Every DAW has its urban legends about how mixes sound different after being bounced down to a final stereo track. Talk of fixed-point versus floating point summing equations is usually introduced as evidence to prove one side of the argument or the other. When there are real differences, and mixes do sound different after being rendered to disc, it usually comes down to one of three factors.
The first and most common is pilot error of some sort, like not having master faders set at unity, unintentionally leaving some plug-ins muted/unmuted, inadvertently leaving a normalize checkbox engaged in the Bounce dialogue box, misunderstanding the various pan law options, or even listening back in an app like iTunes, with some optional EQ enhancements enabled.
The second potential cause is a plug-in that generates random variations that are not the same each time. For example, random LFO modulation in a soft synth, or sample and hold modulation shapes that vary on each play through.
The third, less common variable, is the use of outboard gear in a mixdown. If you are incorporating outboard synth hardware or effects processing, it’s possible their settings haven’t been recalled with precisely the same parameter values each time.
If you’ve ruled all these out, and are concerned your overall dynamics and stereo image of your bounced file indeed differs from your multitrack version within Logic, there is a sure fire way to test whether there is something untoward going on.
In your Logic project, route your entire mix to an available bus. Send that bus directly to the stereo output. Then create a new single track routed to a separate independent bus, which is also, in turn, routed to the stereo output. Place your bounced file on that new track. So now you have the bounce arriving at one bus, and the full arrangement arriving at another.
Put a plug-in that allows you to reverse the phase of the signal on one of the busses. The built-in Gain plug-in is ideal for this task in Logic Pro X. Playback your project with both the bounce and the full mix playing at the same time. Make sure to lower the stereo output temporarily, as the summed signals will no doubt overload the final output stage.
Now flip the phase on the plug-in. By reversing the phase, the signals should cancel each other out, and result in complete silence. If you get full silence, then you know that the two signals (your multitrack mix, and your bounced stereo file) are identical. If you get some leakage, or anything other than full silence, it indicates that there is, in fact, something different in the bounced file. If that’s the case, then it’s time to roll up your sleeves and start hunting down poltergeists.