Jump to content

Cakewalk and Reaper which one and Why?


Aloe Duke

Recommended Posts

2 hours ago, Maestro said:

 

In the past, it was common that some DAWs sounded difference because the audio engines were, then, less developed.  Some DAWs (like Samplitude, for example) were far more advanced than others (like Pro Tools, for example) in that area.  But, times have changed, and there is relative parity there, except - aforementioned - in cases where the DAWs use a processing chain designed to deliver a specific sound character.

And B.I.N.G.O just came out to play. 

This is it.

You've answer my arguement. This was the first thing that came to mind days before I entered this forum. Which i've said/asked many time in my replies.

Gibson might have never worked on this - "This" being the processing chain signal coding on their mix engine. 

Link to comment
Share on other sites

5 minute test then:

Loaded up a 1KHz sine wave in CbB, 1 second on, one second off, duration of 10 seconds.  MCompressor was strapped over the FX bin at the default settings except for the ratio that I set to 4:1 so you can hear it better. Exported this as a WAV.

Replicated this exactly in REAPER and loaded the WAV I exported in CbB and flipped the phase.

Results: The threshold kicks in exactly in the same place, and the difference between the two is consistent with what I saw with the rounding errors I saw above. If you're hearing anything different then it's all placebo.

EVERYONE's ears lie. It's entirely subjective. Null tests like this show us up for the fallible humans that we are.

  • Like 2
Link to comment
Share on other sites

2 hours ago, Lord Tim said:

Had a bit of time before a session tonight and I did a quick test.

NOTE: CbB Pan Law set to 0dB center, sin/cos taper, constant power. Internally I'm using 24 bit depth, not 32 bit or 64 bit double precision.

First we'll do some Self-null tests to ensure CbB isn't creating inconsistent results that don't null with itself, just as a control.

----------------

CbB Self-null test 1:

Input files: 
1KHz Sine - 24 bit, 44.1KHz stereo
White Noise - 24 bit, 44.1KHz stereo
Sine Sweep - 24 bit, 44.1KHz stereo
Drum Loop - 24 bit, 44.1KHz stereo

Output:
24 bit, 44.1KHz stereo files, all exported seprately.
Reimported back into CbB along with the original input file and the phase inverted

Results:
All files entirely nulled to silence.

----------------

CbB Self-null test 2:

Input files (all 24 bit, 44.1Khz mono):
Kick
Ride
Snare
Bass
Hats

Volume for each track set to -6db. Panning set to center.

Output: 24 bit, 44.1Khz stereo mixdown
Reimported this file, inverted the phase of this new track

Results:
Output meter showed -136dB maximum, or practically inaudible difference. I would expect that to be lower or 100% using 32 or 64 internal bit depth.

----------------

CbB Self-null test 3:

Input files (all 24 bit, 44.1Khz mono):
Kick (panned center)
Ride (panned 100% L)
Snare (panned 50% R)
Bass (panned 50% L)
Hats (panned 100% R)

Volume for each track set to -6db. Panning as above.

Output: 24 bit, 44.1Khz stereo mixdown
Reimported this file, inverted the phase of this new track

Results:
Output meter showed -136dB maximum, or practically inaudible difference. I would expect that to be lower or 100% using 32 or 64 internal bit depth.

----------------

CbB Self-null test 4:

Exactly the same setup as test 3, except MCharmVerb was in the FX bin on the kick and snare tracks with the same settings (see attached pic for settings).

Output: 24 bit, 44.1Khz stereo mixdown
Reimported this file, inverted the phase of this new track

Results:
Output meter showed -136dB maximum, or practically inaudible difference. I would expect that to be lower or 100% using 32 or 64 internal bit depth.

----------------

CONCLUSION:

CbB completely nulls with itself for all intents and purposes - as expected - even with MCharmVerb and panning applied.

------------------------------------------------

REAPER vs CbB Null test 1:

Input files: 
1KHz Sine - 24 bit, 44.1KHz stereo
White Noise - 24 bit, 44.1KHz stereo
Sine Sweep - 24 bit, 44.1KHz stereo
Drum Loop - 24 bit, 44.1KHz stereo

Output:
24 bit, 44.1KHz stereo files, all exported seprately.

Imported the CbB exports from CbB Test 1 for each respective file and inverted the phase.

Results:
All files nulled to silence.

----------------

REAPER vs CbB Null test 2:

Repeated CbB test 2 but inside REAPER, same source files, same settings.

Output: 24 bit, 44.1Khz stereo mixdown

Imported the CbB export from CbB Test 2 and inverted the phase.

Results: Output nulled to -48dB maximum. Not completely silent, but close to silent in a real-world situation. I would suggest this is due to rounding differences with the 24 bit file and the internal bit depth of REAPER mixing engine.

