When is a plugin not plugged in? And are DAW's more temperamental than we think?
Two recent experiences were eye (or ear) opening to say the least. The first occured when I decided to re-order how the tracks appear in the DAW, merely to make things look more organized. I had already exported a wav file with the disorganized appearance, ordered the track in more logical groupings, and re-exported a wav file. Now I would swear they sounded different, very subtle, but different.
I am willing to leave some small window of doubt about my observations of the above, but on the second part I am about to describe, I am positive.
I had had a certain mix, and liked it. A few days later, during some experimentation, I made several changes to it, and inadvertently saved it instead of saving as something else, meaning that I changed a mix I did not want to change.
I spent the next few days reverting the settings using notes, other versions of minor difference, and memory. Yet even after restoring everything I could identify, the mix was different, not bad, but still slightly different.
But just now, I remembered one change I forgot to restore. In the first export, I had two plugins installed, a simple limiter and a mild high frequency boost plugin across the two bus, but I exported the file with the bypass switches on, meaning the effects should not be printed to the export file.
On the version I was hoping to restore exactly, I did not have these unused plugins installed. I thought it could not make any difference, but after laboring endlessly to get it exact, I began to wonder if they could indeed have an effect, even if they are bypassed and thus not functioning.
I just installed the two plugins, and bypassed them, and viola, I had the mix restored identically. It seems having plugins installed even if they are bypassed (and thus not functioning) will have a sonic effect. Of this situation, I am certain.
Is this a known phenomenon? Any other thoughts? It is slightly distressing to think that a non-functioning plugin could have an effect, as it adds a whole new order to the already infinite combination of possible settings.
Comments
That's pretty funny because I frequently change the names of fil
That's pretty funny because I frequently change the names of files and I don't have that problem. But then I don't use Cuebase. Heck I don't even know how to spell it.. For instance, I just finished a 3 camera +16 tracks of audio, 4 school cluster choir Christmas show. There was a CD utilized & played back on stage as the kids sang with. Unfortunately, that CD recording was not part of my multitrack capture. No problem, there was also my secondary laptop to capture a stereo PA board mix. But that was inadequate. So I had to ask the choral director for those music CDs to lay back into the production. These CDs were basically like karaoke discs. First time through with vocals as an example, second time through as just a music track. So while I ripped these CDs I also wanted to see what the original included vocal reference track sounded like. Unfortunately it was already mixed with the music. So I labeled it "mixed track". I wasn't certain if the music only track along with the mixed track was produced on a computer? That's easy to find out. So I took the music only track which was the second cut on the CD and combined it with the vocal & music mixed track (labeled differently). After synchronizing them both down to 1/44,100 of a second, I was able to invert phase on the music only track combining it with the vocals with the music and completely canceled the music 100% leaving only a vocal track. Would you like those tracks? Would you like to try that yourself? Both files were labeled differently and both canceled 100% when synchronized by eye, with no perceptible music left behind. So that proved to me that that particular piece of music was produced a computer. The other cuts from the other CD were obviously produced from an analog multitrack machine as this process was not possible on the second CD. This is nothing new to me. You see even if we had used analog machines with SMPTE synchronizers you still couldn't cancel things out because accuracy was only within 30 ms or 1/30 of a second which is a huge difference from 44.1 kHz of a difference. So even if there are residuals it's out of the range that makes any difference and you still can't hear it. I'd be happy to forward those files to you and you can try it in Cuebase and maybe we have the same problem? This is how we would start troubleshooting. Send me your old, your tired, your poor screwed up tracks and I'll fixem' ferya'. Because I just know it's not the software. You might be experiencing something else? I really think perhaps you are. I'm not new at this digital audio stuff and I know what I know. I mean if you inserted a plug-in and bypassed it, it's inserted, it's running, it's quiescent. But digital audio inverted digitally will cancel 100%. Don't forget it's math that works even if I'm not good at math they're just things that compute.
I would really love to see your face when you figure out what you did wrong. Because I know at that point you will be delighted with yourself. As right you should be.
Keep on trackin'!
Mx. Remy Ann David
Hey Remy, how can you be so certain that I cannot export two fil
Hey Remy, how can you be so certain that I cannot export two files sequentially and identically except for one letter change to the name? I am also curious to know what you think is the reason only the drums will not sum to zero. One would think a mistake on my part during export, like inadvertently making some kind of change -say due to a muscle spasm that hurled my arm onto the keyboard without me noticing- would result in a difference residue that contains all mixdown tracks, not just drums. And the situations you describe about your use of phase inversion are more complex indeed, and thus different than what I am doing with a most simple export of identical mixes to a mastering window.
I attribute it to an internal computer or software glitch, because of course there is no logical reason why two exports that differ in name only should produce such an acoustic anomaly. Given the complexity of today's systems, it's a miracle the stuff works as often as it does. ;-) Or are you suggesting the stuff is infallible? :smile:
But then again, it would be entirely unreasonable for me to expe
But then again, it would be entirely unreasonable for me to expect you to consider any possibility other than operator error. It is a natural tendency, and one even I would be inclined towards, if I were on the other side. All of your comments and insight remain appreciated. :smile:
Infallible? It's far more fallible than we are with pencil and p
Infallible? It's far more fallible than we are with pencil and paper. Have you ever noticed those little icons in the lower right-hand corner of your Windows desktop? I have mine set up to show what programs launch at startup. One would think that the order of those icons would remain in the same sequence from turn on to shutdown to turn on. But they don't. So what is logical or consistent about this inconsistent mathematic anomaly? It tells me our computers (operating systems programs and such) are less stable and consistent than they really should be. We're not quite there yet. Remember early car tires weren't pneumatic. Can you imagine what that was like? Thanks to Michelin that all changed. I still can't really figure out why someone needs to shut down a computer before one switches it off? I mean off is off. But it's probably so the operating system can try and remember where it last went to the bathroom? I know that there are at least millions of lines of code to make this all work so, it's a screwed up as our government systems are in the way they operate. Status quo I guess?
