Soft synth resource usage varies quite a lot. With my primary synth instrument (Zebra), one patch can impose a very light load while another patch might bring the CPU to its knees. The same is true - to varying degrees - of other soft synths I've used. Some, such as Z3ta and TTS-1, are relatively consistent (although the TTS-1's CPU usage is fairly high). It's entirely possible that 19 soft synths could either be too much for your CPU, or no problem at all.
 
At any rate, freezing a synth not only removes the CPU overhead, it also reclaims whatever RAM the synth had been using. With all synths frozen, you now have an all-audio project capable of literally hundreds of tracks, even on a very modest system. The downside is you have to un-freeze in order to make edits to the underlying MIDI data or synth configuration.
 
Note that 27% CPU usage is low, nowhere near the danger zone. I have projects that run at 70-80% CPU with no problems. If you're experiencing crashes (real crashes, e.g. GPFs) then something else is awry. It may be that one of those synths is failing internally, which can be an especially common occurrence with many free synths. You might find that when one particular synth is frozen your problems go away, in which case you might consider retiring that one from future use.

I didn't read every post in detail to this point.
 
Here's my take.  They are similar functions. Freezing and Bouncing. How you choose to use them depends on how you work and your computer's CPU power.
 
I will freeze a midi track when I need to have the CPU power doing something else and I still may wish to edit that track before the project is done. If I need to run a CPU heavy plug in or synth with samples, and I think all the midi tracks existing will suck the CPU power to an unacceptably low level, I will freeze the synths in the project at that point before proceeding. Freezing converts the existing synth's output to a wave, prints it to the same track, and suspends the synth without changing it's settings or patch/sample. I can unfreeze at will, and I'm back in the game with that synth.
 
Bouncing, to me is saying that I'm pretty sure I'm done with the track and the synth. I can bounce it which is like freezing except that a totally new track is created to contain the output. I will often archive the synth track and hide it at that point. I'm left with the audio wave output to work with. Bouncing is also handy for use with audio tracks where you may have a vocal track (for example) with reverb, EQ, and compression in the FX bin. All those plugs are using CPU power. Bounce the track to print the FX to audio and you can once again archive the original track taking it and it's FX out of the processing power loop.
 
My newer DAW has plenty of power, and I've never had to worry about this issue with it, but prior to my building it, I worked on a stock Dell laptop and CPU power was a critical issue for me at that time. Using the methods listed  above allowed me to work on big projects, and use the CPU heavy plugs needed in the projects.

thanks for the advice guys
 
i freezed about 7 or so synths and now i hover around 17%-22% which is great
sad to read that bitflipper is suggesting something else is wrong but there's not much i can do about it.
thankfully it works well enough most of the time.
 
question about the frozen tracks:
-if a track has a volume envelope in it.....i'll need to adjust that envelope if i want that track to have more or less volume yes?  last night i grouped every track so that I could move ALL of my sliders up by the same amount.  i'm pretty sure that the frozen tracks with volume envelopes did not change.  i then went into the frozen track and moved the node up a bit.  does this sound about right?   am i losing this change if i were to unfreeze the track?
i guess i just had the thought that once a track is converted to audio the volume envelope changes are written in stone like something in the FX bin would be.  thinking that by moving all of the sliders up by 15%....that this envelope would be raised by the same.  i do not at this point believe that is the case though?
 
-what happens to a frozen track if i insert some blank measures in the middle of the project?
for these tracks with the volume envelopes.....they are now "broken" instead of that continuous WAV/audio file that spans the entire tune.  do i need to unfreeze everything....insert the measures....then refreeze again?
 
 
amazing weekend for my understanding
cheers to you all!

Yeah I think you're right. Freeze the track means that everything in it is frozen including the envelopes.
 
The solution is to bounce the track, archive the source tracks, and then add an envelope to the new bounced track.....or unfreeze, edit the envelope and re-freeze.
 
I used to group my faders, but have not done that in a very very long time. I'm thinking too that freezing a track that has it's fader in a group, may cause that particular track to not respond to the grouping.... I don't recall exactly, but I remember there being some sort of issue when I had done that in the past.
 
I now drop volume envelopes into every single track in my project. Even the bass and drums get envelopes. I may not use them for anything or I may use them simply to delay starting the parts until a certain point in the song. Generally I have between 10 to 16 tracks in a project. Some more, some less.
 
It takes a few minutes to set up the envelopes for the first time, but then, I have the ability to tweeze the volume of each track.
 
Rather than grouping the track faders, which makes them all move together..... I will sometimes assign them to a new stereo buss so that I can use the BUSS FADER to raise or lower them as a group as needed BUT.... I still have the ability to change them in relation to each other in the mix if I need to, for example, bring up a piano or drop the acoustics in a section.
 
the yellow lines are the volume envelopes in the pic below
 

 
hope this helps

definitely helps!
 
