EUREKA YOUR SUMMING WOES ARE OVER!

[illacov]

As I mentioned in a previous post, I've happened upon a routing scheme in the EMU 0404 and all the other EMU cards, whereby you can route your outputs to your inputs INTERNALLY, from a live host to another host at the SAME time.

What I did is play my live mix in Cubase SX and this went to the ASIO outputs in Patchmix. Then I routed the output from Cubase SX in patchmix to my WAVE Host input (stereo).

I still use Cool Edit, and I set it up to record at 192 khz. I thought it wouldnt work simultaneously, but VOILA! It recorded my mix with NO summing, but the best part is the mix went thru my converters so it was actually summed OTB style, but ITB!

The results were INCREDIBLE. I may actually scrap my Low End Summing project and invest my business expenses into DAW upgrades. HOLY SMOKES.

Sound Files will be posted soon!!

Peace
Illumination thumbsup

[octatonic]

Is this High End?

[illacov]

If you can accomplish this with an Apogee or a Lynx AD/DA then you kind of eliminate the whole ITB summing thing, since in essence you recorded you music again through your converters and not thru a summing engine in you daw.

How about theoretically noiseless OTB(ITB?) summing. Thats mighty high end bro. What you think I should bury this in some forum where no one will find it? These kinds of observations need to be shared!
Ok Ill make it high end. You need a really expensive LCD monitor autographed by Bob Katz to see the difference. Is that better?
But its in the High End forum because I want the big dogs to put their two cents in too!!

Peace
Illumination

[octatonic]

Originally Posted by illacov

If you can accomplish this with an Apogee or a Lynx AD/DA then you kind of eliminate the whole ITB summing thing, since in essence you recorded you music again through your converters and not thru a summing engine in you daw.

How about theoretically noiseless OTB(ITB?) summing. Thats mighty high end bro. What you think I should bury this in some forum where no one will find it? These kinds of observations need to be shared!
Ok Ill make it high end. You need a really expensive LCD monitor autographed by Bob Katz to see the difference. Is that better?
But its in the High End forum because I want the big dogs to put their two cents in too!!

Peace
Illumination

I think you are labouring over a few misconception about what analog/digital summing really are.

I've responded to low end theory.

[Kiwiburger]

People interested in high end gear are really the people with good ears who have to have whatever it takes to get good sound, regardless of cost. So yes - a free software solution that sounds better than expensive hardware should qualify as high end. People don't spend big money for no reason - not unless they are stupid.

BUT - I can't see any advantage at all over what you are proposing, compared to mixing ITB. Mixing ITB with Cubase/Nuendo is about as good as it gets. I believe the real issue is that it is perhaps too perfect, and doesn't introduce enough distortion and noise and phase smearing for slutty high end tastes.

Let me get this straight: say this is a 24 track mix, you are sending those 24 tracks to 24 D/A converters, which are cabled to 24 A/D converters, which you then record multitrack? No - that can't be what you mean, because then you would still have to mix in either the first or second DAW. The only thing you would achieve is an analog interface that would introduce some noise, distortion and phase smearing, which might be what you want to hear.

I'm guessing you are really talking about mixing ITB, but sending the 2bus mix from your summing bus to your D/A converters, which are cabled to your A/D converters and re-recorded to a stereo DAW. Now that is basically just mixing ITB, but you've added an analog interface (your converters) just for the hell of it. Maybe a little noise, distortion and phase smearing, but not much.

Or - if you are bypassing your converters totally, and simply mixing ITB and recording that stream to another app, I can't see any real advantage over mixing ITB period.

Is the purpose of this simply to record the realtime mix, as opposed to the rendered mix? That's a good way to miss out on all the benefits of a rendered mix - and possibly get some of the phase anomolies that occur when your CPU is maxed out.

Why not simply mix ITB (render the file) and compare it with what you are achieving. I don't expect much difference really. I think the whole OTB hardware deal is a scam, playing on user incompetence or ego/gear lust.

If there is some euphonic distortion to be obtained in the analog realm, then identify what that is, and if you can't recreate it with a plugin, playing the rendered mix through your best converters through this magic analog device and re-record that with your best converters. Or, if this magic analog device is only required on certain tracks - process those independantly. Or track with it in the first place.

I don't see what the fuss is all about.

[Nahuel]

1- all DAW SUCK (IMO) at rendering Wav files.
2- that's why some people use tapeit or the routing of their card with another app (like wavelab or soundforge) to get a decent mix ITB (btw I heard some protools rendered files and they seem to be nicer than the common "native DAW" results)
3-nice convertors+nice analog summing>nice convertors is (IMO) better than all that.

btw, (I dont really give a ****) this thread is far from being "high end".

[dione]

Originally Posted by illacov

As I mentioned in a previous post, I've happened upon a routing scheme in the EMU 0404 and all the other EMU cards, whereby you can route your outputs to your inputs INTERNALLY, from a live host to another host at the SAME time.

What I did is play my live mix in Cubase SX and this went to the ASIO outputs in Patchmix. Then I routed the output from Cubase SX in patchmix to my WAVE Host input (stereo).

I still use Cool Edit, and I set it up to record at 192 khz. I thought it wouldnt work simultaneously, but VOILA! It recorded my mix with NO summing, but the best part is the mix went thru my converters so it was actually summed OTB style, but ITB!

The results were INCREDIBLE. I may actually scrap my Low End Summing project and invest my business expenses into DAW upgrades. HOLY SMOKES.

Sound Files will be posted soon!!

Peace
Illumination thumbsup

Uhm...and how can you actually use the Asio drivers in both Cool Edit and SX at the same time ?

