The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Ableton - sound quality Virtual Instrument Plugins
Old 1 week ago
  #211
Gear Maniac
As to Diva, it has analog emulation that makes the sound slightly different every time it is played, as that is how real analog hardware supposedly behaves.
Old 1 week ago
  #212
Gear Head
Quote:
Originally Posted by Mage View Post
I guess that "people who argue it’s only 1+1 are ********" would offend a few even though it’s an understatement. I wonder why "Hello World" code samples exist for programming languages while implementing a DAW is only a few codes of line.

This is what happened.

1. I’ve been using Ableton for years. I’ve bought Logic yesterday. The main reason was Alchemy (+ curiosity).
2. Without ever reading about it, I noticed that it sounded better than Ableton.
3. I found this thread and a few others, I learned that 1+1 = 2, so I tested it.

Before I tell you the experiment, I would like to underline how brilliant idea is to test the theory on single and unmodified audio files (no plugins, no gain changes, etc). Surprise, you don’t need a DAW to play original audio files. A music player can do it for you.

So, the test was simple.

1. I imported a track both into Logic and Live. Logic can’t import FLAC which is ridiculous. It’s also ******** that it modifies my source files. The person who is responsible for that should reincarnate as an earthworm.

2. I put the phase invert on the track in Live.

3. I exported the track from Logic, imported it into Live, and played them together. They nullified each other. (I used FabFilter Q 2 on -120 dB to check it).

4. I added FabFilter L2 on the track both in Logic and Live. I used the same (saved) setting that’s a slight modification of "Basic/Modern". Export + Import again. They were nowhere close to null each other. The lows were missing but I could hear the rest of the track.

So, as long as there is no warp, there are no plugins, they can save the original audio file. Wow. As soon as there is a plugin, it’s not 1+1.

My test doesn’t prove that any of the two sounds better than the other. It only proves that they don’t sound the same.

So, as long as your general music production workflow is importing a single audio file into your DAW, and exporting it without plugins, modifications, or tempo changes, the DAW makes little to zero difference. If it’s what you do, then you should not worry about their sound.

By the way, the fact that Logic touches my source audio files without my explicit consent is already enough reason to find a third DAW, even if I use copies (I do).
I've used both Live and logic for 12+ years. An average day for me consists of time spent in both DAWs... The short answer is you either did something wrong or overlooked something.

About your replies later with other tests, a DAW does not color the sound of a plugin, it's the other way around. I'm not trying to be abrasive but something in your test is wrong.

For a true null test you would use one singular uncompressed and unconverted file, identical gain staging (even 1/10 of a dB immediately breaks a null test. Everything has to be identically gain staged end to end), identical plugin types (i.e. AU in Logic and Live), and export to identical uncompressed files, (bit depth and sample rate) with either no dither on export in both DAWs, or dither using the same plugin with the exact same unmodified preset, (and no dither from the export menu of the DAW.) Everything has to be totally identical or it simply is not a null test.

Again, even small fractions of a decibel break a null test. And non-identical plugins settings guarantee the files will not null. This is the case with compressors and limiters, as all compressors and limiters cause distortion. Distortion that would very likely be audible in a null test, and depending on differences in gain staging could be quite audible.
Old 1 week ago
  #213
Lives for gear
Mage, where are your assumptions? You seem to assume a lot of things which renders your results much less interesting. As far as I can see you assume:
  • That the same instrument fed the same MIDI recorded twice would sound exactly the same. For that to succeed you need a very simple instrument with no random activity, i e a very boring instrument. All the greats I like don't even play one note exactly the same. And you chose Diva from U-he? C'mon!
  • That a dynamic plug-in, like a TUBE COMPRESSION plug-in, will react the exact same way fed the same signal. And you're likely not even feeding it the same signal (see previous post), but even if you did emulating tubes will likely not sound the same even in the same DAW.
  • Are you testing recording or mixing? A better approach is to use the same exported tracks from a third tool, like for example Reason. This way you ensure that you are testing mixing, not recording and you also ensure the audio signal is the same in both DAWs.
  • Assuming you can ignore pan law when doing these tests.

Thanks for testing, but it'd be better if this was a community effort and that files are shared. I actually feel this is very important. Of course with clear instructions some tests can be repeated, provided everyone have the same plug-ins. The best way to ensure that is to use free plug-ins. There are great free ones.
Old 1 week ago
  #214
Gear Head
Quote:
Originally Posted by Mikael B View Post
Thanks for testing, but it'd be better if this was a community effort and that files are shared. I actually feel this is very important. Of course with clear instructions some tests can be repeated, provided everyone have the same plug-ins. The best way to ensure that is to use free plug-ins. There are great free ones.
I'm up for it
Old 1 week ago
  #215
Lives for gear
 

Quote:
Originally Posted by Mage View Post
I did another quick null test using NI Supercharger GT.

