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
Is there a sound difference in various DAW's ? Virtual Instrument Plugins
Old 3rd September 2008
  #1
Gear Maniac
 
BushmasterM4's Avatar
 

Is there a sound difference in various DAW's ?

Ive read, where some people say that if you load up the same tracks in different DAW's on the same PC, use the same hardware and monitors and press play, each will sound different. I find it hard to believe. Anyone ever do an actual test ? Its all digital. I can see where different hardware would effect the sound. But not software. Let me know if you know anything to back up such a claim.
Old 3rd September 2008
  #2
Lives for gear
 
Bhang's Avatar
 

I have used quite a few and I think the difference is not noticeable. The real difference is in the EQ's compressors etc. but many use 3rd party plug ins anyway so this is often a non issue. I don't believe that there is a difference in the audio quality pre say. there may be a difference but I doubt anyone could tell in a blind test. Its all 0's and 1's that we're created by your audio interface. the only other area would be in how they bus signals together. In the output there might be a noticeable difference, but which one is better would be a matter of opinion.
Old 3rd September 2008
  #3
Lives for gear
 
ewegogetemtiger's Avatar
omfg

it will never end

ever

utfsf -- this bull**** has been beat to death for eons

stop the madness now.

trolls stay at least 60 feet back

Danger Will Robinson!

Old 3rd September 2008
  #4
Moderator
 
Oroz's Avatar
 

Quote:
Originally Posted by BushmasterM4 View Post
Ive read, where some people say that if you load up the same tracks in different DAW's on the same PC, use the same hardware and monitors and press play, each will sound different. I find it hard to believe. Anyone ever do an actual test ? Its all digital. I can see where different hardware would effect the sound. But not software. Let me know if you know anything to back up such a claim.
I did a test between Digital Performer and Pro Tools and they sounded exactly the same. You can read more about it in this thread:

https://www.gearslutz.com/board/music...-broken-5.html

My first post is #140
Old 3rd September 2008
  #5
Gear Maniac
 
docbop's Avatar
 

Every programer writes code a little different, every compiler generate code slightly different so technically yes there is a difference. For a basic track is it a difference you can hear I would say no. Maybe someone with extraordinary great ears might but most no.

Now you get past basic tracks into EQ, and effects and such then they all have their own formulas and algorithms they use and that is audible. That is why there is such a big market of plug-ins and such.
Old 3rd September 2008
  #6
Lives for gear
 
T_R_S's Avatar
Lynn Fuston put out a DAW CD a while back
Old 3rd September 2008
  #7
Lives for gear
 
Sk106's Avatar
 

I would assume the differences are redundant. (this is assuming without the usage of the plug-ins that came with the DAW application).
People say Reason and Orion have a sound of their own though - haven't tried it myself.

If there would be audiable differences, I would suggest they would come out the most with large track counts and plenty of conversion and/or editing and plugs.

A few canadians guys, calls themselves "infinate wave", did some pretty serious testing of this on a purely techical level. They published their results online - and even made a snazzy interface to look at it through. The test was based purely on software performance, and targeted downsampling a 96 khz rising sine tone to 44,1 khz within many hosts - noting aliasing problems and other things. I'm not sure how valid it is or how one might interpret it to the best way. For example, it seems Logic 8 got a worse result from being run on osx Tiger than osx Leopard. Assuming things like that is a true representation of the SRC of different hosts, it was a bit of an eyeopener to me. (especially since I'm on Steinberg software and it got pretty low score - bastards)

SRC Comparisons
Old 3rd September 2008
  #8
Lives for gear
 
rhizomeman's Avatar
no.
Old 3rd September 2008
  #9
Mgr
Lives for gear
 
Mgr's Avatar
 

Thumbs down

Old 5th September 2008
  #10
Gear Maniac
 
BushmasterM4's Avatar
 

Thanks to those who replied. And to those few who slammed me for not researching. I did research and thats why I asked. Some say yes, some say no. I wanted to know anyone knew of a professional test done. Thanks again
Old 6th September 2008
  #11
Lives for gear
 
allencollins's Avatar
 

Quote:
Originally Posted by T_R_S View Post
Lynn Fuston put out a DAW CD a while back
those test cd's ARE WORTHLESS. they run the same track through the DAW. Sorry that NULL test..........itself should be NULLED.

