Is there a sound problem on mixdown with FL Studio or not?

[Nokenoku]

@frootyloopz
Yes, 64bit resolution is higher than 32bit resolution.
But guess what:
32bit is already high enough to push every errors way below audibility.
So 64bit is a nice marketing-thing, but nothing more.

"When recording at 16 bit, any audio below 48dB (such as the decay of notes or subtle detail in the sound) is actually captured with less than 8 bits"

Now that's one meaningless observation.
What counts are the quantization errors. With 16bit they are at -96dB.

You can't hear signals, which are 96dB below other signals. And even with fade-outs, where your max-level gets more near to the quantization/dither-noisefloor, you won't hear it, if you don't turn up the volume on your listening-system (assuming you had a normal listening volume before).

[Burillo]

Even if there is a difference in sound, which I strongly doubt, it is only that ---a DIFFERENCE!! Not all this --I say that 'X' is 'better' than 'Y' stuff. There is no such thing as a 'better-o-meter' that you can toss in here that says so.

The only thing we can say is that "I work better with 'X' than with 'Y'"

well, frankly, there are some things that might be counted as "better" simply because everyone will tell you so - i highly doubt that if you reduce the bit-depth of your mixes to 8 they will sound "different" - they will sound like crap, that's for sure (but that's not what this topic is about anyway).

Files null, we finally got to that part, so technically the conversation should be considered closed. However, guys such as demzer clearly feel the need to insult certain people with unbacked claims and couldn't even bother to read what people say to them - OK, more power to them

[Burillo]

@frootyloopz
Yes, 64bit resolution is higher than 32bit resolution.
But guess what:
32bit is already high enough to push every errors way below audibility.
So 64bit is a nice marketing-thing, but nothing more.

dammit, people keep confusing those bits... 64-bit DAW processing != 64-bit internal plugin processing != 64-bit system (as in Win7x64) != 64-bit recording != 64-bit mixdown != 64-bit color != 64-bit Nintendo != 64-bit this != 64-bit that!

"When recording at 16 bit, any audio below 48dB (such as the decay of notes or subtle detail in the sound) is actually captured with less than 8 bits"

Now that's one meaningless observation.
What counts are the quantization errors. With 16bit they are at -96dB.

You can't hear signals, which are 96dB below other signals. And even with fade-outs, where your max-level gets more near to the quantization/dither-noisefloor, you won't hear it, if you don't turn up the volume on your listening-system (assuming you had a normal listening volume before).

that is actually a meaningfull observation (although it was meaningless in the context of what he was replying to), 24-bit recording REALLY, PERCEPTIBLY sounds better (or different, if you like) than 16-bit recording, but again what demzer was replying to was about 24-bit mixdown, not 24-bit recording - hence the confusion.

Apparently, demzer either doesn't know that 24-bit recording can be converted to 16-bit without any obvious and perceptible loss of quality (CD is 16-bit - to say that converting 24-bit to 16-bit is a common practice would be a serious understatement) or he doesn't really bother to read what people are saying to him, because his reply about 16-bit vs 24-bit was completely out of line with what he was replying to.

[Nokenoku]

dammit, people keep confusing those bits... 64-bit DAW processing != 64-bit internal plugin processing != 64-bit system (as in Win7x64) != 64-bit recording != 64-bit mixdown != 64-bit color != 64-bit Nintendo != 64-bit this != 64-bit that!

I wasn't confusing anything. I know exactly all those differences.
And I know, that a 64bit engines in a DAW won't bring you anything regarding sound quality.
For certain plugins it might make sense to operate in 64bit resolution, but for the in- and output 32bit is still enough in every case.

[crimsonwarlock]

My point is that the summing engine is not the only aspect of a DAW that contributes to it's sound.

The summing engine doesn't contribute ANYTHING to the sound. It just sums !!!!!!!

Are you a parrot? That has been said countless times in this thread and everywhere else on the internet- I have said it myself.

And yet you still say that a summing engine is contributing to the sound (you actually say it's not the ONLY apsect that contributes to a DAW's sound).

This is what I was pointing out: the summing engine isn't contributing ANYTHING to the sound. That is the whole point of the debate: SUMMING DOES NOT DO ANYTHING TO THE SOUND. No coloring, no compression, NOTHING !