Warping and Flex were off. I used the AU version in Live.

They didn’t null each other at all.
PT HD and Ableton null perfectly with third-party plugins from different companies in this video.

YouTube

Make a test with a third DAW to find out whats going on. It doesn´t matter which plugin format you use as long as something doesn´t work as intended within a DAW.

If a DAW doesn´t null tracks internally it hardly will with other DAWs.
Old 1 week ago
  #216
Lives for gear
 

This is a funny thread.


Btw, whats better sound quality in this context anyway?
Old 1 week ago
  #217
Here for the gear
 

Quote:
Originally Posted by lllubi View Post
PT HD and Ableton null perfectly with third-party plugins from different companies in this video.

YouTube

Make a test with a third DAW to find out whats going on. It doesn´t matter which plugin format you use as long as something doesn´t work as intended within a DAW.

If a DAW doesn´t null tracks internally it hardly will with other DAWs.
I would need to compare every DAWs if I said that "all DAWs sound differently".

I only argued with the statement that "all DAWs must sound the same as it’s only ones and zeros".

And there was a line of code that must have been copied from this repository:
https://github.com/share-for-a-bette...se-uses-to-mix

It’s enough to find one example to show that DAWs don’t sound the same to prove the statements above are wrong. Your suggested method would be needed to figure it out which DAW of the two I tested is "wrong".

There are plugins when it makes little sense to expect they would null each other. A synth such as Diva, or a distortion effect, likely uses random number generators.

On the other hand, I would expect two Kick 2s (without distortion) null each other in two DAWs. And they did.

I would expect the same from a limiter, and they did not. But this happened only when I used -2.5 dBTP. I find it strange.

Considering that both Kick 2 and L2 with 0 dBTP nulled, I would rule out the user-error.

By the way, although my initial impression was that Logic might have been sounded better than Live, additional tests using my ears lead to the conclusion that I can stick with Live. Most likely, I will compare them in the future, just be sure.

Right now, I rather invest my time in learning music theory than comparing two DAWs since I don’t want to use one of them anyway. I don’t expect much good from a DAW that can’t open flac but modifies my wav files. Such details tell me enough about what I should expect from a software.
Old 1 week ago
  #218
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by zedsdeadbaby View Post
I've used both Live and logic for 12+ years. An average day for me consists of time spent in both DAWs... The short answer is you either did something wrong or overlooked something.

About your replies later with other tests, a DAW does not color the sound of a plugin, it's the other way around. I'm not trying to be abrasive but something in your test is wrong.

For a true null test you would use one singular uncompressed and unconverted file, identical gain staging (even 1/10 of a dB immediately breaks a null test. Everything has to be identically gain staged end to end), identical plugin types (i.e. AU in Logic and Live), and export to identical uncompressed files, (bit depth and sample rate) with either no dither on export in both DAWs, or dither using the same plugin with the exact same unmodified preset, (and no dither from the export menu of the DAW.) Everything has to be totally identical or it simply is not a null test.

Again, even small fractions of a decibel break a null test. And non-identical plugins settings guarantee the files will not null. This is the case with compressors and limiters, as all compressors and limiters cause distortion. Distortion that would very likely be audible in a null test, and depending on differences in gain staging could be quite audible.
All good points. Just one detail: Dither is random so even if using the same plugin, two different exports (even in the same DAW) should not null.

Alistair
Old 1 week ago
  #219
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by Mage View Post
And there was a line of code that must have been copied from this repository:
https://github.com/share-for-a-bette...se-uses-to-mix
First, the link doesn't work. Secondly, no, the line I copied to this thread was copied directly from the source code. IF that line has been put up online somewhere else with that title it is probably because I have been saying the same thing in these type of discussions for a very long time and demonstrating what I say by providing the line of code from the source files. (Sometimes with extra lines for more detail. It depends a bit on the discussion). Not because I am getting the idea from anyone else.

Of course someone else might simply have done the same thing as I do in these discussions, back up their arguments with facts. Providing a line of summing code is of course the obvious way to back up the argument. Not exactly rocket surgery...

Quote:
Your suggested method would be needed to figure it out which DAW of the two I tested is "wrong".
See below.

Quote:
I would expect the same from a limiter, and they did not. But this happened only when I used -2.5 dBTP. I find it strange.
Besides the issue of dither (if it was turned on), my guess is that FF L2 uses a numeric problem solving approach. This is getting more and more common in plugin development. That will prevent it from nulling (even with itself) in many cases. It has nothing to do with any DAW being "wrong".

