honestly why do people debate the sound quality of a host?

[eduardo_b]

Sometimes one can choose to be a moron or not. What i don't get is why you would want to choose moron.

I'm assuming the "you" is not a reference to me given that I am agreeing with you. What I'm saying is that those who believe otherwise are choosing to disagree with the basic concept of using maths and testing to prove there are no difference. Their premise is that they can hear differences, so the proposition that it can be proved otherwise goes against what their ears are apparently telling them. In other words, in their minds it is subjective.

[jupiter8]

Sometimes one can choose to be a moron or not. What i don't get is why you would want to choose moron.

I'm assuming the "you" is not a reference to me given that I am agreeing with you. What I'm saying is that those who believe otherwise are choosing to disagree with the basic concept of using maths and testing to prove there are no difference. Their premise is that they can hear differences, so the proposition that it can be proved otherwise goes against what their ears are apparently telling them. In other words, in their minds it is subjective.

I was'nt referring to you. Unless you claim there is a difference.

Anyone remotely interested in music production should know that the ear isn't exact. And to argue it is more exact than maths is a bit daft don't you think ?

If one knows the least about scientific methods one knows that one does'nt start with the conclusion and try to find the proofs to back it up. One does the exact opposite.

[eduardo_b]

Anyone remotely interested in music production should know that the ear isn't exact. And to argue it is more exact than maths is a bit daft don't you think ?

Well, yes, but one must also recognize the quirks of human nature. For example, despite multitudes of carefully modeled A/B tests comparing audio cables, in which it was consistently shown that listeners simply could not reliably hear differences among them or in comparison with normal speaker wire, there are still plenty of customers for these overpriced cables. And they believe there are audible improvements.

Why they do so is open to question, but there it is. And, in the end, it doesn't really matter to those of us who don't feel the need to spend significantly more money for speaker cable or DAWs.

[jupiter8]

Anyone remotely interested in music production should know that the ear isn't exact. And to argue it is more exact than maths is a bit daft don't you think ?

Well, yes, but one must also recognize the quirks of human nature. For example, despite multitudes of carefully modeled A/B tests comparing audio cables, in which it was consistently shown that listeners simply could not reliably hear differences among them or in comparison with normal speaker wire, there are still plenty of customers for these overpriced cables. And they believe there are audible improvements.

Why they do so is open to question, but there it is. And, in the end, it doesn't really matter to those of us who don't feel the need to spend significantly more money for speaker cable or DAWs.

The difference between an AB test on speaker cables and host summing is that one can prove 100 % definitley sure there's no difference. There's absolutley no room for error there since it is digital.

An AB listening test can only be about 99.9 % sure depending on your stamina.

[eduardo_b]

The difference between an AB test on speaker cables and host summing is that one can prove 100 % definitley sure there's no difference. There's absolutley no room for error there since it is digital

So the opposition says, we don't believe host summing tells the whole story. The fact is, human beings are quite adept at ignoring science when it conflicts with other emotions and perceptions. If that weren't so, you wouldn't hear people trying to prove the earth is only 6000 years old or that global warming doesn't exist.

I think there are lots of reasons why hosts might seem to have different sound qualities, but I also think it has nothing to do with the hosts themselves but rather other unseen or unrecognized variables. Accepting this changes the argument from hosts to other things, but then again, the critical element is acceptance. Again.

[DWb]

No it isn't. You can easily prove there's no difference in sound quality between hosts. That's why it is so frustrating to us with at least half a brain to see this nonsense repeated ad nauseum.

Out of interest, is there a web page you can send people to where it explains in words of one syllable a) why this is theoretically the case and b) how they can check for themselves by making sure the hosts are set up the same[1], summing a bunch of wav files and then phase cancelling them. And maybe goes on to explain that yes there are differences between the host, but whether or not 1+1=2 is not one of them.

I hear people saying 'you should use Cubase not Floops because the summing engine is better' from time to time on other forums, but can never be arsed getting into a lengthy argument.

[1] and I'm guessing that this is the step that's going to be 'hardest', ie making sure that you've not got any panning law issues or any processing in the signal path in whatever hosts you're using.

