Hybrid OTB/ITB vs ITB - listen for yourself

[chris_m]

I thought i'd like to share my personal findings on this topic, this is far from being scientific and cannot be compared to complete OTB mixing. Just my little experiment.

The Setup...

I used tracks from a song that's avaible on The Meth Lab. I downloaded these as i don't have my own appropriate material right now. My 2 clips are 15 seconds short and i only used the guitar and drum tracks. If posting these is a problem please contact me and i'll remove the 2 clips asap.


I quickly put a mix of the guitars and drumset together using Reaper 2.55 and my plugins. Since i only have 6 channels of d/a avaible right now i had to keep it down to 3 stereo groups so i could bring these 3 OTB. All leveling was done ITB so i could match the level of each channel perfectly on the console. Then it was rendered directly to 44khz and ITB ended here.

For OTB the 3 stereo groups drums and guitars left/right were routed to my soundcraft ghost console. I got the ghost for my private projects because i use analog synths, outboard gear, don't need recalls and generally don't like sitting at computers in my free time.
Since the ghost is basically still a lowend console, i was curios to see how it holds up to a clean and noisefree ITB mix. This was my primary intention.

I then recorded the output back via my profire 610 (also used for D/A). The levels of both files match very well, around -16.5dB rms.

I don't want to makes this another "guess what is what" topic so i named both files as what they are:

http://www.colorhythm.de/itb.flac

http://www.colorhythm.de/otb.flac


After listening and ABXing in foobar i was surprised how close the OTB mix is to the ITB mix. For me this is a good thing. Since the ghost is basically a lowend console i expected there to be a loss of punch, more veil and harsh highs, whatever you associate with lowend stuff But compared to the ITB mix it actually does very well, and that using a profire 610 for AD/DA and the oldest cabling i could find in my basement.

The OTB mix in my ears enhances what the ITB mix offers, without introducing any negative attributes to the mix. OTB sounds more open, panning goes a bit wider, there is more depth in the drums and guitars. This especially noticeable on headphones where the ITB mix sounds "closed in" compared to OTB.

How far mixing OTB completely could enhance this would be interesting. Using real hardware compressors/eq and 32 D/A outs, grouping on the console etc. thumbsup


With a hybrid approach the differences are undeniably there, but subtle.

[Mr.HOLMES]

Originally Posted by chris_m

I thought i'd like to share my personal findings on this topic, this is far from being scientific and cannot be compared to complete OTB mixing. Just my little experiment.

The Setup...

I used tracks from a song that's avaible on The Meth Lab. I downloaded these as i don't have my own appropriate material right now. My 2 clips are 15 seconds short and i only used the guitar and drum tracks. If posting these is a problem please contact me and i'll remove the 2 clips asap.


I quickly put a mix of the guitars and drumset together using Reaper 2.55 and my plugins. Since i only have 6 channels of d/a avaible right now i had to keep it down to 3 stereo groups so i could bring these 3 OTB. All leveling was done ITB so i could match the level of each channel perfectly on the console. Then it was rendered directly to 44khz and ITB ended here.

For OTB the 3 stereo groups drums and guitars left/right were routed to my soundcraft ghost console. I got the ghost for my private projects because i use analog synths, outboard gear, don't need recalls and generally don't like sitting at computers in my free time.
Since the ghost is basically still a lowend console, i was curios to see how it holds up to a clean and noisefree ITB mix. This was my primary intention.

I then recorded the output back via my profire 610 (also used for D/A). The levels of both files match very well, around -16.5dB rms.

I don't want to makes this another "guess what is what" topic so i named both files as what they are:

http://www.colorhythm.de/itb.flac

http://www.colorhythm.de/otb.flac


After listening and ABXing in foobar i was surprised how close the OTB mix is to the ITB mix. For me this is a good thing. Since the ghost is basically a lowend console i expected there to be a loss of punch, more veil and harsh highs, whatever you associate with lowend stuff But compared to the ITB mix it actually does very well, and that using a profire 610 for AD/DA and the oldest cabling i could find in my basement.

The OTB mix in my ears enhances what the ITB mix offers, without introducing any negative attributes to the mix. OTB sounds more open, panning goes a bit wider, there is more depth in the drums and guitars. This especially noticeable on headphones where the ITB mix sounds "closed in" compared to OTB.

How far mixing OTB completely could enhance this would be interesting. Using real hardware compressors/eq and 32 D/A outs, grouping on the console etc. thumbsup


With a hybrid approach the differences are undeniably there, but subtle.

This is exactly the same topic as with summing and internal bounce.
If you are happy with it stay there.

I just can say for myself if you do two different mixes one into the console or summer and one bounced ITB you will get two different worlds.

The world would be not big in sound but in dynamic range you will feel much more comfortable with the OTB summing and you will mix faster than ever because you have more headroom.

So no need to discuss gear make music with it and listen...listen...listen...listen...listen...listen.........