Quote:
Right now, I rather invest my time in learning music theory than comparing two DAWs
Always a better idea!

Quote:
I don’t expect much good from a DAW that can’t open flac but modifies my wav files. Such details tell me enough about what I should expect from a software.
Although it sucks that Logic doesn't support FLAC, it is purely for commercial reasons, not technical reasons. (Apple want to promote their own proprietary compression formats). One could argue that it a good practical or even ethical reason to avoid Logic (and everything Apple IMO) but not a good technical one.

As for modifying the source files, that is operator error. When you save a project, there is a tick box to save a copy of the imported files into the project folder.

Alistair
Old 1 week ago
  #220
Gear Nut
 

This is very interesting. I went back to Logic recently to do some work and immediately felt a difference. Now I admit I was using logics built in FX. I don't know though if it's the way the virtual instruments work, the quality of the inbuilt EQ works or something but it was much warmer and bigger. I know those are subjective statements, and in no way is it a stab at Ableton - for me it just blossemed quicker. I wish I could say how. Will I be selling of my copy of Ableton. I don't know, I have learnt the DAW a lot and will keep it for a while. It's also the only DAW like Logic that I have that I can run on both PC and Mac - which is an advantage. Logic is a little slower to construct a song in, but the flip side is I work differently in it. I am curious to see what the tests overall reveal, and perhaps I should find the time to do my own tests. What will be interesting is to use some virtual instruments rather than just audio tracks to see if it's a midi timing thing or something to do with the way audio is handled from a virtual instrument or if it's just the love of the Logic plugins. As this thread seems to have evolved I will try and experiment.
Old 6 days ago
  #221
Here for the gear
 

Quote:
Originally Posted by UnderTow View Post
First, the link doesn't work. Secondly, no, the line I copied to this thread was copied directly from the source code.

Yeah, have you read the link? The parts between the slashes were:

share-for-a-better-world

the-line-every-daw-in-the-universe-uses-to-mix


Quote:
Originally Posted by UnderTow View Post
Besides the issue of dither (if it was turned on), my guess is that FF L2 uses a numeric problem solving approach. This is getting more and more common in plugin development. That will prevent it from nulling (even with itself) in many cases. It has nothing to do with any DAW being "wrong".
The questions was not whether we can find an excuse/explanation why the DAWs that never ever produce different results, produce different results.

The question was whether the DAWs that never ever produce different results produce different results. The answer is, yes, they do.

Even if every DAW in the Universe uses the same line to do everything.

Quote:
Originally Posted by UnderTow View Post

As for modifying the source files, that is operator error. When you save a project, there is a tick box to save a copy of the imported files into the project folder.
Alistair
I guess you think you know it better what I did than I do.

Surprise!

I did not save the project. In fact, I did not save anything in Logic ever.

It happened and happens before/without saving. I just tried it again. It’s unacceptable.
Old 6 days ago
  #222
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by Mage View Post
The question was whether the DAWs that never ever produce different results produce different results. The answer is, yes, they do.
It has been clearly explained to you by several people why your tests are not valid yet you continue to make the same claims. The only things we can conclude from this is that you lack the necessary technical understanding and insight to make any useful statement on the subject and that continuing this conversation with you would be a waste of time.

Over and out,

Alistair
Old 6 days ago
  #223
Gear Nut
 

What's become very obvious to me is that the tests are being done by various people - who probably don't really know what they are testing. I think Null tests are invalid really, dithering aside they don't test midi performance, plugin integration, fader movements, panning etc.

there is a lack of understanding I feel from everyone here. I cannot be discounted from that pack and I am not trying to do so. It's just that everyone I know who have used Logic, Ableton, Pro Tools, appears to hear a difference. The argument is that somehow we are doing something wrong, have a technical lack of understanding or are deluded. I think the only ones who could in theory prove one way or another would be the programmers behind the software, which is never going to happen.

I'm not a fanboy of any of the DAWs specifically as they each bring magic and difficulties to the table. I remain convinced however that there is more at play here and some spouted beliefs that it's just 1's and 0's and things null test which is not enough for me. A DAW does far more than just summing 2 wav files together.

I could run null tests but I believe that I need to have a better understanding of the backbone of the software to know what and how to test.

