The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Gear for sale  Search Gearslutz Go Advanced
Q for Paul Frindle Dynamics Plugins
Old 18th May 2006
  #61
Lives for gear
 

Quote:
Originally Posted by innesireinar
Ok, Ok, Ok, Ok...understood.
The only way for setting the digital level in my system is the master fader of PT.
Is this the reason why the master fader of PT has insert slots post fader?
Because from what I've learned once the signal has been dithered and "converted" to 16 bit it's not possible to change the digital level without backing into a high word lengh signal, am I right Paul?
Ok - it's possible to change the level providing the new signal is redithered after the level change. IT does not matter that if it was dithered before - the dithering is destroyed by the level reduction.

So in the case of an output fader, the multiplier in the fader has very high precision and wordlength. A very important rule (!) every time we change anything a new signal is created :-)

The output fader produces a large output word which will need to be redithered to 16bits and then truncated to 16bits, if you are aiming at a CD master.

This is why the dithering and truncation should be done within the W/S application itself as the very final process after the output fader.

The fact that in PT the master inserts are after the fader at least allows you to do the dithering legally using a plug-in (i.e. the dither is not affected by master gain setting). But it's a complete nuisance as well because reducing the master affects your buss processing. It's this last practical problem that makes the master fader a bad choice for monitoring - not dither or resolution loss. And this is also the reason why most of the time you are listening to the converters driven absolutely flat out - and no amount of analogue monitoring control after the event will give the converters a break :-(

This is but one way of many that an incomplete (or naively presented) application can encourage people to hear bad sound unecessarily - even though no 'technical' problem exists (this is a big and deep subject - perhaps for another time).

Normally of course, a mixer app should have separate master and monitor outputs with both being properly dithered for their intended destinations. This should be part of the DAW application itself.. Quite simply, the means to monitor your sound properly is missing from PT..

So since you don't have that luxury, you're better off feeding the output digitally to another DAC 'box' that has an internal digital gain control with it's own legal dithering regime :-)

I should add that the OXF-R3 console had an all digital monitor control..

I hope this makes sense :-)
Old 19th May 2006
  #62
Gear Nut
 
innesireinar's Avatar
 

In fact Digidesign would have made the first 4 slot pre and only the last post in the master fader (or better, let the user to switch all five slots pre/post). In this way an eventually compressor or a limiter with the dither disabled in the first four slots would not be affected by the fader position allowing to move the fader for monitoring task.
Regarding the Sony Limiter a good idea could be to make it by modular sections.
Having the switchable pre/pos slots in PT and having the limiter with separate modules I could insert the limiter in a pre slot without affect this processsing by moving the master fader and by putting the last dithering section in the fifth post slot.
BTW I can do it now by a trip. Bussing all channels to a stereo aux channel, doing in this aux all processing, limiting included (by disabling the dithering) given that all aux slots are pre fader then, routing this aux to the master where could be there another instance of the limiter with all its setting "clear" but with the dithering enabled.
By having the limiter in modular sections could be useful for not disperding unuseful DSP power given that in the last instance only the last dithering step is working.
Your point?
Old 20th May 2006
  #63
Lives for gear
 

Quote:
Originally Posted by innesireinar
In fact Digidesign would have made the first 4 slot pre and only the last post in the master fader (or better, let the user to switch all five slots pre/post). In this way an eventually compressor or a limiter with the dither disabled in the first four slots would not be affected by the fader position allowing to move the fader for monitoring task.
Regarding the Sony Limiter a good idea could be to make it by modular sections.
Having the switchable pre/pos slots in PT and having the limiter with separate modules I could insert the limiter in a pre slot without affect this processsing by moving the master fader and by putting the last dithering section in the fifth post slot.
BTW I can do it now by a trip. Bussing all channels to a stereo aux channel, doing in this aux all processing, limiting included (by disabling the dithering) given that all aux slots are pre fader then, routing this aux to the master where could be there another instance of the limiter with all its setting "clear" but with the dithering enabled.
By having the limiter in modular sections could be useful for not disperding unuseful DSP power given that in the last instance only the last dithering step is working.
Your point?
There are a few points here:

1. Although the bussing procedure you mention gives you a way of using the master fader as a monitoring level control (and make the aux effectively your master), it still doesn't really give you a complete monitoring path that a proper monitoring section would normally give. I.e. it works, but is a bit incomplete and inconvenient.

