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 DAW Software
Old 28th August 2018
  #331
Lives for gear
 

Quote:
Originally Posted by stinkyfingers View Post
i just tested a few levels quickly (Live > soundflower > Reaper) and it seems to be the same whether it's realtime stream or a rendered file. it's not a conclusive test, but i didn't notice anything unusual. i did notice the master output will clip @ 0 dBFS...ie...the stream is only 24 bit from Live.
How does Reaper behave when you try to test its internal routing for nulls?

If you route a signal through a send to a sends return and phase invert the return. What does Reper readout on the master bus?

In pre-fader mode, in my tests, yesterday Ableton nulled to infinity on SPAN only at even figures from starting from 0 to -10.

With uneven figures, SPAN showed -150,5 dBFs or lower.
Old 28th August 2018
  #332
Lives for gear
 
stinkyfingers's Avatar
 

Quote:
Originally Posted by lllubi View Post
How does Reaper behave when you try to test its internal routing for nulls?
Basically there are 2 results, like your test. (Based on gain values used and the error related to)
It’s either a null to “-inf”, or a null to -313.07 dBFS, which is basically the difference of 64 bit float “error”. It can vary a little based on the input signal and actual gain applied, but it’s always under -300 dBFS if it’s not “-inf”

A null figure in the -150 range coincides with 32 bit float “error”.
Old 28th August 2018
  #333
Lives for gear
 
jsblack's Avatar
 

Hey Stinky, where did go your dirty donut..? I miss it quite a bit!
Old 28th August 2018
  #334
Lives for gear
 
stinkyfingers's Avatar
 

I just couldn’t resist anymore and I had to eat it...
(It was a bit stale but still pretty damn good)
Old 28th August 2018
  #335
Lives for gear
 
jsblack's Avatar
 


So it's 50 shades of pink now.. fair enough
Old 28th August 2018
  #336
Gear Guru
 
Karloff70's Avatar
 

Quote:
Originally Posted by jsblack View Post

So it's 50 shades of pink now.. fair enough
That's the blushing after eating the dirty donut.
1
Share
Old 28th August 2018
  #337
Lives for gear
 
jsblack's Avatar
 

Old 28th August 2018
  #338
Lives for gear
Just a reminder for any testers to not forget to check DC Offset. I made a little quick Live vs Reaper comparison test by recording a MIDI clip in Reaper and rendering this as 24bit 44.1kHz undithered audio. I used this recording as the source for a new 1 track Reaper project and a 2 track Live project (used a buss). I then rendered at -5.99dB fader position on both audio tracks and made sure I had the same peak value in the master. I summed both rendered files and was somewhat surprised that I didn't get a complete null.

After viewing the file data I noticed that the left channel of the Live source file had a 0.05 % DC Offset and the Reaper had a 0.01 %. I removed DC Offset in both files and summed them again. Complete null.
Old 29th August 2018
  #339
Lives for gear
 

Quote:
Originally Posted by Mikael B View Post
Just a reminder for any testers to not forget to check DC Offset. I made a little quick Live vs Reaper comparison test by recording a MIDI clip in Reaper and rendering this as 24bit 44.1kHz undithered audio. I used this recording as the source for a new 1 track Reaper project and a 2 track Live project (used a buss). I then rendered at -5.99dB fader position on both audio tracks and made sure I had the same peak value in the master. I summed both rendered files and was somewhat surprised that I didn't get a complete null.

After viewing the file data I noticed that the left channel of the Live source file had a 0.05 % DC Offset and the Reaper had a 0.01 %. I removed DC Offset in both files and summed them again. Complete null.
DC offset didn´t change my results with Ableton´s internal routing.
Looks like Ableton calculates send and returns in 32-bit floating point. On the other hand, they tailored volume fader and send resolutions to null to infinity at some fixed points.
Old 29th August 2018
  #340
Lives for gear
 

playing a song in the finder sounds different than playing the same song in itunes ( with all settings nominal ) as does playing the same song in a daw through the mac IO. every code is written the way some engineer wrote it to get it done... this is not always optimal... i think the manufacturers lie about how well their daws talk to hardware...

