An Amazing piece of software ....

http://www.proaudiodsp.com/

Go here and check out the "Dynamic Spectrum Mapper" there's a main introduction video and one based on
Buss Compression techniques.

I've never seen or heard anything quite like this, mapping the whole frequency content of a piece of music
and doing compression on all bands based on a snapshot, from a good part of that track, or a totally different
track - or vocal or whatever !
It's a really stunning and exciting tool and no I don't work for them BTW !!

I'm gob - smacked !

MM.

another piece of software that will get people thinking they can do it at home and still sound like the big boys...

boo software. Yeah Analog.

Looks very clever...

-But am I alone in thinking that it DESTROYS the 'character' of the drum sound?

Instant 'cardboard-box' drum sound, on the example used for the video.

And the one example where they show the 40dB boost-correction and say "imagine if you got a bad sounding vocal... this could correct it" is rather disingenuous I think... -they "learned" the spectrum BEFORE they screwed up the sound. -If you really did get a screwed-up vocal sound to mix, you wouldn't have the benefit of the "good" sound to "learn"...

It's no doubt very clever, and I'm going to give some serious consideration to thinking of applications where it might be of help, but in the meantime I'm going to urge everyone who watches the demo to CLOSE YOUR EYES and listen to the drum sound before and after... If you just watch the pictures you tend to go "oh, how clever!" -If you LISTEN to the sound, you might go "oh, how disappointing!"

Keith

Another piece of poop to replace actual electronics.

I'm with pucho, analog FTW.

However; it does seem pretty clever as Keef pointed out.

It may have some application to make something sound like something else if rich enough in spectral content. Or perhaps make something sound like it should have from a better template or spectral map.

I learned long ago with analog design that just because you can do something, doesn't mean it will sound good or be useful.

Another tool for better or worse.

JR

Well I thought that in the "Buss compression" video, the treatment on that Jazz track was
stunning ..... but I'm obviously completely deaf it seems ..  : :

Being able to map the "sound" of a vocal and fix all the "bad mike technique" and "off mike"
sections WITH the sound of that individual persons vocal, is just fantastic.

BAH HUMBUG !

Marty, don't get me wrong... Paul Frindle is a VERY good bloke... I have a lot of respect for him

And I just went back and listened to the Jazz example. -It works more interestingly on that piece, but more expecially when the music gets very "sparse", -to my ears at least.

But if you'd indulge me for a minute, and go back to the first video again, and just skip to just before 9 minutes, to the point where he switches it in and out for comparison... I'd be curious to know whether you prefer it in circuit, or in bypass. -I just went back, and I definitely still prefer it in bypass... -probably just a bad example, or completely unsuitable settings; but on a sales-pitch demo, I'd rather hear something which I didn't prefer the sound of in bypass!

THAT's the bit that I'm talking about.

Now I've always been a proponent of clever dynamic EQ, such as the BSS DPR-901, and this might be really good for other tasks, such as film dialog/ADR matching...although using it to do so would automatically 'force' you to use dynamic range reduction... I'd have to actually do a few sessions with it to see if such a restriction is more of a burden than a benefit... before that, I couldn't say.

Keith

I gave a brief listen to some of the examples a while ago. My thought was that for the complete destruction in the examples it sounded amazingly good. That was on my macbook though. I'm on a pee cee for audio so I didn't look at it any closer.

I just downloaded a demo of the Flux Alchemist. This may be the first plugin I actually want to use. It sounds good and has two features that are amazing. I asked around here a while ago if anyone knew of a compressor with a floating threshold. The only thing anyone could come up with was a circuit in an Orban reverb. This thing has a floating threshold feature called "hysteresis". It also has a vari mu style ratio. Increasing ratio with increasing level as well as the inverse. Decreasing ratio as level increases. That's really slick.

I'm not quite sure what a floating threshold even means... It is pretty common to use hysteresis in noise gate thresholds to prevent chatter around a given level.

JR

JohnRoberts said:

I'm not quite sure what a floating threshold even means...

Click to expand...

It monitors an RMS level and keeps a crest factor based on the RMS level. Floating threshold is my term. "Hysteresis" is theirs.

I downloaded the software as it's got a 21 day ilok license.

I'll put it through it's paces for a while and see how I feel about it.

MM

SSLtech said:

I'd be curious to know whether you prefer it in circuit, or in bypass. -I just went back, and I definitely still prefer it in bypass... -probably just a bad example, or completely unsuitable settings; but on a sales-pitch demo, I'd rather hear something which I didn't prefer the sound of in bypass!

THAT's the bit that I'm talking about.

Click to expand...

Keef - OK I'm with you on that bit, which is only 20 secs worth of the demo, it's subtle but there is a bit of
a "Lifeless flatness" when IN circuit, I'm betting that it was just a "Duff" setting in that example.