Argue me and between yourselves all you want, it's not enough. Honest scientific and in depth tests are required to determine what's going on.
Old 6 days ago
  #224
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by RichardNolan View Post
The argument is that somehow we are doing something wrong, have a technical lack of understanding or are deluded.
For clarity sake, I am not saying anyone is deluded. There are many reasons why two DAWs (or even the same DAW) will not null and/or will not sound the same. My arguments in a nutshell:

1) It is not the summing itself
2) It is often due to settings like pan-laws
3) Another culprit can be a plugin misbehaving or simply behaving differently in a different DAW
4) Issues with particular plugins/DAWs and delay compensation
5) DAW specific algorithms like time-stretching and SRC most certainly DO sound different
6) Included plugins also sound different (not to mention included content for those that use it).
7) Different work-flows will usually yield different results
8) Different visual feedback (for instance in metering) will cause people to make different mixing decisions
9) There is a certain amount of placebo and expectation bias in all of us. (Me included).
10) DAW specific idiosyncrasies (Like FL Studio having a limiter on the master).
11) Probably a few things I am forgetting right now

All this amounts to people getting different results from different DAWs. What it does NOT amount to is one DAW sounding better than another DAW. It all depends what one does with any of the DAWs and which workflow and visual look best fits any user's individual style and taste.

Alistair
Old 6 days ago
  #225
Lives for gear
Quote:
Originally Posted by RichardNolan View Post
I think Null tests are invalid really, dithering aside they don't test midi performance, plugin integration, fader movements, panning etc.
Performance is a different subject really, but what null tests do is focusing on what really matters, which is the sound. That's why they are both valid and relevant. Claiming they are not is quite laughable. Likely what you don't like is that null tests prove you're hearing wrong. That's a very inconvenient truth unfortunately. It's also part of learning at any stage of your career.

Quote:
Originally Posted by RichardNolan View Post
I could run null tests but I believe that I need to have a better understanding of the backbone of the software to know what and how to test.
That's a very good first step and I agree one should understand what one is testing. But you don't need a system developer background in order to do that. Some basic understanding of digital audio, MIDI and some system analysis skills should be sufficient to make discovery.

Quote:
Originally Posted by RichardNolan View Post

Argue me and between yourselves all you want, it's not enough. Honest scientific and in depth tests are required to determine what's going on.
Of course. And shared tests so others can repeat and improve them.

Last edited by Mikael B; 6 days ago at 01:34 PM..
Old 6 days ago
  #226
Gear Nut
 

Perhaps needing a coder to verify a test is going a bit far, however as pointed out here are some many things that throw these tests.

I never refer to one DAW sounding better however, although many do which makes these conversations a little tricky sometimes because it upsets one party against another.

In a nutshell though there are many variety that are leading to differences in end results. For me I feel more inspired in Logic to write what I want to write musically. That's not to say I have not been having a lot of fun in Ableton though.

I am glad thought I have a Mac again and able to run Logic (which was my issue before - I only had a PC). So now I can run both!
Old 6 days ago
  #227
Lives for gear
 

Quote:
Originally Posted by RichardNolan View Post
Perhaps needing a coder to verify a test is going a bit far, however as pointed out here are some many things that throw these tests.

I never refer to one DAW sounding better however, although many do which makes these conversations a little tricky sometimes because it upsets one party against another.

In a nutshell though there are many variety that are leading to differences in end results. For me I feel more inspired in Logic to write what I want to write musically. That's not to say I have not been having a lot of fun in Ableton though.

I am glad thought I have a Mac again and able to run Logic (which was my issue before - I only had a PC). So now I can run both!
Plugin developer(s?) have participated on this very thread..
Old 5 days ago
  #228
Lives for gear
Quote:
Originally Posted by Calagan View Post
It's crazy !

So much bad music all around, and so much superhuman engineers with surnatural hearing, knowing without effort when a mix comes from Logic or Live.

...
See. This is what will drive the aliens nuts, when they will come to conquer earth. Everybody will surrender, not the stable genius audio engineers. They will start to discuss complete bull**** until the heads of the aliens will explode.

Audio engineers wills ave us all... after that they will discuss further differences between DAWs and who was it with what frequency at which level that made the aliens explode.



All these guys that "feel" or "experience" these "differences" between DAWs... can we please sent them all to Golgafrincham, where they are desperately neededto develope a new society? Hint, hint...
Old 5 days ago
  #229
Gear Head
 
Calagan's Avatar
 

ah ah ah ah....

I didn't think they could be useful, but in fact they are...
Old 5 days ago
  #230
Lives for gear
Quote:
Originally Posted by RichardNolan View Post
Perhaps needing a coder to verify a test is going a bit far, however as pointed out here are some many things that throw these tests.
Badly set up tests, yes. However, nullifying signals is a basic test that are valid everywhere, also outside of audio in the "analog world" (physical). It's a characteristic of frequency and applied physics and mathematics. Therefore, if the test is found to be properly and reasonably constructed and executed nullifying results can't be explained away as insignificant.

There's nothing wrong having input from a music software developer, or several.

