Skip to main content

60v summing had me and now 120v mixing is opening up new possibilities. What do you think?

Comments

TheJackAttack Fri, 04/29/2011 - 13:09

I think that most electronics deal with microvoltages and milliamps. The actual analog inputs/outputs etc are at most 48v. I think mostly that it is crucial the power supply be very well designed for whatever voltage is specified on the mixer/summing box/preamp/what have you. When folks discuss the value of high voltage rails for a tube preamp circuitry it is for a very specific purpose. So basically you have to know why the designers specified "120v" and what perceived or actual problem it's supposed to solve. Remember, in some circuits voltage is additive and in some it's constant. Make sure you're not heading in a direction with a knob marked "11".

audiokid Fri, 04/29/2011 - 13:26

TheJackAttack, post: 369910 wrote: Make sure you're not heading in a direction with a knob marked "11".

Now that was perfect!

Unless we get the opportunity to hear and experience how much easier it is mixing with juice like this, and without knowing at this point I can only optimistically assume this must sound juicier. It should be that much easier getting there (big and open sound) in less time. I mean, wow, what a difference between the wanna be pre amps and the big boys like my Millennia M-2b and...
I think you've nailed a main question.
I'm going to invite SPL here to hopefully enlighten us as to why they've made a mixer that is described as an emotional experience.

Big K Sat, 04/30/2011 - 05:04

Why 120 volts?
SPL has engineered a completely different approach to HiEnd audio processing. With the HiGain solution the get MANY advantages over the common technology.
SNR 116 dB at 35 (is it called ?) headroom, which gives you a dynamic range of 150 dB!!
That is better than the requrements for PCM 24b /192 kHz or 1 bit DSCat 256 kHz SR,
The usual op amp chips were replaced, because those have additional irrelevant circuitry inside for other industrial applications.
No coupling condensers in the front end to prevent noise. The whole signalpath is kept symmetrical and runs through various stages towars the ultra low noise amp transistors.
Since both lines of the symmetrical signal are independent, the produced noise does not add up. The total noise of the signals is thus lowered.
The Supra amp works on plus/minus 62 volts and the signal path is constructed with the best components available till signal reaches the separate class A amp.
The main gole was to achieve High amplification with lower phase shift and lowest distortion over a bandwith of 200 kHz...

This is a hell of a constuct, giving you pourest audio through top notch engineering and materials... I doubt that there is anything better, at least, not known to me...
Big K ( enjoys the SPL PQ... http://spl.info/en/hardware/eqs-vitalizerr/pq/in-detail.html )

audiokid Sat, 04/30/2011 - 09:54

SASman, post: 370013 wrote: I literally never thing about voltage, good gear sounds good at the voltage it is designed to run on IMO.

I have a well designed Yamaha Console with preamps that don't sound half as good as my Millennia M-2b or ADL600 pre-amp. I don't know what you mean there? I'm pretty certain mixing and summing is about headroom.

audiokid (enjoys the SPL Passeq and wants the NEOS:)

Big K Sat, 04/30/2011 - 10:53

Chris... for God's sake...If you have a safe budget and it is what you need, get it ...!
There is no sence in repeating your ordeal... A mixer that can match these specs I have not seen, yet.
Make sure you have a thorough listening! You won't get any sweetening colorations of a NEVE or SSL consoles, etc.
and 10 Grand is lot for a summing console w/o EQs, etc. Again, we are in the last 2 percent corner ...
:-)

Hello S&S
Any gear must work with the voltage it was designed for.
That does not rule out that some concepts need higher voltage and wouldn't even be possible to design with lower voltages.
120 volt is not a marketing gag, at all... It is an expensive way to HighestEnd audio. If one needs such pristine performance, well... if I
had use for it, I'd probably be very tempted to get the best...

Big K ( enjoying a cold beer, right now..lol)

audiokid Sat, 04/30/2011 - 11:22

Nicely put Big-K!

Well, I'm going to enjoy my MixDream for a good while but if SPL sent me this beast, I certainly would know what to do with it! I'm interested in the 120v rails for one reason, headroom (space) and hopefully my current summing system will help me be able to separate the group of tones better OTB than what I am experiencing as limitation to ITB masters, without adding noise. I'm not looking/ expecting console color or bells and whistles. My choice selection of preamps and the DAW do this very well. But, who knows if its the gospel solution. I'm simply blessed to be able to dig deeper into both realms and never give up trying to sound better. I definitely don't think ALL digital is the Holy Grail to perfect acoustic music.