If you have 5 daws and you record 1 song 5 DIFFERENT TIMES..... ONCE AGAIN RECORD the SAME song 5 Different times there will be a difference. Not including the nuances in playing ofcourse

I've done tests and there is a difference. Subtle but a difference. My theory is it has to do with
the soundcard device drivers and how each daw interacts with the driver. Also I thin kthe way DAW's impelement plugins could be an issue. Again the NULL test DOES NOT WORK it is not a true test. It is irrelevant....... Sure the internal processing of a pcm file should not vary. But with dithering and other variables it is a fact that rounding and truncation errors/differences will occur

Also CPU's may be a problem. Even a CPU has the potential not to process data the same way twice. Even a greater chance 2 cpus/computers. Will have minute differences. I know I know you think that is not possible BUT IT IS. I had a 1+ hour conversation today with a Metrology/process guy at my work on how this can happen. With CPU's and software it all comes down to the amount acceptable errors. Every CPU that comes of the line has errors. however miniscule they may be, there still are errors

Im not implying the differences are big or even matter but there is a difference. The difference may be during realtime playback or during recording. Maybe there is no difference during bounce/flattening but there are so many variables.

Think about it....... How many times has your DAW crashed? I know I know, if you are on a MAC it's never happend (ya right) but let's think about this. How many time has your DAW crashed? Why did it crash? because there is a bug or many bugs? So if it is possible for a daw to crash due to bugs then why wouldn't be possible for one of these bugs to cause a problem with the 'sound' of the DAW? Or the way it process the pcm data.
Old 6th September 2008
  #12
Gear Addict
 
Mike Derrick's Avatar
 

Allen the null tests do work. And you have to (you must) use the same audio files otherwise, like you mention, you bring in other variables such as soundcards etc.

So yes, the null tests are accurate for using the exact same audio files, and in fact that is the only proper way to test it.

Allen, in the tests where you heard differences, were they done as A/B tests or A/B/X? Because it is very easy to fool oneself into hearing differences where there are none with A/B tests. Throw in an X and see if you still hear the differences? If you can pass an A/B/X test with 100% then you can safely say you are hearing differences, otherwise A/B can be deceiving especially between identical audio files.

cheers.
Old 6th September 2008
  #13
Moderator
 
narcoman's Avatar
 

Quote:
Originally Posted by allencollins View Post
those test cd's ARE WORTHLESS. they run the same track through the DAW. Sorry that NULL test..........itself should be NULLED.

If you have 5 daws and you record 1 song 5 DIFFERENT TIMES..... ONCE AGAIN RECORD the SAME song 5 Different times there will be a difference. Not including the nuances in playing ofcourse

I've done tests and there is a difference. Subtle but a difference. My theory is it has to do with
the soundcard device drivers and how each daw interacts with the driver. Also I thin kthe way DAW's impelement plugins could be an issue. Again the NULL test DOES NOT WORK it is not a true test. It is irrelevant....... Sure the internal processing of a pcm file should not vary. But with dithering and other variables it is a fact that rounding and truncation errors/differences will occur

Also CPU's may be a problem. Even a CPU has the potential not to process data the same way twice. Even a greater chance 2 cpus/computers. Will have minute differences. I know I know you think that is not possible BUT IT IS. I had a 1+ hour conversation today with a Metrology/process guy at my work on how this can happen. With CPU's and software it all comes down to the amount acceptable errors. Every CPU that comes of the line has errors. however miniscule they may be, there still are errors

Im not implying the differences are big or even matter but there is a difference. The difference may be during realtime playback or during recording. Maybe there is no difference during bounce/flattening but there are so many variables.