No, there is no logical reason why two exports that differ in name only should produce any acoustic anomaly. That just plain doesn't make sense. So it's probably Windows? XP or 7? Vista? But since you've indicated that since my phasing " tricks" are complicated, this is where your problem is. What I'm doing is so basically simple it's almost criminal. That's why I know there is something else at play here. You're not doing this on purpose I know. It's doing it on purpose to you because it's watching you. It knows what you're trying to do. It's only purpose in life is to thwart your efforts. Resistance is futile. You will be assimilated. I'll tell you brother, I'm there too. I really got tired doing a lot of component level circuit Board repairs but battling with the operating systems & software is definitely taking its toll upon me. It's never the same way twice. But my audio is. Right now you are inverted. Poof! And when you wake up, you can start drinking again.
I don't have to worry about drinking when I get up as I haven't gone to bed yet so I don't need to stop. Iced tea! No not Long Island.
Mx. Remy Ann David
jmm22, post: 360693 wrote: I just finished the experiment. I exp
jmm22, post: 360693 wrote: I just finished the experiment. I exported the two mixes, one with stereo bus plugs bypassed, and the other with the plugs deleted. I then imported the two to a mastering window, and phase reversed one track. This resulted in a very audible drum track, to the exclusion of everything else. The fact that all other instruments vanished indicates the tracks were synchronized and that the phase reversal worked. I then reverted the phase on that track, and changed the phase on the other track, with identical results.
The residue drum track was mostly high frequency content, yet with some bass. If I were to liken it to an equalizer, it would be something like an extreme smiley face.
It's not logical that anything on the 2-bus would selectively affect drums and leave the rest unaltered.
How true. That is why I believe it is an internal error of some
How true. That is why I believe it is an internal error of some kind. If I had made a mistake in the mixdown exports (a reasonable hypothesis) the phase cancellation residue would consist of all instruments, not just drums. I even tried time shifting one of the exported tracks, to see if I could cancel the drums manually, but this proved unsuccessful.
Even with all of the various experiment iterations (no less than four trials) I have the first one as the most compelling evidence for me, the one that prompted the thread, namely that I could not render the accidentally changed mix indistinguishable until I added the two bypassed plugs. I now realize it would be futile to try to convince others that this was the case, and I gladly accept that others will assume it is my error. But these files are not going anywhere, and perhaps a circumstance will arise in the future where I can demonstrate this phenomenon more tangibly, to someone more experienced, who can confirm the said anomaly, or more precisely indicate my mistake, if there is one.
But just when I thought I would have to learn to live with the s
But just when I thought I would have to learn to live with the social stigma of neurological newbieness, I found my second OEM copy of Cubase LE4 that I said I would send to John. So perhaps he's willing to act as an objective experimenter in sorting this anomaly out.
But what I want to know first is whether Remy will be willing to put in writing that I am not "inept" and "clueless" if John can replicate the anomaly.:tongue: Or maybe she wants to up the ante! LOL
But let there be no mistake about my light hearted teasing, at m
But let there be no mistake about my light hearted teasing, at my expense no less. Remy, you do not have to make any concession regardless of outcome. John, pm me your address, and I'll send the program in short order. Hopefully you might find some time to try the experiment. I should also repeat it here with Cubase VSTi drums instead, to rule out Addictive Drums as the culprit, because you will not have AD, to exactly replicate the experiment. :smile:
I was going to conduct a whole series of experiments deleting on
I was going to conduct a whole series of experiments deleting one plug at a time, but on a hunch, I attempted something much more fundamental, and seemingly foolproof, with even more bizarre results. The residue track from phase reversal seems to depend on what name you give to your export!
Now before everyone snickers, here is an explanation of what I did. I took my mix, opened up the export window, gave it a name, "Experimental export 1" and exported it. Once it finished, and without closing the export window, or changing any setting, and while holding my breath for good measure, I merely changed the name to Experimental export 2.
I then opened a mastering window, and brought down two fresh tracks to avoid any possibility of error in duplicating the default one Cubase provides. I imported #1 in the first track, and #2 in the second track. To reiterate, these were identical other than one lone digit in the name.
The phase reversal of one track resulted in exactly the same residue smiley face drum track.
I then reversed the phase on both tracks, with the same result.
I then deleted #2, and imported #1, such that the two tracks in the mastering window were identical not just in details, but in name as well.
The phase reversal was complete, with no residue sound.
I then deleted both #1's and imported two #2's.
The phase reversal was complete, with no residue sound.
So, what say you, mighty masters of the box? :smile: Maybe I have Cubase LE4, The Twilight Zone Edition. Cue music... [[url=http://[/URL]="http://www.youtube…"]YouTube - Twilight Zone intro.[/]="http://www.youtube…"]YouTube - Twilight Zone intro.[/]