----------------

REAPER vs CbB Null test 3:

Repeated CbB test 3 but inside REAPER, same source files, same settings.

Output: 24 bit, 44.1Khz stereo mixdown

Imported the CbB export from CbB Test 3 and inverted the phase.

Results: Tracks panned away from center did NOT null with the default REAPER settings. Overriding the project defaults for each track to dual mono reduced the difference by 6db but the files did not entirely null, and no settings would achieve this.

----------------

REAPER vs CbB Null test 4:

Repeated CbB test 4 but inside REAPER, same source files, same settings.

Output: 24 bit, 44.1Khz stereo mixdown

Imported the CbB export from CbB Test 4 and inverted the phase.

Results: Obviously this will not null as per Test 3 because the panned tracks weren't correctly nulling due to the differing pan laws. Reverb decay did not null on the stereo sides, in a similar way to regular tracks panned off center did not null.

----------------

CONCLUSION:

When tracks are panned center CbB nulls either completely or very close to 100% with REAPER. The minor difference betwen multitrack mixes can be down to rounding (ie: the exported files were 24 bit / 44.1Khz vs an internal 32 bit or 64 bit mixing engine, so the more tracks imported, the more there will be differences as they're mixed together). In modern pop/rock/metal/hiphop/etc music, apart from very quiet sections or the end of reverb tails, in real-world use, this is going to be practically inaudible.

When tracks are panned off-center, the results don't null for any tracks not panned to center. This is due to differing pan laws. This will be audible for those tracks only, even in a modern mix, and can be compensated to match by raising or lowering the track volume.

When there are stereo effects added, the results don't null for the amount of sound on the stereo part of the material. For time based effects, this can make the tails seem either louder or quieter depending on the settings and the width of the effect, and will definitely be audible in a modern mix.

------------------------------------------------

Bonus test:

Importing the exports from CbB and REAPER into Adobe Audition matched the results seen above.

----------------

My take?

I stand by what I said about pan laws being the biggest difference between how these DAWs sound. That doesn't make either one sound better, but just slightly different. The more tracks, the more stuff is panned, the more differences there will be. If I had more time I would repeat this with 64 bit settings to rule out rounding errors with the 24 bit stuff I was using, so don't take this as anything like a definitive scientific test. I'm sure others here could do it better.

Ultimately, if you're finding that one DAW is dramatically better than the other then more power to you - use that DAW. If having slightly louder reverbs and some tracks a little differently balanced in REAPER when you're doing an exact apples to apples comparison with CbB, you'd probably get more benefit by actually listening to what you're mixing rather than comparing the differences between each DAW. Can you hear that -50dB track difference? I doubt it. Is a slightly louder reverb going to sound different? Sure. Will that negatively impact your mix?  If it does, perhaps practicing mixing more than splitting hairs might be a good option.

 

Reverb settings as used above:

1178338196_ReverbSettings.jpg.468914f894e2a4d9df94c18d1f6269ce.jpg

I dont know. 

Link to comment
Share on other sites

14 minutes ago, Will_Kaydo said:

 

So basically what I'm asking, is - Does this have to do with the headroom differences between the two? Cakewalk db boost goes up to +6db, where reaper is at 12db boost.  

 

So why would I start a DAW war when it's my Go-To DAW. If the Mix engine are still Digital based - than obviously things will be unique in their own way. Though, I cannot ignore the fact that hearing is in fact true. My ears don't lie 😂

Hmmm....somebody correct me if I am wrong....

Whatever headroom there is I assume is adjusted for from start of a daw project. If you have +6 dB headroom, you from start adjusted down all levels 6 dB from start, or you would have nowhere to go when boosting. 0 dBFs is ceiling in the end, now adjusted to be -6 dBFs.

But is plugin aware of this - I think it sense whatever comes in from clip and hits it - but in fact is 6 dB adjusted.

Same if +12 dB headroom, that all numbers(or what to call levels) are adjusted down internally to fit 0 dBFs + 12 dB.

 

This is a theory, sombody that knows chime in.

I remember running Cubase, there was a remark in manual for a version about these levels and how they would be handled since sounding louder.

From Cubase manual

"Volume Max
Allows you to specify the maximum fader level. By default, this is set to +12 dB. If you
load projects that were created with Cubase versions older than 5.5, this value is set
to the old default value of +6 dB."

 

Another thing I noticed running some of Waves hardware emulating plugins is that they did not like to be hit too hard from clip, they were like hardware made to be hit with -18 dBFs peaks as I recall. I have to adjust not too hot clips, or just use Gain knob in console. If you were still well within 0 dBFs peaks was not enough. I got overs all the time and no presets were even close to work, had to turn down input level to zero just about.