The NEOS appears to be spot on because it is a headroom and monitoring CHAMPION, both area's that ( to my ears and knowledge) a single pass ITB master is limited and misleading due to what happens with larger and more complex track counts that get crammed into a corner (during stage one).
The combination of organic acoustic music and digital audio doesn't quite fit well to me so... I believe headroom of two kinds will help space it out and make more accurate printing to CD or a stereo master easier.
I'm going to spend this next year learning as much as I can about this because my love and 30 years experience in acoustic and digital music programming is telling me to. When I see headroom, I get excited.

I hope to post examples so we can all move past wondering.

SPL Thu, 05/05/2011 - 04:32

120 V

First of all thanks for the invitation to provide some information - and sorry for the late reply ... easter vacation.

120 volts
BigK has outlined the basics of 120v engineering very good, thanks :)
It is also important to know that for the Neos not only some stages, or OPs, but the whole unit is based upon 120 volts operational voltage.
So why 120 volts? SasMan is right, everything should sound good at its voltage. The proof of good sound is a result, and this must be delivered. As some argued we use 120v technology as a marketing argument, we are not afraid sharing the results we deliver with our 120v devices (for example:  http://spl.info/ind… Bob Ludwig: Sound Performance Lab
The main point with 120v rails: the more energy we have to process signals, the higher the dynamic bandwidth. This simple fact does not mean a unit sound good because it runs on 120v rails, but it is a decent base to work on. We all know, achieving good results with a complex device such as summing mixer comes from many aspects, but here I focus on specific 120v issues rather than repeating common knowledge.

The advantages of high dynamic bandwidths, as a matter of principle, are less and later distortion. With a standard +/- 15 v circuitry, overload resistance may be at around 20 dB. With the Neos, we are clearly above 30 dB (limit of measuring equipment). A high dynamic bandwidth enhances in two ways: upwards we have more headroom, downwards we have more distance towards noise (= better signal to noise ratio).
Now, if a converter delivers 24dB, you do not have to lower its output signal (which would reduce bit resolution). Currently there are no units at all delivering levels which could overload a Neos input. Internally, the high level is running on, and when you add (up to) 6 dB, you still have a good headroom. So whatever you feed, whatever you add - no distortion. This allows for "free mixing" without even thinking about level control - and therefore, no distraction from the creative part of mixing life. And it allows to keep dynamic levels, and their differences, in the maximum available bandwidth you can process nowadays.

Analog Summing
We are in close contact with the German physicist Ralf Koschnicke who proved advantages of analog summing against ITB mixes. His main point: While 24 bit is a sufficient dynamic resolution, digital resolution on the TIME SCALE is not sufficient.
Ralf Koschnicke will publish his work in english soon, for now I can only link to the German version of his work on this (PDF, 2,4 MB):
http://acousense.de…
Here is a quote on this from his website [url=http:// http://www.acousens… ] http://www.acousens…
"Contrary to the view enshrined in the development of the CD and specification of the industry digital standard of around 25 years ago, we know today that human auditory perception can evaluate many more minute details in the time domain than can be detected within the frequency transmission range of standard digital technology (the smaller the details the higher the frequencies needed)."
http://acousense.de…

BTW, I recommend the fantastic Acousense HD recordings.

We have read in some pro ITB statements that "analog distortion" may deliver sound results preferred by some engineers for some genre. Sometimes also analog equipment is associated with adding noise.
While this may apply to some equipment, we can prove for all SPL 60v or 120v analog summing devices that both distortion and noise values are far away from being perceptible. With the Neos we are heading towards the boundaries of measuring equipment. Our analog summing solutions do not add sound-coloring distortion or noise - they are by definition, concept and specification no effect processors.
We, and also many customers, have always associated analog summing with benefits especially in terms of localization and spatial imaging (transparence and depth). Our approach has always been based upon empirical methods. Now Ralf Koschnicke's scientific analysis exactly supports our impressions. We are looking forward to spreading the news :)

The Neos is a suggestion for a hybrid digital/analog studio solution to combine the best of both worlds, regarding sound quality, workflow, and costs.
We think in combination with a DAW, the Neos is an interesting alternative to strict ITB or strict analog concepts. We are not against anything, but we are looking for improvement. Hence we offer another choice and we would be glad if engineers are interested in getting to know it.

