Go Back   Cockos Incorporated Forums > REAPER Forums > REAPER General Discussion Forum

Reply
 
Thread Tools Display Modes
Old 05-07-2012, 09:39 PM   #1
Lucero del Alba
Human being with feelings
 
Join Date: Apr 2012
Location: Buenos Aires, Argentina
Posts: 8
Default REAPER sum: mixdown render vs bus routing recording

Hi,

I've heard several Pro Tools engineers recommending you should rather bus all of your steams (drums, bass, guitars, ...) to a mix bus before hitting the master bus, and then use that mix bus to record your mixdowns in order to achieve better results, when compared to what you get when you bounce/render. I'm fairly new to REAPER (and to DAWs actually, and to music production, and to... well, who cares anyway) and I was wondering how you guys deal with summing on REAPER to get the best out of it, or say, to get the most "transparent" mixdown, or say, the most similar to what you hear while mixing.

Here's what I noticed on REAPER.

While trying different compressors on drums I decided to render some mixes to import them as different takes on a "mixes" track for better A/B'ing (I know you can freeze tracks, or snapshot the whole mix with SWS, but this worked better for me here). It didn't take long before I noticed the renders where not _exactly_ the same to what I was monitoring. So I decided to route the parent folder of the drums in which I was inserting the compressors, and the reverb of the snare to an armed mix track that I set Record: output (stereo), and then I recorded there these drums mixdows, which I was at the same time comparing against the renders I've previously made.

First thing I notice is that even before turning any compressor on, the waveforms are slightly bigger for the renders than for the routed recording, and the metering shows some .1 ~ .4 dB difference, in favor of renders; it's certainly not much but what is more interesting (or baffling) is that renders and routed mixdowns recorded in-the-box sound different! I compared 5 compressors, and it seems to me that renders were in general kind of boxy on the snare tail (release), so the routed mixdowns sounded better to me for this reason... though not always! ¡¿?! We are, of course, talking about surgical sound differences, not day/night things, but still noticeable.

Could be related to REAPER's lack of VCA's (DCA's, if you may) support? Would using JS plugins vca_master and vca_slave plusing to route as VCA steams would make any difference? (notice below I was carefully routing the reverb as well). Mind you, my track setup was like this:

Code:
Drums [with a 0dB send to a mix track only when recording mixdowns]
 ├ MIDI [w/multichannel routing to the tracks below]
 ├ Kick (folder w/o audio)
 │  ├ in
 │  ├ out
 │  └ sub
 ├ Snare (folder w/o audio)
 │  ├ top
 │  ├ bottom [with a send to reverb]
 │  └ 1176
 ├ Hats (audio)
 ├ Toms (folder w/o audio)
 │  ├ rt1
 │  └ rt2
 ├ Overheads (audio)
 └ Ambient (folder w/o audio)
    └ close
Reverb [with a 0dB send to a mix track only when recording mixdowns]
Also, I'm both recording and rendering 88.2 kHz 24-bit FLAC files (no resampling nor dithering at any point since it's the both the soundcard and session setting).

Hell, I have no idea. I'm pretty confident I'm not taking into account something relevant... hopefully here's where some pro gets in to save the day, and to share what's best for A/B'ing complex scenarios, and to mix down! And seriously, that stuff sounds different! How come??
Lucero del Alba is offline   Reply With Quote
Old 05-07-2012, 09:51 PM   #2
FarBeyondMetal
Human being with feelings
 
Join Date: Dec 2007
Location: Los Angeles, CA
Posts: 390
Default

When you render, are you using the same sample rate and bit depth as your project? Also, are you rendering to WAV?

There theoretically shouldn't be an audible difference but I've never tested this myself so I am definitely interested.