[illacov]

YA dont

See the beauty is, Cool Edit uses WDM drivers if Im not mistaken and I have an old enough version that supports a high sample rate, but also WDM drivers with no problem. These two drivers (ASIO and WDM) DONT conflict.

Cool huh?

In theory you could run Cubase and Sonar at the same time if Im not mistaken since they both use two different drivers.

I dont know if it would work with audacity or kristal, but i think audacity may run on mme drivers (dont quote me on that though).

Peace
Illumination

[dione]

Originally Posted by illacov

YA dont

See the beauty is, Cool Edit uses WDM drivers if Im not mistaken and I have an old enough version that supports a high sample rate, but also WDM drivers with no problem. These two drivers DONT conflict.

Cool huh?


I dont know if it would work with audacity or kristal, but i think audacity may run on mme drivers (dont quote me on that though).

Peace
Illumination

okay...but the WDM drivers don't give the same results as the Asio drivers...i've tested it in Wavelab and the Asio drivers give much better and clearer results as the WDM drivers when using an E-Mu card (i have the 1820m)...

so if you think your solution works for you,why not just send that Asio channel back to SX and record it in SX ? So you can still use the better drivers and you don't have to work with different programs at the same time...

[nathanvacha]

Yeah i think the entire point of summing is the analog processing though... that's why it's called summing as opposed to bouncing. I realize you're doing the same thing technically, combining all your tracks into stereo, but i think in all fairness we should differentiate between itb and otb in this application... people seem to be confused already enough with what this new "summing" thing is. So re-rerouting itb is just a more complicated way of bouncing in real time, like the way ProTools does "bounce to disk."

[illacov]

The whole point is the fact that my version of Cubase doesnt do 192khz and for arguments sake i wanted to see if theres a real difference in sound quality. The answer is yes people.

Cool Edit is the only program i had that would do 192khz i havent got my hands on SX3 yet.

Yeah asio drivers are more stable, but im only recording one stereo track.

Its a non issue if you got the cpu power.

I played back a full mix at 100ms latency with Cool Edit recording at 192khz all on a XP 2000 (1.67ghz) machine with 768 MB of ram. This machine is a dinosaur, its actually our tracking machine.

I mix at my house.

But theres the answer....thats why i did it that way. When I have SX3, i wont need to use cool edit in that capacity, but for now it comes thru in a pinch.


Peace
Illumination

[illacov]

Rendering a track is an approximation of what those sounds actually sound like in one stereo file. But a recording of them, no matter where you get it is different than a rendering of it.

bouncing to disk is not the same as recording what you would have bounced.

Whats really interesting is that I find that actually recording your individual tracks instead of rendering or bouncing them increases their clarity in cubase. Call me cooky.

Peace
Illumination

[Kiwiburger]

Illacov - I'm still unclear exactly what you are proposing. From what I can tell, you are simply endorsing the re-recording (at a higher sample rate) of the master bus stream from a realtime running DAW project. As such - it is perhaps a little better than using TapeIt, which would be at the DAW project's sample rate.

It seems to me that all you are proving is that recording at 192 Hz sounds a little better than 44.1 Hz. That shouldn't be a surprise.

It seems that you are still mixing ITB - just with the different sample rate. From what I can tell, your audio never actually goes through the D/A - A/D process - you are just routing the digital stream, and capturing it at a higher sampling rate. Then you downsample it anyway. My guessing is that this process can't restore any lost frequency content from the original project rate, but it probably averages out the digital jitter - by effectively oversampling. You might be onto something if this is what is happening - however high end users have superior word clock stability, and probably don't need this step.

Or - if i've misunderstood what you are saying, and your audio actually goes through D/A - A/D process - there is a very logical reason why you would hear a difference.

The theoretical dynamic range of 24 bits is 144 dB. But cheap converters only handle 100 dB, and the best converters handle 114 dB. Where is the missing 30 dB? You have to think of converters as being a fundamentally analog device. The analog circuitry only has a (best case) 114dB range, which means the extreme ends of the range are colored with noise or distortion.

If you - like many DAW operators - insist on squeezing your mix into the entire 24 bit range, by approaching 0dBFS, then you are most definately going to get noise and coloration from the converters.

So if you have a 2buss mix that is peaking at even as low as -6dBF - IF you route this through a D/A converter, a cable, and back into the box with a A/D converter - GUESS WHAT - you have introduced a lot of noise and coloration. Compared to the ITB mix, with 144dB resolution or better - you have reduced your mix to 114 dB (or whatever dynamic range your converters have). Will you notice this? Hell yes!

Maybe you will like the distortion. The thing about deep bass is that a deep sinewave at 40Hz or whatever is very boring. It gets interesting when you introduce some harmonics.

This - I believe - is what summing OTB is all about. Introducing noise, distortion and phase smearing.

Also - rendering a file should be as mathematically perfect as the software allows. Re-recording a DAW's playback stream is likely to be less perfect, due to phase issues caused by CPU saving tricks taken by the coders. The closer you get to maximum CPU load, the 'worse' these effect might become. But our ears might actually prefer this!

From what I can tell - your approach still involves literally 'summing in the box' because there is no way you can get a 2bus mix without summing all your tracks in the digital domain. So I think the choice of the word summing is erroneous. I think you are talking about processing or re-recording the previously summed 2buss mix. Correct me if i've misunderstood you.

For my money - I prefer to render, because I believe it is mathematically perfect. If I want distortion, I prefer to choose it on my own terms.

[respirator]

The discussion is continued here: http://www.kvraudio.com/forum/viewtopic.php?t=122450

If you need a laugh...

[octatonic]

Originally Posted by respirator

The discussion is continued here: http://www.kvraudio.com/forum/viewtopic.php?t=122450

If you need a laugh...

I got one page in and then couldn't continue.
Christ on a bike... some people just don't get it.