Turn off reverb on plugins before mixing?

I usually turn off all reverb for a rough mix and then I introduce one reverbn on a bus into the mix later. But I don't always do it the same way and I don't think there are any hard and fast rules on it.
 
 I have been known to mix and match reverbs in a mix. After the mix stage and further into the fx part of a mix some mixers stage reverbs. They might take one that they like the early reflexions in and add another for more space.....but this takes experience or trial and error. The guys who are good at it can make it work but the guys still learning the tools usually end up with mud. I don't necessarily claim to be in the first group :) The fewer instruments involved the better if mixing/staging reverbs. They can either augment or fight one another.
I might sometimes use a reverb that came with a particular synth because it might bring out the very best in that particular sound. Most noobs tend to err on using way too much of it.Since I compose some etherial music I might use it more liberally than say...a guy writing funk.

Most virtual and hardware synths have reverb built into the voices, along with a bunch of other effects.  And it is true that too much overuse of reverb can smear things up.  On the other hand, not everyone is trying to present a "live and true" document, as if all the instruments were played live at one time in one room and need a single room reverb to complete the illusion.
 
You will come up with your own "mix' (pun intended) of techniques and tricks.  Both ways of working are valid.
 
@

There is no right answer to this.  It also depends a lot on what type of music you are producing as well.  For example if you are producing electronic music and you are using a variety of synths (hardware or software) I have actually found things can sound better when you use the built in reverbs/effects of many different instruments.
 
Earlier on I used to switch off a lot of built in reverbs/effects but I have found in many cases you want these parts to be more separate in nature.  It helps to set them apart from each other and built in effects is an excellent way to do it.  It will save CPU power too in the process.
 
All of the hardware synths I use feature built in effects.  I have learned to program them more deeply and as a result they are all adding their own quite distinct character all helping to separate the instruments much more so.  Especially once you start toning them way down too and put them back into subtle mode rather than swimming mode.  Before you right them off spend some time playing and programming the effects to the max.  Often you end up creating nice processing that may just surprise you. 
 
Many people turn the effects off these things not knowing enough about the effects themselves.  With some care and programming very effective effects can be applied within the instruments.  In many cases the built in effects are very integral to the patch. (eg the chorus in a Roland Juno 106. You won’t get that effect anywhere, period) Take them off and they are gone.
 
But this may apply to electronic music for example.  It may not work so well in an acoustic or rock track though. Then it may be better to switch a few of them off and put more than one part into a suitable effect or space as per the OP.  eg to create the effect of several things being in the same space.  It may work for some things but it can hinder other situations.  You need to evaluate. 
 
Some synths have killer reverbs and effects, way better than a lot of DAW effects too.  And lots of em. There are 16 effects processors in my Kurzweil PC3K and 56 in the Novation Supernova II. Yes folks 56!  And they sound bloody fabulous too.  You would be foolish to switch these off!
 
 
 

In a sparse mix I would be more inclined to turn the built in FX off and use an external reverb. A sparse mix is when the reverbs are going to be more exposed so you will more likely to hear them.  In a dense mix once you edit the effects (see below) then you need external reverbs less so.  You will barely hear the reverbs then and the fact each synth has its own reverb character is going to be a plus in the long run. 
 
Also using delays is a good idea as well.
 
Don't forget that with many built in synth reverbs all you have to do is edit the effect.  Make the room smaller and alter the wet/dry mix so there is a lot less of it and restore the patch somewhere so the new effect mix comes up everytime.  Still sounds like people are too lazy to edit the built in synth effects.  That is the problem here.

Jeff Evans
There is no right answer to this. 

+1... great way to start your post Jeff.
 
Do what sounds good, and what sounds right for the song... it doesn't matter how you get there. I generally strip all reverb, but I'll leave any other effect in place - providing it fits. During mix I'll use 2 - 3 reverbs on sends and that helps to keep my "space" sounding pretty even. I should state that I am after natural sounding mixes - which is the biggest reason I do it that way. It's certainly not the only way to get things done.
 
 
 
 

Thanks to all for the input. You've got me thinking about aspects that I haven't considered. Normally I'm going for a natural sounding mix but my prog rock influences have me throwing in curve balls occasionally.
 
Adam, Mark (sorry), thanks for taking the time to listen to some of my songs. Most of those are pretty old and I don't have the raw tracks, so I can't go back and try a remix. Plus, I'd rather look forward. (BTW, feeding the cats is also my job!)
 
Right now I'm trying to finish up a brand new song that's a much simpler arrangement- (Cakewalk-SI) drums, rhythm guitar, (synth) flute, and lead vocal. In keeping with my usual "can't leave well enough alone" habit, I also recorded a bunch of harmony and background vocals for the last verse, chorus, and fadeout. I'm debating whether I want to keep them or not. The song has a nice casual feel throughout without them, but when they come in I can't decide whether it sounds really cool or really dated (think the band America, who I loved back in the 70's.) I think I'll post both versions and "let the people decide".
 
Anyway, I think I'll try freezing those synths dry and using a common reverb for all of the elements to try and get a live recording feel.
 
A question: I'm also just now exploring the virtues of using busses, so correct me if this is wrong: The way to do this would be to have a bus just for the "room" verb (with the mix set to 100% wet) and then route the other busses to the room bus using the aux sends, post fader? That way I can control the overall amount of reverb with that one fader independent of everything else?
 
 
Doug