ProSoundWeb Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 ... 7 8 [9] 10 11 ... 22   Go Down

Author Topic: Possible X32 firmware rumored. Your suggestions?  (Read 12016 times)

Scott Holtzman

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 5613
  • Ghost AV - Avon Lake, OH
    • Ghost Audio Visual Systems, LLC
Re: Possible X32 firmware rumored. Your suggestions?
« Reply #80 on: September 14, 2019, 01:01:22 am »

Hey Scott, how so? It's a stateless (point) process and, so far as I know, only gets interesting if the numerical representation (number of bits, fixed, float, etc.) runs out of range. These days that's not likely to be a problem. There used to be a dilemma in summing a large number of digital or analog signals which was how to scale the output in the face of limited range. The most conservative approach is to assume full-scale on all inputs at the same instant, but this requires you to chop more off the bottom of the sum, raising the quantization (digital) or physical (analog) noise floor. In practice a large number of uncorrelated signals are unlikely to peak all at once. I remember having this discussion with one of Digidesign's engineers years ago. Like I said, not a problem anymore. If I'm wrong I'd like to know in what way. --Frank

Well my actual practical experience is limited to telephony and the open source package Asterisk.  The conference bridge module seemed to be a nightmare for the devs, was incredibly sensitive to timing (really didn't work without an external bits clock until rewritten in 2015 to take advantage of PC architecture changes to derive a precision timing reference) and until aforementioned rewrite was a CPU hog.  It also never sounded very good.  No transcoding was going on.  That took place in another module as everything on the software "crossbar" as we called it was transcoded to ma bell standard 8khz x 8bits PCM. 

So Frank here is where I am confused, summing is not a logical or boolean operation, it's mathematical so to me is an analog construct.  A1+B1=C1 but in the digital world where amplitude and frequency are coded in both the bit depth and the data.  IE: I get that if you load a 16bit audio word into a register and shift left or shift right you increase or decrease the volume.  From this I infer that summing multiple encoded streams (assuming lossless) you have all the usual problems of floating point arithmetic on a binary device mainly precision errors. 

I probably am way over complicating this. 
Logged
Scott AKA "Skyking" Holtzman
River Delta Audio is now:

Ghost Audio Visual Solutions, LLC
Cleveland OH
www.ghostav.rocks

Frank Koenig

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 882
Re: Possible X32 firmware rumored. Your suggestions?
« Reply #81 on: September 14, 2019, 01:50:18 am »

So Frank here is where I am confused, summing is not a logical or boolean operation, it's mathematical so to me is an analog construct.  A1+B1=C1 but in the digital world where amplitude and frequency are coded in both the bit depth and the data.  IE: I get that if you load a 16bit audio word into a register and shift left or shift right you increase or decrease the volume.  From this I infer that summing multiple encoded streams (assuming lossless) you have all the usual problems of floating point arithmetic on a binary device mainly precision errors. 

I probably am way over complicating this.

Summing linear pulse code modulation (PCM) signals, as we use in digital audio, is pretty straight forward. Each output sample is the numerical sum of the values of the input samples at that time. As long as nothing overflows that's all there is to it. We're summing waveforms in the time domain, just like analog.

y(n) = x1(n) + x2(n) + ... + xN(n)

Now telecom makes me think of a standard (CCITT?) for non-linear amplitude quantized PCM. This was done to achieve a greater dynamic range from a small word size, 8 bits, I think. They were good at building non-linear A-to-Ds and D-to-As and that worked fine. But summing, filtering, or doing pretty much any other processing on those signals was damn near impossible. So far as I know the signals were always converted to linear PCM internally for any processing (most likely using a look-up table).

I find it interesting that to this day pretty much all digital signal processing is performed on linearly amplitude quantized and uniformly time quantized signals. No one appears to have cracked the nut of non-linear processing. Fortunately, floating point processors, humongous word sizes, and fast processing keep linear processing viable. And, of course, signal-specific data compression, such as MP3 or JPEG, takes care of the transmission/storage problem obviating non-linear amplitude quantization.

I'm sleepy. I better shut up and go to bed now.

 --Frank
Logged
"Nature abhors a vacuum tube." -- John Pierce, Bell Labs

Scott Holtzman

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 5613
  • Ghost AV - Avon Lake, OH
    • Ghost Audio Visual Systems, LLC
Re: Possible X32 firmware rumored. Your suggestions?
« Reply #82 on: September 14, 2019, 03:14:36 am »

How about some updates for the M/X32 app's?  Maybe some custom type layout options?

Read,yje whole thread.
Logged
Scott AKA "Skyking" Holtzman
River Delta Audio is now:

Ghost Audio Visual Solutions, LLC
Cleveland OH
www.ghostav.rocks

Helge A Bentsen

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 771
  • Oslo, Norway.
Re: Possible X32 firmware rumored. Your suggestions?
« Reply #83 on: September 14, 2019, 03:18:03 am »

Can it be a case of "different consoles sum audio differently?

I had a discussion with a colleague last night, he mixed the headliner on his choice of console, I mixed the opener on a different console. Both well-respected consoles. We ran into a issue on one console, it sounded like the mix hit a point where it compressed. So we had a look around, input levels were ok, bus levels were ok, no processing on the master etc. but for some reason once you hit the first yellow light on the master bus in one of the consoles, it sounded like it compressed. That console was routed through the other one by AES, and going straight through. No processing in the second console, we verified that also. So we ended up reducing the master level on the "compressing console" a bit and raising the level on the input fader on the second console and it was like the mix "came alive".