[lerone]

OTB is great for summing, it does generally sound a bit richer in all aspects. A healthy combo of plugins (think urs csp and psp squad) + a nice summing box is the way for me nowadays. I miss the MTA 990 tho :(

[chris_m]

Originally Posted by Mr.HOLMES

The world would be not big in sound but in dynamic range you will feel much more comfortable with the OTB summing and you will mix faster than ever because you have more headroom.

So no need to discuss gear make music with it and listen...listen...listen...listen...listen...listen.........

This is a forum to discuss gear after all .... The added headroom is a joy to work with. But for now i'll keep working ITB for other peoples songs.



I was also really surprised by the perfomance of the profire 610 ad/da conversion. Thats a great little unit. Build quality is fantastic too.

[illacov]

Originally Posted by chris_m

I thought i'd like to share my personal findings on this topic, this is far from being scientific and cannot be compared to complete OTB mixing. Just my little experiment.

The Setup...

I used tracks from a song that's avaible on The Meth Lab. I downloaded these as i don't have my own appropriate material right now. My 2 clips are 15 seconds short and i only used the guitar and drum tracks. If posting these is a problem please contact me and i'll remove the 2 clips asap.


I quickly put a mix of the guitars and drumset together using Reaper 2.55 and my plugins. Since i only have 6 channels of d/a avaible right now i had to keep it down to 3 stereo groups so i could bring these 3 OTB. All leveling was done ITB so i could match the level of each channel perfectly on the console. Then it was rendered directly to 44khz and ITB ended here.

For OTB the 3 stereo groups drums and guitars left/right were routed to my soundcraft ghost console. I got the ghost for my private projects because i use analog synths, outboard gear, don't need recalls and generally don't like sitting at computers in my free time.
Since the ghost is basically still a lowend console, i was curios to see how it holds up to a clean and noisefree ITB mix. This was my primary intention.

I then recorded the output back via my profire 610 (also used for D/A). The levels of both files match very well, around -16.5dB rms.

I don't want to makes this another "guess what is what" topic so i named both files as what they are:

http://www.colorhythm.de/itb.flac

http://www.colorhythm.de/otb.flac


After listening and ABXing in foobar i was surprised how close the OTB mix is to the ITB mix. For me this is a good thing. Since the ghost is basically a lowend console i expected there to be a loss of punch, more veil and harsh highs, whatever you associate with lowend stuff But compared to the ITB mix it actually does very well, and that using a profire 610 for AD/DA and the oldest cabling i could find in my basement.

The OTB mix in my ears enhances what the ITB mix offers, without introducing any negative attributes to the mix. OTB sounds more open, panning goes a bit wider, there is more depth in the drums and guitars. This especially noticeable on headphones where the ITB mix sounds "closed in" compared to OTB.

How far mixing OTB completely could enhance this would be interesting. Using real hardware compressors/eq and 32 D/A outs, grouping on the console etc. thumbsup


With a hybrid approach the differences are undeniably there, but subtle.

Until you expand to at least 12 to 16 channels you are really hearing the ITB summing with that few channels.

If you had run 16 channels to that ghost console, even with the faders on the ghost at unity, lowered the output of the mixer like -10db or so and followed the outputs of the ghost with a really nice 2 channel preamp, you would DEFINITELY have a different response for this thread. Do you have any hardware compressors? Use them as inserts on the ghost channels, (whole different world!!)

You will see. I guarantee it.

I mix with 24 channels out of the box. Whole different world. My summing unit has steel Cinemag transformers on the outputs. Lots of vibe and mojo.

Peace
Illumination

[DeadPoet]

I did a comparable test: 16channels summed in Cubase4 and the same channels summed on my Ghost.

The difference is *very* small.

As for added headroom: I tried upping the levels on the Ghost and driving the mixbuss harder but everything turned into mashed potatoes.


Conclusion for me: will continue summing ITB because of easier recall.



Herwig

[illacov]

Originally Posted by DeadPoet

I did a comparable test: 16channels summed in Cubase4 and the same channels summed on my Ghost.

The difference is *very* small.

As for added headroom: I tried upping the levels on the Ghost and driving the mixbuss harder but everything turned into mashed potatoes.


Conclusion for me: will continue summing ITB because of easier recall.



Herwig

How many mono channels??
How many stereo??

And by this I mean, actually using a single output for mono, versus telling a stereo channel to act mono. As a former SX3 user, I remember how I almost fell out of my chair when I first started using DEDICATED mono outputs versus trying to route a mono signal inside Cubase to a stereo output and expecting mono behavior on that output. World of difference. Seriously.
For stereo you used 2 outputs to two channels on the console and panned them hard left and hard right? Or used a stereo channel on the Ghost?

Any hardware compression?

One of the ideas is to try mixing with the faders at unity and lowering the gain of the master buss and then increasing its output by strapping a 2 channel compressor (transparent to flavored is your call) to make up the gain.

I find it interesting that people have such mixed results from system to system. Did you wordclock your system?
I used to mix without wordclock for my channels and once I started using it, the stereo image increased 25 percent and there was better transients. OTB I mean.