Quote:
Originally Posted by RichardNolan View Post
I feel more inspired in Logic to write what I want to write musically.
And that is a fine reason to choose a DAW.
Old 5 days ago
  #231
Gear Nut
 

Quote:
Originally Posted by whiteaxxxe View Post
See. This is what will drive the aliens nuts, when they will come to conquer earth. Everybody will surrender, not the stable genius audio engineers. They will start to discuss complete bull**** until the heads of the aliens will explode.

Audio engineers wills ave us all... after that they will discuss further differences between DAWs and who was it with what frequency at which level that made the aliens explode.



All these guys that "feel" or "experience" these "differences" between DAWs... can we please sent them all to Golgafrincham, where they are desperately neededto develope a new society? Hint, hint...
No that is precisely some sort of stab into the very thing that shouldn't be ignored. This isn't a flat earth theory or something. I don't choose to believe in some difference - it is obvious to myself and to many others that something is different. Null tests are useful yes, however (and I haven't searched a lot) there are many things to "null" test, not just some tracks of audio.


The one thing I feel comes across from some is that it's some sort of fanboy challenge. I have maintained across this thread that I use all of them, but I do hear a difference.

I honestly would like to know what it is that we all hear. I think only some very controlled aspect of several tests would help to shed some light on the facts. It would be a good magazine type of article - but only if done with appreciation of all the main differences that can occur.
Old 5 days ago
  #232
I know I said this before, and was ignored.....
But when you are recording into a DAW and working on your music you are balancing, eq'ing and compressing as you work etc.
I'm never dropping stereo files into my DAW and trying to null it.
You change things based on what you are hearing. Therefore, when I have finished some music in Ableton Live, Logic or Pro Tools it sounds equally good. Because I have been working with the sounds from the first note played to the last mix adjustment.
It's an interesting academic debate about whether DAWs sound different when playing back a raw stereo file, but NO ONE uses daws that way. Everyone builds up multiple tracks of midi and audio, eq'ing, sculpting the sounds as they go along.
That's EXACTLY why many great sounding hit records are produced and released using all the DAWs mentioned in this thread.
Great sounding mixes produced using Live, Pro Tools and Logic.
As a musician, my goal is to improve my work methods so I can produce the best recordings in any DAW, like other professionals. I am not interested in proving one DAW has a minute sound difference to another. That's immaterial when you are making sounds with your drums and synths, and sculpting those sounds as you work.
Old 5 days ago
  #233
Gear Guru
 

Quote:
Originally Posted by RichardNolan View Post
I think Null tests are invalid really, dithering aside they don't test midi performance, plugin integration, fader movements, panning etc.
Developers who have participated in these discussions say that a plug in will perform the exact same operation when presented with the exact same audio. In any DAW. That's just a fact. Midi performance and pan law will not make your DAW sound "thin". Get a grip. And who moves the fader? Well YOU do. If you moved the fader then you are what made your mix "different".

Quote:
there is a lack of understanding I feel from everyone here.
No, people understand you perfectly, we just don't believe you. In the entire history of these threads - and there are dozens of them - not one person has ever even accepted the challenge to listen to a bunch of different files and tell us "which DAW" made each one. That's because they don't sound any different unless you set them different.

Quote:
It's just that everyone I know who have used Logic, Ableton, Pro Tools, appears to hear a difference.
That's strange, because Nobody that I know personally thinks there is a difference in the sound. In the workflow sure. And I know lots of professional engineers. Of course they know how a blindfold works. Admit it, you and your friends are looking at the screen, at the colors, at the GUI and that's when you say: "gee this DAW sounds kind of thin compared to the other one"

Quote:
The argument is that somehow we are doing something wrong, have a technical lack of understanding or are deluded.
If someone else gets a null, and you don't, you are doing something wrong! Do you have any idea what the odds are against getting a null "by accident"? It makes the Powerball look like a sure thing. Also, everyone is susceptible to expectation bias. Everyone, no exceptions. But if you think placebo is the same as "hallucinating" - then of course you are going to reject the very idea.

Quote:
I think the only ones who could in theory prove one way or another would be the programmers behind the software, which is never going to happen.
Some of these programmers are right here on this website and they have already weighed in, you choose to ignore them and concoct a Conspiracy Theory instead! And then you wonder why you aren't being taken seriously? DAWs did not fall out of the cargo door of a Flying Saucer. They were designed and built by human beings, and there are human beings who understand how they work and don't work.

You just aren't one of them.
Quote:
A DAW does far more than just summing 2 wav files together.
That's really all the DAW does by itself, everything else the DAW does is really done by you, the user. You turn this knob, you turn that knob, you insert this plug in. You get a different sound by how you interact with the DAW. That is why every Difference-Hearer hears a different difference!