I tried it last night and you DO have to be careful which bit you "capture" ie: a bar without a cymbal is no good
as some very high detail is missing when you then "apply" it to the track.

The comparison of heavy comp with this system and heavy comp with another "Famous" plugin is interesting though
( I'm guessing it's a URS or UAD plugin ... of that quality ) that, I found to be useful and accurate IMO - FWIW

Good ears my man !!

MM.

Gold said:

JohnRoberts said:

I'm not quite sure what a floating threshold even means...

Click to expand...

It monitors an RMS level and keeps a crest factor based on the RMS level. Floating threshold is my term. "Hysteresis" is theirs.

Click to expand...

That I can understand... (maybe I should have gone to the link). It actually sounds useful in some cases to target the results rather than means.. Kind of like cruise control on an automobile, instead of pushing the gas pedal to drive at a constant speed.

I have speculated for some time how typical consoles/mixers suffer from the same means vs. results disconnect. If instead of console EQ controlling boost/cut, it targeted a final spectral balance, you could tell it where you want to end up.

I see this target based audio mixing, more useful for live performance when you know where you want to end up, than recording where you are still figuring out on the fly where you want to go.

This could be the ultimate Karaoke application, where you plug in a backing track and targets for vocal. A smart EQ and smart dynamics could get you part way there (throw in a smart pitch fixer and you're even closer.

This is ironically of more utility to entry level or amateurs, who are probably less inclined to pay the price for that much smarts. With the constant downward spiral of DSP costs, it will happen, eventually... or not.. 

JR

Karaoke??? -Kindly do not use curse words like that again!

The Orban 'parasound' spring reverb did indeed have a limiter on the input, with a switch on the front which allowed the threshold to be "fixed" or "floating", as I recall.

It was a fantastic application; controlling a spring reverb, -because it only limited transients, which are exactly the things which make springs 'splatter' and 'crash'...

We sold the parasound, and later on we sold the MicMix spring reverb also... both were very GOOD examples of where clever (but moderately simple) engineering solutions had been applied to a fundamental problem. -Of course we were prosperous enough to replace them with EMT plates, so even though they were pretty good... we were still upgrading!

I'm going to show this to a dubbing engineer friend... -I think he'd like to play with it.

Keith

Sorry, please excuse my sordid past (I have designed a Karaoke box, a DJ mixer, and probably other unmentionables).  Perhaps I should keep that stuff secret. 

The market for audio technology is far bigger than studio tricks or things like DSP chips wouldn't exist. We all need to be a little thankful for the dreaded consumer market that drives so much new technology.

JR

The automatic sidechain frequency capture got me thinking. It's not the same as this tool, but it's a pretty neat idea: You could capture the inverted response of an audio track using something like Voxengo Curve EQ (use white noise as reference), and then feed this to the sidechain input of your analog compressor (like the GSSL). This way compression would affect the frequency band of your track a lot more equally than usually. Will it be any good? Would have to try...

I would find this useful for disk cutting. Tone arms hate large crest factors with significant low frequency content. Clicky bass drums are really hard to cut. It's the crest factor more than the absolute level that matters. This can fix that.

It is impossible to do with a normal compressor with a fixed threshold.

Gold said:

I would find this useful for disk cutting. Tone arms hate large crest factors with significant low frequency content. Clicky bass drums are really hard to cut. It's the crest factor more than the absolute level that matters. This can fix that.

It is impossible to do with a normal compressor with a fixed threshold.

Click to expand...

I like that, a mastering engineer worth his salt should NOT be afraid of good software

MM

I want this plug-in, not for buss work or anything of that sort, but personally feel it will be the best software de-esser out there.

Currently I like the Sonalksis CQ-1 for that job, but will demo this one from Paul F asap.

I think it is very clever.

-T

Gold said:

I would find this useful for disk cutting. Tone arms hate large crest factors with significant low frequency content. Clicky bass drums are really hard to cut. It's the crest factor more than the absolute level that matters. This can fix that.

It is impossible to do with a normal compressor with a fixed threshold.

Click to expand...

I don't follow the physics behind tone arm mistracking and sensitivity to crest factor. Why does it care how quiet it was in the context of the peaks?. Yes I've heard they don't like large LF, stereo signals but I always thought it was a simple peak amplitude limitation. Am I missing something?

I recall seeing at least one old cutting lathe with an inductor across the cutting head to shunt the stereo LF signal to mono.

What you may be seeing in the low crest factor situations is less pure stereo L-R peaks. The L peak is the same but there is R information in the average part of peak-to-average ratio which reduces the peak stereo amplitude of L-R. 

Sorry about the thread hijack.

JR