I understand a couple of things about this debate:
1) Many people listen to MIDI files on the $0.99 cent synthesizer chip on their sound card. No wonder they dislike the MIDI sounds as the Synth they use to play the sounds is cheesy sounding (What do you expect for a chip that costs less than a dollar?)
2) Others listen on software synths, many of which are designed to put the minimum load on the computer's CPU (especially the affordable ones). They compromise the line between taxing the computer and making good sounds. The VSC that used to come with BiaB is a prime example. The original Sound Canvas, Roland SC55 sounds much better than the VSC ever dreamed of sounding -- and it's latency is somewhere in the neighborhood of 5ms (or for all practical purposes, nada).
3) Playing music for a recording session and playing music for a live performance are two entirely different skills and require two entirely different approaches and playing styles. That is why so many groups use studio musicians to cut their records (of course they don't always tell you that). But people like Carol Kaye have played on over 10,000 sessions and were the musicians in countless band and single artist recordings. And that is just the "wrecking crew" group in California.
I don't know if I posted this account before in this thread or a related thread, so please excuse me if it's redundant.
I am doing a "Friday the 13th" party in January, so I learned three songs for the gig, Stevie Wonder's "Superstition", Albert King's "Born Under A Bad Sign" and the Etta James version of The Temptations' "Shakey Ground".
We have a regular Tuesday afternoon gig and we always get there early to set up and run through new songs on the PA system (we have a smaller system in our studio, but it doesn't always sound the same}.
A couple of weeks ago, a friend who is an excellent guitarist and songwriter came early while we were going through "Shakey Ground". He wasn't familiar with the song so I played a WAV file of the Etta James version that recorded from my Vinyl LP and put on my computer (from "7 Year Itch" a masterpiece of an album). Sample of the song in question here:
https://www.youtube.com/watch?v=eEqfbITeaXcWhen it was done he remarked that
my backing track had a lot more punch than the recording. Believe me, it's not because I'm a better musician than the session guys on that album, they are monsters. It is because of a couple of things:
1) MIDI instruments have more separation and more dynamic response than a mastered/compressed recording. This is the main reason why audio loops sound so "karaoke-ish".
2) As I mentioned before, playing live is different from playing in a studio. There are things you do live that don't work on a recording and things on a recording that don't work live. Balance, groove, dynamics, etc., are all different on stage than they are when you are looking through the glass of the fish tank.
I didn't do the song with BiaB because I am so in love with the Etta James version that I copied the bass, guitar, and drum part to the best of my ability. I did a slight variation on the horn part, and left holes for Leilani and I to play live instruments on top of the track (I don't want the MIDI musicians to have all the fun).
But basically it is the clear, separation of MIDI instruments and the ability to edit a MIDI file that make my backing track sound better than the recording. If I could have found a karaoke file of that arrangement (I didn't look so I don't even know if it is available) the karaoke version would have sounded like I was playing karaoke. The instruments would have been played, mastered and blended for a studio performance and not a live performance.
Now playing at an ungodly loud volume would help the recording. That's why discos get people moving. But if you hear a live band, a MIDI band and the recording played at the same volume, the live band would have the most punch, followed by the MIDI band and the recording would have the least.
It's just the nature of the beast.
Insights and incites by Notes