EDIT: Just read the end of your post. I am out of ideas...
FarBeyondMetal is offline   Reply With Quote
Old 05-08-2012, 10:25 AM   #3
Lucero del Alba
Human being with feelings
 
Join Date: Apr 2012
Location: Buenos Aires, Argentina
Posts: 8
Default

Hi FarBeyondMetal,

I re-run everything without the reverb this time, and I see no coloration like last time, that is no "boxyness" on the snare tail anymore; which leads me to think that the internal routing for the reverb was different when rendering vs recording on the setup I made, very surely a mistake I'm making somewhere (I have the feeling VCAs would have prevented it anyway).

I should have also taken into account that I was using Superior Drummer 2, which has a pool of random samples on drums, meaning that you can expect some slight differences on levels and even sonic performance if you look in very detail; having said that, I do notice higher levels *on average* for renders when compered to routed recording to a track, which I hardly hear anywhere but on the kick, and weirdly enough I seem to notice more fullness on the bottom end (I insist, extremely slight but noticeable enough) on recorded tracks (which are actually less loud!) than on rendered files.

In short, still lost. Any comments on how to properly deal these scenarios (that is, rendering vs routing to a recording armed track) would be greatly appreciated.
Lucero del Alba is offline   Reply With Quote
Old 05-08-2012, 10:35 AM   #4
Link0126
Human being with feelings
 
Link0126's Avatar
 
Join Date: Jul 2010
Posts: 1,268
Default

Make sure both tracks are perfectly lined up. Reverse the phase of one. See if they null. If not, you will hear the difference.
Link0126 is offline   Reply With Quote
Old 05-08-2012, 10:38 AM   #5
sonnie
Human being with feelings
 
sonnie's Avatar
 
Join Date: Jan 2010
Posts: 652
Default

Can you upload these different tracks?
Do you record within Reaper, or via ReWire/Soundflower/Jack into any other DAW?
sonnie is offline   Reply With Quote
Old 05-08-2012, 11:13 AM   #6
Mercado_Negro
Moderator
 
Mercado_Negro's Avatar
 
Join Date: Aug 2007
Location: Caracas, Venezuela
Posts: 8,676
Default

I'll give this a shot later but the first suspicious things I see are:

FLAC (there is some coding/encoding going on which may affect results/levels)
I think the lack of VCA groups has nothing to do with this
There are reverbs involved, things won't sound the same
__________________
Pressure is what turns coal into diamonds - Michael a.k.a. Runaway
Mercado_Negro is offline   Reply With Quote
Old 05-08-2012, 11:24 AM   #7
Lucero del Alba
Human being with feelings
 
Join Date: Apr 2012
Location: Buenos Aires, Argentina
Posts: 8
Default

Quote:
Originally Posted by Link0126 View Post
Make sure both tracks are perfectly lined up. Reverse the phase of one. See if they null. If not, you will hear the difference.
I do indeed, they cancel each other big time but not completely when I reverse the phase of a duplicated track; however, apart from this gain slight difference there are also random samples on every mixdown because the drum software uses random samples, so it's not really possible to cancel completely, not even on two identical setups. I'll circumvent this out by consolidating the drum set and giving it another try.

Quote:
Originally Posted by sonnie View Post
Can you upload these different tracks?
Do you record within Reaper, or via ReWire/Soundflower/Jack into any other DAW?
Let me give it another try with a consolidated drum set this time, both with FLAC and WAV files, and I'll upload it. Also, I record in REAPER yes -- I got the VSTi drum software inserted on a multichannel track with MIDI, and that routed to different tracks as explained, no need for ReWire.

Quote:
Originally Posted by Mercado_Negro View Post
I'll give this a shot later but the first suspicious things I see are:

FLAC (there is some coding/encoding going on which may affect results/levels)
I think the lack of VCA groups has nothing to do with this
There are reverbs involved, things won't sound the same
I see. If that's the case with FLAC it's good keep that in mind, especially after normalizing and while mastering with little headroom. Still clueless anyway about the bottom end response on the kick.