Think about it....... How many times has your DAW crashed? I know I know, if you are on a MAC it's never happend (ya right) but let's think about this. How many time has your DAW crashed? Why did it crash? because there is a bug or many bugs? So if it is possible for a daw to crash due to bugs then why wouldn't be possible for one of these bugs to cause a problem with the 'sound' of the DAW? Or the way it process the pcm data.
A - the main reason the track will come out five different ways would be down to the operator. The engines perform the same.
Old 6th September 2008
  #14
Gear Head
 

can you tell the difference between something that was mixed in pro tools vs cubase vs logic vs anything else? can anyone? in a blind test? just find the platform/environment that you like best and trust that the digital summing (or whatever you want to call it) is fine.
Old 6th September 2008
  #15
Gear Addict
 
hackenslash's Avatar
 

Quote:
Originally Posted by allencollins View Post
If you have 5 daws and you record 1 song 5 DIFFERENT TIMES..... ONCE AGAIN RECORD the SAME song 5 Different times there will be a difference. Not including the nuances in playing ofcourse
Sorry, but if you record 1 song 5 different times into the same DAW it will be different.

There is no difference.
Old 7th September 2008
  #16
Lives for gear
 
ewegogetemtiger's Avatar
Quote:
Originally Posted by BushmasterM4 View Post
Thanks to those who replied. And to those few who slammed me for not researching. I did research and thats why I asked. Some say yes, some say no. I wanted to know anyone knew of a professional test done. Thanks again
Yes, some say there is a difference but what you will notice is that not one of them can ever offer the slightest shred of evidence to support their claim. Period.
Old 7th September 2008
  #17
This entire post I've quoted here is hogwash. Pure unadulterated hogwash.

Sorry to be so blunt, but the truth hurts sometimes.

To prove it, we ran different versions of PT on five different computers running an assortment of OS's and received bit identical results. That means that out of hundreds of millions of values, not a single bit was different.

Different computers=different results? Grab a bunch of computers and run some calculations on them and see how much numerical difference there is in the answers. We all know that different banks using different computers come up with different bank account balances daily. I use the bank with the computers that round up.

Read this entire post again and believe the opposite is true of everything it says.



Quote:
Originally Posted by allencollins View Post
those test cd's ARE WORTHLESS. they run the same track through the DAW. Sorry that NULL test..........itself should be NULLED.

If you have 5 daws and you record 1 song 5 DIFFERENT TIMES..... ONCE AGAIN RECORD the SAME song 5 Different times there will be a difference. Not including the nuances in playing ofcourse

I've done tests and there is a difference. Subtle but a difference. My theory is it has to do with
the soundcard device drivers and how each daw interacts with the driver. Also I thin kthe way DAW's impelement plugins could be an issue. Again the NULL test DOES NOT WORK it is not a true test. It is irrelevant....... Sure the internal processing of a pcm file should not vary. But with dithering and other variables it is a fact that rounding and truncation errors/differences will occur

Also CPU's may be a problem. Even a CPU has the potential not to process data the same way twice. Even a greater chance 2 cpus/computers. Will have minute differences. I know I know you think that is not possible BUT IT IS. I had a 1+ hour conversation today with a Metrology/process guy at my work on how this can happen. With CPU's and software it all comes down to the amount acceptable errors. Every CPU that comes of the line has errors. however miniscule they may be, there still are errors

Im not implying the differences are big or even matter but there is a difference. The difference may be during realtime playback or during recording. Maybe there is no difference during bounce/flattening but there are so many variables.

Think about it....... How many times has your DAW crashed? I know I know, if you are on a MAC it's never happend (ya right) but let's think about this. How many time has your DAW crashed? Why did it crash? because there is a bug or many bugs? So if it is possible for a daw to crash due to bugs then why wouldn't be possible for one of these bugs to cause a problem with the 'sound' of the DAW? Or the way it process the pcm data.
Old 7th September 2008
  #18
Quote:
Originally Posted by ewegogetemtiger View Post
Yes, some say there is a difference but what you will notice is that not one of them can ever offer the slightest shred of evidence to support their claim. Period.
Which is why I drafted several dozen people from all over the globe to try to get to the bottom of this issue. We tested 17 different DAW/computer/OS alternates in order to hopefully put this issue to rest.

We didn't determine what causes the differences that we observed between some, but we did find that most sum identically. The others which wouldn't were broken.

If anyone is unfamiliar with the tests that TRS and Allen referenced above, there's a great deal of discussion here:

Awesome DAW-SUM Comparison - 3dB

And the files are available here if you want to do your own listening.