[deep'n'dark]

I have made one test between Live and Reason by using same audio files.
http://www.freewebtown.com/nitrostatic/ ... alized.wav
http://www.freewebtown.com/nitrostatic/ ... 20mini.wav

[eduardo_b]

There's absolutley no room for error there since it is digital.

The other thing is that when people do test to compare DAWs, there's no way to know if the comparisons are correctly modeled and executed. If there are apparent audible differences, the reasons may have nothing at all to do with the hosts but rather with the comparison itself.

I used to work on a very large scientific journal, and the referees were constantly rejecting papers for invalid models and insufficient data. So comparison testing hosts is not likely to be scientific enough to offer meaningful (as opposed to anecdotal) results.

[Resonant Serpent]

So...

If I load fifty tracks of audio in a host A, then load fifty tracks of audio into host B, then render both projects to stereo, they will be exactly the same? This experiment, of course, would be done without plugs, and all settings at exactly the same for both hosts.

I did this experiment back when I was trying to decide between Sonar 5 and SX3. At the time I clearly preferred Sonar, and my spectrum analysis showed that the files were different.

Anyone have any proof one way or the other? I could understand the files coming out the same if there was only one track, but dozens of tracks?

When I've talked to host manufacturers, even they talk about their superior summing engines, and it's listed as a feature enhancement for certain host updates. I've even talked about it with professional mastering engineers, and they think there is a difference also.

While it may only be 1% difference, some want that 1%.

Does that 1% actually exist?

[jupiter8]

There's absolutley no room for error there since it is digital.

The other thing is that when people do test to compare DAWs, there's no way to know if the comparisons are correctly modeled and executed. If there are apparent audible differences, the reasons may have nothing at all to do with the hosts but rather with the comparison itself.

I used to work on a very large scientific journal, and the referees were constantly rejecting papers for invalid models and insufficient data. So comparison testing hosts is not likely to be scientific enough to offer meaningful (as opposed to anecdotal) results.

There is too a way to know if they did it right or not. It is not a matter of opinion or statistics. If 999 can't get their files to cancel out and 1 can it means one did the test correct and 999 screwed something up.
This is third grade maths we're talking about. You can hardly have an opinion wether 1-1=0 or not.

When I've talked to host manufacturers, even they talk about their superior summing engines, and it's listed as a feature enhancement for certain host updates.

You could you just point me to one mention of enhanced sound in any host. And upgrade to 64 bit does'nt count. Just one link is enough.

[eduardo_b]

While it may only be 1% difference, some want that 1%.

Does that 1% actually exist?

It's certainly possible it exists, and perhaps it may be audible if one closely examines the resulting files, but it's basically anecdotal observation versus digital processing. To me, the question is whether any of this matters. I say it doesn't, and certainly not before workflow and other features are evaluated.

A/B tests are interesting but really have no analogue in real life, in which music is played using an endless variety of electronic devices and speakers in all kinds of listening environments. There's nothing but variables.

This means nuances are altered or simply lost once the music leaves the system it's mixed on. I often wonder if this is widely recognized by those who are doing music at home. Every finished project is going to be a compromise in terms of how is sounds anywhere else.

So one could rightfully say that the host is irrelevant to the final product unless it's only heard on that system. For perfectionists, this could be bad news.

[Innominandum]

Well there are a lot of factors that could affect the sound quality of a host...

- Mixing (32-float, 64-float, 24-fixed).
- Internal resampling quality; this could be done many times in a single project if the user isn't careful.
- And probably most importantly; the included instruments and plug-ins.

I suppose the interface could affect the user's perception of the sound too.

[lotus2035]

honestly why do people debate the sound quality of a host?

...because they have nothing better to do... ...and that goes for both sides of the fence..

...at times in the past I have been sucked into the madness of these kinds of arguments but really whats the point?..you might as well do this :

[hibidy]

nuendo, 20-40% better sound engine

AND it's expensive, so it has to be good!

[bullshark]

you need, say, a Studer Vista.

I checked it out, but it doesn't go to 11 and has no glowing tube so I'll pass.