It could be as subtle as a driver difference between the daw and the interface. many years ago we did a test at our spot between nuendo and protools. same everything else. the sound was noticeably different. nuendo was better.

that being said i came up on a Sonic Solutions rig, the biggest one on the east coast at the time ( 24 full digital I/O, 5 nubus cards on a quadra 950 )>? and although it was boasted ( and cursed ) for having the most lines of code in any software at the time. it SOUNDED RIGHT. audio felt good coming out. I have never had that feeling from Ableton. I mix things for people from all kinds of sources and sometimes even on their own computers integrated into my room. ableton has never sounded good to me. one time my intern did a test where he set the buffer to different settings and bounced the same simple project and they sounded not the same. if i can get him to provide files for your enjoyment i will put them in Gdrive for you...

Yelling commence in 3...2....1...
Old 29th August 2018
  #341
Lives for gear
Quote:
Originally Posted by NewAllianceEast! View Post
playing a song in the finder sounds different than playing the same song in itunes ( with all settings nominal ) as does playing the same song in a daw through the mac IO. every code is written the way some engineer wrote it to get it done... this is not always optimal... i think the manufacturers lie about how well their daws talk to hardware...…
DAWs do not talk to hardware. Audio drivers do. All of what you suggest above is not true, except for sample rate differences as not all apps ask for the audio driver to adapt to the current stream rate. So real time resampling may sound different. This is a driver issue, not a DAW issue and is not relevant unless you can show a specific DAW, Live in this case, fails to make the audio driver adapt to the current sample rate when another do not fail.

Quote:
Originally Posted by NewAllianceEast! View Post
audio felt good coming out. I have never had that feeling from Ableton.
I understand this feeling but I think it's about audio problems you might have had. Like for instance if Live is struggling with playback there may be small artefacts (I'm not talking about stuttering). This creates a not so nice feeling of that something is wrong. But you are quite unspecific here and talk about really old things and Live is at version 10 now. Any old feelings are actually not relevant.

Also if you like to work with a DAW it stands to reason you like what you create. This is not relevant either here, except that one should of course choose tools one likes. That's personal preference and not sound quality which could be proven to be different or even detrimental.

There are four main areas where sound quality comes into play which is
  • Recording
  • Playback
  • Audio processing
  • Summing

All of those can be measured and compared between DAWs and for processing different methods within a DAW can be compared as well (native devices can of course never be compared operating in another host). We should not conflate these though.

My suggestion is to re-read login's observations two pages back.

Last edited by Mikael B; 29th August 2018 at 10:11 PM..
1
Share
Old 29th August 2018
  #342
Lives for gear
 

Quote:
Originally Posted by NewAllianceEast! View Post
it SOUNDED RIGHT. audio felt good coming out. I have never had that feeling from Ableton.
I have the same impression when it comes to playing live MIDI within Ableton. Feels closer to how I experience my own MIDI performance compared to other DAWs.
Old 29th August 2018
  #343
Gear Maniac
Quote:
Originally Posted by NewAllianceEast! View Post
playing a song in the finder sounds different than playing the same song in itunes ( with all settings nominal ) as does playing the same song in a daw through the mac IO. every code is written the way some engineer wrote it to get it done... this is not always optimal... i think the manufacturers lie about how well their daws talk to hardware...

It could be as subtle as a driver difference between the daw and the interface. many years ago we did a test at our spot between nuendo and protools. same everything else. the sound was noticeably different. nuendo was better.

that being said i came up on a Sonic Solutions rig, the biggest one on the east coast at the time ( 24 full digital I/O, 5 nubus cards on a quadra 950 )>? and although it was boasted ( and cursed ) for having the most lines of code in any software at the time. it SOUNDED RIGHT. audio felt good coming out. I have never had that feeling from Ableton. I mix things for people from all kinds of sources and sometimes even on their own computers integrated into my room. ableton has never sounded good to me. one time my intern did a test where he set the buffer to different settings and bounced the same simple project and they sounded not the same. if i can get him to provide files for your enjoyment i will put them in Gdrive for you...

Yelling commence in 3...2....1...
Since testing is now frowned upon on this thread, I'll keep it simple: PLACEBO EFFECT
1
Share
Old 29th August 2018
  #344
Gear Guru
 

Quote:
Originally Posted by NewAllianceEast! View Post
i think the manufacturers lie about how well their daws talk to hardware...
I think people on the Internet lie when they talk about how carefully double-blinded their listening is.

Quote:
... Sonic Solutions ... audio felt good coming out. I have never had that feeling from Ableton....[
Quote:
Originally Posted by lllubi View Post
I have the same impression when it comes to playing live MIDI within Ableton. Feels closer to how I experience my own MIDI performance compared to other DAWs.
So one person says Ableton sounds bad and another person thinks Ableton sounds good? Why it's almost as if people were talking about subjective impressions and personal results. It's almost as if it's user dependent and that there's no such thing as the DAW automatically "forcing" a bad sound on you! It's almost as if (if the DAW is a comfortable fit) you could get a good sound in any DAW!

