Ok, time to enter the lions den.
 
If I am so wrong about recording hot, why then do at least two interfaces have a function built in that emulates audio hitting tape and or headroom on a analouge desk??
 
Here is MOTU's offering: http://www.motu.com/products/motuaudio/828mk3 this is the MOTU 828 mk3 and included is their Cuemix software, which allows you to print with compression, EQ and Reverb.  Here is the blurb:
 
Apply hardware DSP-driven compression, EQ and reverb to every input and output, independent of your host computer. When recording from the mic/guitar inputs, the 828mk3’s signal overload protection gives you an extra 12 dB of headroom above zero with no digital clipping or harsh artifacts.
 
The MOTU series comes with DSP EQ and Compression on every channel, including a LA2A emualtor, so to be able to drive this unit slighty hot is very importent.  Otherwise you would not get the full benefits of the emualtion.
 
The 2nd unit that has a "softclip" effect is Sonic-Core or Creamwares XCITE unit http://sonic-core.net/joomla.soniccore/index.php?option=com_content&view=article&id=28&Itemid=18&lang=us I use to own Creamware cards and regulary used the softclip effect whilst recording into Sonar.  Or if you like printing the audio with the effects applied, rather than recording clean.  Here is the blurb from the Scope effects manual:
 

Soft Clip M/S

The SoftClip module permits the

loudness of a signal to be boosted

without introducing digital overload

distorion (hard clipping). Even with

signals which are already at the

maximum level, an apparent loudness

increase can be produced. In addition,

the sound takes on a warmer, more

"analog" character. Both mono and

stereo versions of the effect are provided.

Controls

Drive

Adjusts the intensity of the effect. An

orange LED lights when the effect begins

working.

Level

If you wish to alter merely the sound of a

signal and not its loudness, you can use

this control to counteract the loudness

increase produced by the effect.
 
This is what I like the most:
 
Even with

signals which are already at the

maximum level

So can anyone explain why on two audio interfaces we have over protectors and or Softclip, if there wasn't an element of truth to what I have been purporting, that is there can be some benefit to recording hot and this could be as in the MOTU units, where you have an emulation of the LA2A on every channel and overdriving the LA2A is part of that particular units sound.

 

Or in the case of The Scope Fusion platform, SoftClip, they are trying to emualte 2nd level harmonic distortion on the way into your DAW.

 

Any opinions on this idea???

You're reading what you want to read from this. None of this 'evidence' suggests anything of which you claim. You're reading between the lines and extrapolating your own ideas. If I were to read between the lines (which doesn't mean I'm right), I would extrapolate all these to be mainly marketing tools - making the unit do more than it really needs to do and to maybe look more suitable for people who think that hotter sounds better so they sell more units to a wider audience. It's certainly got you hooked ;)

Can this end already?

Two separate issues, Ben. Hitting tape hard does have its benefits, but it's got nothing in common with digital recording. 

First of all, with tape you need a fairly hot signal just to assure an acceptable SNR because the dynamic range of tape isn't as wide as digital. Second, if you hit the tape hard enough you can push it into saturation, where the tape loses linearity and imposes a smooth and (usually) pleasant compression. None of this is applicable to digital recording.

So, the device uses compression on the signal if it's hot enough to clip. Lots of interfaces and other devices do that. It's great protection for people who try to record too hot.

But any significant audio benefit can only be applied in the Analog domain.

This is what Ben, in my opinion, is getting confused about by "recording hot"

If you clip anything in the digital domain the result will not be musically related to the data that's been clipped, and will NOT produce pleasant sounding 2nd or 3rd harmonic distortion

End of.

Bristol_Jonesey


...End of.

If only....

Apogee was the first to offer soft clip, I believe.  Here is the blurb.

"APOGEE DUET 2 – SOFT LIMIT

Soft Limit can be heard on countless platinum selling records around the world. It employs an analogue design to prevent digital clipping by rounding off transient peaks in real-time before they pass through the analogue-to-digital converter. In essence, this allows you to record several more decibels of apparent level whilst subtly providing an analogue-style warmth to your sound! This really is some impressive technology! "

I've highlighted the revelant part.  It is an analog process before the convertor.  Some digital processing also mimic this analog saturation/distortion.  I think CraneSong was one of the first, but in every case I'm familiar w/ it is an analog-type of effect done in the digital realm.  I can't think of one product that replicates digital clipping, since it is easy enough to do that yourself and the cheaper the convertor the easier it clips.  You don't need any fancy, expensive hardware to enjoy digital distortion, only an ability to turn everything up to 11.

@

The fact we are talking about DSP based effects mean we are in the digital world and have already gone through our converters. Driving HOT does not necessarily mean pushing your A to D close on the way in.

