yorolpal
Yea...it's kinda like wanting to have a detailed understanding of quantum "membrane" theory before you go to Starbucks for fear you'll drive into another universe on the way there. Or is it?? :-)

You jest, but it's no laughing matter. I've done exactly that, except it wasn't Starbucks, it was Guitar Center.

mike_mccue
 
Sometimes I wonder...
 
How does a simple technical question turn into a series of pedantic and remedial lectures about gain staging?
 
 
Goofy.
Slightly Frustrating as it seems condescending.
Goofy.
 
 
all the best,
mike

Though I sympathize with you here Mike, I should point out that sometimes people might be doing their gain staging evangelizing and whatnot for potential newbs that might be lurking here.
 
But it is indeed off topic and does indeed come off as pedantic and condescending when it isn't made it clear that the information is addressed to newbs rather than experienced peers.

Hey everyone! :-)
 
Thank you for being so open minded about my posting.
 
I want you all to know that I value everyone who has taken part in this thread and that I am always eager to discover and learn stuff from you.
 
It is really helpful, for me, to be able to bounce ideas around with people that are so well experienced.
 
Thank you all very much!
 
all the best,
mike

drewfx1
 
Um, because in the real world it's already a moot point with 32 bits?
 
The truth is the only reason people think 32 bit is an issue is because some marketing folks used carefully worded language to imply that it is.
 
But I bet if you go back and carefully parse what they actually say, you'll see that they don't actually ever say that there's an audible difference - instead their very careful wording talks about errors in the abstract and leaves it to the reader to jump to the conclusion that those errors are a problem.
 
Now here's the question for you: If this stuff was really a problem, then why would they use that very careful, manipulative wording?

 
Folks made records with 16Bit digital recording.  Marketed to the public as being squeaky clean/clear/quiet and accurate  
With 16Bit audio, rounding error (from multiple generations of destructive processing) was certainly audible... and it sounded nasty.
 
If the extra resolution resulted in a significant performance hit... I could see the debate.  
When it comes at virtually no CPU load, I'll take the extra resolution... and never give summing accuracy another thought.   
 
 

Jim Roseberry
When it comes at virtually no CPU load, I'll take the extra resolution... and never give summing accuracy another thought.   

But you see this is the beauty of it - someone who's concerned that there might be a CPU hit under load can just do the reasonable thing and leave 64bit turned off because there's no real world impact on summing accuracy, and then they never have to give the potential of an adverse CPU impact another thought! 

I think its important to point out that although 64 bits give us a lot of freedom not all plugins handle overloads well. Bit and Jeff have a very important point about gain staging that applies to this point too. My view don't abuse the freedom. 

Hardly a new topic, Pro Fools have been in a dither over this for years and not much has really changed. For example:
 
http://repforums.prosoundweb.com/index.php/topic,6955.0.html
 
http://duc.avid.com/showthread.php?t=320440
 
There are actually valid architectural reasons why the choice of math impementation can differ very significantly between processes to be executed using a general-purpose CPU and processes to be executed using specialized DSP chips or a GPU or an FPGA, e.g.:
 
http://www.mpi-inf.mpg.de/~strzodka/projects/double/
 
Anyway, even though Sonar's audio engine has been able to employ 64-bit double precision floating point on 32- and 64-bit systems since at least Sonar 5 iirc (Ron Kuper days?), there are still places where audio data streams need to be converted bit-depth-wise and between fixed/integer and floating-point representations, so rather than fixating only on a mix engine's precision it is also rather important whether these necessary conversion operations are optimized for the hardware employed. See for example:
 
http://software.intel.com/en-us/articles/utilizing-intel-avx-with-cakewalk-sonar-x1
 
http://www.tomshardware.com/reviews/fx-8150-zambezi-bulldozer-990fx,3043-5.html
 
http://software.intel.com/sites/billboard/article/cakewalk-intel-and-windows-8-bring-high-performance-touch-enabled-mobile-workflows-musicians
 
Btw, for plug-in experimentation outside a DAW's mix engine, the freeware VSThost is available in both 64-bit double precision and 32-bit floating point versions:
 
http://www.hermannseib.com/english/vsthost.htm
 
Finally, what one hears may have nothing to do with a DAW's internal precision and everything to do with its dithering (or, in the case of the original SAW DAW software, lack of any dithering at all-- just do a search for Bob Lentini and Bob Katz and dithering and truncation).