Balancing sound volume in combi

[dhjdhj]

I created my first real combi (i.e, one I will actually use live) with piano, strings and choir, the idea to be able to fade in and out the latter two as needed.

I found however that the strings and choir are as loud as the piano when the piano volume is on full and the strings and/or choir are barely on.

So it's impossible to get any decent dynamic range out of the piano, there's no headroom left.

Is there some other internal setting that's causing the final output of the strings and choir to be way higher than the piano (or vice versa)?

D

[steve m]

you could go the the original programs you used to creat your Combi and edit the oscillator section to give you more gain. Alternatively you could insert a Compressor into the timbres you want to adjust and tweak the make-up gain in the effect page.

[mwiik]

One great trick is to use IFX, just as described above.

I often use Master Limiter. Just lower the treshold to about -3 -4. That will do it for any sound that is to low in a combi.

Regards
MW

[tcornishmn]

The other way to do this is to go to the timbre parameter page and adjust the EQ trim. This is my go to option because I don't have to modify the original programs.

[dhjdhj]

This worked quite well for me. Thanks very much for the suggestion.
D

One great trick is to use IFX, just as described above.

I often use Master Limiter. Just lower the treshold to about -3 -4. That will do it for any sound that is to low in a combi.

Regards
MW

[Ray]

I created my first real combi (i.e, one I will actually use live) with piano, strings and choir, the idea to be able to fade in and out the latter two as needed.

I found however that the strings and choir are as loud as the piano when the piano volume is on full and the strings and/or choir are barely on.

D

I use over 50 combis live and don't have any problems with volume differences after I have initially set up the combi volume settings. If anything, I find the individual piano programs to be louder than the string programs.

Which program sounds are you using that have such a difference in volume?

Ray.

[EJ2]

I guess it's a matter of programmers' preferrence what should come first in establishing your "mix". With a potential 16 programs to slot in a combi, I tend to work from the "Program Select/Mix" tab and the Timbre/Track button followed by using the sliders to make my initial adjustments for each program/timbre. That comes first. I then follow up with setting Timbre EQ, IFX, etc.

[dhjdhj]

Interesting - it may be that I'm missing some steps. I simply choose some programs (Deep Piano, Legato Strings, and one of the basic choir patches) and threw them into a new combo. I did not configure any "combi volume settings" and was simply using the physical sliders to control the volumes.

Seems to me it would be really nice if there was a function you could apply to a combo that would do the equivalent of "normalizing" the volumes so that all programs on the combi have the same volume ranges.

I use over 50 combis live and don't have any problems with volume differences after I have initially set up the combi volume settings. If anything, I find the individual piano programs to be louder than the string programs.

Which program sounds are you using that have such a difference in volume?

Ray.

[Ray]

Interesting - it may be that I'm missing some steps. I simply choose some programs (Deep Piano, Legato Strings, and one of the basic choir patches) and threw them into a new combo. I did not configure any "combi volume settings" and was simply using the physical sliders to control the volumes.

Seems to me it would be really nice if there was a function you could apply to a combo that would do the equivalent of "normalizing" the volumes so that all programs on the combi have the same volume ranges.

I mean setting up the volumes happens automatically when you have altered them using the physical sliders and saved to HDD. You are obviously doing that anyway.

As I say, I don't have a problem with volumes except for some of my own samples which I now add +5 to +10 db to make sure they are loud enough. So don't understand why you have the problem.

Happy to do a phone call to discuss/compare combi settings if you are in the UK.

Ray.

[EJ2]

Yes, there is a lot more to creating a decent "COMBI" (short for "combination") than just throwing a few timbres together. To be sure, with the host of programs available and their unique characteristics - from in-your-face leads to mellow motion synths - you are going to need to get in there and adjust sounds to your "tastes". This latter notion will be different for each user and be highly dependent on the purpose of each sound. So, in my opinion a "Normalize" function is not going to be an effective tool. I suspect intensive programmers would in most cases over rule any normalize function anyway.

There are a variety of ways to adjust/set your mix in combi mode. Right from the "Program Select/Mix" tab, you can simply touch the volume level of any particular timbre, then use the "Value" slider, or the 'INC/DEC" button, or the dial wheel to set your mix. It's not that difficult.

[Ray]

I've just tried a combi with Bosdorf Piano, Legato Strings and Classic Choir programs.

Strings/choir are indeed louder than the piano. With piano at 127, strings and choir needed to be at 70 for a decent mix. But OK to use I think.

Ray.

[dhjdhj]

There shouldn't have to be!

Yes, there is a lot more to creating a decent "COMBI" (short for "combination") than just throwing a few timbres together.

I'm not talking about adjusting sounds for taste - I'm talking about basic initialization. I am arguing that starting from a point where all programs have the same volume range is simply more convenient initialization and often useful without any more effort or expertise required.

To be sure, with the host of programs available and their unique characteristics - from in-your-face leads to mellow motion synths - you are going to need to get in there and adjust sounds to your "tastes". This latter notion will be different for each user and be highly dependent on the purpose of each sound. So, in my opinion a "Normalize" function is not going to be an effective tool.

Is your opinion that the typical Oasys musician must be an intensive programmer?

I suspect intensive programmers would in most cases over rule any normalize function anyway.

I understand - I'm proposing another way!
[quote="EJ2]
There are a variety of ways to adjust/set your mix in combi mode. Right from the "Program Select/Mix" tab, you can simply touch the volume level of any particular timbre, then use the "Value" slider, or the 'INC/DEC" button, or the dial wheel to set your mix. It's not that difficult.[/quote]

[EJ2]

There shouldn't have to be!