On a side note if you are using hardware inserts in your mix setup at all, it would serve you best to use them with an actual mixer or use them inline to your summing solution. They sound different being used this way versus being used inside the DAW before you send the channels out.

Ive heard some good mixes OTB on Soundcraft Ghosts.

Wonder what the issue may be here.

My summing unit has 8 mono channels (dedicated) and 8 stereo channels (dedicated). I absolutely love it.

BTW I use Reaper 2.55 as well version 3 is smashingly promising.


Peace
Illumination

[DeadPoet]

Originally Posted by illacov

How many mono channels??
How many stereo??

16 mono channels going to 16 mono outputs going to 16 mono tape return inputs on a stereo desk
Some panning was done in the ITB mix which I eyeballed on the desk sum, that might actually be the only difference there was between the two files. Can't post the result as the music I tried this with was no my own.

Any hardware compression?

no, just summing.

Did you wordclock your system?

? I don't understand the question. Yes, there is a wordclock present in my setup, but this has nothing to do with any results.

I used to mix without wordclock for my channels and once I started using it, the stereo image increased 25 percent and there was better transients. OTB I mean.

I think you mean 'dedicated wordclock' instead of just 'wordclock'.
Again, nothing to do with comparing this kind of summing.

Ive heard some good mixes OTB on Soundcraft Ghosts.

I know there are. I'm actually very happy with the desk. My comparison just meant TO ME that I don't have a sonic benefit in going through the hassle of setting up the desk, recalling, bouncing, etc.
Only benefit would be setting general levels with real faders instead of having to use a controller (BCF in my case)


Herwig

[illacov]

Originally Posted by DeadPoet

16 mono channels going to 16 mono outputs going to 16 mono tape return inputs on a stereo desk

no, just summing.

? I don't understand the question. Yes, there is a wordclock present in my setup, but this has nothing to do with any results.

Herwig

Few things. Your master converter has 16 analog outputs? If it doesn't and you're using additional converters to get those 16 outputs then what's syncing them? Are you synced thru adat clock? In this scenario wordclock would matter for summing due to your master interface having the best clock for sync. This would affect sonics to no end. It wouldn't destroy them but it would be audible.

If im correct you went into tape returns?

Why didn't you go into the actual input channel strips? You bypassed the best reason for the console which is the headroom those preamps can provide when they boost your signal. Or the headroom you can create (preamp dependent) by raising that input trim a touch so u have some headroom. The tape returns on consoles would possibly have if at all a lineamp and it would be next to nothing compared to the input block pres as far as clarity and detail. Plus you lose eq and for some inserts and additional bussing. I say try again. U will get way different results.

If your system was clocked properly that's fine I was just curious. You don't think wordclock has anything to do with summing comparisons? Even if you didn't mix otb and had mutiple hardware comps and eqs and inserts in your daw then depending on how you have your ad/da setup you would benefit from clocking to your master interface (again that's if you have more than one interface) the whole external clock business is another story altogether. Im not speaking on that simple remarking that sync is very important when multiple converters are used.

Good luck with sorting it all out if that's your wish. I wouldn't return to ITB unless they figured out a way to integrate the physical architecture of a real console with a ad/da interface. Something like a 2U rackmount unit that only required 1 special cable to hook up to a card in your tower. Maybe it would have jacks for hardware compressors and eqs and reverbs. But it would need no cables for inputs or outputs All electronic switching internally. The unti would be the heart of the DAW. It would eliminate lightning fast renders for mixdowns but would be pretty cool. Real physical channels with DAW functionality.
Imagine it with trannies on the "virtual" ins and trannies
on the virtual 2 buss out utilizing a super quiet preamp circuit to recover gain from summing.

Peace
Illumination.

[DeadPoet]

Originally Posted by illacov

Few things. Your master converter has 16 analog outputs? If it doesn't and you're using additional converters to get those 16 outputs then what's syncing them? Are you synced thru adat clock? In this scenario wordclock would matter for summing due to your master interface having the best clock for sync. This would affect sonics to no end. It wouldn't destroy them but it would be audible.

The main DAW has an RME 9652 in an Alesis HD24 - 24 outs. Clocked by a Mutec SmartClock

If im correct you went into tape returns?

Why didn't you go into the actual input channel strips? You bypassed the best reason for the console which is the headroom those preamps can provide when they boost your signal. Or the headroom you can create (preamp dependent) by raising that input trim a touch so u have some headroom. The tape returns on consoles would possibly have if at all a lineamp and it would be next to nothing compared to the input block pres as far as clarity and detail. Plus you lose eq and for some inserts and additional bussing. I say try again. U will get way different results.

A few thoughts on your paragraph:
- I think you confuse headroom with basic audio levels. (and I don't think the pres in the Ghost are best known for their headroom or crunch).
- I do not lose EQ or routing options by going in the tape returns instead of the line inputs. Hit the 'Rev' switch at the top of the channel and your tape return is on the big faders. EQ, Aux and all.
- Going in the tape returns is how my setup and patchbay is configured. I actually prefer to go through as few electronics as possible (tape inputs vs line inputs).
- I don't think the line inputs pass through the preamps but I'll have to check the block diagram for that.

