final mix sound quality

alabama · Post by **alabama** » Mon Aug 02, 2004 4:12 pm

Greetings,

I have a Pulsar II card and Cubase SX. I'm still trying to perfect my tools and have a query that I've wondered about for a while.

I'm working on a project at 32 bits/44.1kHz. I'm porting sound from the Scope environment to Cubase via ASIO drivers, and then from Cubase back to Scope. I have several tracks of audio recorded in Cubase.

I then do a master mixdown to 2 tracks (Cubase multiple tracks --> Scope --> Cubase 2 tracks) to add in Scope effects. The master mix sounds kind of flat compared to the multiple tracks. I'm not doing any other processing to the final mix other than adding effects. Maybe this is just a headroom issue, where each signal has its own individual pipe, but then pushing multiple signals into one pipe fills up the frequency/amplitude space and causes you to lose some fidelity.

I'm just curious what other tools/implementations people use to get a nice sounding final mix.

Thanks

blazesboylan · Post by **blazesboylan** » Mon Aug 02, 2004 6:52 pm

Are you using compression? On each individual track, I mean?

You might also want to check out J9K's tape saturation plug in the Devices forum.

Subtle effects placed on the 8 buses can also fatten up the sound a lot. E.g.: mult the buses back out into mixer inputs, strap in some reverbs, and keep the levels on those inputs just high enough to thicken the mix. For what it's worth I usually put vox -> bus 1, bass -> bus2, drums -> bus 3/4, guitars -> bus 5/6, anything else -> bus 7/8. I also compress the heck out of the buses.

Dunno if you've already tried all of the above but hope it helps...

Cheers Alabama,

Johann

Basic Pitch · Post by **Basic Pitch** » Tue Aug 03, 2004 10:10 am

I would have to agree,

In my long journy to try and get better at mixing I have found it all comes down to dynamics, not just compression alone but try and think of the larger spectrum, sound placement, EQ, effects like shorus, verbs to open the sonic spectrum, panning and so forth its like a paining almost.

Here is a link to an artical and site I found invaluable, it was not untill I read into this site when I really started to understand about mixing and using different dynamics to pain a picture.

http://www.tweakheadz.com/perfect_mix.html

There was another site I wanted to link but it seems to be no longer, read that artical, not just that one only but that site has alot of interesting information regarding EQ technics compression, mastering, mixing, production and lots.

Hope that helps some, I know it did for me.

Cheers!

alabama · Post by **alabama** » Tue Aug 03, 2004 12:06 pm

Thanks. And good article. I guess I get a little frustrated with how nice everything sounds on multiple tracks, but bringing it to a master mix seems to lessen the quality. Part of it, I suppose, is that my ears get used to hearing it a certain way, and any change sounds foreign and out of place.

I've never done much with compression/gate/limiter. I've thought some about getting the optimaster. I know there are many threads on that subject. From a hobbyist point-of-view, I have to start thinking that there is more work to do once all the tracks are in place.

I've thought that optimaster might be a quick-fix for getting a nice sounding final mix, but it seems there's a good bit of work to be done before using optimaster. I might be disappointed if I put the money into that and expected it to voila, create a nice final mix for me.

blazesboylan · Post by **blazesboylan** » Tue Aug 03, 2004 2:24 pm

I suspect everyone else here will disagree with me, but IMO the Optimaster is a bit too complex to be usable.

On the other hand I highly recommend the Vinco!

A lot of folks will disagree with me on this one too, but the stock compressor + limiter work really nicely IMHO. I use them to really crush the kick and snare and often the bass, too.

One option worth considering for a mix bus compressor is the FMR Really Nice Compressor. It will do what no plugin does: add a bit of colouring to your sound. It's also a lot easier to dial in settings if you haven't used compression a whole lot.

BTW I consider EQ to be a last resort. This is just a preference thing, but it might be worth noting. These days I rarely use EQ on anything but kick. Then again I'm only working on my own source material. If you're recording "live" instruments you do have to spend a lot of time with your mics in order to get the EQ right at source. Once you have everything mic'ed nicely then all you need is some compression to really get a full sound out of each instrument.

Incidentally have you tried posting your mix in the Music forum? The folks there are always helpful.

$0.02,

Johann

astroman · Post by **astroman** » Tue Aug 03, 2004 5:03 pm

On 2004-08-02 17:12, alabama wrote:
... I then do a master mixdown to 2 tracks (Cubase multiple tracks --> Scope --> Cubase 2 tracks) to add in Scope effects. The master mix sounds kind of flat compared to the multiple tracks. ...

I understand this in the way you have your monitors on the Scope mixer' master outs, but let Cubase do an 'internal' mixdown, right ?
Otherwise it would be really strange, unless you have a significant difference in output level.
Ears are nonlinear devices, so a different level WILL produce a different frequency response, which might lead to a false judgement of the mixes' content.