Neos
... is not "only a summing mixer". We think its advantages and the package justify the price. It provides:
- a 24mono/12 stereo into 2 summing mixer
- with faders
- an ultimate quality monitoring controller

Price
Compared with a high-end collection of a summing amp with fader box plus monitoring controller, the price is just competitive - not even mentioned the unique qualities of 120v technology and manufacturing (selected/handmade components, handmade production in Germany).
Apart from the isolated view on package and competition, some other considerations may also be interesting for a correct assessment, for example:
- The Neos is made for, you guessed it, professional production environments. A decent acoustic room planning and installation starts at investments of about 20.000 Euros/USD (open end, of course). If a studio once is conceived as a DAW-based environment, the way back to a full-sized analog desk is not an option as it would destroy the acoustics. With the Neos, you keep everything, mount three processors in a side rack and place it into your desk - it integrates seamlessly into a DAW studio.
- Compared to a full-sized analog desk, power consumption of a DAW/Neos setup safes a lot of money - depending on local energy costs, it can pay back pretty soon. CO2 footstamp not mentioned :) (standby power consumption: 75 watts)
- The Neos is a very space-saving solution for mobile recording etc. (19 inch/7U)

Emotion
We were excited about the first reactions of engineers who tried the Neos prototype at the Musikmesse in Frankfurt, as they all reacted emotional. But we prefer to let others talk about the results when we have delivered them, so we are really looking forward now to the first user reports and their productions.

Cheers,
Paul

SPL Fri, 05/06/2011 - 01:17

Yes Boswell, I mixed that up a bit, sorry.
Our 120 V current runs on +/- 60 V rails.
+/- 15V results in 30 Volt current - common for pro audio.
Some good desks use +/- 18V, we know +/- 24V from older discrete designs.

I have found the terms "overload resistance" in a dictionary some time ago; it was suggested for German "Uebersteuerungsfestigkeit" - got it? :)
I guess you simply say max. input level/headroom.

IIRs Sun, 05/22/2011 - 04:37

SPL, post: 370615 wrote: Here's the english version of Ralf Koschnicke's essay (in my post 19 I only had the link to the German version)
[="http://www.acousenc…"]Aspects_of_Audio_Transmission.pdf
[/]

I'm not convinced.

Lets accept your assertion that a 48Khz samplerate has too low a "temporal resolution": in what way does converting that signal to analog for the summing stage help? This is like suggesting that converting all your 128kbps mp3s to 24 bit wavs will restore the lost detail!

Surely the only way to improve the "temporal resolution" would be to use higher samplerates..?

SPL Mon, 05/23/2011 - 00:06

You are right, 48 kHz is not sufficient. But exactly this is explained - and why 96 kHz provides a sufficient sample rate in RECORDING or STORING audio, while only higher sample rates may keep all the detail in MIXING/SUMMING that can be assumed if you proceed from our hearing capabilities reg. localization.
According to the idea of this essay, you have to do with very fine signal structures in summing. Keeping small structures requires accordingly high resolution - a physical law. When you mix many channels, complexity can get higher than in recording (whatever you record - it is already summed ;-). Therefore, perfect summing may require even higher sample rates than recording.
This is one aspect - resolution on the time scale. The other one is possible error propagation in digital summing. Both aspects may interesting for ITB/OTB summing theories.
As I already stated before - usually, we only read about "analog dirt" (noise, distortion) in such discussions, and we wanted to enrich this discussion with a few other aspects.

IIRs Mon, 05/23/2011 - 03:05

SPL, post: 371578 wrote: Therefore, perfect summing may require even higher sample rates than recording.

This is the main problem I have: as far as I can tell digital summing is already perfect. Its one of the few aspects of mixing ITB which will not introduce aliasing, and (as I understand it) would therefore not benefit from oversampling at all.

My maths isn't up to proving it, but I strongly suspect that (assuming the same bandlimited source signals, and perfect linear phase up and down sampling filters) oversampled summing would null perfectly with a non-oversampled version.