If your system was clocked properly that's fine I was just curious. You don't think wordclock has anything to do with summing comparisons?

The digital clock was the same in my ITB and OTB summing test. So the quality of that clock is of no importance to the difference between that particular test.





Herwig

[Doc Mixwell]

Originally Posted by DeadPoet

The digital clock was the same in my ITB and OTB summing test. So the quality of that clock is of no importance to the difference between that particular test.

Herwig

Are you sure about that? The way I see it; if you guys are using these files; the ITB mix has nothing to do with your master clock, as there is no more AD/DA conversion happening other than your monitoring [which you are basing your choices from]. If they were tracked; with your external clock, synced to the AD converter used, then it would be a different story. Fact is; using your master clock to sync the DAC, DOES affect the resulting sonics of the DAC, and then it all spills down hill to the analog electronics.

[illacov]

Originally Posted by DeadPoet

The main DAW has an RME 9652 in an Alesis HD24 - 24 outs. Clocked by a Mutec SmartClock



A few thoughts on your paragraph:
- I think you confuse headroom with basic audio levels. (and I don't think the pres in the Ghost are best known for their headroom or crunch).
- I do not lose EQ or routing options by going in the tape returns instead of the line inputs. Hit the 'Rev' switch at the top of the channel and your tape return is on the big faders. EQ, Aux and all.
- Going in the tape returns is how my setup and patchbay is configured. I actually prefer to go through as few electronics as possible (tape inputs vs line inputs).
- I don't think the line inputs pass through the preamps but I'll have to check the block diagram for that.



The digital clock was the same in my ITB and OTB summing test. So the quality of that clock is of no importance to the difference between that particular test.





Herwig

And going thru tape returns and then routing back to the faders on the main channels via a "switch" bypasses little electronics. Think of it this way.
You are going thru the opamps, ics and resistors, capacitors for the tape returns and are routing them to the input channels. What electronics did you skip.

BTW I was inferring that you go thru the actual XLR inputs on your Ghost

Take a quick gander at the dynamic range of the XLR pre versus the line amps and you'll notice pretty different figures. Hence different sonics. I assumed you were already hip to this I apologize. The input channels via the XLR inputs.

I understand you're being a bit leery of my logic but you've been teetering on being dismissive for a while with this thread.

Its not cut and dry and despite the answers you're giving. Your master clock affects the performance of the DAC. Point blank. No other way to spin it. This in turn as Adam mentioned affects the analog which in turn affects your results aka your mix OTB. I don't quite know your reason for the Mutec over the RME clock, I actually heard the RME clocks are decent and its always best to clock from the master interface.

However this is still a small piece of the puzzle. And we'll leave it at that.

Secondarily. And most importantly, you're the first person I've encountered in a long time to discount the value of the Ghost mic pres, especially for small duties like maybe 10db to 15db of gain. Now the Ghost power supply is another story.

I was in no way saying that you get sonic flavor by turning up the input trim on your Ghost pres. I was saying you are giving yourself some level to PLAY with (which allows you to back down on the DAW end) since the Ghost does more of the gain staging for you.
Sort of like having a high wattage amplifier with a mixer. Lower wattage means more gain from the mixer to get louder levels, higher wattage means a little does alot aka headroom (quick and dirty not clinically correct).

Obviously though you feel I have no clue about any of this stuff and its pointless to go back and forth.

But I'm certain you're beyond being convinced, at least by me. You have your findings I have mine. Somebody would have to come into your setup and show you the difference. Then maybe perhaps then maybe you might say Oh I see.

Let's leave it at that.


(Adam thanks for the clarification.)

Peace
Illumination

[DeadPoet]

@ Adam: whoops, you're right. Sorry.

@Illacov:

I understand you're being a bit leery of my logic but you've been teetering on being dismissive for a while with this thread.

Would you explain this please, I'm not native English speaking, I didn't understand this part.

I don't quite know your reason for the Mutec over the RME clock, I actually heard the RME clocks are decent and its always best to clock from the master interface.

I have 2 DAWs feeding several digital interfaces. Easiest with a house clock (especially at the price I found it for).

....<snip> But I'm certain you're beyond being convinced, at least by me. You have your findings I have mine. Somebody would have to come into your setup and show you the difference. Then maybe perhaps then maybe you might say Oh I see.

Your eloquent explanation makes me itch to go in the studio and try feeding my DAWs outputs into the XLR inputs (which is physically impossible in my setup without resoldering a bunch).
I have my TRS line inputs on my patches as well. Would they make a big difference?


Herwig

[The dman]

Have you checked out the sonic differences in the clocks? In my setup my HD24xr sounds better when clocked to internal than it did when I had it clocked to a lucid Genx. More open in the top end so yes it does make a difference

[chris_m]

The line and xlr inputs share the same signal path on the ghost, i remember reading this in the manual. The mix b is has it's own bypassing the preamp. In my tests using the line/xlr inputs provides a warmer but also slightly less dynamic sound. From these 2 I use whatever fits the sound i have in mind, clean or "warm". It's great to have both as an option. I didnt notice a big difference in headroom on both inputs.