There is no reason why you cannot just set your incoming rms levels down at a ref point such as -20 dB FS. That is where the soft clip comes in on some A to D. They are just catching the signal  to keep things under control and prevent going into odB FS. It would be a form of limiting I would say. And 12 dB of headroom might just be another way of saying the signal can go 12 dB above normal and it will be contained down before it eventually may clip. You don't have to come anywhere near it. We are cruising right down at -20 with 20 dB of headroom before any of these things start to apply.

Once into the digital world the signal flows into the pre DSP that Ben is talking about. Nothing to stop you driving the input of that plugin hard and really making it work. But then the final output of any plugin or process can be reset back down to -20 dB FS or ref level. Only driving hot within the plug to get the best out of the DSP process, then clean lower level digtial from that point on again. Onto a track to be recorded etc.

My EMU 1212 setup has got onboard DSP processing from a raft of effects and they sound pretty good actually. Not my first choice but very useable. So the effects can be applied, heard and recorded right from there. They also pop up inside your DAW as VST plugins which is another great way to access them. You don't have to use a lot of stuff on the way in but sometimes some gain, downward expansion and just very loud catch transients limiting can make a big improvement to the quality of the track you are recording.

AT


.... It employs an analogue design to prevent digital clipping by rounding off transient peaks in real-time before they pass through the analogue-to-digital converter. In essence, this allows you to record several more decibels of apparent level whilst subtly providing an analogue-style warmth to your sound! This really is some impressive technology! "

I've highlighted the revelant part.  It is an analog process before the convertor.....

Exactly. Soft Clip, Soft Limit (Limiter), whatever they call it... is still an effect that processes the signal before it goes into the A/D converters. It is little different than running an inline compressor/limiter/saturator/whatever in between a signal source and the interface. If you can crank up the output of those compressor/limiters so that they reach, or exceed, 0db,  the results will not be pretty and for sure won't add anything musical to the end product.

Ok lets get a few things straight, I never said digital clipping or overs was a good idea.  I said a level of -6 to -3 db was the desirable level to aim for when setting input levels.  Even Danny D says he sets a level of -6db as his input level.
 
What we have been talking about just so we're clear is a)resolution.  So to recap, resolution was whether we lost any audio information if we recorded the input signal too low.  Most have said this is not true but it may have been the case when we only had 16 bit converters.
 
Then b) and I may have gone on a tangient on this one and confused readers, 2nd and 3rd harmonic distortion and how to achieve this.  I argued that to get this most desirable effect, we had to record somewhat hot.  Now this is where it got confusing and I never suggested that I was the only one who did it: was this idea of hybrid musician's and the use of real and virtual instruments and this perhaps because of audio interfaces rather than a mixing desk needed another think about how we set levels and record into the ubitiquious black box.  This was to achieve 2nd and 3rd harmonic distortion.
 
Now if my little knowledge of electronics are correct, each pre-amp whether that be transitor based or tube based has optiumum operating level and somewhere around this level we start to get harmonic distortion 2nd for Tube based pre-amps and 3rd for Tranistor based pre-amps.
 
Now I have added c) Softclip which it seems quite a few of these audio interfaces now employ.  Softclip, Soft Limit and Overload protection or whatever you call it suggests that there is an element of truth to what I am saying, that is to get 2nd and 3rd level harmonics on the analouge side of your convertors you may need a fairly hot signal.  Now what is also cool about these DSP effects is that they emualate some form of saturation, whether that is Tape or Tube.
 
Now I never mentioned anything about clipping your convertors, ok!! I understand this is bad.  I did however suggest that once the audio passes into the convertor, who know's what happens.  And in a link to a GearSlutz, Paul Frindle's based Digital Myths thread he said there might be something going on in terms of digital harmonics.  Now I am not going to pretend I understand this or say it's true because it's just not possible to know this without doing some proper testing.
 
This is where people were getting confused, people thought I was talking about clipping, when in actual fact I was talking about this concept and wether this was possible and or desirable??
 
Here is where Danny D might be correct, I can write a lot of babble and it can be a little hard to decipher but I hope the above statement clears up what I have been talking about.
 
Ok Jeff's above statement is partially true but the only Soft Limiter (which is what these devices are) that actually allows you to change the level is Creamwares/Sonic-Core's offering.  Which I have used but not in the way Jeff has suggested.
 
Here is an example: http://soundcloud.com/aaudiomystiks/spacemonkey the vocal on this track was recorded super hot infact if you go to the link you will see a picture showing you that indeed there is no head room.
 
When I did the new mix, I did nothing to the vocal, it was super present, needed no compression and or EQ, I just turned down the gain by about 6db and mixed the track.
 
This is what I am talking about when I talk about recording hot and trying to get 2nd and or 3rd level harmonic distortion.
 
I hope I have cleared up some of my earlier ramblings.
 
Neb    
 
The term Digital Warmth has to do with these three intrinsically linked concepts.  My terminology sucks, I got to chapter 25 of the rule book and threw it away.  I am not an audio engineer I am a music technologist, this is where we are getting confused.  Perhaps I am talking about ideas and less about audio engineering theory one O one!!!