2. You are right that using 2 limiters (one doing the comp and the other providing the dither) would achieve the right result (please note that the first limiter will have 24bit dither operational as well). But setting the limiter to dither only is a bit difficult as it will start compressing if monitoring levels get high enough.. Currently there is no single control to set the limiter with the compression switched completely out. It would be better/safer is to use a separate dither plug-in instead.

BTW to set the limiter so that there is no compression:

- Set input gain below 0dB (only signal above 0dB on input meter gets compressed)
- Set soft knee control to minimum (to prevent limiting for signals below 0dBfs)
- Set safe mode off
- Set enhance to minimum
- Set auto comp (in the meter section) to off.

3. Making the limiter itself in parts (to save unecessary processing on TDM and giving it the advantage of dynamic processing loading as in RTAS and LE) is a bit problematic as the parts would not work together in the same way and quality would suffer - for instance the internal headroom could not pass between sections etc.. Again, if you want just dither it's better to use a good dither plug-in.

There is perhaps a case for the reconstruction meter (maybe without the auto correct processing) and dither sections to a be offered as separate plug-ins? IMO this would be useful to people? But now I cannot get this made myself.

4. I agree that having both pre and post inserts on the master is a must IMVHO. I find the post only inserts on the PT master a real limitation in use :-(
Old 16th July 2006
  #64
Lives for gear
 

Paul, please let me ask you a question. Someone on Geraslutz adviced me to ask you.

Is there any outboard unit doing actually the same thing (or close) to the Sony Oxford Inflator plugin ?
Old 17th July 2006
  #65
Lives for gear
 
minister's Avatar
i just want to say, THANK YOU !! to Mr. Paul Frindle. Not only for his fantastic plugs, but for so freely sharing his knowledge, insight and experience in well written, (reasonably) easy to understand posts. he uses well-chosen, hard-working words to trenchantly describe complex topics. frankly, since i read his post where he first 'set the cat amongst the pigeons' about a year and a half ago, he has really OPENED MY EYES!

i have a deeper understanding of what is 'under the hood' which actually translates to my mixes being better now, my assistants mixes are better now... THANKS PAUL!! yer ACES!

you have to read what he wrote a few times and really work hard at understanding what he is saying. at least i try to ... in my own small way.
Old 17th July 2006
  #66
Lives for gear
 
jdjustice's Avatar
Quote:
Originally Posted by minister
i just want to say, THANK YOU !! to Mr. Paul Frindle. Not only for his fantastic plugs, but for so freely sharing his knowledge, insight and experience in well written, (reasonably) easy to understand posts. he uses well-chosen, hard-working words to trenchantly describe complex topics. frankly, since i read his post where he first 'set the cat amongst the pigeons' about a year and a half ago, he has really OPENED MY EYES!
I heartily second that thanks. This thread is one of the most informative I have read in quite a while. It means a lot to us slutz when someone as talented and intelligent as Mr. Frindle takes time out of his busy day to help neophyte engineers gain some truly invaluable knowledge.

Cheers and thanks again!!

J.D.
Old 17th July 2006
  #67
Lives for gear
 

Quote:
Originally Posted by yareck
Paul, please let me ask you a question. Someone on Geraslutz adviced me to ask you.

Is there any outboard unit doing actually the same thing (or close) to the Sony Oxford Inflator plugin ?
Not as far as I know :-(

The Inflator was not based on any H/W I have seen or made before. It was based mostly on experience I had decades ago regarding how tube amps produce much more sound for your watts than solid state amps, because (if the design allows) there is a region of quite heavy but tolerable distortion at high levels. Much of the design skill in making tube amps was to get this right by various means.
I was just aiming for that sound I remembered in my head from the late 60's and after much experimenting ended up with the Inflator. So basically it's a distortion generator.

The most likely outboard gear that would have something like a similar effect would be those using push-pull tube output stages that didn't try too hard to be very linear (i.e. not using too much -ve feedback in the design etc). Although the Inflator isn't modelled on (or resembling) a tube as such, you may find a loosely similar kind of effect.

I am thinking in particular that Manley make some units that actually have a -ve feedback control - aimed at giving you some control over this kind of tube effect?
Old 17th July 2006
  #68
Lives for gear
 

Quote:
Originally Posted by jdjustice
I heartily second that thanks. This thread is one of the most informative I have read in quite a while. It means a lot to us slutz when someone as talented and intelligent as Mr. Frindle takes time out of his busy day to help neophyte engineers gain some truly invaluable knowledge.