 

Sounding different, could it be here some plugins are more sensitive which levels they are hit with?

In particular if emulating analog where level significantly affect what harmonics you get.

Some plugin overlooked this adjustment daws make with different headroom.

Correct me if I am wrong.....

Link to comment
Share on other sites

1 hour ago, Will_Kaydo said:

The reverb sounded more darker projecting a "fuller" sound - in Cakewalk there's a roll-off in the low end of the reverb and boost in highend giving it a thinner-but-shimmering "aka" gloss/air sound to it. Compression are more sensitive. 

So basically what I'm asking, is - Does this have to do with the headroom differences between the two? Cakewalk db boost goes up to +6db, where reaper is at 12db boost.  

Maybe something like this could provide you with an answer: Gainmatch (I discovered it through a feedback request post from feedback request post from Niko Panunggal)

The website lists several common problems:

  • Does It Sound Better, Or Just Louder?
  • Have you ever seen a plugin that sounds so much better when you put it on?
    And later discovered that it just adds 1 dB of gain on everything?
  • Or maybe a new compressor, that has a fixed 2 dB steps output knob?
    So that comparing before and after is a pain?
  • Or maybe a saturation plugin with no output gain, so that it's almost unusable?

   Seems to me that it might address many of the discussed issues in this topic

Edited by Teegarden
Link to comment
Share on other sites

1 hour ago, LarsF said:

Hmmm....somebody correct me if I am wrong....

Whatever headroom there is I assume is adjusted for from start of a daw project. If you have +6 dB headroom, you from start adjusted down all levels 6 dB from start, or you would have nowhere to go when boosting. 0 dBFs is ceiling in the end, now adjusted to be -6 dBFs.

From 0db up to the highest DB on the meter reading. In cakewalk this goes up to 6db - that's your headroom. 

Edited by Will_Kaydo
Link to comment
Share on other sites

1 hour ago, Will_Kaydo said:

From 0db up to the highest DB on the meter reading. In cakewalk this goes up to 6db - that's your headroom. 

Yes, and it's not just about setting a number on the faders or meters - internally you have to adjust since digital has a max number limit.

12 dB that is 4 times up from 0 dBFs. So number cranking inside has to handle this.

If you have a 24-bit file of samples, that peak at 0dBFs(meaning integers inside are +/- 8 miljon), and load that together with others you cannot boost that 6 dB and store again(it would be +/- 16 miljon). So some clever processing is going on how you treat it. My assumption was that it is treated as being +/- 4 miljon(down 6 dB if headroom is 6 dB), which leaves room for 6 dB boost without overs.

I described what I think they do internally - and how it might affect what goes into plugins as well - which could change sound. Especially if some non-linear thing is going on.

 

But as many stated - just having a loudness difference of 1-2 dB is enough for ears to perceive differently in frequency content. So my theory might be off regarding headroom thingy.....

Edited by LarsF
Link to comment
Share on other sites

1 hour ago, pwalpwal said:

i think this stuff only becomes an issue when you start comparing daws... "if it sounds right it is right" - -if you prefer the sound of one over another, use the one you prefer :) life's too short

But, but, but this is the internet. It was born so folks could hide out in the basement and argue about the little of stuff and call people names ;)
Like I want to use Reaper but I'm afraid of "The Reaper"  :D  

  • Like 2
  • Haha 1
Link to comment
Share on other sites

CONCLUSION: (Quoted from Lord Tim)

When tracks are panned center CbB nulls either completely or very close to 100% with REAPER. The minor difference betwen multitrack mixes can be down to rounding (ie: the exported files were 24 bit / 44.1Khz vs an internal 32 bit or 64 bit mixing engine, so the more tracks imported, the more there will be differences as they're mixed together). In modern pop/rock/metal/hiphop/etc music, apart from very quiet sections or the end of reverb tails, in real-world use, this is going to be practically inaudible.

This then could be the reason for the discrepancy between how my large orchestral project sounded in Reaper vs. CbB. when you're doing 50 or so midi tracks, those small differences can really add up.

Link to comment
Share on other sites

Yeah, they absolutely do! :) This is kind of how the Console Emulator works in the Pro Channel in CbB. By itself it seems like it does nothing, but throw that over 30 or 40 tracks and you really start to notice things sounding different - the subtle distortion, especially on panned tracks really can widen a mix out. It's kind of the idea behind Harrison Mixbus, in fact - that's made to sound like CbB does with Console Emulator engaged on every track by default.

