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

[sorohanro]

by some half taught musicians !!!

Discussion was about concepts not about people, but if you brought this up so elegantly, here, have a reading while I make the test again using different plugins:

EDUCATION

1982-1986 Music School, Cluj-Napoca, Romania
1986-1993 Sigismund Toduta Music High School, Cluj-Napoca, Romania
1993-1998 Gheorghe Dima Music Academy (trumpet), Cluj-Napoca, Romania
1995-1997 Bach Academy Master-classes (Baroque music), Cluj-Napoca, Romania
1996-1998 Summer Orchestra and Chamber Music classes, Nirbator, Hungary

WORK EXPERIENCE

1997-2003 State Philharmonic Orchestra, Tirgu-Mures, Romania
2003-2006 Romanian National Opera, Cluj-Napoca, Romania
2007-2008 State Philharmonic Orchestra, Tirgu-Mures, Romania
2009-2010 NOA Opera Festival, Vilnius, Lithuania

to name a few.
Also participating to Music Laboratory Network "The Process" - Sound Recording Laboratory, a international master class on recording/ mixing/ mastering of classical and opera music. Main lecturer was prof.Moritz Bergfeld.

[sorohanro]

Ok, took a break from working and made the test again.
Took everything from beginning, opened a new file, made sure that there are no effects and all the knobs and sliders are to default/ 0 bd.
Imported the same midi file (so all the timing and velocities are as similar as possible). Changed the Hammond with a electric piano (no more Leslie rotary speaker problem) and put all the drums with "single sample" option, so hi-hats and snare triggers the exactly same samples.

The files not normalized look like this:

The Mu.Lab file is louder.

Still, we are professionals and we don't get fooled by "louder=better".
Normalizing:



Still, the Mu.Lab is more "Fat". There is something strange, the Mu.Lab file have a bit of latency noticeable in the beginning.

Finally. here are the files, both out of same midi, same plugins, same settings, no effects, both rendered at 32 bit float.
http://uploading.com/files/2f415m69/FL% ... test2.zip/

Same midi file that you have in first test and Glue Reeds instead of Organized Trio.
http://www.mokafix.com/glue-reeds.html

[nucleon]

The same tracks I uploaded also render the same from Live 8 and Reaper btw.

[sorohanro]

The same tracks I uploaded also render the same from Live 8 and Reaper btw.

by the way, quite a nice track

can you also try with VSTi instead of wav files ? like same midi file played through same VSTi ?

On my computer is just not the same. Both hosts use the same soundcard driver.

[nucleon]

Thanks

Im downloading now, tho uploading.com is having troubles.

[sorohanro]

I hope that my intervention in this thread didn't ruin Tony/ Gol's day.
I do admire his patience but mostly I admire and appreciate his work. FL, as I sad on other DAW's sections of the forum is a dream from the point of view of a composer. The technical details don't stay between musician and his music.

Also I appreciate nucleon's work, we are on the road to find if I'm having hallucinations or not

[Carbonboy]

For the record, all DAW software developers see 100's of these posts every year.

Image-Line have a great summary article on it in their manual -

http://flstudio.image-line.com/help/html/app_audio.htm

Note that it doesn't say you won't find differences. BUT the differences relate to settings not 'secret' knowledge on the part of the developers.

Carb

[nucleon]

Heres a test using a kick, snare and hat midi pattern with Addictive Drums, with each drum set to 'single sample' and both reverbs turned off.

http://uploading.com/files/m3245b9e/FL% ... BTest.zip/

From MuLab it renders a little louder than from FL, but if you normalize both, they cancel when you invert one and add them together.

It appears MuLab is playing back different VST's at different output levels.

Audio clips in both apps are the same level as shown in my previous test.

[nucleon]

Are you playing back the sax with the MuLabs sampler or as an audio track btw?

[nucleon]

Heres a phase cancel test in FL with your recent renders.

http://uploading.com/files/d3acb13m/can ... Btest.zip/

It suggests different plugins are outputing at different levels between apps.

First a section loops from just the MuLabs version.

Then with the same section from FL, but delayed by 64 samples and inverted in the mixer, the drums and most of the bass cancel.

The same section repeats again but with the FL version increased in gain a bit and this time the sax cancels instead.

[rifftrax]

Honestly, have to hand to Gol for being like, one of select few DAW programmers in existence that will actually step users through problems like this. That says a helluva lot honestly. Only other guy I can think of who's involved like that is Justin from Reaper and it's funny because those are the only two DAWs I actually use.

[tony tony chopper]

Still, the Mu.Lab is more "Fat". There is something strange, the Mu.Lab file have a bit of latency noticeable in the beginning.

you shouldn't judge little differences in displayed waveforms especially in this case where a wav is slightly longer than the other, shrinked to the same view length there will be differences.

If by fat you mean a spectral change, you have to understand that, for this to happen, a host would have to filter the output to color it. Again, a change in color isn't something you don't do, it's something you do, it can't be done by mistake. Taking 2 plugin outputs to a wav file is a simple addition.
If a host was helping a plugin to process, that could even make sense, but VST's receive no help from the host. There's no DSP library provided by the host that could make a plugin sound different in different hosts, a plugin is on its own.

[trimph1]

so the DAW is just the blank canvas then?

if that is such then what is going on with the spectral differences here--?

questions from a n00b---

[LeVzi]

Stop shooting the fella down just cos he has noticed something... ppl are allowed opinions, gol is explaining to him.

And Nucleon, dont think I seen you around these parts before How you keepin ?

[tony tony chopper]

if that is such then what is going on with the spectral differences here--?

but there are none