PRR

gml opamps
« Reply #100 on: July 03, 2004, 02:57:33 AM »
One key difference between feed-forward (FF) and feed-back (FB) AGC: almost any darn detector and gain cell can be "made to work" in FB, but FF requires that the detector and gain cell track very well.

It is suggested that Gyraf's vari-Mu limiter is FF. I would like to hear from someone who knows what is inside (Jakob?), but I doubt it. The Voltage/GR curve for tubes is not any simple law, and no ordinary detector could give the right signal to work in FF mode. Especially with tube variations and replacements. The advertising text does say "no feedback", but I bet it means "no audio feedback". While it is possible to devise a FF vari-Mu (I've played with it), it is not easy, and really wants a matched tube in the GR path, which this unit does not have.

Is FB "unstable"? At modest Ratio, no, or not so you'd notice, unless you have a very unsuitable choice of time constants. It comes down to basic feedback theory: error and overshoot are easily understood and often can be tamed. It does get harder at high ratio, and impossible to get "infinite compression". But how often do we hear a difference between 10:1 and infinity:1 (neglecting artifacts)? The ear is not such a precise judge of level, and musical signals are highly dynamic and imperfectly predictable.

I don't think there is anything basically wrong with FB for ordinary music limiting. And it does seem that its theoretical weaknesses can be musically useful.

If you are going to mess with levels, aiming for less dynamics, you probably do not want to do "perfect limiting". The ear is not a great judge of level, but when every peak is -exactly- the same, on sounds that should be varied, the ear senses that this is "not normal". You get that very bland flat sound that some of the older DBXes gave. The minor "errors" of a slightly bobbly FB limiter let the peaks vary a bit, leaving some life in the sound while allowing much higher average for the same peaks. The perfect FF limiter could maybe get a couple dB higher average at the same peak, but the ruler-flat peaks lose "2dB of liveliness" and may not "sound better" or significantly louder.

For a PWM gain cell, with a linear ramp/triangle switcher drive, the detector must be linear. If we actually want 40dB range (a lot), and the top is 10V, then the rectifier must be linear down to 0.1V with error less than about 10mV for 1dB error. That is far from impossible but much more precision than you need in a FF limiter.

Because DBX works in the exponential realm, the problem is less. Their error budget is still on the order of a few mV, but the max GR signal is only about 1V, and error distributes smoothly over the whole 40dB range. All you do is trim offset or unity-gain. But exponential action gets odd when you use R-C networks for time constants; DBX uses a wacky "non linear capacitor" to give a useful (and oddly specified) time-domain action.

If everybody here had milliVoltmeters and knew what they said, we might have some fun with FF. But for solder-pokers without a wall of test gear, a FB limiter is fairly self-correcting without precision action.


SonsOfThunder

gml opamps
« Reply #101 on: July 03, 2004, 03:16:09 AM »
PRR just called us all "solder-pokers".  Is that good? I'm not sure I've ever been called that before... :grin:

DBX uses a wacky "non linear capacitor"
Can you explain, PRR?

Charlie
"The sow would rather have her ear than a purse." - PRR

PRR

gml opamps
« Reply #102 on: July 03, 2004, 03:27:38 AM »
> Can you explain, PRR?

See THAT AN103, page 6-8.

rafafredd

gml opamps
« Reply #103 on: July 03, 2004, 11:39:54 AM »
Quote
But for solder-pokers without a wall of test gear, a FB limiter is fairly self-correcting without precision action.


 :oops:

I bet he´s talking about me.

SonsOfThunder

gml opamps
« Reply #104 on: July 03, 2004, 04:37:27 PM »
Quote from: "PRR"
>
See THAT AN103, page 6-8.


Ooo...that much math makes my head hurt these days. :grin:
But I get the point. I learned something new today!  Thanks PRR!

Charlie
"The sow would rather have her ear than a purse." - PRR

Learner

gml opamps
« Reply #105 on: July 04, 2004, 10:35:55 AM »
Wow!~

Thanks PRR!!! :thumb:

Donno what we'll do without cha! :guinness:
Still into audio gears!