There's many ways the OTB mix could be improved and this was just quickly put together because i was curious if the ghost could keep up to ITB. This wasn't meant to be contra ITB, OTB or the ghost console in particulary, i love mine

What are the going prices for a ghost these days? I got mine for 1200 Euro, 32ch with the meterbridge and cps275. It needed some serious cleaning but now it looks like new and is troublefree. This was the best money i spend in ages for creativity, fun and sound.

[cowudders]

As I'm slowly putting my Ghost back into action via an SSL alphalink - nice thread

On a current project, using the Ghost's preamps via the insert-out directly into the converter, I'm very happy with the sound - a real surprise. I also have an API 3124 and two GT Bricks.

I'll also mix this project (indie-experimental stuff) OTB and look forward to the fun and surprises (!!) when mixing on a board again.

Originally Posted by chris_m

What are the going prices for a ghost these days? I got mine for 1200 Euro, 32ch with the meterbridge and cps275. It needed some serious cleaning but now it looks like new and is troublefree. This was the best money i spend in ages for creativity, fun and sound.

I'm thinking all the time about selling mine, but as soon as I do something with it, I change my mind again. Really weird.

OT: How did you clean your pots, as I'm sure they were crackling (like mine, after letting it sit mostly idle for 5 years)? The switches? As you are from Germany, I'd be happy to learn about the EU equivalent to US' de-oxit, or whatever it's called. thanks.

cheers from Austria,
bernd

[illacov]

Originally Posted by chris_m

The line and xlr inputs share the same signal path on the ghost, i remember reading this in the manual. The mix b is has it's own bypassing the preamp. In my tests using the line/xlr inputs provides a warmer but also slightly less dynamic sound. From these 2 I use whatever fits the sound i have in mind, clean or "warm". It's great to have both as an option. I didnt notice a big difference in headroom on both inputs.

There's many ways the OTB mix could be improved and this was just quickly put together because i was curious if the ghost could keep up to ITB. This wasn't meant to be contra ITB, OTB or the ghost console in particulary, i love mine

What are the going prices for a ghost these days? I got mine for 1200 Euro, 32ch with the meterbridge and cps275. It needed some serious cleaning but now it looks like new and is troublefree. This was the best money i spend in ages for creativity, fun and sound.

Herwig went thru the tape ins. That should in many respects be very different than line inputs. On my k1 board the wiring is very different for line inputs at least per the schematics. I think theres an input transformer and opamps for the mic pre versus just opamps etc... dynamic range may be similar on ghost line ins and xlr channels but real world experiemce would help.

To the OP your test proves nothing regarding otb or itb. you mix 3 stems on a console. That's not a full mix. If we test consoles like that nobody would buy em.

Herwig would be tickled if another guy or gal comes to his ghost console. Bumps each mic pre 10db and puts the faders at unity. Set the daw levels to the console so they aren't clipping the d/a. Find digital 0 for your system with this calibration.

Now drop all daw sends to your console 15db. Strap a 2 channel mic pre across your master buss outputs. Input gain 15db output gain should peak at -6db under full mix if possible. Adjust master buss fader not to clip. Try bumping input gain on the preamp to compensate for lost gain on the master fader. Now mix to taste. Notice how far forward you can brink vocals and drums. If done properly you should at least see some differences that aren't suubtle. Again if I could get these results itb then I would have way less outboard and way less headaches. But engineering is an artisans path like finishing marble. Its a pain sometimes. But worth it.

Peace
Illumination

[DeadPoet]

Hey Illacov, you have tickled me

I have a few spare hours tomorrow, will try some tests then.



Herwig

[illacov]

Originally Posted by DeadPoet

Hey Illacov, you have tickled me

I have a few spare hours tomorrow, will try some tests then.



Herwig

if you can at least just patch right into the line inputs versus the tape ins that would be better Im thinkin?

Try the pre across the outputs on the console. Can't hurt.

I've always done it this way. Never really like the gain circuit on the master buss for most consoles. Always preferred to use as little gain or fader level on the master fader and then a nice clean pre on the outputs to boost. I use a passive summing mixer now with my console on the direct outs. Sounds bananas.

In my otb adventures I've always noticed how plugin compression sounds different versus it being itb. Maybe being on its own separate dedicated d/a clears up the transients?

Good luck.

Btw do you know about Creation Audio Labs here in the USA? They mod Ghosts. Channels as well as master section.

Let me know if you are interested in a passive sidecar summing matrix. Its an oldschool design that you would hook up to your ghost direct outputs and bypass your master section with. You would need a preamp to make up lost gain but its customizable. Mine has cinemag output transformers with steel cores. Very colored and aggressive when pushed. Very different sound from your mixes. Mine is 24 channels. 8 mono 8 stereo.