Or haven't you noticed, that for every guy who says Abelton is "thin" , another guy says "no, Pro Tools is the 'thin one ... Abelton is "thick". No, you're both wrong, Pro Tools is the Woody One. ("Woody". BTW, is an exact quote from one of your fellow Believers) Among the people who claim to hear a difference, there is a total lack of agreement about what that difference sounds like! This simple fact renders all your anecdotal 'evidence' moot.

Quote:
Honest scientific and in depth tests are required to determine what's going on.
the first such test that I was aware of, was conducted back in 2003 on 20 different DAWs. They have been going on ever since and people like you just keep on ignoring them and keep on believing what you want to believe. Flat-Earther indeed!

IMO, you really have some nerve going on a forum in 2018 and saying "somebody should do some tests"

These tests have been done to exhaustion. Null tests, blindfold A/B/X. Done and done and done. You may be unaware of the history of this discussion and how many tests have already been performed, but you have been given this opportunity to learn from people whose knowledge exceeds your own, and you have refused. This puts it back on you.

Here is the bottom line: The burden of proof is now on you. You Believers have to prove to rest of us that Pro Tools is the "thin one". Or whatever ridiculous thing you believe. I don't have to prove there is no Bigfoot. Can't prove a negative anyway. You have to bring us a Bigfoot - a live one, a dead one, a Bigfoot skeleton. Something anything. Not excuses, not hypotheses. Evidence. You can't, because you don't have the will or skills to do real testing, (where you would get nulls anyway) nor could you listen to a file blindfolded and ever tell us which DAW made that file. You've got nothing but your 'feelings'. Feelings are great - for everything except scientific proof.
Old 5 days ago
  #234
Here for the gear
 

Quote:
Originally Posted by UnderTow View Post
. The only things we can conclude from this is that you lack the necessary technical understanding and insight to make any useful statement on the subject
Says the person who:

1. believes that every DAW in the Universe mixes tracks by the single line of code you found somewhere In The Daw Source Code

2. makes false assumtions about what I did and what I did not

3. fails to understand that although I, at first glance, condered Logic might have sounded better, I decided to stick with Ableton because of my tests

Please keep your promise and never answer me.
Old 5 days ago
  #235
Here for the gear
 

Quote:
Originally Posted by pottering View Post
As to Diva, it has analog emulation that makes the sound slightly different every time it is played, as that is how real analog hardware supposedly behaves.
Yes. I guess a lot of synths may use random generators at various places (distortion, saturation, etc.)

It also made sense that Kick 2, without distortion and harmonics, behaved the same way in both DAWs.

Anyway, I’m at the beginning of the journey of music production, and I can see the truth in "the fewer plugins and the less signal processing, the better results".

I spent days on recreating psytrance kicks as close as possible to 100% accuracy. (The purpose was not copying them but understanding the synthesis.)

I zoomed on the tracks I liked beyond 1/16384. I zoomed until I could see the dots on the wave. I tweaked the envelops in Kick 2 and Massive as long as I had to do it to produce the same waveform.

The harmonics were missing, but otherwise, the result is way better than anything I could produce by adding saturation, reverb, filter, compressor, etc.

In the end, I liked my kick more than the original sample. And that’s from one of my favorite tracks. Yet there is not a single EQ on my kick.

The bassline isn’t ready yet but I had the same experience. The less effect I added and the more I tweaked the envelops, the better it sounded.

My point is, I started believing (and seeing) it’s possible in Ableton to produce the same or even better quality sound than I hear in my favorite tracks if I do it right.

I find Ableton’s workflow pleasant. In the end, I could remove even the s(M)exoscope plugin as using Ableton’s record function was better for comparing the generated waveforms.

Mixing seems is super difficult. So far, when I tried to mix the bass with the kick I created, it never sounded right. I could have blamed Ableton but mixing my kick with the bass from a track sounded fine.

I can imagine that the mixer in Ableton is more "raw", and the one in Logic does some magic. Many years ago, I tried to mix tracks in FL Studio. I remember I also had the initial impression that it sounded better than in Ableton, at least without tweaking the tracks (EQ, volume).

If it’s true, then maybe Live gives us more control in the mixing. That’s how I meant "raw".

A few weeks ago, I remastered my last two mixes in Ableton. I tried everything I could. The best result came from removing the compressor from the master, and adjusting the volume levels as long as I had to.

I used L2 but not for limiting (much). I checked the loudness.