Now I'm not saying REAPER has console emulation or anything like that running, but just the subtle gain differences in both the rounding and the pan law can really add up over a couple of dozen tracks, so I'm not surprised to hear that. :) Of course, it's rudimentary to match that in CbB just by adjusting track volumes, etc. so it's not inherently better as such, just different.

 

  • Like 2
Link to comment
Share on other sites

Really if I was younger with more brain cells I would want to tackle Reaper more in depth for all it has to offer. But alas the little free time I have I rather actually be a musician and not an audio engineer.  Reaper is one heck of a DAW. Mean and lean is Reaper and I have to give Justin and company credit but it doesn't seem to click for me.

Link to comment
Share on other sites

Yep, agree. I've lurked on the REAPER forums since the very beginning and seen it grow over the years into a truly fantastic DAW. It absolutely does NOT click with me at all, though! HAHA! I spend more time just wondering why things don't work how I feel like they should rather than doing any actual work on it - and in REAPER's defence, that's *my* failing to a degree, there's lots of people who work super fast on it, but I'm definitely not one of them. CbB just fits how I like to work.

Link to comment
Share on other sites

3 hours ago, Will_Kaydo said:

This is lost in argument.

I don't even know what my original question was anymore. 

This was your argument:

"Am I mistakenly to find that Reaper projects the sound differently than Cakewalk? My plugins sounds way fuller in Reaper. I really don't know how to explain it - but have anyone experience something like this? I hear compression better, the plugins are more sensitive there as what it is in CbB. The same with a Reverb - it sounds more true and natural than what it does in CbB. This is the same for every plugin. 

I use the same plugins in both and in CbB and there's a highly noticeable difference."

I merely pointed you to a possible tool that could help you figure out what is really going on i.e. different DAWs have different settings which also goes for the plugins in those DAWs.

Just try a tool like this (there are similar tools as well) and maybe the other DAWs and the plugins in those DAWs don't sound so different after all. And this kind of tool seems to help getting faster and better mixing results.

Link to comment
Share on other sites

3 hours ago, mdiemer said:

CONCLUSION: (Quoted from Lord Tim)

When tracks are panned center CbB nulls either completely or very close to 100% with REAPER. The minor difference betwen multitrack mixes can be down to rounding (ie: the exported files were 24 bit / 44.1Khz vs an internal 32 bit or 64 bit mixing engine, so the more tracks imported, the more there will be differences as they're mixed together). In modern pop/rock/metal/hiphop/etc music, apart from very quiet sections or the end of reverb tails, in real-world use, this is going to be practically inaudible.

This then could be the reason for the discrepancy between how my large orchestral project sounded in Reaper vs. CbB. when you're doing 50 or so midi tracks, those small differences can really add up.

Not knowing the exact amount CbB attenuates by, but could it be the difference (when adding up multiple tracks) between 3dB attenuation and a more accurate 3.01dB?

db = 20 * log10(amplitude)

But it can be cheaper (in terms of CPU) to calculate 6.02 * log2(amplitude), where 6.02 ~= 20 * log10(2)) *

* log10(2) = 0.301029996, 20 x log10(2) = 6.020599913

Or different tapers on the pan law (remember CbB - and SONAR before it - can do sin/cos as well as square root)?

  • Like 1
Link to comment
Share on other sites

13 hours ago, Will_Kaydo said:

Yep. I always do this. 

😅 I've checked for these too. I have my fair share of knowledge. 

Try it yourself.


As you can guess, when I was writing ReaCWP I was also checking the result. So on my own I always could Null (down to reasonable level) whatever I have thought has to null.

Your question can be solved scientific way: upload 2 projects, so CWP and RPP, with the same example audio and free compressor which sound/does something differently. And let people explain the difference.

If you can (have time/internet/have the place), also render and upload the results which are different. It can be something apart from projects producing that (but don't be surprised rendered output is different from what you hear, in that case please try to "render" playback using audio loopback).

Just to make it clear. I am convinced plug-ins are working differently for you,  it is not imagination nor "0.1dB". I simply try to pin from where it comes, in case you are interested. But there are so many variables that guessing in the thread is not productive.

 

Link to comment
Share on other sites

3 hours ago, Kevin Perry said:

Not knowing the exact amount CbB attenuates by, but could it be the difference (when adding up multiple tracks) between 3dB attenuation and a more accurate 3.01dB?

db = 20 * log10(amplitude)

But it can be cheaper (in terms of CPU) to calculate 6.02 * log2(amplitude), where 6.02 ~= 20 * log10(2)) *

* log10(2) = 0.301029996, 20 x log10(2) = 6.020599913

Or different tapers on the pan law (remember CbB - and SONAR before it - can do sin/cos as well as square root)?

Uh, okay, if you say so. (In other words, this is way over my head).

Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...