Cheers and thanks again!!

J.D.
Guy's it very kind of you to say this :-) I actually really gain personally from posting (just like everyone else on the forums), it forces me to think about and review what (I thought) I already knew :-)

Many times I have learnt much from what people are saying and having to think about issues and verbalise them in ways that make sense to me - and hopefully others as well. I have come to exciting new and interesting realisations even from the most seemingly elementary topics of discussion. We are all trying to understand and make better audio..

And right now I am 'between employment' so I have the time :-)
Old 17th July 2006
  #69
Gear Nut
 

Hmmm... that last sentence made me think...

Does that mean you are no longer working for Sony Oxford, pray tell?

Cheerio
Roger
Old 17th July 2006
  #70
Lives for gear
 

Quote:
Originally Posted by theRog
Hmmm... that last sentence made me think...

Does that mean you are no longer working for Sony Oxford, pray tell?

Cheerio
Roger
That's right, I officially left in April. As a corporate 'restructuring scheme' Sony E.U. started offering voluntary early retirement for staff over 50 years old. Since I was increasingly experiencing worrying health issues (mostly due to stress) and my wife has had a long term debilitating illness for several years, my performance was becoming increasingly below that which might be expected of me. I therefore considered that taking the early retirement offer was the fairest thing to do for everyone concerned.

I am now concentrating on regaining my health and composure, helping to bring up our 2 children and considering what I might do to continue to contribute to the art I love in future :-)
Old 17th July 2006
  #71
Lives for gear
 
minister's Avatar
well, thanks for publicly sharing that Paul and i wish you all the best in your future as you try to get on physical terra firma. ..many of us know the ravages of stress....


...though, selfishly...i am sad that you are not going to be developing more great OXFORD plugs. but, alas, there is more to life!

Oh, but now that you have time on your hands, i was wondering....what's the deal with 'audio'? :-)

cheers paul!
Old 18th July 2006
  #72
Lives for gear
 

Quote:
Originally Posted by minister
Oh, but now that you have time on your hands, i was wondering....what's the deal with 'audio'? :-)

cheers paul!
There will be more 'audio' I am sure :-)
Old 19th July 2006
  #73
Gear Addict
 
bonne's Avatar
 

Quote:
Originally Posted by Paul Frindle
One such thread was the so called '96K' thread on PSW which even inspired one person to write a book! Let me try as briefly as possible:
I've been trying to find this thread, Paul. Do you have link for it?

Your input on the forum is much appreciated.

Thanks!

Jørn Bonne
Old 19th July 2006
  #74
Lives for gear
 

Quote:
Originally Posted by bonne
I've been trying to find this thread, Paul. Do you have link for it?

Your input on the forum is much appreciated.

Thanks!

Jørn Bonne
Yes you are right - just had a look and it's gone. It was originally on the George Massenburg forum that has now turned into the Reason In Audio forum following George's departure, but the archives seem to have gone? I did a search using my name in the Forum home and there are no entries before Tue, 27 April 2004 03:34. This was of course after this particular thread ended.

I don't know if there's another way to extract archives before then?

I'll try again later..
Old 19th July 2006
  #75
Lives for gear
 

Quote:
Originally Posted by Paul Frindle
Yes you are right - just had a look and it's gone. It was originally on the George Massenburg forum that has now turned into the Reason In Audio forum following George's departure, but the archives seem to have gone? I did a search using my name in the Forum home and there are no entries before Tue, 27 April 2004 03:34. This was of course after this particular thread ended.

I don't know if there's another way to extract archives before then?

I'll try again later..
BTW there's another longish thread that they have retained as a 'sticky' where some useful conversation went on regarding mixing ITB versus OTB. But whilst this is interesting stuff, it doesn't go as far into the nuts a bolts as the original 96K thread kicked off by Nika Aldridge.

http://recforums.prosoundweb.com/index.php/t/4918/2578/
Old 19th July 2006
  #76
Lives for gear
Quote:
Originally Posted by minister
i just want to say, THANK YOU !! to Mr. Paul Frindle.
+1.
Old 19th July 2006
  #77
Lives for gear
 
minister's Avatar
Quote:
Originally Posted by Paul Frindle
Yes you are right - just had a look and it's gone. It was originally on the George Massenburg forum that has now turned into the Reason In Audio forum following George's departure, but the archives seem to have gone? I did a search using my name in the Forum home and there are no entries before Tue, 27 April 2004 03:34. This was of course after this particular thread ended.