SSLtech

gml opamps
« Reply #106 on: July 04, 2004, 01:06:02 PM »
I don't know who Master Po is/was, so I'm lost there...

The fact is that there are national and racial stereotypes. While I never noticed it when I wrote it, Shherlock Holmes is an example of the other way that 'national profiling' works. The fact is that speaking with a British accent in the US has always been an 'unfair advantage' in this field. I do find that there are times when people ask questions and accept an opinion from me as stone-tablet fact. Occasionally I'll take a guess at an answer and people will act like the AES published a ruling! -I can't say that it's ever really worked against me in my field of work. It even works over the phone. The Internet is the one place where the playing field is a bit more level.

It's so ridiculous that someone might ask a question of NewYorkDave, Dale Manquen, Kev or even PRR, and then come and ask me. If we all gave different answers, the stereotype of the "Intelligent Brit" will weight my answer, occasionally to such an extent that people will sometimes accept a guess of mine over the experience of someone as smart as PRR, simply because they spoke with a regional accent.

However...

If I had a strong provincial Black accent, I know that people would seek second opinions rather than accept what I said without question.

That's just not right. It happens, but it's wrong: very wrong.

The same prejudice that elevates me a little represses others a lot.

While I intended no disrespect at all (and I trust that Learner accepts and appresiates that) I do appreciate that some people have grown up in parts of the world where some physical attribute of theirs -whether it's the colour of their skin, the shape of their eyes, a stammer or a limp- has led others to either mock them openly "just in fun... no offence meant" or to discriminate against them, either intentionally or unconsciously. People get fed up with that, and rightly so.

When someone takes offence at something that someone said without deliberate intent to hurt, there are usually two responses: either "I'm sorry, thanks for pointing that out, I'll make the effort to be better, Let me buy you a drink" or "What's wrong with you? -If you think I meant to inslt you you must be stupid!" I believe that only one of these responses is ever appropriate.

I've never had to live with negative prejudice. On the contrary, in many ways my life has been a charmed and a fortunate one. While I may be an opinionated old git, I sincerely hope that as I grow older I don't grow even more ignorant than I already am! I'm sure that between now and whenever they finally dig a hole to throw me in, I'll accidentally offend more people; that's probably inevitable. However, if I didn't apologise, or if I suggested that they shouldn't be as sensitive, then those would be failings on my part.

The thought of people feeling insulted accidentally mortifies me.

No... if I mean to insult someone, they'll KNOW[/u] that I mean to do it! :twisted:

Keith
"A waist is a terrible thing to mind"
Quote from: PRR
Ah, but that was 1999; we don't party like that any more.

NewYorkDave

gml opamps
« Reply #107 on: July 04, 2004, 07:00:34 PM »
Nah, when I hear a British accent, I just think, "drunken bastard."


 :roll:

I kid! I kid!  :grin:

mnats

gml opamps
« Reply #108 on: July 04, 2004, 08:07:30 PM »
Both Sherlock Holmes and Master Po are examples of stereotypes. One was a old, blind, wise monk and the other a clever, slightly myopic detective. As I recall, Master Po was played with great dignity by Keye Luke, a Chinese born American who spoke - even in that role - with a flawless American accent. So if any crime was commited it was that SSLtech gave him a Chinese accent - understandable since apparently he didn't know who he was quoting.

But it was clear from SSLtech's original post and subsequent retration that his intention was not to offend. English is spoken all over the globe with different accents. Why is a Chinese accent more offensive than a French one? Or a German one? I think the answer might lie within the beholder. There is also the question of the intention behind it. Should we laugh at Peter Sellars playing Inspector Closeau or storm out of the room enraged?

I don't see the relevance of the question but I am of Asian origin. If that somehow changes someone's opinion of my views I suggest prejudice might be tainting your perceptions.

Finally, I wish the energy expended here could be directed back to DIY. If you want to respond to my post, please do it by PM or take it to The Brewery.