Perhaps you could suggest a test I could run to verify your assertion that conventional digital summing is indeed flawed?

SPL, post: 371578 wrote: This is one aspect - resolution on the time scale. The other one is possible error propagation in digital summing.

I tested this one: Reaper can sum hundreds of channels with error rates down at around -140dBFS approx. If this was really significant, surely 16 bit dither noise at > -90dBFS would be totally disastrous?

IIRs Thu, 05/26/2011 - 02:14

Ok, so I attempted to test the oversampled summing thing myself.

I started by importing 4 drum loops into Reaper (44.1KHz), dropping their gains by 12dB, and rendering the mix to a new file, which I will call "A".

I then loaded all 4 files into Soundforge, and upsampled them to 192KHz (using the iZotope 64 bit algo with linear phase settings). I then loaded them back into Reaper again, set Reaper's rate to 192KHz, and rendered file "B".

Obviously A and B are at different samplerates, so I had to convert in order to compare. I tried downsampling file B back to 44.1KHz (using the same iZotope algo): the two files nulled in the low frequencies, but showed significant differences at the top end.

So I then tried upsampling file A to 192KHz: this time I got quite a good null, down below -80dBFS.

Of course, the only really conclusive result would have been a 'perfect' null. But it seems to me most likely that the differences in the results are just artifacts from the rate conversions...

SPL Thu, 05/26/2011 - 06:16

IIR,

thanks for your interest in this, and sorry that I currently cannot participate with quick replies ... busy, and we all discuss the whole issue here, too :)

First let me please try to avoid possible misunderstandings (also for other readers of this thread).

1. We - or the author of the paper we have linked to - are not recommending oversampling from 44,1 or 48. In contrast to that, the paper gives arguments for recording and storing with at least 96kHz. And there are arguments to mix/sum with even higher sample rates. One main idea is abput It is about keeping smallest structures we are capable to perceive, and which conditions/specs may be required.

2. We know, of course, the arguments in this paper are questioning accepted wisdom. This challenges controversial comments, or attitudes. However, we do not do that in order to prove others wrong in the first place. We are searching for explanations of quality differences we - and many engineers - have identified between analog and digital summing.
So we are not coming from a theory to practice. We are coming from empirical findings and we are looking for explanations beyond strange statements about analog technology. We have hi-end machines here, dead silent with headroom hitting the ceiling, we have wonderful results - and that is supposed to be "analog coloration"?!

On the "error propagation" idea - I do not think that the described oversampling scenario would be a valid test routine, as it is self-referential.
They key question is: what happens if you mix many channels before a final D/A conversion, instead of summing with an analog machine which is fed by single previously A/D converted channels? And of course we are comparing to a result as close as possible to the original content - which is not a signal recorded in 44,1/48.
Nowadays, A/D and D/A conversion in 24/96 gives a very good representation of an original waveform. But as a matter of principle, interpolation and measuring tolerances can produce errors. As I just said, not too much at 96 for a single channel.
But there is the law of error propagation - and you can not only add good things in summing, you have to add everything, also errors. And so the summed signal may consist of a higher error rate than each single channel. Now comes D/A conversion, and the converter deals with a remarkably higher error rate for the summed signal than with every single channel conversion.
Please note I try to repeat in a short, comprehensible form - please check again the details on page 18.

The author did not do further research to specify the dimension of such possible errors. But there is one further thought on this: summing at 48 or 96 kHz results in 10 microseconds difference reg. time scale resolution (large parts of the essay are underlining the importance of resolution on the time scale for keeping finest structures, and the importance of such fine structures especially in summing). Therefore, mixing in 48 could produce deviations due to error propagation in a comparable dimension. On the other hand, jitter is measured in nanoseconds and good values are given in one-digit numbers.

Again, the part about error propagation obviously is an idea, or hypothesis. But we thought there is quite some evidence, and both aspects, resolution on the time scale and error propagation, may explain something about the differences between digital summing and high-end analog summing.

IIRs Thu, 05/26/2011 - 07:50

Thanks for the reply.

SPL, post: 371754 wrote: 1. We - or the author of the paper we have linked to - are not recommending oversampling from 44,1 or 48. In contrast to that, the paper gives arguments for recording and storing with at least 96kHz. And there are arguments to mix/sum with even higher sample rates.