I might tend to agree with you. And for most synth/workstations, one could argue that one shouldn't need to be an intensive programmer to develop one's own sound (combination, performance, multi - whatever it's called). However, the problem for creating your own combination for the old red Karma, OASYS, or M3, is the myriad parameters involved in a fully KARMA-fied combi. It's far more complicated than working with a Triton or a Fantom or Motif for that matter, because KARMA Technology and the Real Time Controls take the user into a labyrinth of exponentially endless possibilities.

I've said this time and again. With devices as deep as the Karma, M3 and OASYS, you have to commit to a steep learning process. By analogy, it's near impossible to find a fortune in gold or diamonds just lying there waiting to be picked up. You have to dig for it in an intensive mining operation.

In less than an hour, I have whipped up a performance/multi on various synths - Fantom S (now sold), V-synth XT, JD 800, JD 990, Access Virus (also gone), Yamaha CS6X (sold a long time ago), Waldorfs, General Musics, Ensoniqs, etc. etc. all long gone. Now, when I started learning to develop my own combis for the Karma, I made tons of mistakes. As with you, I was totally perplexed. BUT, unlike most users, I didn't quit. I stuck with it. I worked with the manuals, the FAQs, the tutorials right at the keyboard, hour upon hour, day in day out. In fact, I am still learning. However, that kind of dedication paid dividends - carry over to the OASYS and M3, not to mention monetary rewards for my combis.

As I said, I can churn out a decent sound on my other synths in a few hours. Building an OASYS combi, on the other hand, takes me weeks or months (depending on the complexity I'm after), and I've spent years of intense learning to program/develop fully KARMA-fied combis.

Just the same, I can appreciate where you are coming from. The V-synth XT has a neat function providing "designer" templates - jumping points to develop new patches. But, again, there's not much to compare vis a vis a Karma, M3, or OASYS combi.

My best advice is to stick with it. Don't give up. Eventually, you will make a series of quantum leaps in your programming. I know it's a pain in the ass. But, if you persist, you will get to the point where you will really enjoy your own creations.

Cheers,
Eric

[ski]

Hey Guys,

There are lots of reasons why it's simply not possible or even practical to set all programs to the "same level" so that, in Combi mode, all programs are "equally balanced". Actually, in some respects a LOT of effort is made to achieve this (explained below) but... well, read on...

Any kind of "absolute level" is purely an electrical/electronic attribute of an audio signal and one which often has little to do with the way people perceive a sound's loudness. This is an important and very real concept to understand:

A full string ensemble program and an oboe program, to name two, are programmed such that the perceived loudness is relatively the same as you're scrolling through programs on the program level: intuitively you wouldn't think that a string section and an oboe would carry the same "weight" in most arrangements, but preparing sounds for a synthesizer is a different ballgame entirely. Getting the levels of disparate sounds relatively close to one another is done intentionally so that you're not having to constantly ride the volume control up and down as you're auditioning patches. Think about it... say you play that string patch and it was soft. You crank the volume, right? Then you scroll to the oboe wave and it's screaming f-ing loud and blows your head off. Not a desirable situation.

The same goes for Combi's.

Now you may indeed find some exceptions here and there where the level from one program to another (or one combi to another) aren't quite "equal", but the overall game plan is to get the perceived levels relatively close.

Hopefully I've made it clear that the idea of "initializing" or "normalizing" a level may have little bearing on the desired result --- "even" levels --- because of the idea of "perceived loudness", something which simply isn't equal from any one sound to any other sound.

Now, to building Combi's... Say someone wants to layer piano and a pad. How loud should each one be before they're combined? Maybe that person only wants a hint of the pad, to fatten up the sound a little bit or provide some kind of whispy ethereal sustain. It goes without saying, then, that there's no way to predict this in advance for a multitude of users. So IMO, the idea of making programs (on the program level) relatively equal in perceived loudness makes the most sense.

Anyway, adjusting the relative balance between Timbres is what the faders (level controls) are for in the Combi. If all you're going for is a layer of piano, strings, and choir, you assign a Timbre to each one of those and then adjust the balance accordingly. I have to ask, how difficult is that to do? There's no learning curve involved -- just combine and balance. Oh yes, and save!

Finally, dhjdhj, in terms of your own samples needing to be boosted +5 or +10, that sounds to me like it's potentially due to a problem with your sampling technique. Are you sure you aren't recording samples at levels that are just too low? Or perhaps their timbre is such that they can't "compete" against other sounds in terms of perceived loudness?

One more thought... there's the idea of using EQ to make certain sounds "pop" more than others in a Combi... For example, if you're combining a pad, a choir, and strings, you're talking about three "fuzzy" sounds. There's no way that the character of each of those will come across as distinct without some degree of EQ'ing involved. Many times, sonic separation can be achieved simply by using the 3-band EQ on each Timbre and cutting/boosting frequency bands accordingly. But if that EQ isn't enough, any Timbre can be processed through a Parametric or Graphic EQ IFX.

[dhjdhj]

By levels, I did indeed mean perceived levels.....and I wanted the volume sliders ranges to start so that lowest and highest would match (within reason) the perceived levels.

And I haven't done any sampling --- all I was doing was creating a combi using three existing programs.

Hey Guys,


Finally, dhjdhj, in terms of your own samples needing to be boosted +5 or +10, that sounds to me like it's potentially due to a problem with your sampling technique. Are you sure you aren't recording samples at levels that are just too low? Or perhaps their timbre is such that they can't "compete" against other sounds in terms of perceived loudness?