Edit: Seems that learner deleted his posts which contained charges of racism directed toward SSLtech after he adopted an Asian accent in one of his posts. Ironically, this is what learner posted earlier:
Quote from: "Learner"
Vhere ist de iron??? Vhy ist dist unit hav no iron but still sounds good?

I guess a German accent is politically correct but an Asian one is not :roll:

gml opamps
« Reply #109 on: July 04, 2004, 08:19:45 PM »
Quote from: "NewYorkDave"
Nah, when I hear a British accent, I just think, "drunken bastard."
:


That's odd Dave..(hic)..we always speak very highly of New Yorkers..!!

 :green:
http://www.lazpro.com
I reserve the right to not sell to idiots


gml opamps
« Reply #110 on: July 04, 2004, 08:21:27 PM »
Quote from: "NewYorkDave"
Nah, when I hear a British accent, I just think, "drunken bastard."
schnip..   .hic..    .  .   .   ..
                                                   



 :shock:   Constanoon afterble :?   :oops:  No...I'm not under the affluence of inkahol   :?    :guinness:  hic
If you want to sound crazy please tell me about THE DEEP STATE.

gml opamps
« Reply #111 on: July 04, 2004, 08:39:58 PM »
:green:  :green:  :guinness:  :guinness: burp
http://www.lazpro.com
I reserve the right to not sell to idiots

SSLtech

gml opamps
« Reply #112 on: July 04, 2004, 10:30:53 PM »
... I haven't had a c*nt all day, drinkstable...

 :guinness:  :green:  :guinness:  :green:  :guinness:  :green:  :guinness:  :green: :guinness:  :green:  :guinness:  :green:  :guinness:  :green:  :guinness:  :green: :guinness:  :green:  :guinness:  :green:  :guinness:  :green:  :guinness:  :green:

By the way... this is the only day of the year when it's terrible bring British...

"Oh yeah man.... 4yh of July... that's when we celebrate kickin your asses, man...."  :roll:

...it gets old.

-I think I'll sue! :green:

It's all good! :thumb:

Keef
"A waist is a terrible thing to mind"
Quote from: PRR
Ah, but that was 1999; we don't party like that any more.

NewYorkDave

gml opamps
« Reply #113 on: July 05, 2004, 12:59:12 AM »
Quote
By the way... this is the only day of the year when it's terrible bring British...

"Oh yeah man.... 4yh of July... that's when we celebrate kickin your asses, man...."  :roll:


I don't know... That's just one military/historical event. I honestly don't think most Americans are thinking of that as they celebrate the fourth. I think those of us who can be bothered to consider it beyond an excuse to get loaded and set off fireworks, are inclined to see it more as a celebration of the ideals on which the nation was founded, and toward which we continue to strive, in our better moments. To me, it's not really a commemoration of a particular historical event. The exception to this was the bicentennial in 1976, when you could NOT f**king get away from people marching around in three-cornered hats  :wink:

gyraf

gml opamps
« Reply #114 on: July 05, 2004, 07:55:03 AM »
Quote from: "PRR"
(..) It is suggested that Gyraf's vari-Mu limiter is FF. I would like to hear from someone who knows what is inside (Jakob?), but I doubt it. The Voltage/GR curve for tubes is not any simple law, and no ordinary detector could give the right signal to work in FF mode. Especially with tube variations and replacements. The advertising text does say "no feedback", but I bet it means "no audio feedback". While it is possible to devise a FF vari-Mu (I've played with it), it is not easy, and really wants a matched tube in the GR path, which this unit does not have. (..)


Yup, the G10 is a straight feed-forward sidechain design alright, as Jaakko mentions.

When initially testing various topologies with our panel of golden ear "testflyers" there was great consensus that the FF design acted much more musical and gave much more intuitive control than the (technically more precise) feed-back topologies.

It turned out that the technically "wrong" unlinearity of ratio with compression amount - stemming from the much less than linear gain-versus-controlvoltage - was actually percieved as a strength rather than a flaw in real-life-use. The only serious drawback is that at very extreme settings you possibly run into ratios higher than inf:1, resulting in bizarre behaviour. But in use, this was never considered a problem by any of our test-flyers.

It should be noted that in this topology it is not possible to build a simple and predictable set-and-forget limiter a'la the classic radio-transmitter protectors - but this is far from the scope of modern-day dynamic controllers. Somehow it's become a sculpturing tool rather than a safeguard..

And yes, you're right - this way of using the sidechain calls for very minute selection of the remote-cutoff tubes used. This is one of the really work-heavy sides of manufacturing out G10's. Currently we have around a 1:10 rate of useable tubes.

Jakob E.
..note to self: don't let Harman run your company..

CJ

gml opamps
« Reply #115 on: July 05, 2004, 07:57:21 PM »
Wasn't it Father Poe?
That was my friends Uncle. He has an autographed picture of him.
They used egg whites to blank out his eyes!
Very "Americanized" chap ya know.
Just some more useless info from cj!
 :guinness:
If I can't fix it, I can fix it so nobody else can!
Frank's Tube Page: www.mif.pg.gda.pl/homepages/frank/vs.html
Guitar Amps: http://bmamps.com/Tech_sch.html

gm

gml opamps
« Reply #116 on: July 06, 2004, 06:18:40 PM »
Quote from: "mnats"
DBX did make feedback compressors. They were used in the encoder side of their noise reduction units. The reason is interesting: when paired with a feedforward expander and placed back to back the two sides detect and respond to the same signal. Logically, it makes sense but the problem is that they were never hooked up back to back but were passed through the analog recording medium. Interesting to think of the two schemes as essentially cancelling each other out though.


Yes, you're absolutely correct.  But they were not in the strictest sense sold as a "compressor".

George

mnats

gml opamps
« Reply #117 on: July 06, 2004, 08:19:37 PM »
Quote from: "gm"
Yes, you're absolutely correct. But they were not in the strictest sense sold as a "compressor".

I've read, not only here but in other places, the pros and cons of feedback and feedforward and the unpredicability of response of either in different contexts (usually, but not always, depending on what was being sold). So I found it fascinating that this wild, untamed force could be used in something which by its nature demanded control and precision.

My DBX noise reduction unit is long since gone but I'm guessing that if hooked up back to back encode to decode it would sound a lot more (I hate this word but it fits here) transparent than if hooked up in the usual way with an analog recorder in the middle. Would that be because the two complementary arrangements cancel each other out somehow or because at these low ratios there aren't many artifacts to begin with?

Igor

gml opamps
« Reply #118 on: October 09, 2006, 09:17:18 PM »
Hi to all, I just was reading this thread, and found this:
Quote

What about an adjustable FF/FBCK control, ganged with ratio: Low ratios feed forward, and higher ratios gradually more feedback, ending in feedback-only at limiting. Two rectifier circuits would be needed before the mixing, but timing, threshold, and soforth could still be common.

Just to share other way to get some benefits of FF timing.
I was experimenting with changing the time constants by signal/compression.
Attack timing circuit feed from forward (input) detector,
and release timing circuit feed from feedback (output) detector,
but the whole compressor is feed-back.
Guy who very comfortable with 8200 found this compressor
more flexible, and love them both for sound:
8200 is clean surgery, my vary-mu he call "jakuzi for electrons" :-)
It working nice some years btw...
http://www.geocities.com/igor_jazz/mast_comp
I like to see ho attack/release leds changes with the signal,
more compression-faster attack and so on.
Next thing will be tranzienT Desighner which controlls attack/release of vary-mu....too much ideas, never have time on this.
http://www.ij-audio.com
(new site hopefully will be up in July)


 

Related Topics

  Subject / Started by Replies Last post
5 Replies
2410 Views
Last post February 16, 2007, 12:22:04 PM
by chrissugar
4 Replies
6108 Views
Last post December 31, 2008, 08:08:38 PM
by joe-electro
0 Replies
1310 Views
Last post June 16, 2009, 06:54:07 PM
by mcfate
7 Replies
2019 Views
Last post December 28, 2012, 12:44:07 PM
by r2d2