In the end, it’s better to be in control.
Old 5 days ago
  #236
Here for the gear
 

Quote:
Originally Posted by zedsdeadbaby View Post
I've used both Live and logic for 12+ years. An average day for me consists of time spent in both DAWs... The short answer is you either did something wrong or overlooked something.
I’ve spent 12 hours a day on average under the Sun and another 12 hours under the stars. I’ve been doing it for decades. The short answer is if you believe the Earth is orbiting around the Sun you either did something wrong or overlooked something. That’s my argument.

Quote:
Originally Posted by zedsdeadbaby View Post


For a true null test you would use one singular uncompressed and unconverted file,
So, your 12+ years with DAWs didn’t help much. It makes no sense what you wrote. First of all, you can use compressed source files. Either lossy or lossless.

Secondly, you can use compressed export if it’s lossless compression.

Finally, you can do it without source files as well. The goal was not comparing how DAWs import and export audi files but to see if they sound the same or not.

Quote:
Originally Posted by zedsdeadbaby View Post

identical gain staging (even 1/10 of a dB immediately breaks a null test.
Unless you want to test how they behave when they add/remove gain.

Quote:
Originally Posted by zedsdeadbaby View Post
Everything has to be identically gain staged end to end), identical plugin types (i.e. AU in Logic and Live),
This is not true. I considered it, but unlike you, I tested it as well. The plugins I tested and that nulled in the two DAWs also nulled in the same DAW AU vs. VST.
There is a difference between testing and guessing.

Quote:
Originally Posted by zedsdeadbaby View Post

and export to identical uncompressed files,
The issue you are talking about is not called "compressed vs. uncompressed" but "lossy vs. lossless".

Quote:
Originally Posted by zedsdeadbaby View Post

Everything has to be totally identical or it simply is not a null test.
Again, not true. For example, 0ms vs. 5ms lookahead did not make noticable difference with L2. We are back at guessing vs. testing or using words (everything vs. something).

Quote:
Originally Posted by zedsdeadbaby View Post
And non-identical plugins settings guarantee the files will not null. This is the case with compressors and limiters, as all compressors and limiters cause distortion. Distortion that would very likely be audible in a null test, and depending on differences in gain staging could be quite audible.
OMG. Why would I try to test with different settings? I can’t believe you wrote it down.

Besides, "distortion" does not have to be different unless random number generators are involved. I don’t see why a limiter would do that.

Again, you wrote "all compressors and limiters" cause distortion.

Most of the should not when the signal is below the threshold.

"Distortion" is a subjective word. When we talk about null test, it makes sense to use the word "change". Most plugins meant to cause changes. That’s their purpose. Creating music is about changing silence to sound. Trust me, it’s a huge change.

When one uses null test to compare DAWs, it’s not written in stone to use only audio files. (It’s wrong to do only that). It makes no sense to compare sounds without applied changes. The point is to test whether they behave the same way or not when the changes are expected to be the same.

For example, when a synth like Diva "changes" silence to sound, it should not always null. It could, depending on the settings (and the internals of the plugin).

For example, Kick 2 nulls (in certain cases).

I would expect L2 null regardless of whether the sound level reached the treshold or not. It did null with 0 dBTP but not with -2.5 dBTP. Even if no limiting was applied.
Old 5 days ago
  #237
Here for the gear
 

Quote:
Originally Posted by Mikael B View Post
Mage, where are your assumptions? You seem to assume a lot of things which renders your results much less interesting. As far as I can see you assume:
  • That the same instrument fed the same MIDI recorded twice would sound exactly the same. For that to succeed you need a very simple instrument with no random activity, i e a very boring instrument. All the greats I like don't even play one note exactly the same. And you chose Diva from U-he? C'mon!
  • That a dynamic plug-in, like a TUBE COMPRESSION plug-in, will react the exact same way fed the same signal. And you're likely not even feeding it the same signal (see previous post), but even if you did emulating tubes will likely not sound the same even in the same DAW.
  • Are you testing recording or mixing? A better approach is to use the same exported tracks from a third tool, like for example Reason. This way you ensure that you are testing mixing, not recording and you also ensure the audio signal is the same in both DAWs.
  • Assuming you can ignore pan law when doing these tests.

Thanks for testing, but it'd be better if this was a community effort and that files are shared. I actually feel this is very important. Of course with clear instructions some tests can be repeated, provided everyone have the same plug-ins. The best way to ensure that is to use free plug-ins. There are great free ones.
I also wrote it multiple times that some sounds and effect are supposed to be different.

It’s normal and recommended to do tests that are expected to fail. It’s a double check or counter-test to eliminate user error. In the software industry, it’s called "test driven development".

I used both audio files and MIDI instruments.

When I used instruments, I exported the MIDI file from Live and imported it into Logic.