Awesome DAWSUM Samples - 3dB

NOTE: The files are not available for free. If you don't want to hear them, you don't have to pay for them.
Old 7th September 2008
  #19
Lives for gear
 
feck's Avatar
Quote:
Originally Posted by Lynn Fuston View Post

Read this entire post again and believe the opposite is true of everything it says.
I love it!
Old 7th September 2008
  #20
Lives for gear
 
allencollins's Avatar
 

Quote:
Originally Posted by Lynn Fuston View Post
This entire post I've quoted here is hogwash. Pure unadulterated hogwash.

Sorry to be so blunt, but the truth hurts sometimes.

To prove it, we ran different versions of PT on five different computers running an assortment of OS's and received bit identical results. That means that out of hundreds of millions of values, not a single bit was different.

Different computers=different results? Grab a bunch of computers and run some calculations on them and see how much numerical difference there is in the answers. We all know that different banks using different computers come up with different bank account balances daily. I use the bank with the computers that round up.

Read this entire post again and believe the opposite is true of everything it says.

Do you have any idea what is the required floating point precision when calculating money? Or should I say
what precision is NOT required. Bad example.
Old 7th September 2008
  #21
Gear Head
 

Quote:
Originally Posted by allencollins View Post
Do you have any idea what is the required floating point precision when calculating money? Or should I say
what precision is NOT required. Bad example.

Get over yourself.


Apart from DAW software that is broken ( i.e Cubase VST which is years old anyway ) they all sound the same.

Lynn Fuston has put more effort into this than anyone. And in the early days when many of us were unsure we also spent hours testing it.

They sound the same.

Go and make some music
Old 7th September 2008
  #22
Lives for gear
hmmm....

ever heard of samplitude ...... I kid.
Old 7th September 2008
  #23
Lives for gear
 

why do i think that they sound different?

cause they´re different to work with and routing/mixing is sometimes very different.

therefore the result has to sound different. comps, reverbs, eqs don´t act exactly the same in different routing-ways.

when recording one channel, you can´t notice a difference.

when you playback one channel without editing/routing, you can´t notice a difference.
Old 7th September 2008
  #24
Moderator
 
narcoman's Avatar
 

Quote:
Originally Posted by allencollins View Post
Do you have any idea what is the required floating point precision when calculating money? Or should I say
what precision is NOT required. Bad example.
I do. And you're still not right.

I'll give you a clue (and you'll have to do some research here !hah!) - the Virtuosi project in the 90's.
Old 7th September 2008
  #25
Quote:
Originally Posted by allencollins View Post
Do you have any idea what is the required floating point precision when calculating money? Or should I say
what precision is NOT required. Bad example.
Like any analogy, mine is imperfect. But barring that, I stand by my statements.
Old 7th September 2008
  #26
Quote:
Originally Posted by eddierodriguez View Post
when you playback one channel without editing/routing, you can´t notice a difference.
Likewise when you play back 22 channels like in my test. You can't notice a difference. Because there is none.

Let me be very specific though. I'm not saying all DAWs sound alike. All my testing proved was that summing ONLY, when done correctly, is the same on all the DAWs (that weren't broken) tested.

Throw plugins into the mix and all bets are off.
Old 7th September 2008
  #27
Lives for gear
 
allencollins's Avatar
 

Quote:
Originally Posted by Lynn Fuston View Post
Like any analogy, mine is imperfect. But barring that, I stand by my statements.
You stand by your statement?.... ooooh.

All I can say is incident angle is a killer when rounding. Deposition and beam strike will always make you wonder how these cpu devices can even work.

Despite what you think you know. The Manufacturing of cpu's still has a little bit of luck involved. This is the difference, it's not all science as you would think. There is still some finessing involved. In the industry it is known as 'tuning'. Same with software there is always some tweaking involved. A compromise if you will.
A comprimise that can leave you with errors and flaws.

When you write code you assume in good faith that the linker and source to source translator have no bugs and are 100% error free. When you buy an operating system you expect the same level of consistency. We know this isn't the case. MS and Apple put out 'hot' fixes and sp's EVERYDAY. The Intel compiler I use
everday was written in 2006. But yet I'm developing source for Vista and XP both came out after 06
Tell me there can't be errors?