I don't know if there's another way to extract archives before then?

I'll try again later..
do you mean this thread?

http://recforums.prosoundweb.com/ind.../288/0//11603/

there's a ton of posts by nika in there...
Old 20th July 2006
  #78
Lives for gear
 

Quote:
Originally Posted by minister
do you mean this thread?

http://recforums.prosoundweb.com/ind.../288/0//11603/

there's a ton of posts by nika in there...
No, it's an earlier thread where Nika himself was extracting the information rather than re-applying it.
By raising the controversial topic of sample rates he very effectively did everyone a service by charting his own education to this stuff publically by asking the salient questions live in the forum - without embarrassment. He then methodically and diligently tapped up everyone he could find 'in the know' and wrote a book based on that info - along with his own opinions and conclusions from his personal experiences of having made the journey.
Because Nika and I shared the same passion for jargon busting and bringing the truth out of the mire of industry spin and myth, it provided a great platform for discussion and learning. Questions from Nika and all the other people provoked a need to provide readable and understandable answers. I spent ages and ages on it :-)

By the time the 192KHz thread you have suggested was running, he had no further need of my input. He had basically taken up the chalice and was running with it - which was great. So I very much took the back seat on the basic underlying theory and was concentrating more on higher level application issues in other threads :-)

The 'sticky' thread on Reason In Audio (the link I posted earlier) is an example of that - i.e. what actually occurs when this stuff is applied to real situations with real kit people are actually using. The ITB/OTB issue unleashes a technical can of worms which lended itself well to issues not covered in previous threads. I included some interesting and perplexing examples people could try for themselves to prove various points etc..

IMO, both these threads illustrate very graphically just how amazingly useful forums can actually be - if threads are directed sensibly and can generally stay on course :-) The question and answer exchange is an extremely powerful learning tool, because the responsibility of honesty, correctness and transparency to people at all levels forces you to revisit stuff you take for granted as a designer - question it thoroughly and present it in a form anyone can understand.

I revelled in this. Since I am completely lacking in formal education, self taught and thus without a single qualification - and not of genius intelligence either - if I can grasp it, then so could almost anyone else :-) In reality - in the cold light of day - with the spin, jargon, marketing sophistry and 'religion and magic' taken away, nothing in it is so complex it can't be explained in real language and understood by anyone with the inclination to think about it logically and can heavily resist being intimidated by others into thinking they are somehow 'not welcome within the club of honoured intellectuals' :-)
Old 21st July 2006
  #79
Gear Addict
 
bonne's Avatar
 

Quote:
Originally Posted by Paul Frindle
I revelled in this. Since I am completely lacking in formal education, self taught and thus without a single qualification - and not of genius intelligence either - if I can grasp it, then so could almost anyone else :-) In reality - in the cold light of day - with the spin, jargon, marketing sophistry and 'religion and magic' taken away, nothing in it is so complex it can't be explained in real language and understood by anyone with the inclination to think about it logically and can heavily resist being intimidated by others into thinking they are somehow 'not welcome within the club of honoured intellectuals' :-)
This is just what a lot of us need, Paul. I'll read up on the "sticky" thread and may have some more "Q's for Paul Frindle" then.

Hope someone is able to retrieve the "96K" thread. Is Brad listening?
Old 25th July 2006
  #80
Gear Addict
 
bonne's Avatar
 

Quote:
Originally Posted by Paul Frindle
- If I your are mastering it yourself (i.e. for client approval and stuff), use a meter (or comp/limiter) that lets you see the actual reconstructed signal.

This to avoid making a file that could inexplicably start sounding rubbish on your client's (or other people's) systems!
Paul, another Q for you, if you have the time.

The Sony Oxford Limiter (not available in VST yet) has this ability and so does the TC 6000 Brickwall Limiter (not within reach for many of us).

For the many people using the software Waves L2 or similar what do you recommend to guard against these illegal peaks?

Thanks

Jørn
Old 25th July 2006
  #81
Lives for gear
 

Quote:
Originally Posted by bonne
Paul, another Q for you, if you have the time.

