Discussion in 'Mixing, Post-Production, and Effects' started by musicalweather, Jun 12, 2018.
^^ This. So true, for some reason.
I like to bounce in realtime for the same reason @charlieclouser explained. But on big projects most of the time it is offline, cause my CPU is in too much pain and has trouble to play everything without cracks and pops
I've always thought some quality compromise/trade-off might be happening as off-line renders can take longer than real-time, but I'm happy to go with what I hear when I'm writing.
that's the idea
Oh really? That's interesting. Do you think there's a noticeable difference? Say, you would bounce the same session online and offline?
I always bounce in real time - call it paranoia, call it I learned to monitor the "product" from an early age, and it is just a habit.
I have heard that real-time vs off-line can be different, but I've yet to demonstrate it to myself. Sonar offers realtime, offline, and even "fast bounce". I've taken the same project (obviously without external effects) and used all three methods. While the checksum differs, a "null test" has yet to show any significant differences. And by significant, I mean the differences are typically 100 dB down (or more). That is, for all intents and purposes, inaudible.
I have never had two mixes cancel completely, but in fact I'd expect some differences, a synth starts a waveform slightly differently, which affects how any processors process that signal, and so on.
If the differences were significant (i.e. audible) then I'd have to listen to both to decide which one I liked better<G>!
That's really interesting. While I don't watch the clock, I've always thought offline bounces happened in less time. Now I guess I need to check!
I record my mixes in real time to do a last listen, then master and bounce out the file. If I have time I'll usually listen to the final audio file output to double for errors, and to gloat and bask in the glory of my composing brilliance.
it is not so much that you save or loose a lot of time. it is about when the calculating power peaks. the peaks are usually short (that's why you don't hear a drastic difference). i guess if you'd look at the transients of the waveforms you would see a difference though.
bouncing offline can take longer if you use instruments or effects which allow you to select a better quality when rendering offline. synths do that, some reverbs and other effects as well.
The new version does but I have an old v7 TDM rig due to full length PCIe cards and hardware. Sounds great however and use as audio playback, stemming and mastering. All my midi rendering is done in Reaper unless I need to master, then I use PRO Tools HD2 and render in real time.
I always bounce offline and then listen back to check the results, for the reason babylonwaves stated, and particularly because a dozen years ago I worked on developing both a VST host and plugins for an audio company and came to understand firsthand how much easier it is for a plugin to handle an offline render.
Except that the printed file can always have issues. To QC, you should always listen to all final files before they go out. I always do offline because I don't want to have to sit there listening and then sit there listening yet again. When I'm creating masters in different formats, I'll normally just listen to the most processed formats because any issues in the other files will carry through to those.
The other thing to keep in mind is that if you do it offline you don't have to worry about any clicks or pops that you might get from any instruments. People who have systems that sometimes have clicks and pops (like with a busy cue) will have to sit there all day trying to get it to print cleanly.
I've heard that Pro Tools used to be horrible for printing anything so I know several people who still use a digital out to a hard disk recorded just to print their mixes. I think nowadays it's not an issue.
Always in real-time. If I don't, there occasionally artifacts (pops, clicks) that are a direct result off offline rendering.
Almost always in realtime, expect for game music.
I render mine in fake time. My whole life is rendered in fake time.
I also do offline because of speed and that most of the time it works just fine, but I've never thought of using the real-time as a last listen. I think I may try this in the future.
Interesting how people have experienced issues with both realtime and offline bounces. One school of thought is to listen during a real time bounce – but it doesn’t seem likely you are hearing the actual resulting file. Either way, I’m in the camp of those who listen to every resulting mix for QC issues. I have also experienced many plug-ins that have errors in offline bounces – some Play libraries still have problems with this for me, as do some Kontakt-based. So for anything that really counts, I tend to do realtime bounces...
I prefer to listen to the final render in a run-of-the-mill media player rather than the DAW because I find that the change of scenery helps give me a bit of mental distance from the mixing process which changes the way I listen. I don't know if it actually improves my error-checking abilities, but it sure feels like it does.
Totally depends. If I'm using outboard and hardware synths it's realtime, if not it's offline. Never noticed a difference really. Final files need to be listened to anyway.
Separate names with a comma.