I used ~80Mb .wav files so it would take much space on my Google Drive even if I had a licence that allows me to share those audio files. But I do not have such license.

Since everyone has different instruments, mine can be missing from your DAW. I don’t know what should I have uploaded in addition to a 1 bar lenght MIDI file.
Old 5 days ago
  #238
Here for the gear
 

Quote:
Originally Posted by RichardNolan View Post
What's become very obvious to me is that the tests are being done by various people - who probably don't really know what they are testing.
You are right.

Quote:
Originally Posted by RichardNolan View Post

I think Null tests are invalid really, dithering aside they don't test midi performance, plugin integration, fader movements, panning etc.
I did not use dithering.

I don’t see why null test would not measure differences in the midi performance. They do, and it was one of the things I wanted to test. It is inluded into the "whether the DAWs sound different" question.

Although I used a simple midi file, with one note on it eight times, it nulled win Logic and Live when the plugin was accurate and without random numbers. It confirms that such testing is possible. It would be interesting to use a more complicated MIDI file to test mixing, for example.

Quote:
Originally Posted by RichardNolan View Post

there is a lack of understanding I feel from everyone here. I cannot be discounted from that pack and I am not trying to do so. It's just that everyone I know who have used Logic, Ableton, Pro Tools, appears to hear a difference.
Count me in. The reason I found this thread is that I could hear a difference. As for Logic vs. Live, I still believe there is, and most likely, in the mixing. See the failed L2 null test.

However, to me, the question was whether I can stick with Live or not.

Quote:
Originally Posted by RichardNolan View Post


The argument is that somehow we are doing something wrong, have a technical lack of understanding or are deluded. I think the only ones who could in theory prove one way or another would be the programmers behind the software, which is never going to happen.
In psyhology, it’s called cognitive dissonance. Many people want to believe they were right otherwise they feel uncomfortable. So, instead of trying to test what I tested, they write "You did it wrong".

I find it sad when people who were not there believe they know what I did. It would be better to ask "Did you use dithering?" instead of stating I must have done something like that. But again, many people want to believe they are right.

Others, like me, and likely you, want to know the truth.

In the end, you don’t need to see the programming. Seeing the programming is only needed if you want to answer the question "Is the programming the same?"

The question was "whether they sound the same?". The only way to test it is testing how they sound. This may sound weird but it happens all the time that people make wrong conclusions on false assumptions.

If you want to know the truth about a subject, you need to test the subject. Nothing else matters. If one DAW was written in C++, and another was not written in any programming language, but it grew from an organic material that’s from another planet, it can still happen they sound the same. (Most likely, they won’t but it can happen).

Of course, there are details one should know. For example, certain plugins use random number generators. But you don’t need to see the programming to compare the sounds.

Quote:
Originally Posted by RichardNolan View Post


I'm not a fanboy of any of the DAWs specifically as they each bring magic and difficulties to the table. I remain convinced however that there is more at play here and some spouted beliefs that it's just 1's and 0's and things null test which is not enough for me. A DAW does far more than just summing 2 wav files together.

I could run null tests but I believe that I need to have a better understanding of the backbone of the software to know what and how to test.

Argue me and between yourselves all you want, it's not enough. Honest scientific and in depth tests are required to determine what's going on.
You have many valid points, and I think your approach is correct even if it’s a bit different than mine.

My ultimate question was whether I should use Logic or Live. Of course, it would have been easier to answer it if they always sounded the same when expected.

Finding the answers is usually way easier than finding the correct question (and focusing on it).

It’s also great to reconsider the results once in a while. Most likely, I will do more testing later. Right now, I think Live is the right choice for me.
Old 4 days ago
  #239
Gear Guru
 
UnderTow's Avatar
Quote:
Originally Posted by Mage View Post
Anyway, I’m at the beginning of the journey of music production
And yet here you are arguing and insulting the very people that could help you...

For instance with stuff like:

Quote:
Mixing seems is super difficult.
Anyway, good luck with your journey.

Alistair
Old 4 days ago
  #240
Here for the gear
 

Quote:
Originally Posted by UnderTow View Post
And yet here you are arguing and insulting the very people that could help you...

For instance with stuff like:



Anyway, good luck with your journey.

Alistair
No, I haven’t insulted anyone who helped me or were interested. I was friendly with them.

It’s not true for everyone though.

This thread is about Ableton’s sound quality. I wrote in that subject. And then, as always happen on the internet, some people went personal, and when I answer that part, they "cry" or tell me that I am not friendly.

I posted many details about what plugins I tested and how (L2 with basic/modern with added lookahead).

The fact I haven’t yet created music does not mean I have no idea of audio or testing software.
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump