person's motives may not be to retain the original "musical message." Perhaps the producer is giving people what they think they want to hear. Perhaps there is a house sound to be preserved.

You mention the possibility of mic'ing a trombone different than a flute. So how should it be done? Who's deciding? what is their criteria? are they interested in preserving the essence of the composer's orchestration? I hate orchestral recordings in which the woodwinds sound bigger and more present than the brass. That really distorts the point of orchestration. Yet it's frequently done. I own many such CD's.

I know that the average classical listener would not be happy with a super-wide dynamic range, in which the flutes are a lot quieter than the brass. So maybe the producers are just giving people what they want to hear or expect, rather than deciding what makes musical sense.

I hate modern issues of old recordings which are thoroughly de-hissed. Obviously this is done because they think the modern listener is allergic to hiss, but it kills the life of the recording.

I don't mean to say I have the one true perspective. I'm pointing out that at least my perspective is based on the desire to accurately convey the musical message, and I'm not sure that's a very common perspective in modern recording.

Sheffield Lab is a major exception. They know music!

 

[gregorio]

[1] L can go into the monitor room and listen both to R1 and R2.
[2] L might be able to make a judgement then, about how each of them is alike the live sound, and how it differs.
[2a] ... If appropriate in this circumstance, L might be able to judge which one is closer to the live sound in all the important ways...

You haven't answered my question but for now I'll proceed under the assumption that you do want to know how things actually work and for the benefit of anyone else reading this thread who does.

1. Let's forget about R1 and R2 for a minute. There are both fewer and far more variables than that, the cabling and ADC are not variables for example.

2. How can L make a judgement, how does L know what the live sound is? Neither the live sound nor the performance is a constant, both are themselves highly significant variables. Let's assume T is a very good, experienced, professional trumpeter. T has practised his/her solo piece in a practise room, a rather small and probably quite acoustically dead room. However, T will NOT have practised all the factors you mentioned to create a good performance for that practise room, T will practise a performance which will (hopefully) sound good in a much larger and more acoustically live performance venue. For example, the performance T has practised will be more staccato and less legato than is optimal for a practise room and most of the factors will be altered and non-optimal for the practise room, particularly the dynamics.

2a. If L is familiar with this live sound (in the practise room), just about the least useful judgement is how closely the recording matches that live sound! If L is T's teacher though, L will be aware of all this and will not want the recording to match the live sound. L, like T, will be thinking in terms of some idealised sound for the recording, a much bigger, less dry acoustic just for starters. This leads us on to:

I'm not talking so much about recreating the "live sound" accurately, as there is no simple way to recreate and sound field -- and no need to recreate the sound field -- in order to do what I'm talking about -- presenting the musical message accurately.

Now we're getting somewhere! L is rather irrelevant now, because now we're talking about the accuracy of the musical message/intention rather than about the accuracy of actual sound/field, and NO ONE knows the musical intention better than T. However, the musical intention/message is the only thing T (or L for that matter) knows about but as you state, we're not talking just about the musical intention/message, we're talking about accurately presenting it to listeners. We are therefore now in the realm of nuances sound/audio and outside the realm of T or L's expertise/knowledge. For example, let's say we eventually arrive at a recording which both L and T are happy with, that accurately represents the musical message, we'll call this R3. Is my job as engineer done, can I just send R3 off to the pressing plant? Nope, if I did, I would be incompetent! The first thing I'm going to do is change what L and T we're happy with and create R4. In a comparison, L and T will almost certainly prefer R3 to R4 and probably be unhappy that I've even created R4 but at this point, L and T's perception is largely invalid/irrelevant because they are inexperienced and somewhat ignorant of the nuances of sound/audio! What L and T are perceiving is a representation of the "musical message" in my studio but consumers/listeners are not going to be in my studio, they are going to be presented with something quite different. As an engineer, I've (hopefully) got a good idea of the nuances of sound/audio and will create R4 with this difference in mind, to hopefully more accurately present the musical message to the consumers/listeners with their systems. At some stage before the process is finalised, I'll make a temporary master (typically called a "rough mix") of R4, give it to L and T who'll listen to it in their cars and home systems. If I were to also give them R3 for comparison, assuming I'd done my job well, they'd now prefer R4, although what they'd like the best might still be R3 in my studio but of course we can't distribute R3 + my studio! This is just one of numerous situations/decisions where the musician's perception is too limited by ignorance of the wider issues of sound/audio.

Of course, the example you've given is very uncommon, there are relatively few recordings of solo, unaccompanied trumpet. The vast majority of trumpet recordings are of a trumpet in an ensemble of some sort and then we're into the problems highlighted by the film analogy and what bigshot is talking about.

[1] Someone is calling the shots, and that's person's motives may not be to retain the original "musical message."
[1a] Perhaps the producer is giving people what they think they want to hear.
[2] You mention the possibility of mic'ing a trombone different than a flute. So how should it be done? [2a] Who's deciding? [2b] what is their criteria? [2c] are they interested in preserving the essence of the composer's orchestration?
[3] I hate orchestral recordings in which the woodwinds sound bigger and more present than the brass. That really distorts the point of orchestration. Yet it's frequently done. I own many such CD's.

1. With popular music genres that's often the case to a greater or lesser degree. With say classical music, I don't know of a single case of that occurring, except with classical/orchestral film scores.
1a. Not really. Firstly, as bigshot stated, it's a collaborative process, typically the conductor would have a very significant input/influence over what the producer wants the listeners to hear. Secondly, presenting the musical message accurately to consumers/listeners depends largely on who those listeners are, the target demographic. Is the target demographic audiophiles or is it a more general public? If it's the latter then the accuracy as perceived by audiophiles may have to be compromised in order to improve accuracy for the general public.

2. There is no one right way for how it should be done, it depends on the circumstances of each individual case.
2a. The chief recording engineer.
2b. To capture the best possible recording in a manner that allows the producer the flexibility stipulated.
2c. No, that's not the recording engineer's job. It's not even really the producers job. The musical message is defined by the conductor, and how or if the essence of the composer's orchestration is preserved is a function of the conductor's individual interpretation of the score. Of course though, on top of this is this producer's job, to produce a product which is an accurate presentation of that musical message to the target listeners.

3. The woodwinds often should sound more present than the brass, in an orchestra the woodwinds are often closer to the audience than the brass and some brass instruments, the french horns and tuba for example, are only ever heard indirectly by the audience. Generally, the woodwinds should not sound bigger/louder than the brass, they should sound smaller/quieter, although of course that depends on the situation, there are parts of some pieces where the brass is supposed to be quieter than the woodwinds. Certainly, the dynamic difference between the woodwind and brass sections is sometimes reduced on CDs, which, on a high quality reproduction system/environment, is not ideal and does distort the point of the orchestration. However, on a poorer quality reproduction system/environment, reducing the difference between woodwind and brass actually improves the orchestration! A poorer system/environment will have a significantly lower dynamic range and although a reduced difference between woodwinds and brass distorts the point of the orchestration, that's far more preferable than simply not hearing the woodwinds at all (due to being below the listeners' noise floor), which would utterly destroys any notion of the orchestration! This brings us back to who the producer's target listeners/consumers are.

G

 

[bigshot]

Sorry Johncarm. I'm just telling you about how music is produced. It isn't about capturing a performance. It's about sculpting one in the studio that is better than just capturing one.