Peace
Illumination

[pjk]

Originally Posted by illacov

Why didn't you go into the actual input channel strips? You bypassed the best reason for the console which is the headroom those preamps can provide when they boost your signal. Or the headroom you can create (preamp dependent) by raising that input trim a touch so u have some headroom. The tape returns on consoles would possibly have if at all a lineamp and it would be next to nothing compared to the input block pres as far as clarity and detail. Plus you lose eq and for some inserts and additional bussing. I say try again. U will get way different results.

I dont understand this. I thought you would have to go into the line ins or tape return because they recieve line level signals (from interface outs). If you go into preamps arent you lowering headroom because you are putting a line level signal (too large for a preamp) into a preamp that is used to seeing much smaller signals (~30-40db) ? I suppose you could reduce the levels in the DAW before sending out, but this will ony increase noise due to more amplification being needed (lowering a line level signal and then raising it back up).
I thought headroom was a function of the preamp when using mic level signals and not line - and by raising trim gain you are loosing headroom as you are going towards the headroom of the circuit ? surely this is why you need line level signals (even if they are just pads) - if not then i dont understand the point of line ins (other than different connector type).
This is what i was tought, but i may have got it wrong. Could you give me an explanation as im confused and wondering about this now thanks.

[illacov]

Originally Posted by pjk

I dont understand this. I thought you would have to go into the line ins or tape return because they recieve line level signals (from interface outs). If you go into preamps arent you lowering headroom because you are putting a line level signal into a preamp that is used to seeing much smaller signals (~30-40db) which then get amplified ? I suppose you could reduce the levels in the DAW before sending out, but this will ony increase noise due to more amplification being needed (lowering a line level signal and then raising it back up).
I thought headroom was a function of the preamp when using mic level signals and not line - and by raising trim gain you are loosing headroom as you are going towards the headroom of the circuit ? surely this is why you need line level signals (even if they are just pads) - if not then i dont understand the point of line ins (other than different connector type).
This is what i was tought, but i may have got it wrong. Could you give me an explaination as im confused and wondering about this now thanks.

Call me crazy but as much as that looks right on paper, it hasn't worked out that way in my experience.

Technically speaking, the only thing stopping you from using a mic preamp is the connector. As far as levels, your preamp is the determinant for all that. Think about it. Line level outs from a keyboard usually people DI it. I just use a 1/4" to XLR cable and boom it works. Ive used a DI before as well, but I barely reach for it to preamplify a synth nowadays.

When I do layback to tape, I run out of my interfaces outputs (XLR) into my PM1000s and PM700s. Now technically this is line level stuff we're talking. Not a single problem with signal quality, headroom, noise etc..

So while it doesn't sound right: What's stopping someone from just plugging into the XLRs?? Im no Ghost expert but on plenty of consoles you'll see the line amps don't crank out as much gain as the mic pres. Now this isn't to say you wouldn't have a line amp that can't do alot of gain. But Ive always followed this principle to much success. If you can go into mic pres for summing on individual channels then do it. Usually guarantees more headroom. Set up the preamps with a nice bump in gain like 15db or so (nothing major for a decent preamp so pretty good sound quality) and then calibrate your DAs.

What you do in essence is make it so you don't drive your DAs at all and you end up with less chances for smeared, distorted or clipped audio. That's how I figure it. You're basically just shifting some duties from the D/A to the console. Not a bad idea in my book.

Peace
Illumination

[illacov]

Just to add on. This just came to me.

Think about the old days of recording. If you went into a tube based reel to reel or something straight from the console, then there were a few possibilities.

First off...Mic to onboard pres on console. Console has sends, maybe inserts so to speak of.

These are "line level" correct?

Well they either had to go into line inputs on the reel to reel (if there were any) or they just went into the mic inputs, which were going right thru the preamps on the reel to reel. While it seems a non issue to most, bypassing those preamps would eliminate alot of flavor going to tape due to the transformer coupled stages on those tube reel to reels.

Just because something is line level or not doesn't mean one can't experiment.

Secondarily, even if you JUST go to the line inputs on a Ghost or any other console are you turning up the input trim on those channels to give you some headroom and if not, why not??

This is one of my long running projects is creating that hardware bump between the D/As and summing. Its nice to have that extra headroom. Its not there to warm up or smash your audio, simply give you some headroom.

I do this with my pres on the PM1000. For example.

I can attenuate the signal down to say something silly like -20db going to my D/A output and then crank up the inputs on my PM1000 (recapped and clean PSU = no noise on higher gain). Then I route the output of my preamp to my Tascam 38. For what its worth this process is punchier than using some different D/As I have without a preamp. Without the preamp inline they distort, go figure.

But you can definitely get something similar out of a well put together console.

I can give you other reasons why having that headroom is cool for someone in my shoes too.

That extra gain on those signals (remember mine go to my direct outs post fader) allows me to punch the transformers (in my summing unit) right in the face and they say "WE OBEY." Which in turn gets me better results on my mixes.

Peace
Illumination

[pjk]

illacov

Thanks for spending the time to explain that-it is very interesting and something ive been wondering about for a while regarding how it all used to be done with tubes, transformers tape etc. I think your right here. This is exactly what im after when i finally get my garage studio finished. Ive got an old tweed pre/eq (mariner transformers) that im wanting to push on the mix buss. (not tried them yet so no doubt they will need recapping but heres hoping!)