So by implication, if you recorded at 44.1/48 there would be no benefit in going out of the box for summing either..?

SPL, post: 371754 wrote: One main idea is abput It is about keeping smallest structures we are capable to perceive, and which conditions/specs may be required.

But, once you've lost them, you've lost them, correct? Upsampling won't put them back, nor will converting to analog.

SPL, post: 371754 wrote: We are searching for explanations of quality differences we - and many engineers - have identified between analog and digital summing.
So we are not coming from a theory to practice. We are coming from empirical findings and we are looking for explanations beyond strange statements about analog technology.

It seems like you are saying that you started from the assumption that analog summing is technically superior to digital. Did you test that assumption at all, and if so what technical deficiencies did you discover with digital summing?

This is the fundamental problem: no-one seems to have yet been able to demonstrate any actual tangible flaws with digital summing. As far as I can see adding up strings of numbers is one thing that computers do really well, without adding any nasty aliasing. Would it not make more sense to mix entirely analog, then go back into the box for the summing stage?

SPL Fri, 05/27/2011 - 01:51

As I understand the essay, it is of main importance to separate recording specs and summing specs on the one hand, and the two aspects of resolution on the time scale and error propagation.
If there is a loss by recording with 44.1/48, you certainly cannot get it back by oversampling. Oversampling as such is not an issue in these thoughts at all. But it is explained what could be preserved by recording with at least 96 kHz - or what we lose with the "CD format". Time resolution is a main issue here.
In summing, the problem of error propagation may add to time resolution aspects. Error propagation would also affect summing of signals recorded in 44.1/48, as explained in my last post.

We are saying that high-end (!) analog summing results are superior in quality over digital summing. This can be proven empirically by results - such recordings do exist :)
The author of the essay, Ralf Koschnicke, does classical and jazz recordings with his label Acousense. He offers comparison sets with CD, 24/96 DVD, and 192 FLAC files of the same recordings. Stunning differences.
All SPL tech staff, founder, product manager … are musicians or engineers, too - we all know about the differences. We are not fooled by effects, as we use high-end units that do not introduce perceivable effects - this can be proven by measurements. And … where is the successful, big facility without analog console? How many award-winning productions are not done with analog desks?
Our impression has always been that hi-end analog summing is superior especially in terms of localization, transparency, spatial imaging, the way how instruments "blend" from one to another, how they are embedded in a mix. There is quite some evidence that these qualities may really be related to aspects of time resolution and error propagation.

Sure, a computer is good in adding numbers. But, as I tried to explain before, and as explained in the article, you may lose information at two stages before the box actually starts calculating: insufficient resolution, error propagation in complex operations introduced by sampling (=measuring with tolerances). This would also happen if you sum digitally after mixing in analog - though the last step, adding the numbers, probably works out :)
A combination of DAW and hi-end summing mixer can change the game in many DAW-based studios with a focus on audio quality. That is why we are really looking forward now to the first productions made with the Neos.

IIRs Fri, 05/27/2011 - 06:15

SPL, post: 371796 wrote: We are saying that high-end (!) analog summing results are superior in quality over digital summing. This can be proven empirically by results - such recordings do exist :)

Are these recordings available online anywhere? Could you provide a link?

Of course, for these recordings to prove anything there would need to be a comparison: a digitally summed version of the mix, plus a (precisely calibrated and level matched) analog version. Ideally there would also be a third file: the digital mix bounced through two channels of the analog device in question, using the same converters as the summing test.

SPL, post: 371796 wrote: The author of the essay, Ralf Koschnicke, does classical and jazz recordings with his label Acousense. He offers comparison sets with CD, 24/96 DVD, and 192 FLAC files of the same recordings. Stunning differences.

In what way does that relate to summing?

SPL, post: 371796 wrote: All SPL tech staff, founder, product manager … are musicians or engineers, too - we all know about the differences. We are not fooled by effects, as we use high-end units that do not introduce perceivable effects - this can be proven by measurements..

A modern DAW's mix bus does not introduce perceivable effects - this can be proven by measurements.

SPL, post: 371796 wrote: And … where is the successful, big facility without analog console? How many award-winning productions are not done with analog desks?.