The Sony Oxford Limiter (not available in VST yet) has this ability and so does the TC 6000 Brickwall Limiter (not within reach for many of us).

For the many people using the software Waves L2 or similar what do you recommend to guard against these illegal peaks?

Thanks

Jørn
This is difficult because what you really need is a signal reconstruction meter that runs on a VST host platform - you then need to insert this at the end of your mix and processing.. If this was a good quality device it would show you the 'intersample' errors.
But even then you would need to correct them by reducing the level manually when needed. This is tedious and because it's not a dynamic process it's not synonymous with reaching the max legal modulation levels (people want) at all times :-(

Anyway I have done a quick web search for any such metering device on VST and cannot find one. (Maybe I should make one).

The other (rather tortuous) way is to actually decode the PCM with a real DAC and put a calibrated analogue fast peak meter after it! But you must be careful because this will only work IF the DAC itself does not clip with the intersample errors. If the DAC does clip, then you would need to reduce the meter calibration to something below max (say red light for -0.5dB wrt flat out) and aim to keep the red light out. This will potentially lose you 0.5dB of max possible modulation (because you would be aiming at -0.5dB average peak), but it would at least avoid the problem.

Ok - now if you've followed this so far, it is clear that if the DAC above does not clip with intersample peaks and DOES indeed give out legal analogue levels above normal flat out when intersample errors occur, you could elect to control these dynamically by inserting a well calibrated and very fast attack analogue limiter after the DAC to reduce the overmodulation and then reconvert it back to digital again afterwards!!

Now this last idea seems a bit mad as you are applying 2 lots of converter generation loss to the actual programme itself (i.e. a DAC and an ADC), but this is indeed what many people are achieving by mixing out of the box. If you read the sticky thread on PSW 'Reason in Audio' you will see that I have discussed this effect as part of the reasons why (in the absence of a reconstruction limiter/meter combination) mixing OTB may apparently give better sounding results to overmodulated programme, despite the obvious converter generation loss.

I hope this makes sense and is helpful :-)
Old 25th July 2006
  #82
Gear Addict
 
bonne's Avatar
 

Quote:
Originally Posted by Paul Frindle
This is difficult because what you really need is a signal reconstruction meter that runs on a VST host platform - you then need to insert this at the end of your mix and processing.. If this was a good quality device it would show you the 'intersample' errors.
But even then you would need to correct them by reducing the level manually when needed. This is tedious and because it's not a dynamic process it's not synonymous with reaching the max legal modulation levels (people want) at all times :-(
What you wrote makes sense, Paul. A roundtrip to analog could solve this problem.

For us dedicated ITB people having to go via analog to get this solved it's not good news.

Regarding your suggestion to manually reduce level, is it likely that the most troublesome illegal overshoots will always be in close connection to the highest level sample peaks as seen on the GUI of your wave editor? In which case finding those, zooming in on them and attenuating these samples would bring down the overshoots as well, yes?

Or could troublesome illegal peaks also be created where the surrounding samples were of a more moderate level, after your typical mastering processing with eq filters etc ?
Old 25th July 2006
  #83
Lives for gear
 
Nordenstam's Avatar
 

Verified Member
Hi Jørn!

I'm not Frindle, but I hope this answer may help a little bit anyway.

The trick of the 'true' digital metering is the oversampling. A digital oscilloscope needs at least 10x the signal frequency to make an accurate representation. I'd guess the same rule goes for an oversampled peak meter. You could simulate this to a certain extent by lowering the level by a fixed amount and oversampling this to the highest rate you can find in your editor. That's probably 192kHz, but 384kHz would be better for this purpose if any program is supporting it. Lowpass filter this with a very steep filter at 20kHz and you'll probably find that the peak level have been raised by the previously hidden intersample peaks. It would at least give you a hint of the extent of the problem.

On the other hand, using the RME metering shows and abundance of intersample peaks on most any CD from the stores. If everyone else does it.. (but I usually don't!)


>Regarding your suggestion of manually reducing level, is it likely that the illegal overshoots will always be in close connection to the highest level sample peaks as seen on the GUI of your wave editor?