Just to clarify - i would be reducing the signals in the DAW (to -20db) before they hit the D/As and then boosting with desk/outboard pres ?

Lastly, you mention raising trim gain on the pres or line ins on the desk to get more headroom, but i thought the more you raise the gain the closer you get to max. headroom and so are loosing headroom by doing this ?
Are you doing it because it gives more headroom to the signal than if you had simply come from the line outs of interface/d/a converter ? If so i think im starting to understand as the interface line outs must also have line amps and these wont be as good as using the boards - is this the theory ?

Many thanks

[illacov]

Originally Posted by pjk

illacov

Thanks for spending the time to explain that-it is very interesting and something ive been wondering about for a while regarding how it all used to be done with tubes, transformers tape etc. I think your right here. This is exactly what im after when i finally get my garage studio finished. Ive got an old tweed pre/eq (mariner transformers) that im wanting to push on the mix buss. (not tried them yet so no doubt they will need recapping but heres hoping!)

Just to clarify - i would be reducing the signals in the DAW (to -20db) before they hit the D/As and then boosting with desk/outboard pres ?

Lastly, you mention raising trim gain on the pres or line ins on the desk to get more headroom, but i thought the more you raise the gain the closer you get to max. headroom and so are loosing headroom by doing this ?
Are you doing it because it gives more headroom to the signal than if you had simply come from the line outs of interface/d/a converter ? If so i think im starting to understand as the interface line outs must also have line amps and these wont be as good as using the boards - is this the theory ?

Many thanks

Uh not exactly (IMHO).

You see, by adding the extra gain at the preamps on the mixer and backing down on the D/A outputs, you're avoiding hitting the ceiling on the D/A. So in essence you can get closer to 0 without worrying about clipping the D/A.

But really lets look at this.

If you turn down the D/As (via the DAW or DSP mixer) to -20db and then turn the mixers preamps to +15 db, we have a difference of 5 db before we hit 0. Could be complete horseshit but lets check our facts here. 5 db of headroom is alot, but we have to remember the summation of all those signals into the matrix on any console and how combining those signals is going to get you up to 0 pretty quick with multiple signals. So the reality is that while my examples were just to get your beak wet, you have to play around more with it. Like Turn the D/A down to something like -32db and then boost 20db for something like 12 db of headroom per channel. Most times you won't be using all 12 db of headroom and since we're using either a summing mixer to mix the signals together or maybe we're strapping a preamp across the outputs of the mixer (so we can turn the master buss down and turn the preamp up = more headroom), we can pretty much tweak the whole setup to avoid any clipping but get more "give" on our channels. On the really good consoles they have good summing sections and may not require the workarounds or bandaids.

But these are the techniques I use in my studio all the time and they work out great. If you have a mid level mixer maybe from a Mackie up to say the lesser Midas stuff, it would be well worth your while to try an alternative way to sum the signals off the mixer. You can design your summing unit to taste, transformers or not, multiple pathways, parallel mix inputs etc...

shoot me a PM if you want some tech contacts.

Peace
Illumination

[pjk]

Thats great thanks man. Got you now. Seems like a good idea now youve explained it.
I will definately give this a shot on my mackie onyx - cant afford a better board at moment but in time, in time (as Linda Blair said).
Now back to finishing this studio - god i hate physical work.

[deuc647]

Originally Posted by DeadPoet

I did a comparable test: 16channels summed in Cubase4 and the same channels summed on my Ghost.

The difference is *very* small.

As for added headroom: I tried upping the levels on the Ghost and driving the mixbuss harder but everything turned into mashed potatoes.


Conclusion for me: will continue summing ITB because of easier recall.



Herwig

I cant believe no one caught this, the ghost is a nice console, but cannot be pushed, it has decent headroom but shouldnt be overdriven, if you get the mixbuss redone, im sure it will yield much wider and deeper results, the preamp on the buss out ive always wondered about but never put it into effect, damn it, guess i got more money to spend now

[illacov]

Originally Posted by deuc647

I cant believe no one caught this, the ghost is a nice console, but cannot be pushed, it has decent headroom but shouldnt be overdriven, if you get the mixbuss redone, im sure it will yield much wider and deeper results, the preamp on the buss out ive always wondered about but never put it into effect, damn it, guess i got more money to spend now

I'd recommend something like a really clean preamp like a stereo SC1 from Five Fish Audio (70 db of clean gain or something wild like that) or if you want something colored the skies the limit.

Peace
Illumination

[chris_m]

Hi Illumination thanks alot for your interesting posts. I can definately see how your technic works for you but in the end it comes down to just adding as much noise (or better colorartion) through analog amplification stages as possible. That this will sound very different from an ITB mix should be obvious to everyone

[chris_m]

Originally Posted by cowudders

I'm thinking all the time about selling mine, but as soon as I do something with it, I change my mind again. Really weird.

