bitflipper
Lynn, have you tried just turning the 64-bit engine off and listening?
 
Or, if you want to get scientific about it, export the full mix with and without the 64-bit engine enabled and do a blind ABX test.
 

Null test too?

if you don't touch your faders , you will notice nothing ;)

You can try and do a null test too. Render a project in 32-bit mode, then rerender in 64-bit more. Take those two, load in to a new project. Invert one, and play back. Any differences will show on the meter as well as the level. So if you get digital silence, then no difference. If you get stuff down in the -140dB range, then way below audibility, and below the noise floor of any DAC. If you get levels of -30dB (you won't) then there's a problem.
 
Realistically not only is 64-bit unneeded headroom in pretty much any realistic project, it isn't all that helpful since most plugins aren't double precision. Your synths, FX, etc are probably mostly single precision, hence they are all rounding everything to 32-bit.
 
I'm not saying don't take double precision when you can get it but I would so not worry about it.

Zo
if you don't touch your faders , you will notice nothing ;)

He's not kidding! : )

Well, I think 64 bits should be a good thing on x64 OS.
Yes you won't hear any difference.
But, today CPU use 64 bits processors with 64bits registers. Using 64 bits engine should have an influence on the way data is retreived in memory and "might" speed up a bit data access, at the expanse of memory size. 64 bit processors access memory using 64 bits chunks and when storing lower bit nbr, there's one more operation to perform to get the needed value.

Lynn

bitflipper
Lynn, have you tried just turning the 64-bit engine off and listening?
 
Or, if you want to get scientific about it, export the full mix with and without the 64-bit engine enabled and do a blind ABX test.

Dave, I'm going to download x3c again and give this a whirl.  I don't have golden ears, so I don't expect to hear any difference.  I'm sure CW will have this fixed in no time.

Therein lies the heart of the ongoing dilemma: the presumption that if we don't do everything just right and don't use the best gear, that someone else will hear shortcomings that we ourselves can't perceive.
 
Hence the ongoing forum questions: which interface is most accurate, what sample rate/bit depth to use, which reverb/compressor/limiter sounds best, what dither algorithm is better, which MP3 bitrate is acceptable, do all equalizers sound the same, does the 64-bit engine make a difference?
 
Whenever such queries are posed publicly (especially on Gearslutz), you can count on somebody replying that X made a "night and day difference", or that it was "like a veil being lifted". But think about it: if the differences are really so profound, then why do these questions repeatedly get asked in the first place? Because everybody fears that there are serious flaws in their gear and/or methodology that everybody else but them can hear.
 
Granted, with time and practice we do get better at listening, and everybody's hearing acuity is naturally a little different. Some are tone-deaf while others have perfect pitch, and high-frequency sensitivity drops with age and abuse. But the Golden Ear syndrome is largely a myth, or is at least irrelevant.
 
If you can't hear a difference between method X and method Y, try again. If you still can't hear it, then just let it go - chances are no one else can, either. 
 
 
 

Sycraft
You can try and do a null test too. Render a project in 32-bit mode, then rerender in 64-bit more. Take those two, load in to a new project. Invert one, and play back. Any differences will show on the meter as well as the level. So if you get digital silence, then no difference. If you get stuff down in the -140dB range, then way below audibility, and below the noise floor of any DAC. If you get levels of -30dB (you won't) then there's a problem.

 
You have to be careful here - if you have any random stuff going on, it won't null properly. So you need to turn off all dither and freeze any synths or FX doing random processing in your tracks. This might include anything with random modulation, synths with free running oscillators and such. The best way to test for this is to export using 64bit twice to verify that you get a perfect null that way first.
 
Also keep in mind that Sonar's meters only go to about -100 dBFS.
 

Realistically not only is 64-bit unneeded headroom in pretty much any realistic project, it isn't all that helpful since most plugins aren't double precision. Your synths, FX, etc are probably mostly single precision, hence they are all rounding everything to 32-bit.

 
In terms of headroom, 32 bit floating point is impossible to clip without intentionally trying very, very, very hard just to prove you can. In terms of precision, given competent programmers, plugins will do calculations at an appropriate bit depth internally regardless of Sonar's engine setting. Generally 64 bit double precision is appropriate mostly for recursive calculations (i.e. calculations that include a feedback loop where the errors might accumulate over thousands of iterations).

Use a cassette recorder you won't regret it...

bitflipper

Lynn

bitflipper
Lynn, have you tried just turning the 64-bit engine off and listening?
 
Or, if you want to get scientific about it, export the full mix with and without the 64-bit engine enabled and do a blind ABX test.

Dave, I'm going to download x3c again and give this a whirl.  I don't have golden ears, so I don't expect to hear any difference.  I'm sure CW will have this fixed in no time.

Therein lies the heart of the ongoing dilemma: the presumption that if we don't do everything just right and don't use the best gear, that someone else will hear shortcomings that we ourselves can't perceive.
 
Hence the ongoing forum questions: which interface is most accurate, what sample rate/bit depth to use, which reverb/compressor/limiter sounds best, what dither algorithm is better, which MP3 bitrate is acceptable, do all equalizers sound the same, does the 64-bit engine make a difference?
 
Whenever such queries are posed publicly (especially on Gearslutz), you can count on somebody replying that X made a "night and day difference", or that it was "like a veil being lifted". But think about it: if the differences are really so profound, then why do these questions repeatedly get asked in the first place? Because everybody fears that there are serious flaws in their gear and/or methodology that everybody else but them can hear.
 
Granted, with time and practice we do get better at listening, and everybody's hearing acuity is naturally a little different. Some are tone-deaf while others have perfect pitch, and high-frequency sensitivity drops with age and abuse. But the Golden Ear syndrome is largely a myth, or is at least irrelevant.
 
If you can't hear a difference between method X and method Y, try again. If you still can't hear it, then just let it go - chances are no one else can, either. 
 
 
 

It just confirm that many "nay say"' don't know anything what they talking about.
 
Actual you hear the different. And if you took the time and also made a null test/--->"phase test" you would also hear the errors between x32bit VS x64 double precision audio engine. That is why all majority of the plugins use x64 double precision "oversampling" today example UAD2, WAVES, NI, Spectrasonics, EastWest, Voxengo, etc..
 
 
Or we can be more scientific about it if you all like? Even the math back it up that it is a different.
 
It is a BIG difference even for those who think its just a "gimmick".
https://www.youtube.com/watch?v=-9EeW9WhNWA
 
Hope it help kill some of the myths out there.
Best Regards
Freddie

Freddie, perhaps you can explain the math to us and how the errors accumulate in your own words?