Mixdown export sounds muffled, boxed-in. What am I doing wro

Since we are talking about mixdowns, something else that always bothered me is the Pan Law which I still believe is set to default as EQUAL POWER in 7.0.6. Changing it to the “0” parameter retains the true panning spectrum. The default setting messes with original source material especially if you’re recording organic sources in stereo as well as your stereo plug-ins. The EQUAL POWER setting is not going to accurately playback what you recorded.

But note that the panning law isn’t applied when panned hard L/R, which is probably how you’d have them for a stereo recorded source, so it would play back accurately in this situation. It will only be applied when panned elsewhere, gradually up to 100% when reaching the centre position.

Mike.

Yes that is what I am talking about…elsewhere! I don’t always hard pan stereo… it depends how it sounds when I’m mixing.

I too have noticed mixdown acts a bit weird. For example, mixing down the same track 6 times gave me different results almost each time. It’s possible some plugin was free running but it never sounded like that when just playing it in the mix. I think real-time export reduced that, but it’s still surprising. I mean, I always believed Cubase could simulate accelerated time for mixdowns to be quicker, that doesn’t seem the case.

What I’m not sure about, and might be the problem here, is whether the summing algorithm is different when mixing down or while playing.
Maybe when playing, it goes full 32-bit through the mixer from all the channels to the stereo and its FX and just then clipped and downsampled. And when mixing down, it “exports” all the tracks in memory, with clipping and all and only then sums them, clipping again, and performing the final mixing. I surely hope this is not the case.
Also it doesn’t seem your tracks are anywhere near the clipping edge.

Try a test…

Create a mix with all tracks routed to Stereo Out

Create another mix with all tracks routed to a single Group Bus

Create a new project with no plug-ins

Drag in both files to the same start point

Invert the phase of one file

Turn down one fader and slowly bring it up

When both files are at 0db you should hear nothing if they are identical

If they aren’t identical you should hear what that difference is

NZR

1 Like

I tried variations of the Phase Reverse test I suggested, with interesting results. Like everyone else here, I’d like to know with as much certainty as possible whether there’s anything to the Group or Stereo Out mixdown difference. I don’t propose these results as conclusive but only as a starting point. If others will test or some one with technical expertise could explain, I know we’d all appreciate the clarification.

I used a song with multiple audio tracks and 1 Instrument Track (Arturia Mini V, playing a temp bass part)

  1. All tracks routed to Stereo Out then Export/Audio Mixdown (48/24)
  2. All tracks routed to a stereo Group then Export/Audio Mixdown using the Stereo out

I opened a new project and dragged the 2 files in at the same start point and Phase Reversed one of the files. I turned one fader down and slowly brought it up. When I reached 0db everything went away except the Arturia Mini V part. i realized this was because the Mini V is running in realtime so it’s different every time you play it.

I did another pair of mixes with the Arturia muted and the same routing setups. With 1 file phase Reversed and both at 0db, I heard absolutely nothing.

Next I opened a Waves EQ insert at the Stereo Out on the Group mixdown. I made a narrow Q midrange boost. When both files were at 0db, all I heard is the boosted frequencies as if using a narrow bandpass filter.

This suggests to me that there is no difference between doing an Export/Audio Mixdown through the Stereo Out vs the Sub Group-Stereo Out.

What about the same thing but using Export/Audio Mixdown versus Audio/Bounce Selection?

NZR

your testing results sounds logical and the way it should be… !
the real question is if the exported mixdown(high quality wav file) is the same quality as the actual sound when playing the project through cubase…!
Edit: maybe a real test would be mixing the same project with cubase and other DAW with same 3rd party plugins with exact same settings …! ( actually some one did it already but not compering export quality vs playing the project but DAW comparison… cubase, logic, protools… it was mixed in logic initially and cubase had the most muddy sounded result :astonished: …but im not sure its a good or right comparison !!!)

I’m new to Cubase and on my first export (Master channel), I realised that my mixdown is “dull” compared to what I hear when playing it live. So I then send all my busses to a single group channel. I then exported this group channel instead of the MASTER and it sounds PERFECT.
This has an added advantage because after my limiter I have my earphone calibration software and a gain vst to save my ears. My mastering FX and limiter are all on the group channel now and the earphone vst is now on the master. I don’t have to disable them anymore when exporting my group channel.