Strange.
Logged

Branimir Bozak

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 60
  • Stara metalika
Re: Possible X32 firmware rumored. Your suggestions?
« Reply #84 on: September 14, 2019, 09:33:48 am »

There is nothing wrong with x32 or m32 preamps.

You tell me why on Pro series consoles low end is much more present, defined and not lost in the mix, compared to x32/m32.

You could hook up a DL251 to a X32, but the preamp and ad da quality will not improve the overall limitations of whatever the software is doing to actually sum the channels.

Both Vi and Si Soundcrafts have the same Lexicon chip/engine for reverbs, and the reverbs sound much better in the mix on the Vi, than on Si, given the opportunity to mix a lot on a Vi1 and using the Expression with compact stagebox (same preamps and adda as vi1), so I don't buy the whole summing thing is the same theory.

Take Ardour/Harrison Mixbus and Cubase. There is definately something going on differently with these two softwares, even when pushing just a couple of channels to sum.

Biggest limitation of these cheap consoles are their softwares, X32 + DL251 does not beat Pro2, even when playbacking a regular cd player on two channels.  8)

Fit the X32 with a Dante card and send the signal direct outed from a CL/Rio stagebox, yes you get nicer signal, more headroom, but still the mix is limited by the x32 software. Call it summing, call it crappy eq, crappy compressor, it still sound "small", "mp3" etc.

The Si Soundcraft has a nice sounding musical eq and really good compressor. Mixes still can't sound huge and separated, like on some other upper range console.

Summing.

Enter word clock quality/accuracy next!  ;D
« Last Edit: September 14, 2019, 09:48:09 am by Branimir Bozak »
Logged
If you scoop, it sounds like poop!

Tim McCulloch

  • SR Forums
  • Hero Member
  • *
  • Offline Offline
  • Posts: 20521
  • Wichita, Kansas USA
Re: Possible X32 firmware rumored. Your suggestions?
« Reply #85 on: September 14, 2019, 10:35:28 am »

Word clock has ZERO affect on the *actual* quality of audio.  It has a huge placebo effect on some soundpersons, though.

Sort of like saying a different alarm clock makes the sunrise look nicer.
Logged
"Practicing an art, no matter how well or badly, is a way to make your soul grow, for heaven's sake. Sing in the shower. Dance to the radio. Tell stories. Write a poem to a friend, even a lousy poem. Do it as well as you possible can. You will get an enormous reward. You will have created something."  - Kurt Vonnegut

Frank Koenig

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 882
Re: Possible X32 firmware rumored. Your suggestions?
« Reply #86 on: September 14, 2019, 11:08:36 am »

Word clock has ZERO affect on the *actual* quality of audio.  It has a huge placebo effect on some soundpersons, though.

Sort of like saying a different alarm clock makes the sunrise look nicer.

Indeed.

1. Remove all effects from the signal chain.
2. Make sure nothing is near clipping. Have > 15 dB headroom.
3. Match levels exactly using nulling.
4. Do a BLIND listening test where you don't know which box you're listening to.

99% (maybe 100) of this BS will go away .

This is not a audiophile/fool forum. That's why I'm here.

--Frank
Logged
"Nature abhors a vacuum tube." -- John Pierce, Bell Labs

Kevin Maxwell

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1247
Re: Possible X32 firmware rumored. Your suggestions?
« Reply #87 on: September 14, 2019, 11:43:33 am »

Since I am the OP I would like to request that the responses stay on topic. I know this is the pot calling the kettle tarnished  ;D .

I like reading some of the off topic stuff but please start another thread for that. Thank you.
Logged

Scott Bolt

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1370
Re: Possible X32 firmware rumored. Your suggestions?
« Reply #88 on: September 14, 2019, 12:29:42 pm »

Since I am the OP I would like to request that the responses stay on topic. I know this is the pot calling the kettle tarnished  ;D .

I like reading some of the off topic stuff but please start another thread for that. Thank you.
Hi Kevin,

In keeping with your thread .....

The current firmware is already phase coherent.  There is no need for a change in this firmware to make "summing" better since it already has perfect signal summing.

Perhaps the algorithms used for eq and compression could be revisited to make a better sounding signal path.

My personal request though is that now that Music Group holds TC Helicon, could they (pretty please with sugar on top) give us an M3000 verb plugin?  I would even settle for other limitations on processing the channels using this effect to have this in my vocal chains.

My request priorities:

1)  M3000 reverb plug-in
2)  Better compression and eq algorithms
3)  Everything else :)
Logged

Tim McCulloch

  • SR Forums
  • Hero Member
  • *
  • Offline Offline
  • Posts: 20521
  • Wichita, Kansas USA
Re: Possible X32 firmware rumored. Your suggestions?
« Reply #89 on: September 14, 2019, 03:46:15 pm »

Scott, the TC FX are part of the new Midas Heritage D.
Logged
"Practicing an art, no matter how well or badly, is a way to make your soul grow, for heaven's sake. Sing in the shower. Dance to the radio. Tell stories. Write a poem to a friend, even a lousy poem. Do it as well as you possible can. You will get an enormous reward. You will have created something."  - Kurt Vonnegut

ProSoundWeb Community

Re: Possible X32 firmware rumored. Your suggestions?
« Reply #89 on: September 14, 2019, 03:46:15 pm »


Pages: 1 ... 7 8 [9] 10 11 ... 22   Go Up
 



Page created in 0.042 seconds with 24 queries.