OT: How did you clean your pots, as I'm sure they were crackling (like mine, after letting it sit mostly idle for 5 years)? The switches? As you are from Germany, I'd be happy to learn about the EU equivalent to US' de-oxit, or whatever it's called. thanks.

cheers from Austria,
bernd

I got lucky and mine weren't crackling at all, just the crm level has a small position where it crackles a little bit, probably oxidation. I'd like to know too

[illacov]

Originally Posted by chris_m

Hi Illumination thanks alot for your interesting posts. I can definately see how your technic works for you but in the end it comes down to just adding as much noise (or better colorartion) through analog amplification stages as possible. That this will sound very different from an ITB mix should be obvious to everyone

While this is correct...

there are many strategies one can try to see if summing OTB or using preamps on the D/A give headroom or better results than summing ITB.

For example, there's the good ole throw a 20 db pad on the outputs of a stereo DA and feed the output into a preamp. Supposedly this should be the same as summing out the box with a summing mixer with a preamp.

There's the method of just feeding the outputs into a preamp without a pad and this is supposed to be just as good as using a summing unit with a preamp following it.

The whole thing about summing for me was not color, it was extended freqency response and less crowding in a full mix.

While we have these wonderful posts about frequency masking and buildup etc..

You have to go back to the source before we say summing is really us adding color or looking for mojo.

Let's quickly go back 40 years to the Beatles.

They had very limited equipment, very little control over eq options, used tape as a recording medium, only 4 tracks to record with and only so many kinds of mics and if I'm not mistaken no outboard pres. So one pre to rule them all so to speak.

As much as its a romantic notion that you can use a mic like an EQ, when you consider that all the mics are the same damn mic, what eq are we using?

There was still plenty of clarity even in the Beatles recordings with 4 tracks, a colored medium and considering the lack of a zoom feature or spectrum analyzers for their recording setup, only their ears and the available EQ bands on their mixers to clear up frequencies.

Point being, as clean, transparent and out of the way as ITB mixing is supposed to be, it requires a ton more tweaking to create open lanes in frequencies (in my experience) than simply making the gentle cuts that are needed here and there and moving onto the next track.

To me being inside the computer is like turning down cheap volume pots on a starter Strat. The tone changes even on something simple like the volume, which should really only make things quiet or loud, versus bright and loud or dark and quiet.


Perhaps its the DA I use to monitor when I mix, perhaps its my monitors, but you have to admit that while I can't get these incredible insane results ITB with those same monitors, I can get galaxies better results on those same monitors and same DA.

Perhaps its worth mentioning that using multiple DA with no colored comps on the outputs, no inline preamps for color and only going into my summing mixer is not some warmish mojo process. You literally mix different. In fact while people will try to argue logic about what DAWs do etc... Its funny to me that the things that we keep trying to do ITB work so much better OTB. My goodness, lets not get started about the action on compressors, their tone etc.. all of that changes when you break out your mix on those many many channels of DA.

I'm getting another 12 stereo channels added to my setup for a total of 48 channels OTB come late May. Its definitely not about color.

While I can turn up my bass ITB and get results that some might say well that's fine enough, I know better. Take that same mix, throw it on 16 to 24 channels alone and watch the bass response completely change, the punch completely change.

I wish I was developing my unit because its cheap and it works.

But its not for sale.

Again I'm sorry you didn't get the results you want, but again not to be a jerk but 3 stereo stems on a console ain't giving you any results that are worth it.

If you think its not worth it to try mixing on a console and maybe you use it as a big tracking monitoring station thats perfectly acceptable. But your results aren't proof of anything except that the earlier summing mixing attempts were just as flawed because they completely under utilized the mixers in the FIRST PLACE.

Guys were sending 4 stereo stems out on 32 channel IO systems to a 56 input console and then proclaiming hmm I can't hear the difference.

No really??


To be honest, if you got the available DA I say try again.

This next statement aint about you, but Ive met quite a few lazy engineers. Quite a few of these guys know that they can get their mixes flown out to the console and get some great results, but they don't want to get issues with recall or don't want to spend the time with getting their land legs back.

Ah well..

Anyhoo, however you get to your happy place with it all I wish you the best.

I would be glad to show you some comparisons that were right on here on gearslutz but these damn servers keep losing the files I posted.

I will try and find you some snippets to post to show you the difference between 24 channels OTB and an ITB mix of the exact same song.

I believe I have some floating on my hard drive. There's one particular mix where its all plug ins, no hardware on the channels and the difference floored a ton of guys on the hip hop forum because they didn't realize how much kick gets swallowed on these ITB setups. Better DA for an ITB setup is going to help you pick out hash and rubs, but for crying out loud if I hi pass something at 35 hz, it shouldn't sound like I hipassed it at 50 hz. I track alot of electric guitar for my hip hop band and they sound really dead and flat ITB, then you fly it out and its like you're hearing them live again. Sorry to be so sweeping and general but yeah, its like you just rewound history and got the amp mic'd up again and dude is playing his heart out.

I'm going to stop here.

Just hope I was able to fill in some gaps for people.

Peace
Illumination