Thanks for your replies guys, I'll give it another try and get back later.
Lucero del Alba is offline   Reply With Quote
Old 05-08-2012, 01:18 PM   #8
Banned
Human being with feelings
 
Banned's Avatar
 
Join Date: Mar 2008
Location: Unwired (probably in the proximity of Amsterdam)
Posts: 4,868
Default

I posted an example of how to get REAPER to have a VCA-like mixing feature. You may want to try that to see if it makes any difference. My guess: it won't.
__________________
˙lɐd 'ʎɐʍ ƃuoɹʍ ǝɥʇ ǝɔıʌǝp ʇɐɥʇ ƃuıploɥ ǝɹ,noʎ
Banned is offline   Reply With Quote
Old 05-08-2012, 01:21 PM   #9
Gerry P
Human being with feelings
 
Gerry P's Avatar
 
Join Date: Jan 2009
Location: Ontario, Canada
Posts: 1,459
Default

Not sure why you started with FLAC as your standard...but I'm anxious to hear your latest results...
__________________
The future ain't what it used to be. Yogi Berra
Gerry P is offline   Reply With Quote
Old 05-08-2012, 02:04 PM   #10
Lucero del Alba
Human being with feelings
 
Join Date: Apr 2012
Location: Buenos Aires, Argentina
Posts: 8
Default

I figured out!

Recording through a routed track vs rendering is just the same in REAPER (and it should be on any other DAW since digital summing is not magic voltages)... and if you do it right, which I didn't, and here's why.

So I had this analog warmer subtly working on the master tack, and the fx was being printed every time I was rendering (since what you render is the master), but it wasn't being printed when I recorded routed tracks since the track I was recording on wasn't YET hitting the inserts at the master channel. In other words:

Code:
Master with FX [what you get when rendering]
 ├ armed track [where I was recording]
 ├ drum set [routed to the armed track]
 │  ├ kick
 │  ├ snare
 │  ├ ...
So, I moved the FX on the master to the armed track et voilà! -- complete cancellation between rendered and recorded files on the armed track.

While testing it I also verified that FLAC and WAV renders/recordings completely cancel each other. FLAC decoding doesn't affect levels in any way whatsoever (which makes sense, 'cause FLAC is lossless by definition, and decoding a FLAC these days is pretty easy on the CPU so it shouldn't interfere with the processing; but it's cool to "prove" it and be with absolute peace of mind while working with FLAC on a pro level though).

It was also suggested above -if I understood it right- that if you use reverbs you won't accomplish same results every time, therefore there would be complete no phase cancellation; but truth is that any digital process that doesn't have a random seed behaves the same every time unless you touch the parameters (just as every single digital reverb I would guess, and even every single "vintage" reverb that's still a piece of software). This wasn't the case with Superior Drummer 2 since it randomly picks samples when playing, just like many drum software that do in order to prevent the "machine gun" effect specially when you program fills.


So, I started this topic asking about the convenience of routing vs rendering, and I just realized that if you pay attention to your routing scheme is just the same... so what's the point in routing all that crap while rendering is faster anyway (and who's buying that Pro Tools summing is better 'cause they do it in real time, yeah real time summing on the digital world my ass).

Finally, remember I said "this thing has more bottom end" when recorded? Well, the vintage warmer was actually "warming out" the kick (kind of awkward, but well).

Sorry for the abuse of bold, but I feel great about understanding and demystifying this to myself, hopefully it will help somebody else!

edit: and my bad about the VCAs! I though that the boxy snare tail could be related of the lacking of the possibility to link send/returns with the level of the tracks you're sending from (which you can safely do with VCA faders) and that tha was something affecting the routing.

Last edited by Lucero del Alba; 05-08-2012 at 02:09 PM.
Lucero del Alba is offline   Reply With Quote
Old 05-08-2012, 02:41 PM   #11
Banned
Human being with feelings
 
Banned's Avatar
 
Join Date: Mar 2008
Location: Unwired (probably in the proximity of Amsterdam)
Posts: 4,868
Default

So, in conclusion: you were probably just hearing PT users talk about how to use PT to avoid some of its specific issues (e.g. internal resolution limits), which are not relevant to REAPER.