Open your eyes man take a computer class or something. From a guy that compares the required floating point presicion of the banking industry to IEEE floating point precision required by a DAW, is suspect.
Old 7th September 2008
  #28
Maybe you need to take a class in computers Allen, seriously there is no need to be so aggressive on this, the fact of the matter is that floating point precision is handled by the processors FPU, it's not something that is down to the application, every app on that one machine will get the same results for single or double precision floating point operations (barring interference from neutrinos, electric spikes etc), for single we're well outside of the range of human hearing before we get anywhere near the limit and they all use single precision floating point operations currently.

The result is the same, the same data is sent to the ports and retrieved from the ports, the parsing of the data is handled in a static and standardized way (there's only so many ways you can do this stuff), coding style has zero impact there unless your coding style includes wrong.

Summing itself is a basic thing, a literal addition. Even Pan laws are fairly basic and thus should be the same across multiple DAWS, but any plugin any EQ anything else involved that convolutes the sound will sound different. However if it's the same plugin used on both daws the results should be the same, it would be interesting to test that and make sure this is the case.
Old 8th September 2008
  #29
Lives for gear
 
allencollins's Avatar
 

Quote:
Originally Posted by mdme_sadie View Post
Maybe you need to take a class in computers Allen, seriously there is no need to be so aggressive on this,
Im not aggressive I made a suggestion as you did .Take a computer class?
okeedokee?

Quote:
Originally Posted by mdme_sadie View Post
the fact of the matter is that floating point precision is handled by the processors FPU, it's not something that is down to the application, every app on that one machine will get the same results for single or double precision floating point operations (barring interference from neutrinos, electric spikes etc),
really what is an fpu? is that kinda like a 287 or a 387? huh? oh the memories thanks for 'pointing' that out too. fpu just can't cut it on it's own? Thank gawd for SSE. Man them extra 128 stack registers
are a godsend. An is if the 8 that mmx gave wasn't enough.

mov( 1, eax ); movd( reg32, mmi ); huh? what? fpu? punpcklbw? psubusw? pmulhuw?



Quote:
Originally Posted by mdme_sadie View Post
for single we're well outside of the range of human hearing before we get anywhere near the limit and they all use single precision floating point operations currently.
thx again

Quote:
Originally Posted by mdme_sadie View Post
The result is the same, the same data is sent to the ports and retrieved from the ports, the parsing of the data is handled in a static and standardized way (there's only so many ways you can do this stuff), coding style has zero impact there unless your coding style includes wrong.
really since when can an app talk directly to a 'port' as you say? Ever hear of protected mode? I guess the words 'device driver' are meaningless to you. Oh you must use DOS. Real mode anyone? After this I know Im replying to someone who has never written code other than batch file or a vbs script. Maybe he was talking about a tcp port? in any event it wasn't the same port a digital i/o talks to that's for certain. Could it be as system call via Kernel Mode ? Like with MMX or SSE. Fpu? well peeeuuuuu something does .......



Quote:
Originally Posted by mdme_sadie View Post
Summing itself is a basic thing, a literal addition. Even Pan laws are fairly basic and thus should be the same across multiple DAWS, but any plugin any EQ anything else involved that convolutes the sound will sound different.
convolute is the word. Boy is that convoluted. Did you actually ever see a data structure? Summing is rather simple. Simple when you have a mackie console with two monster cables summing your mix into your ipod speakers.


Quote:
Originally Posted by mdme_sadie View Post
However if it's the same plugin used on both daws the results should be the same, it would be interesting to test that and make sure this is the case.
the word here is 'should' be. Did you ever implement a real time vst engine in
a daw? Please explain to me just how you know this. Steinberg just gave the source code I guess. Now anyone could implement vst.............

Do you subscribe to pc world? just curious
Old 9th September 2008
  #30
Moderator
 
narcoman's Avatar
 

Quote:
Originally Posted by allencollins View Post
Im not aggressive I made a suggestion as you did .Take a computer class?
okeedokee?
haha ! Nice!!



Quote:
Originally Posted by allencollins View Post
After this I know Im replying to someone who has never written code other than batch file or a vbs script.
you can play with me if you like.....
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
Similar Threads
Thread
Thread Starter / Forum
Replies
Alex.T / Electronic Music Instruments and Electronic Music Production
65
IzzyRock / So much gear, so little time
292
gsilbers / Music Computers
17
Keyplayer / So much gear, so little time
19

Forum Jump
Forum Jump