Digital Audio and SAW

[Burkeville]

Hi Bob and whoever can explain this to me,
Someone asked me why I thought SAWStudio sounded better than other workstations out there and I did not have a technical answer. I know what my ears tell me but I don't understand about floating point math, etc.
Can someone explain this to me?
Thank you

[Bob L]

I settled on integer math 12 years ago for all of SAW's processing for two main reasons... better performance and I personally thought the reults sounded better.

The sound can definitley be argued as a subjective matter, but its pretty obvious about the performance. :)

Today, SAWStudio's engine is still all integer math, except for the EQ algorithm, which is using special hand coded assembly language floating point. The performance and audio integrity has been maintained by the way the algorithm has been coded.

Here is a good article explaining some issues about floating point math in computers:

www.rane.com/pdf/note153.pdf (http://www.rane.com/pdf/note153.pdf)

In the end... we all have to make our own decisions and choices... use your ears... put away the numbers and the details of articles and just listen and choose for yourselves.

Bob L

[matt]

As far as I know there is another factor involved. Programs like Sonar, Nuendo, Logic... use a form of data compression (or destruction) to achieve faster playback. Ever notice how when you loop in the multitrack in other apps that it is seemless. SAW on the other hand has a short delay, but it gives you the straight goods. In my opinion, looping the multitrack is not nearly as important as sound quality. Think about what was involved making a loop with 2" :p
Bob, is this floating point math the "compression" I have referenced?

[Yura]

Programs like Sonar, Nuendo, Logic... use a form of data compression (or destruction) to achieve faster playback.

It is VERY interesting......
is this true? so, where are proofs of that?
in other side, this is not well corresponds to your words, because, realy,
we observe an evident "SLOWER Playback" ( performance) of pointed programs in compare to SAW.

[AudioAstronomer]

Other hosts to my knowledge write a temporary file to disk on the first playback of the loop, then open that temp file from start when beginning the loop over. This is how it becomes seamless.

As to other hosts methods of data manipulation... these is indeed clear fact some sonically disturbing things are done. But I know anyone using that host will say to their death you cant "Really hear it". Sure -110 noise floor for the faders is great... till you have 24 tracks with some gain up here and there.. post fader compressors etc.... Man I could really go on with these discussions. I don tthink it necassary for the end user though.

Burkville, if you would like to chat on the subject I can give you a very detailed look at how other hosts work and you can get your information on saw here... Now if this makes one host "sound" better, I have my own opinions, and others have theres. I do not wish to discuss most of it here, because this is a quick way to start endless forum arguements. There are certainly things many different hosts do differently... its up to the user to decide if this makes a difference to them... even if there's technically a "negative" difference, some people love it... just like how I absolutely adore noise :)

you can email me @ saturn9@tampabay.rr.com or contact me on AIM: The Found Vertex. If you want, after 9pm EST Im always up for a chat on the phone, free long distance here.

:)

[matt]

AudioAstronomer,

Please share with the rest of us. Any in-depth info about the other apps could benefit us all! I personally would like to know why I wasted countless hours trying to make a semi-pro app give me a professional product. :D
I wouldn't worry about starting a never-ending debate on the subject because that's what we are here for. Discovery.

[AudioAstronomer]

I personally think this is a better subject for realtime communication. In even the msot friendly forums Ive seen this turn into a huge debate. The people with the facts state the facts... the people with the insecurities begin to spout out why the facts cant be true because their software is really really cool.

Maybe an AIM chat, trip to the pub, conference call... only way Ill do it :) You can chat with me on AIM almost anytime. Jesse and I had a wonderful time chatting away one night, it's fun :)

[Bob L]

Robert,

I appreciate your position on not wanting to blast out all kinds of things about all kinds of apps and the way they do things.

I feel users can download demos and decide for themselves if they like one app's methods better than another... SAWStudio included.

Thank you for attempting to keep this forum clear of such discussions that invariably step on one application over another.

We are here to talk about how SAWStudio works and how you can get the most out of it's design... hopefully you end up liking the design enough to purchase and then spread the word to others.

Bob L

[AudioAstronomer]

Agreed. I am glad to speak with anyone on the subject of other DAW's as I have spoke with nearly every developer, and being a dev myself I know many of the obstacles... Saw is entirely way above my head still, concepts used here are way beyond anything I, or the entire audio industry has yet to touch on. Perhaps in another year or so of intense discussions here, and research resulting from the discussions I may feel comfortable speaking of SAW in a more intimate manner. For now I can only offer my knowledge on hosts that use standard/quasi-standard techniques and various intricate differences therein.... Once again, not on the forum dedicated to sawstudio :) I really do encourage people to chat with me! Not as much for me to blab, but I am honestly hungering after any knowledge that can be gained by 2 heads gaining understanding of a concept simultaneously. Just look what happens here when 2 people suddenly understand a concept.. then stat bouncing ideas off each other and... and...

anyways. these things excite me, it can be hard to contain ones self. :)

[Carl G.]

I settled on integer math 12 years ago for all of SAW's processing for two main reasons... better performance and I personally thought the reults sounded better. www.rane.com/pdf/note153.pdf (http://www.rane.com/pdf/note153.pdf)

Superior!
Excellent paper.
It's clear the performance is there and the quality.
Carl

[AudioAstronomer]

James Moorer's paper on the subject is a better read imho. The Rane articles tries to summarize his article by sucking out the "mumbo jumbo", which I think for anyone with decent math skills is very interesting :)

[Naturally Digital]

James Moorer's paper on the subject is a better read imho. The Rane articles tries to summarize his article by sucking out the "mumbo jumbo", which I think for anyone with decent math skills is very interesting :)

Agreed.
Dave.