I have no idea! To be honest I would guess quite a lot these days... but that's not really the point: if you mix a record on a large format console with expensive analog outboard processing, and it sounds better than a mix done ITB with plugins, in what way does that prove that the SUMMING stage made the difference?

SPL, post: 371796 wrote: Our impression has always been that hi-end analog summing is superior especially in terms of localization, transparency, spatial imaging, the way how instruments "blend" from one to another, how they are embedded in a mix. There is quite some evidence that these qualities may really be related to aspects of time resolution and error propagation.

But have you ever attempted to verify that impression? I'm talking about a test like the one I describe above, which eliminates all variables except the summing bus?

I also have to call you out on this time resolution thing: you repeatedly imply that the samplerate of a recording limits the "temporal resolution" in some way. But this is blatantly not the case, otherwise (surely) it would be impossible to implement a digital delay shorter than 1 sample. But fractional delays are absolutely possible: check out the free Voxengo Sound Delay plug for example, which will happily delay your signal in 0.01ms steps at any samplerate. Or the well established industry standard BSS Omnidrive PA system controller which also allows time alignment in 10 microsecond steps despite an internal samplerate of 48KHz.

So, while my maths isn't up to arguing Fourier theory with you, I know for a fact that digital sampling systems can represent time differences smaller than a single sample.

SPL Fri, 05/27/2011 - 08:38

IIRs, post: 371799 wrote: Are these recordings available online anywhere? Could you provide a link?
Of course, for these recordings to prove anything there would need to be a comparison: a digitally summed version of the mix, plus a (precisely calibrated and level matched) analog version. Ideally there would also be a third file: the digital mix bounced through two channels of the analog device in question, using the same converters as the summing test.

Strange somehow that we have to prove what works - and is published - since decades ... but you describe a convincing demonstration indeed, thank you for that. I will suggest that here.

In what way does that relate to summing?

Well, I was listing examples ... these comparison sets are not demonstrating summing differences exclusively, of course, but differences between storing formats.

A modern DAW's mix bus does not introduce perceivable effects - this can be proven by measurements.

... ok, but as long as you just compare what happens after and before conversion, such measurements could be self-referential, as stated in my previous post.

if you mix a record on a large format console with expensive analog outboard processing, and it sounds better than a mix done ITB with plugins, in what way does that prove that the SUMMING stage made the difference?

In what way it does not? I think I have repeatedly described possible explanations here. What do you think why analog sounds better?

But have you ever attempted to verify that impression? I'm talking about a test like the one I describe above, which eliminates all variables except the summing bus?

Sure! We even have just built a product to do so every time we turn it on. That is a pure summing mixer, and as I said: we are looking forward to demonstrate exactly this - pure analog summing.

...
So, while my maths isn't up to arguing Fourier theory with you, I know for a fact that digital sampling systems can represent time differences smaller than a single sample.

Sure you can delay a signal at smaller rates than 1 sample. But delay and reproduction of smallest signal structures is not the same. Reproduction is not possible below a half wave length of the max. frequency. See pages 16 and 17 of the PDF.

IIRs Fri, 05/27/2011 - 09:12

SPL, post: 371804 wrote: Strange somehow that we have to prove what works - and is published - since decades ... but you describe a convincing demonstration indeed, thank you for that. I will suggest that here.

If I were to set up such a test, for publication in a well known music technology magazine, would you be interested in participating?

SPL, post: 371804 wrote: What do you think why analog sounds better?.

There are well established, non-voodoo reasons why analog compression or EQ might be superior to digital equivalents: analog processing does not introduce aliasing, while digital processing often does. In the scenario above it seems infinitely more likely that the superiority of the 'big studio' mix is down to better processing, better monitoring, and probably a better engineer. Concluding that all the other differences are irrelevent and that it must be down to summing alone seems rather a leap of logic to me.

And anyway, I don't think analog does sound better! Analogue and digital both have their strengths and their weaknesses, and I believe in taking the best of both. Analogue gear is great for adding 'mojo': funky non-linearities and quirks when driven hard can add character and interest, and no special measures need to be taken to avoid aliasing. Digital gear is not so good for adding 'mojo', unless great care is taken designing clever anti-aliased algos, or lots of cpu resources are thrown at oversampling. But it is really good at adding a bunch of channels together with practically infinite headroom, and a noise floor way below that of the DA converters required to sum OTB.

audiokid Fri, 05/27/2011 - 11:24

IIRs, post: 371807 wrote: If I were to set up such a test, for publication in a well known music technology magazine, would you be interested in participating?