IIRC, Cool Edit Pro shows an oversampled drawing of the waveform. You could perhaps try using this to spot them. Or just get an RME card for the metering. =)

It's actually not that hard to make a coarse guesstimate where the intersample peaks will be. Two adjacent samples with near similar values(a flat peak) will definitely not be converted to a square wave when they leave the DAC!


Andreas Nordenstam

PS: There's an article on my Norwegian mastering webpage titled 'digitale signalnivåer og dithering' that adresses this issue with several pictures. May be of some interest. =)
Old 25th July 2006
  #84
Mastering
 

Quote:
Originally Posted by Paul Frindle

The other (rather tortuous) way is to actually decode the PCM with a real DAC and put a calibrated analogue fast peak meter after it! But you must be careful because this will only work IF the DAC itself does not clip with the intersample errors.
You could drop the level into the DAC and recalibrate vis a vis the analog output. But in the end... so much work. It would be cheaper and easier to get an RME card and use their Digicheck facility, which interpolates intersample overs.
Old 25th July 2006
  #85
Lives for gear
 

Quote:
Originally Posted by bonne
What you wrote makes sense, Paul. A roundtrip to analog could solve this problem.

For dedicated ITB people, like myself and my partner, it's not great news.

Regarding your suggestion to manually reduce level, is it likely that the most troublesome illegal overshoots will always be in close connection to the highest level sample peaks as seen on the GUI of your wave editor? In which case finding those, zooming in on them and attenuating these samples would bring down the overshoots as well, yes?

Or could troublesome illegal peaks also be created where the surrounding samples were of a more moderate level, after your typical mastering processing with eq filters etc ?
Ok - yes you are broadly right that the overs on reconstruction are more likely to happen during strings of large value samples (if only because for instance they cannot happen if no sample ever gets bigger than say 10%) - however it is considerably more complex than that unfortunately..

It is not a given by any means that the presence of large samples alone indicate an over - so if you hit them by lowering their value you could be unnecessarily reducing the levels - AND - by doing so you may actually create an error that was not there in the first place!. Generally messing with undecoded sample values by visual analysis is not effective - because your sample display and eyes cannot reconstruct the band limited signal that will exist after the decoder filter.. Remember that the filter required to reconstruct PCM is the equivalent of a suite of successively accumulated calculations taking into account upwards of 200 sample positions on your wave editor screen.

The ONLY reliable way to know if a succession of samples (and their history) will cause overs on reconstruction is to actually do the math equivalent of actually reconstructing the signal. To do this you need the filter that would be required by the DAC (which is NOT just simple interpolation - as some people assume).

In fact you do not need to actually oversample to get the required math result of the reconstruction filter followed by a peak value detector - it can be done at base rate sampling (as it is done in the Oxford Limiter) - it's just that however you do it, it's a bit costly on processing.. This is why you won't find such metering included all over your W/S application as standard any time soon..
Old 25th July 2006
  #86
Paul... one more vote here for a modular version of the Oxford Limiter... a dither section that you can use by itself as a plugin with a reconstruction over meter would be great (no limiter section). I'd put it on all of my master faders on the way out to the summing box.
Old 25th July 2006
  #87
Gear Addict
 
bonne's Avatar
 

Quote:
Originally Posted by Lupo
IIRC, Cool Edit Pro shows an oversampled drawing of the waveform. You could perhaps try using this to spot them. Or just get an RME card for the metering. =)
Hi Andreas,

Thanks for your tips. I'll look into both of these options. Thanks also for the link to your Norwegian mastering site. Interesting! Will take a closer look tomorrow when I have some more free time.

Cheers

Jørn
Old 25th July 2006
  #88
Gear Addict
 
bonne's Avatar
 

Quote:
Originally Posted by Paul Frindle
It is not a given by any means that the presence of large samples alone indicate an over - so if you hit them by lowering their value you could be unnecessarily reducing the levels - AND - by doing so you may actually create an error that was not there in the first place!. Generally messing with undecoded sample values by visual analysis is not effective - because your sample display and eyes cannot reconstruct the band limited signal that will exist after the decoder filter.. Remember that the filter required to reconstruct PCM is the equivalent of a suite of successively accumulated calculations taking into account upwards of 200 sample positions on your wave editor screen.

The ONLY reliable way to know if a succession of samples (and their history) will cause overs on reconstruction is to actually do the math equivalent of actually reconstructing the signal. To do this you need the filter that would be required by the DAC (which is NOT just simple interpolation - as some people assume).