Feel free to check out that VCA mixing groups example nevertheless. (I'm still hoping to get a bit more feedback on that, especially on whether it affects send/receive levels correctly.)
__________________
˙lɐd 'ʎɐʍ ƃuoɹʍ ǝɥʇ ǝɔıʌǝp ʇɐɥʇ ƃuıploɥ ǝɹ,noʎ
Banned is offline   Reply With Quote
Old 05-08-2012, 03:44 PM   #12
plush2
Human being with feelings
 
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,110
Default

You will do well here Lucero del Alba. Welcome and all the best on future explorations. The ways in which Reaper struggles compared to PT are well documented here...maybe unfortunately, the many ways in which Reaper exceeds PT are not as well documented. After all, we're a user group, not a PR department.
plush2 is offline   Reply With Quote
Old 05-08-2012, 04:36 PM   #13
Mercado_Negro
Moderator
 
Mercado_Negro's Avatar
 
Join Date: Aug 2007
Location: Caracas, Venezuela
Posts: 8,676
Default

Quote:
Originally Posted by Lucero del Alba View Post
While testing it I also verified that FLAC and WAV renders/recordings completely cancel each other. FLAC decoding doesn't affect levels in any way whatsoever (which makes sense, 'cause FLAC is lossless by definition, and decoding a FLAC these days is pretty easy on the CPU so it shouldn't interfere with the processing; but it's cool to "prove" it and be with absolute peace of mind while working with FLAC on a pro level though).
Good to know. I knew it's a lossless format but I wasn't sure about its decoding process and possible side-effects. Thanks for checking that out.

Quote:
Originally Posted by Lucero del Alba View Post
It was also suggested above -if I understood it right- that if you use reverbs you won't accomplish same results every time, therefore there would be complete no phase cancellation; but truth is that any digital process that doesn't have a random seed behaves the same every time unless you touch the parameters (just as every single digital reverb I would guess, and even every single "vintage" reverb that's still a piece of software). This wasn't the case with Superior Drummer 2 since it randomly picks samples when playing, just like many drum software that do in order to prevent the "machine gun" effect specially when you program fills.
No, even these digital reverbs have different results on each pass. It's almost imperceptible in most cases but tails (and probably firs reflections) aren't the same in some reverbs like ValhallaRoom, Lexicon MPX, Eos, Breeze and a couple of others I've tried and own. Try it. Insert a track, import a media file, add a reverb to that track, fully wet or whatever and render it twice. They won't null (you can actually try it with 4 tracks and flipping the polarity on 2 of them to hear it without rendering).
__________________
Pressure is what turns coal into diamonds - Michael a.k.a. Runaway
Mercado_Negro is offline   Reply With Quote
Old 05-08-2012, 05:24 PM   #14
Lucero del Alba
Human being with feelings
 
Join Date: Apr 2012
Location: Buenos Aires, Argentina
Posts: 8
Default

Quote:
Originally Posted by Banned View Post
So, in conclusion: you were probably just hearing PT users talk about how to use PT to avoid some of its specific issues (...)
Quit clearly the case, my friend.

Quote:
Originally Posted by Banned View Post
Feel free to check out that VCA mixing groups example nevertheless. (I'm still hoping to get a bit more feedback on that, especially on whether it affects send/receive levels correctly.)
Will do, please PM me this post you mentioned with any additional comments you may have so I give it a try.

Quote:
Originally Posted by plush2 View Post
You will do well here Lucero del Alba. Welcome and all the best on future explorations. (...)
Hey, thanks for that!

Quote:
Originally Posted by Mercado_Negro View Post
No, even these digital reverbs have different results on each pass. It's almost imperceptible in most cases but tails (and probably firs reflections) aren't the same in some reverbs like ValhallaRoom. Try it.
I just did with ValhllaRoom and you got it right man, I was clearly guessing wrong about reverbs. I find it very interesting actually, how they intended this unpredictability on these products you mentioned; it's good to know, thanks for pointing it out.
Lucero del Alba is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 05:51 AM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.