If Cubase does the final job you might find some answers (or food for thought) in the Well, I have seen the light thread

my 2 cents, Tom

[ This Message was edited by: astroman on 2004-08-03 18:08 ]

Post by **hubird** » Tue Aug 03, 2004 5:22 pm

Well yesterday I tried the Optomaster, after reading the manual, which is pretty good for what it has to be.

I was surprised that the Wizzard created a pretty balanced and usable mix!
I fed the wizzard with a part of the track containing the end of a full speed part followed by a break with only space synth sounds, not really load.
You could say the ultimate test

That's why I disagree with BlasesB, I think it's the perfect thing for you Alabama:-)
Without knowing anything you'll get a quite impressive result.
If you save that preset, you can feed the wizzard with another selection of the track, and see if that fits more to your taste.
Most important thing is that the wizzard balances all parameters much better that you and I probably can.

In the meantime you can study the Vinco, to learn the basics of compression:-)

BTW I think it's true what Astro says, your mix shouldn't differ from the tracks played.
I would build up the effects in Cubase but specially in Scope if you use any, and see where your sound quality leaves.

Yes, a boring proces, but you can take the same short piece

[ This Message was edited by: hubird on 2004-08-04 09:00 ]

blazesboylan · Post by **blazesboylan** » Tue Aug 03, 2004 8:44 pm

Well I'll be darned. I've never heard of this wizard named Manuel, but I'll certainly talk to him and see if he can work some magic on Optimaster for me.

As far as getting comfy with compression: for an absolutely newbie you can't beat the stock compressor. It does exactly what you tell it to -- your peak meter will show you what your threshold / ratio / gain calculations tell you to expect. Vinco is a little more "fuzzy"...

Cheers,

Johann

alabama · Post by **alabama** » Wed Aug 04, 2004 12:58 pm

On 2004-08-03 18:03, astroman wrote:
I understand this in the way you have your monitors on the Scope mixer' master outs, but let Cubase do an 'internal' mixdown, right ?
Otherwise it would be really strange, unless you have a significant difference in output level.
Ears are nonlinear devices, so a different level WILL produce a different frequency response, which might lead to a false judgement of the mixes' content.

If Cubase does the final job you might find some answers (or food for thought) in the Well, I have seen the light thread

my 2 cents, Tom

[ This Message was edited by: astroman on 2004-08-03 18:08 ]

Astro,

That's pretty darn interesting (the 'see the light' thread). I've often wished that everything could be done in the Scope environment. Of course this is a whole other topic... It would just make it easier and more transparent to operate in Scope without interfacing in and out of another program. I've often thought the 'interface' or the 'other program' could be an issue...and part of what makes using Scope a little tough.

Anyway, YES, I'm mixing in Cubase, sending the 2 master outs from Cubase back to Scope where I'm adding Scope effects, then taking the master L/R outs from the Scope mixer, and porting back to to audio channels in Cubase as my final mix. It's these final 2 tracks that definitely lose something.

OK, so now I will use Cubase as mainly a "pass-through", outputting each Cubase channel to a single channel on the Scope mixer (with it's own ASIO driver into Scope). I'll do this to add in effects.

All of this begs the question...once I have my final 'wet' tracks into Cubase, and do the final mixdown to a .wav file, won't I get the same performance hit by the Cubase mixer?

Is there another way to output your final .wav file without it getting moshed by the Cubase mixdown process? (No offense to Cubase intended ... just following the logical progression here from the previous 'see the light' thread) Maybe VDAT? Is VDAT the light?

Too bad the light costs more money...but I digress.

Thanks.

blazesboylan · Post by **blazesboylan** » Wed Aug 04, 2004 1:48 pm

There was another thread on how great VDAT sounds.

However here's more or less what I do:

Basic Pitch · Post by **Basic Pitch** » Wed Aug 04, 2004 2:03 pm

Well here is another approach,

If you have a wave editor such as Wavelab, Soundforge, Adobe etc etc you could then just open that device and set it to record the wave desitnation, hit play in Cubase and record the direct output of the scope card into the editor completely bypassing the recording done in Cubase.

If not another idea would be to just set the master fader in the STM mixer to an ASIO dest channel and set an audio track in Cubase to use that as its input, simple arm that track for record and hit play and watch your audio track in realtime.

Either way should do fine, you could set your optimaster as a main insert so it effects the audio routed to Cubase, and then do any and all additional proicessing in the Cubase editor.

Just another thought

Cheers!

alabama · Post by **alabama** » Wed Aug 04, 2004 6:03 pm

blazes,

Nice picture. So you then export your stereo mix from Cubase to a .wav file? You're happy with the quality?

I guess mixing down in SFP that way prevents mixing down with your Cubase mixer. Then I suppose the Cubase export process just dithers to CD-format and creates a stereo file. This might fix my problem if the Cubase mixer/Scope combination is the problem.