In fact you do not need to actually oversample to get the required math result of the reconstruction filter followed by a peak value detector - it can be done at base rate sampling (as it is done in the Oxford Limiter) - it's just that however you do it, it's a bit costly on processing.. This is why you won't find such metering included all over your W/S application as standard any time soon..
It's a pity the Oxford Limiter is not available on the VST platform. Just what the doctor ordered for ITB mastering.

In the meantime I'll have to find another way of doing this: detecting illegal peaks and getting rid of them.

You advise against the idea of using a wave editor to find these suspects and simply dropping the level on them. I guess using Cool Edit Pro's wave editor, that according to Lupo shows an oversampled drawing of the waveform, would not change that position?

CEP might enable you to get at very short, instantanious peaks (loud samlpes AND illegal intersample peaks alike). You might be able to reduce the damage to below one millisecond of duration in many cases. Not an ideal solution, but distortions under one millisecond should not be audible.

What you said about reconstructing PCM tells me that this probably still wouln't be inaudible though. Am I right?
Old 26th July 2006
  #89
Lives for gear
 

Quote:
Originally Posted by yeloocproducer
Paul... one more vote here for a modular version of the Oxford Limiter... a dither section that you can use by itself as a plugin with a reconstruction over meter would be great (no limiter section). I'd put it on all of my master faders on the way out to the summing box.
I understand, but you must put this suggestion to them as I no longer work for Sony. It would be relatively easy for them to make from a subset of the existing app if they considered it to be a viable comercial product in it's own right. But personally I would doubt that it really is?

I'm not sure how many master faders you are using to feed your summing box, but please be aware that this non-limiting plug-in would still be expensive on processing. The recon meter, compensation section and dither sections are actually more than half of the total processing cost of the whole limiter plug-in. You might as well have the whole limiter and set it to a position where no limiting occurs. At least then you have a top class limiter as well for relatively little more processing cost - which might come in very useful as well?

BTW (apart from DC offset removal), the Limiter is transparent if the input gain is set anywhere below 0dBr, the safe mode is OFF, the soft curve is set to minimum and the auto reconstruction compensation is OFF :-) But bear in mind that DC offset removal may fractionally increase the peak sample value of the signal for very low frequency events.
Old 26th July 2006
  #90
Lives for gear
 

Quote:
Originally Posted by bonne
It's a pity the Oxford Limiter is not available on the VST platform. Just what the doctor ordered for ITB mastering.

In the meantime I'll have to find another way of doing this: detecting illegal peaks and getting rid of them.

You advise against the idea of using a wave editor to find these suspects and simply dropping the level on them. I guess using Cool Edit Pro's wave editor, that according to Lupo shows an oversampled drawing of the waveform, would not change that position?

CEP might enable you to get at very short, instatanious peaks (loud samlpes AND illegal intersample peaks alike). You might be able to reduce the damage to below one millisecond of duration in many cases. Not an ideal solution, but distortions under one millisecond should not be audible.

What you said about reconstructing PCM tells me that this probably still wouln't be inaudible though. Am I right?
If cooledit truly displays an upsampled wavefore that is derived after reconstruction it would possibly be more useful in the way you describe. But I can't confirm that it actually does though? But one thing to remeber is that the errors appear only at the DAC reconstruction filter - so an upsampling process intended for other reasons may not have the correct response characteristics to accurately display the problems.

Short single sample peaks are unlikely to cause overs - it's the action of what happens over many samples that makes a recon over.. The relationship is mostly far too complex to be seen directly. That's why the expensive math processing is needed.

IME the greatest cause of recon errors are actually quiet parts of the music that have been boosted in the mix or by comps and limiters. E.g. Sparse portions of Shania Twain produce much larger and more frequent overs (if boosted) than Green Day at full pelt!! The corellation between the overs and what one would expect from the material itself is not at all intuitive at first sight.

Unfortunately a 1mS burst of something nasty is actually very audible indeed. The Oxford recon compensation overcomes this by dealing with the errors before they arrive at the output (i.e. using a fairly long look ahead). This is why it has quite a long latency..
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
Elsteve9 / So much gear, so little time
5
djui5 / Gear free zone - shoot the breeze
3
slantbar / Q+A with Mike Shipley
2
AudioGaff / The Moan Zone
1

Forum Jump
Forum Jump