Nah!
2
Share
Old 30th August 2018
  #345
Quote:
Originally Posted by NewAllianceEast! View Post
come back to the convo when you get some.
I used Logic for years. For the last ten years I’ve used Ableton Live and PT (either or) on a weekly basis.
I think it’s ok for people to really drill down into the detail by carefully set up and controlled testing - if that’s what floats their boat. Alarm bells go off for me when anecdotal experiences turn into damning negatives about companies ‘lying’.
I play midi parts, record audio, mix and bounce out using more than one DAW regularly and I can’t hear any difference.
If there is, it has to be slight. If someone is playing a midi part and immediately feeling it doesn’t sound good, it’s either user error or there’s something psychological going on.
1
Share
Old 30th August 2018
  #346
Lives for gear
 

Quote:
Originally Posted by chrisso View Post
I used Logic for years. For the last ten years I’ve used Ableton Live and PT (either or) on a weekly basis.
I think it’s ok for people to really drill down into the detail by carefully set up and controlled testing - if that’s what floats their boat. Alarm bells go off for me when anecdotal experiences turn into damning negatives about companies ‘lying’.
I play midi parts, record audio, mix and bounce out using more than one DAW regularly and I can’t hear any difference.
If there is, it has to be slight. If someone is playing a midi part and immediately feeling it doesn’t sound good, it’s either user error or there’s something psychological going on.
We are definitely into subtle territories especially these days when all DAWS sound good in general. i just think it's funny how people would rather defend some corporation than discuss the possibility that they benefit from fudging their specs and numbers to look better. WTF? VW literally programmed their cars to lie to us about emissions ( this is only one of hundreds of examples i can sight that are currently ongoing ). if anyone thinks it's not normal operating procedure then they are the fool.

'caveat emptor' exists for a reason.
Old 30th August 2018
  #347
Yeah fine, all I'm saying is nothing is proven as yet. It's a giant leap from just starting to test for differences to already calling out companies for supposedly 'lying'.
When someone reasonably challenges your claim (in a reasoned and mild way) you write they are 'a dick'.

It's becoming a bit of a pattern - software companies are liars and anyone who disagrees with your wild claim is 'a dick'.

Err, OK.
Old 30th August 2018
  #348
Lives for gear
 

Quote:
Originally Posted by chrisso View Post
I used Logic for years. For the last ten years I’ve used Ableton Live and PT (either or) on a weekly basis.
I think it’s ok for people to really drill down into the detail by carefully set up and controlled testing - if that’s what floats their boat. Alarm bells go off for me when anecdotal experiences turn into damning negatives about companies ‘lying’.
I play midi parts, record audio, mix and bounce out using more than one DAW regularly and I can’t hear any difference.
If there is, it has to be slight. If someone is playing a midi part and immediately feeling it doesn’t sound good, it’s either user error or there’s something psychological going on.
What it really comes down to is this: The OS ( OSX for me ) is constantly doing **** under the hood that we don't know about. and since most people are using core audio to route DAWS to interfaces they are all subject to these potential issues:

the OS may decide to up sample or down sample in the background differently at different times with different apps and audio interfaces. This is why itunes can sound different than the finder with the same file. it's going through different code and the OS may make different decisions based on this. deniers and digital apologists can get all down on me for bringing this up. but it's the reality of every computer you own. and you can't easily control it.

macos - Controlling audio output's sample/bit rate on OSX from command prompt - Ask Different

HOW does osx go between my 32 bit session and the 24 bit interface? who decides when and how those last 8 bits are removed? does core audio truncate them? maybe if your hardware driver sucks. GET IT?
Old 30th August 2018
  #349
Lives for gear
Quote:
Originally Posted by NewAllianceEast! View Post
What it really comes down to is this: The OS ( OSX for me ) is constantly doing **** under the hood that we don't know about. and since most people are using core audio to route DAWS to interfaces they are all subject to these potential issues:

the OS may decide to up sample or down sample in the background differently at different times with different apps and audio interfaces. This is why itunes can sound different than the finder with the same file. it's going through different code and the OS may make different decisions based on this. deniers and digital apologists can get all down on me for bringing this up. but it's the reality of every computer you own. and you can't easily control it.

macos - Controlling audio output's sample/bit rate on OSX from command prompt - Ask Different