man this is all so inspiring.....every time i get a chance to devote 3 or so hours to this I end up having a breakthrough;  either in my understanding or in learning something new.....or even musically with what i'm currently constructing.
last night i bounced a snippet of a track, reversed it....and instantly discovered what will be the bridge melody between the first and 2nd half of the tune :)
 
 
fyi moving all of the faders was an effort to just get them all up around the middle or upper half of their slots on the mixer board
 
i have two tracks that for whatever reason need to be MUCH louder than the rest and so for quite some time all of my tracks are VERY low and i have to turn the volume up on my speakers just to hear my tune.
i went into the groovesynth to make sure the level was all the way up and i also maxed the velocity and that enabled me to drop this slider and raise the sliders on everything else a bit but there still is quite a gap between these two tracks and the rest of the project.
 
i'm not sure if there is a problem per se with most of my sliders being so low....i suppose i could boost the whole track when i'm at the mixing/mastering stage?
or perhaps if the ideal is to have all sliders hovering around 1/2 or 3/4 then i'll just need to find a way to boost these two tracks or perhaps limit the rest of the tune.
i'm about 90% finished my garrison book.....which means i haven't reached the mixing/mastering chapter yet so i'm just brainstorming out loud :)
 
 
thanks guitarhacker hope yall are thawing out down there!
 
 
 

Glad you're having fun, seed! That's what it's all about. Fun and creativity. As the mysteries of the DAW gradually dissipate, the technology becomes more and more transparent until you eventually pretty much stop thinking about it and can concentrate entirely on the music.
 
Level management, however, is one area that requires ongoing vigilance. Get your levels too high and you get into trouble, too low and you might end up with excess noise. The good news is the usable range is HUGE, and all you normally have to worry about is getting too close to the forbidden 0dB mark.
 
The quick 'n dirty solution for your two quiet tracks is a destructive gain adjustment. Don't worry about the ugly word "destructive". It just means that once you've saved and closed your project you won't be able to un-do it. Go to Process -> Gain and add however much you need to bring those tracks up to the general level of the others.

yeah knowledge surely is power!
 
the "forbidden 0db mark".....that's the slider raised all the way up in the red correct?  and will cause clipping?
if so yeah i've managed to keep everything away from there thus far.
thanks for the advice on the gain adjust - i'm hoping it will be just that simple
 
 
referencing your first paragraph again.......how to you guys sort the bazillion synths and samples we have at our fingertips?
i spent a few DAYS just going through pads and leads for this track i'm making.  i can't imagine doing that process every time so i'm thinking i'll have to devise some way of taking efficient notes so i can find the sounds i want quicker.  i know this is a common issue for people who use this type of software and that some of it just can't be avoided.  i have to believe there are ways to simplify this process tho!
 
 

bump about the 0db mark
 
 
weird that i read plenty of sites saying i should aim for this but then today i'm seeing where it says CD players can have clipping if too much or certain sounds are hitting 0db?
 
so is it ok to sometimes go OVER 0db or is the rule to simply stay under?

If you are worried about that "destructive" thing Bitflipper (Dave) spoke about,,,,, simply clone or copy the track first. Then mute the clone and work on the original. (or vice versa)
 
Besides, the destructive "can't ever go back" only exists after you save and close the project. Until that time, you have EDIT>UNDO to bail you out of trouble.
 
0db? .... pffft... rules are for sissies 
 
Yeah, I try not to violate that by keeping it all out of the red.... but the occasional red flash isn't going to kill you.... you certainly don't want to run solid red.
 
On the tracks that you say are very low in volume... reference the pic I posted above.... are the track waves nice and full and fat looking like the wave in track 14 or are they looking more like a thin weak looking straight line for the most part? You may not have sufficient gain on the signal going in to provide you with a nice full wave.
 
Midi, for the most part doesn't have that sort of level issue.... at least not for me. Most times they are nice and full when converted. 
 
Yeah, have a look at the wave form in the track. It should be taking up a large area top to bottom ... 75% or so..... up to just shy of 100%.. you don't want "overs".  When I first started I was having 10% waves and that's a matter of getting the inputs set properly to you get better levels into the inputs. In my experience,  (see track 11 in my pic) even a thinner wave will sound good. Although it will not be as loud, there are things that can be done should you find that a given track is not loud enough. In most cases, I will bring the other tracks down rather than try to pump a weaker track up to their level. Not always, but certainly most of the time.  As long as you're not getting close to the levels where the noise becomes audible.... it's all good.
 
I normally end up with a weaker looking track when I track my acoustic instruments.....acoustic guitar and mandolin. But again, the wave picture is only part of the story..... you have to listen to it in the mix before you start trying to pump it up or apply gain to it. Some of my song tracks with mandolin in them have a pathetic looking (according to me) wave, but they sound just fine.
 
If you have a weak looking and low volume sounding wave..... then copy it or close it, and mute one copy like above with  my first paragraph...then for grins.... use the process audio> Gain> +3db on that track.... this increases the gain/volume by 3 db every time you apply it. You may need to apply it several times to get it up to where it needs to be.... you can also reduce the same way if something is too loud.   OR.... go for broke and simply apply the NORMALIZE function. This takes you right to the top with that one click. I would set the level at 95% or so... this leaves some head room in the track and going to 95% is generally more than enough. When I use that I then have to pull my faders down about half way which is fine. Realize that the down side to either of these volume increasing methods is that the noise in the track is also increased by the same amount as the music. Use these only as a last resort. The goal is to get a nice full wave that doesn't need any help with the volume. Not always possible but that is the goal...at least around here in this studio.
 
Glad to hear that you are having fun.....