[trimph1]

There is a HUGE difference between saying that summing is not the ONLY thing that influences sound and that it DOES NOT influence the sound---sheeesh

[VSTJuNkiE]

64-bit Nintendo !

Yes!! Anybody up for some Mario Party 2? Goldeneye?

[tony tony chopper]

You can really test what I can hear? Over the internet? I won't ask you to bother doing it "again" but if you would like to show me where you have done it before that would be great.

Blind testing.. (making sure to fit the 16bit audio parts in the same 24bit file)

Incidently, why did you choose to give the user the option of choosing the bit depth when exporting from Fruity Loops, if no one can hear the difference?

Marketing mainly (for 24bit). I don't have anything against adding something that's not useful, if it also doesn't hurt. It was also on demand for Pro Tools users (although apparently the latest one can finally read 32bit float wavs).
Exporting to 32bit float is useful however, as you can export the original, full-quality unclipped data, in order to edit it further. It's talking about what the end-user listens to that I say that anything above 16bit doesn't matter.

"When recording at 16 bit, any audio below 48dB (such as the decay of notes or subtle detail in the sound) is actually captured with less than 8 bits"

yes, and also producing noise at a level you won't hear.
If your -whole- song is peaking at -48dB, and you're asking the listener to crank up his listening level by +48dB (meaning that he will blow his ears when he will listen to another song), then yes, it matters. But you're then not rendering 16bit audio, you're rendering 8bit audio in a 16bit file, which is quite different. You could render the same 8bit audio in a 24bit file, and make the same claim about quality.

[spaceman]

17bit has a nice ring to it

[tony tony chopper]

Rifftrax said he couldn't tell the difference between 16 and 24 bit and then someone who makes a computer toy called FLStudio said he had a test to show that I couldn't either.

It's pretty easy to prove your point, just attach a piece of 24bit audio, and then the same one converted to 16bit. And then everyone will hear the obvious difference, right?
Note (again): asking the listener to crank up his volume to hear the inaudible is forbidden.

[mauseoleum]

"When recording at 16 bit, any audio below 48dB (such as the decay of notes or subtle detail in the sound) is actually captured with less than 8 bits"

yes, and also producing noise at a level you won't hear.
If your -whole- song is peaking at -48dB, and you're asking the listener to crank up his listening level by +48dB (meaning that he will blow his ears when he will listen to another song), then yes, it matters. But you're then not rendering 16bit audio, you're rendering 8bit audio in a 16bit file, which is quite different. You could render the same 8bit audio in a 24bit file, and make the same claim about quality.

Hmmm - not so fast!

Recently I was working on a theme that has a certain slap bass-guitar part.

I recorded bass at 16bit - for demo, on a semi-pro audio card that has relatively decent noise performance.

In a mix, said bass needs heavy limiting to "sit" properly and in context.

So there was a problem with "quiet below hearing treshold" garbage resurfacing in a mix. It needed old-school "manual" cleanup in an audio editor (sculpting around phrasing). A noise gate = not works.

This was only a demo, not a production (and not yet rehearsed in minute details), so not so much problem.

But, in a mix context, I can see this being a problem, esp. on acoustic instruments, i.e. miked stuff that is getting recorded at not-really-hot levels.

[Meffy]

17bit has a nice ring to it

Is that a chunk of satire or a bit of parity?

[tony tony chopper]

So there was a problem with "quiet below hearing treshold" garbage resurfacing in a mix.

of course, but to start with I'd more suspect the noise you had enhanced wasn't due to precision, but more analog noise "before" the recording. It amazes me when I see 24bit audio files generally filled with ambient noise way above 16bit, in a spectrogram.
And still, we're talking about rendered results for the end listener. As long as you're still working on audio, it's best to keep its original format.

[mauseoleum]

I'd more suspect the noise you had enhanced wasn't due to precision, but more analog noise "before" the recording.

No, it was definitely an "digital" artifact.

Whether this was the soundcard's "fault" or something else, I can't tell.

But I also know well how the true "analog" noise sounds - in it's many forms.

Well, enough OT ...

[Burillo]

it's only logical to blame 16-bit when the real problem is recording levels not being setup properly