HOW does osx go between my 32 bit session and the 24 bit interface? who decides when and how those last 8 bits are removed? does core audio truncate them? maybe if your hardware driver sucks. GET IT?
I've already covered this possibility and without that tone of yours. You do sound as if macOS willy-nilly would change sample rate/bit depth suddenly when you write "the OS may decide to up sample or down sample in the background differently at different times". That's uncorroborated as far as I can see. As you also say apps in macOS can ask the audio device to adapt sample rate and bit depth. If the interface can change this it's likely to adhere to this, but you can't be sure without testing this happens I suppose. Though why anyone here at gearsslutz would use bad drivers and not know how to match Live's audio settings with their audio interface's does sound far-fetched.

If you don't know how a 32bit audio stream is handled in a 24-bit audio interface, maybe it's time for some studying? Maybe read a manual.

It remains to be heard if anyone can hear these alleged sound differences with apps asking for different audio settings. I think this being highly unlikely unless the app asks for something unreasonable.
1
Share
Old 30th August 2018
  #350
Quote:
Originally Posted by NewAllianceEast! View Post
HOW does osx go between my 32 bit session and the 24 bit interface? who decides when and how those last 8 bits are removed? does core audio truncate them? maybe if your hardware driver sucks. GET IT?
Then all DAWs on a Mac are screwed, not just Ableton.
Old 30th August 2018
  #351
Lives for gear
 

Smile

Quote:
Originally Posted by chrisso View Post
Then all DAWs on a Mac are screwed, not just Ableton.
this is my point. although someone else just told me to go read a manual cuz they apparently know all about DAWS.
Old 30th August 2018
  #352
Quote:
Originally Posted by NewAllianceEast! View Post
this is my point.
OK, then at least you are consistent.
Old 30th August 2018
  #353
Lives for gear
 
stinkyfingers's Avatar
 

Mac users can use HALLab to view what's actually going on with core audio and also change settings...
1
Share
Old 30th August 2018
  #354
Lives for gear
 
jsblack's Avatar
 

Need to install Xcode right..?
Old 30th August 2018
  #355
Lives for gear
Quote:
Originally Posted by NewAllianceEast! View Post
OK i'll be sure to read your manual for you! say hi to your mom for me!
I read manuals every single day. I make tests every day (for anything in my job or for audio). It's way better than belief in magic.
2
Share
Old 30th August 2018
  #356
Lives for gear
Quote:
Originally Posted by jsblack View Post
Need to install Xcode right..?
No. It's an app and is part of "Additional Tools for XCode". Make sure you download the version fitting your OS version. You need to create a developer account at the Mac Developer site, which I believe is free.
1
Share
Old 30th August 2018
  #357
Lives for gear
 
login's Avatar
Quote:
Originally Posted by NewAllianceEast! View Post
OK i'll be sure to read your manual for you! say hi to your mom for me!
Read a book on general computing, then one on maths and lastly one on the scientific method. Getting some literacy in your life doesn't hurt.
3
Share
Old 31st August 2018
  #358
Gear Maniac
 

Well - I agree it looks bad that those of us who feel they can hear a difference cannot produce any evidence. I also think that those who have researched this are probably right - I bet there is no difference. All I have wondered is what it might be that causes the feeling that there is one just due to some many sensing it. Its such a minefield when you look at what could be/ might be creating the feeling that something is different that I think even if you could convince me you might not convince others. I do feel like I need to test this myself - and I have been seriously meaning to but have been so busy, but I will when I have that time just give it all a good go to understand for myself. If I do get any interesting results I will of course share.
Old 31st August 2018
  #359
I question that 'so many' feel there is a difference. This kind of thing is amplified by online forums.
I think hundreds of thousands of people are happily using DAWs based on work flow and feature set decisions, not on sound quality.
Here's the thing though......
If you feel one DAW sounds better than another, you absolutely have a right to use the DAW you feel sounds better, whether there is any evidence to back it up or not.
Posting online that one DAW sounds better, or even different, to another one as if it's a fact? Then you'd better have evidence.
3
Share
Old 31st August 2018
  #360
Lives for gear
 

Quote:
Originally Posted by chrisso View Post
Posting online that one DAW sounds better, or even different, to another one as if it's a fact? Then you'd better have evidence.
Testing MIDI is not popular. As are MIDI devices which are more or less accurate. People in such threads always speak about sound quality but hardly investigate or speak about the tools they make their music with.
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump