[feature request] 64bit floating audio engine

[majkel]

woow, guys discussion is going to conclusion -they all wrong but we right- because its only marketing.
in the real pro studios they use those 'bullshit' products made only with 'marketing' intend... and in yours opinion they all wrong..?
we talking about to equate bitwig to industry standard.
its a commercial daw (not cheap)and i have not a causes to not except from bitwig features which are in competetive products.
step forward we would like to see improved sampler, mixer, metering...
dont go in ableton live way... where they bet 'only' for marketing 'bullshit' and still have not a mixer, vst3 etc.
btw. im also about 20years in a computer based production .. but now we have a 2018.

[majkel]

[antic604]

woow, guys discussion is going to conclusion -they all wrong but we right- because its only marketing.
in the real pro studios they use those 'bullshit' products made only with 'marketing' intend... and in yours opinion they all wrong..?
we talking about to equate bitwig to industry standard.

If you think your music will be better because of 64-bit mixing, then by all means move to Reaper or Cubase (or wherever else). But in the end it's like trying to prove that Mona Lisa wouldn't be considered a masterpiece if one of the brush strokes was pointing in a different direction or there was a hair stuck in the paint... None of such deficiencies matter, when quality of your composition, the excitement of the arrangement and storytelling or the energy and emotion you're trying to convey work. I know that personally I prefer an imperfect, but soulful track to te pristinely clean & sculpted one, that doesn't speak to me... Also, why does it even matter if you have (practically) infinite headroom for summing, if along the way you'd apply "vintage modelled" plugins that emulate certain hardware with all their imperfection or mixing desks wih crosstalk, vinyl crackles, or any other type of warmth / sound coloration?

Today, ANY DAW gives you the tools to create great, professionally sounding music and if you can't achieve this by now, no 64-bit summing will help. I don't even think it matters outside of complex, live-recording setups like tracking a full orchestra, where there's certain way the things *should* sound. But who'd use Bitwig for that?

So I'd rather stick to the tools that inspire me because of their look, workflow and innovative features - like Bitwig. If numbers are so important for you, then use a spreadsheet

[Cyoon]

woow, guys discussion is going to conclusion -they all wrong but we right- because its only marketing.
in the real pro studios they use those 'bullshit' products made only with 'marketing' intend... and in yours opinion they all wrong..?
we talking about to equate bitwig to industry standard.
its a commercial daw (not cheap)and i have not a causes to not except from bitwig features which are in competetive products.
step forward we would like to see improved sampler, mixer, metering...
dont go in ableton live way... where they bet 'only' for marketing 'bullshit' and still have not a mixer, vst3 etc.
btw. im also about 20years in a computer based production .. but now we have a 2018.

64bit float might mathematically improve (reduce) rounding errors, but i doubt it has audible consequences, thus most of it is marketing hype imo.

still, this was an interesting read: https://www.audiorecording.me/advantage ... ation.html#

[sensej81]

Hello , i've been mixing music since 15 years professionally . I have used bitwig for about 2-3 years now.
And there is really something in the air when it comes to mixing quality in bitwig.
My mixes didn't sound that crisp and selective as it sounds in reaper which i used for few years now.
So i did a few comparisons just to prove myself is it something wrong with my perception or is it really a big difference between these two daws.

I have attached few screeshots just to prove that these daws isn't sound the same.

new project > 3 same samples > 4 instances of waves cla76 (same preset) >24bit wav render without dithering.
I did a null test (phase cancelation) in reaper using build in spectrum analyser.
I wasn't suprised that there is a difference.
Bitwig is loosing transients which i always heard in my mixes,they were really moody and increased in bottom end.

reaper bez szumu.jpg

bitwig bez szumu.jpg

comp.jpg

As you can see there is a difference when transients are hiting.
There's just 3 simple tracks with one plugin, imagine what would it be when you mix a 50 track project with 100-200 plugins inside.

[loungepanda]

Could you post the audio files? I want to hear this "moody" difference.

[Hez]

Hello , i've been mixing music since 15 years professionally . I have used bitwig for about 2-3 years now.
And there is really something in the air when it comes to mixing quality in bitwig.
My mixes didn't sound that crisp and selective as it sounds in reaper which i used for few years now.
So i did a few comparisons just to prove myself is it something wrong with my perception or is it really a big difference between these two daws.

I have attached few screeshots just to prove that these daws isn't sound the same.

new project > 3 same samples > 4 instances of waves cla76 (same preset) >24bit wav render without dithering.
I did a null test (phase cancelation) in reaper using build in spectrum analyser.
I wasn't suprised that there is a difference.
Bitwig is loosing transients which i always heard in my mixes,they were really moody and increased in bottom end.

As you can see there is a difference when transients are hiting.
There's just 3 simple tracks with one plugin, imagine what would it be when you mix a 50 track project with 100-200 plugins inside.

It's always important to test things like this because it improves your understanding of how things work under the hood, but it's also important to keep perspective on the results.

What bit depth did you export the Bitwig mix as? Did you dither it on export? Depending on the attributes of the exported Bitwig file, is Reaper performing sample rate conversion on the Bitwig clip to play it back in the Reaper project? Are there definitely no fades on the Bitwig clips? No time stretching?

If all of the above IS accounted for:

Can you hear the results of the null test at normal monitoring volume? Can you hear the results of the null test with your speakers on full volume? Have you properly blind-test AB'd the bounced Reaper mix and the bounced Bitwig mix, several times, outside Reaper (i.e. both as rendered WAVs) and if so, can you reliably identify which one is from Bitwig and which is from Reaper?

I don't know a whole lot about digital audio so I don't want to comment on the results you're showing (i.e. is that an expected level of residual noise resulting from different summing depths?), but you definitely need to make sure all the above is accounted for before assuming there is an audible difference between the summing engines.

I'm always interested in these posts because I have seen situations in the past where people with ridiculously acute ears have noticed an issue in certain processes etc. IIRC a guy on Gearslutz identified some crazy -60dB noise in the sub frequency spectrum of the initial Moog Sub37 release which turned out to be a legitimate issue that Moog subsequently fixed (may have got most of the facts wrong there but he was the only dude that seemed to be able to hear it out of a whole thread of people and he was right in the end). However, most of the time people haven't accounted for idiosyncrasies in the signal/process chain so aren't actually listening to what they think they are.

[majkel]

hmm.. a test which prove something but like as usually a fanboys will not see any difference

[Hez]

hmm.. a test which prove something but like as usually a fanboys will not see any difference

It's a thin line between fanboy & conspiracy theorist ; )

I don't think anybody is saying that there isn't a measurable difference between 64bit summing and 32bit summing using the correct tools, the debate is over whether that difference is audible (or detrimental, for that matter).

Analog summing is measurably 'worse' than digital summing (i.e. it introduces noise and distortion), but the general consensus currently is that people like a bit of noise and distortion when listening to music.

Maybe the -100dB 32 bit summing noise (if that indeed is what it is) that sensej81 has found is actually preferable to the 'cold, lifeless' digital world of 64-bit summing

But basically at the end of the day I'd rather Bitwig build me a nice phat new Sampler than introduce 64-bit summing, that's my only stance on this really.

[majkel]

when you read whole topic you will detuct thats a 'fanboys' was very very sure there is no any difference and its only marketing 'bullshit'
and now we have a measurably test.. we have a final conclusion ..
and what will be a next argue against this 'conspiracy theory'??? aahhh we have a next: no audible effect.
thats so poor argue because we are not talking about free - open source product but about commercial , expensive software with pro aspirations..
better sampler - yes
better audio engine - yes
and many more..

[Cyoon]

fanboy or lead dev..who knows

[JHernandez]

Waiting on an actual scientific test. I'd really like to know what's what.

[majkel]

the best test it, when you feels how hard is pushing forward your big project final mix in ableton or bitwig, and how easier its going with major Daw...but you need to basic knowledge about mixing audio,dynamic compression etc.
thats things what sensej81 was talking about.

[shadiradio]

You can't even make cuts (with no fades) and export a sample accurate result:
http://www.kvraudio.com/forum/viewtopic ... 9&t=482943

[Cyoon]

You can't even make cuts (with no fades) and export a sample accurate result:
http://www.kvraudio.com/forum/viewtopic ... 9&t=482943

do you imply that an objective test is not possible because of that or how does this relate to the topic?