I think I would probably be happiest if I had VDAT and could just keep the final mix in the Scope environment. However, I don't know if VDAT can create a stereo .wav file.

At any rate, it might not be worth the extra money to get VDAT, and the export process in Cubase, once you have your final mix there, might be just as good.

basic, thanks for your ideas as well.

astroman · Post by **astroman** » Wed Aug 04, 2004 6:55 pm

hi Alabama,

I referred to the VDAT thread mainly because people obviously do hear differences in sound once the Cubase processing engine is involved.
I don't have Cubase, so I can't comment the difference.

VDAT is an app with a very specific mode of operation and it's usability depends a lot on personal preferences.

But as both BasicPitch and BlazeBoylan's suggestions show, it's not difficult to circumvent the Cubase engine even without VDAT.

If you do the final mix in SFP and connect both your monitors and a stereo wave device to the master out, you can record with any application.
It doesn't matter how much latency the wave driver introduces, as your listening 'before' the tape

But in any case you should yield a significat improvement in sound quality.

cheers, Tom

bassdude · Post by **bassdude** » Wed Aug 04, 2004 11:05 pm

Just to clarify, you won't be able to do a Cubase export if you choose to work as described in blazesboylan's picture above as SFP is real-time only (think of SFP as outboard hardware). So you actually have to play the track in real-time and record the stereo mix back into cubase, wavelab, Cooledit or whatever. It's slower than doing an audio export in cubase and for some this may be a little too inconvenient. For me the advantages outweigh any disadvantages.

Also, if you want to automate the SFP mixers then you have to start sending CC's over midi between Cubase and SFP.

alabama · Post by **alabama** » Thu Aug 05, 2004 9:27 am

Well, I've learned something important. Let SFP do your mixdowns! And, basically just do as much as you can in SFP and use Cubase as a sequencer/recorder/wave editor program (not as a mastering tool!).

I gave this new setup a try last night, and it's much cleaner to work with.

Thanks all!!

at0m · Post by **at0m** » Fri Aug 06, 2004 7:34 am

Anyone noticed only Cubase users prefer CW mixing over native, there's no Logic users in this topic :>

I mix both in SX and SFP, depending on where I need to ehum, mix the channels, which in turn depends on what groups I want to apply which effects to etc. Next, the pc spits out 10 busses (STM2448 mix & its busses) to an analog mixer from which I usually record. It's warmer than digital mixing, and no bitdepth loss there

gedas · Post by **gedas** » Fri Aug 06, 2004 7:44 am

Atomic,
What kind of a board are you using? I'm looking into some analog summing way as well. It's between an analog summing box (like Folcrom) or an analog mixer which would give some nice eq and preamps for tracking, though a decent one may be quite pricey.

Basic Pitch · Post by **Basic Pitch** » Fri Aug 06, 2004 11:39 am

Interesting,

I never even bothered to think about analog summing, I think I will continue to mix entirely through SFP, then render the mix, rout it out to an analog desk just for a pass though and back into wavelab, I wonder if it would warm up the mix any by simply passing through the analog mixer?

Cheers!

valis · Post by **valis** » Fri Aug 06, 2004 2:29 pm

The shorter the signal chain the better, generally speaking. Meaning if its color you're after you don't necessarily need that giant radio antennae hanging off your DAW to achieve it. There are quite a few preamps, compressors, channel strips, mix processors (fairchild anyone?) etc on the market that would add any color you're after without all the added components in an analog mixer.

Another way to look at it is that an analog mixer is basically a collection of channel strips and a master summing bus. If you're only processing 2 stereo channels then you don't really need the summing bus, and you'll have an awful lot of extra channels sitting there gathering noise & RF, even muted they'll have some crosstalk into the master bus and the channels you're mixing with. More importantly, if you were to spend the money on an 'ok' analog mixer with 24 channels then you would have individual channel strips worth about 1/24th of the total cost of the mixer. If you buy 2 channel strips (or a device that processes in stereo) then you can spend the same amount and get higher quality gear. Think of the difference between having a Mackie or Allen&Heath eq & preamp with no dynamics etc, and having a really nice pair of SSL or TL Audio channel strips with expander/eq/compressor/limiter and perhaps even valve etc.

Then again if you have a large number of outboard synths then the analog mixer may make more sense....

blazesboylan · Post by **blazesboylan** » Fri Aug 06, 2004 3:45 pm

I agree with Valis, there's no point in adding colour at the summing stage. Do it to your individual tracks whereever you need to, and to the final mixdown.

However you don't need to spend $30,000 on a Fairchild 670 to get a nice stereo "coloured" mix. There are lots of options out there, and in many cases the cheap ones add nice colour. I still stand by the RNC as being great across a mix bus.

BTW I used to mix everything outboard and it's a royal pain without automation!

Not all consoles add colour, and Mackies in particular are quite clean.

Cheers,

Johann