IIRs, I was expecting this for some time and could sense you were leading this topic seeking opportunity. I'm not sure I like your ulterior magazine motives from this point forward though, but I do very much appreciate learning.
I've been interested in hybrid DAW systems for years. I've been a DAW user for over 30 years and have never liked certain digital sounds as much as analog. I keep trying to get that fat analog synth sound out of VSTi and look at digital recording in the same way.
Only until the last few years, when I could actually afford quality built analog gear did I get into hybrid audio recording like I am now. It definitely costs money and it is a learning process combining digital and analog together. I have a lot to learn. Whether one is better than the other isn't really why I am going this direction. Its about the sound I'm after.

Taking this discussion away from here, using RO to lure this debate over to a magazine to give yourself a name, I see why you are engaging this now and its somewhat disappointing. On the other hand, where ever I can gain more knowledge about analog recording techniques excites me. I hope we keep Paul ,SPL and even get Ralf here to share more on what they are learning. I think its fascinating and I think our reader like it here too.

That being said, I'm interested in analog for a variety of reasons and do not think one or the other is better. I don't think SPL is saying this either. I do think the combination of digital and analog (hybrid) is definitely very interesting and has benefits that we are discovering. Hybrid Audio is only in the infant stages. We have a lot to learn,

Paul, I hope we can get Ralf here some time too and continue this discussion here?

IIRs Fri, 05/27/2011 - 11:41

audiokid, post: 0 wrote: Taking this discussion away from here, using RO to lure this debate over to a magazine to give yourself a name

That wasn't my motive at all. Sorry if I was tactless to suggest it: I just participated in a microphone shootout for the magazine, and it occurred to me this was also something worth testing. Don't know if they are even interested in the idea...

Perhaps recording.org could organize such a test instead? All we need is a common set of stems really... obviously we would all need to calibrate carefully, but we could print tones at the start of each stem to double-check.

Davedog Fri, 05/27/2011 - 14:15

I, for one, am more than interested in such a test. This is a fascinating discussion and I feel we are blessed here at R.O. to have such a discussion being conducted in a gentlemanly way while still being pointed and aggressive for results. Most other forums where things like this have occurred, usually result in insults and name-calling without any real conclusion other than opinionated fal-der-all.

Carry on gents.

Guitarfreak Sun, 05/29/2011 - 20:12

It seems that the most compelling arguments that have been made deal with differing workflows rather than the amount of headroom in the final stage of signal flow. I do believe that there is a difference between the sounds of analog and digital gear, but what that difference is is largely dependent on the circuits/program coding being compared. Whether or not one is "better" than the other depends on preference. To me, the typical sound of "analog" is more compression and warmer tonality with more coloration, while the digital systems are colder and more matter of fact. If you were to increase the headroom on the piece of analog gear in question and subsequently design a circuit which took advantage of this aspect, wouldn't you be heading in the direction of emulating the accuracy of a digital system?

References keep being made to those "sweet sounding vintage recordings with analog gear" as a staple of reference. I'll tell you one thing, they didn't use analog summing mixers which ran on 120v for maximum headroom without coloration back then. If you want that sound, use the gear which made it.

Davedog Sun, 05/29/2011 - 20:42

Havent we been 'summing' analog since the beginning of the digital revolution? I thought it was a given in the big studios that you ALWAYS ran the signals out of your Mix or even HD system back through your console for the old 'warm up'. And the tracking in analog only to dump into PT or other DAW medium directly after tracking to save passes and lossy issues . And now theres CLASP. Which seems to be the do-all be-all if you still own a tape recorder.

audiokid Sun, 05/29/2011 - 21:58

From what I can tell, the big switch took hold when Pro Tools HD came on. PT Mix was still rasp so there was more OTB summing then. Many studios that were summing OTB, I'm hearing have sold those beautiful consoles or they are sitting in hallways because they take up too much real estate in the control room now and no one wants them.
After this 5 year run on PT HD I see some are finding the way back a bit and where SPL is developing. Ten years ago you couldn't convince me to do this, but it makes more sense to me now.