Signal Level in miniDSP x-over - Page 2 - diyAudio
Go Back   Home > Forums > Commercial Sector > Manufacturers > miniDSP

miniDSP Low cost, modular Digital Signal Processor (DSP) kits for the DIYer from miniDSP.

Reply
 
Thread Tools Search this Thread
Old 25th February 2013, 02:19 PM   #11
diyAudio Member
 
Join Date: Aug 2010
The meters might be calibrated to random noise signals and not for sinusoids

Are you able to repeat the test with random noise? (Maybe I can do so this evening)

Also not sure where actual rms calculation takes place? In the DSP, PIC or PC? Being a bit slow might be a problem with data-transfer between miniDSP and plugin. I agree it would be nice to have a peak indicator!

Last edited by curryman; 25th February 2013 at 02:24 PM.
  Reply With Quote
Old 25th February 2013, 08:48 PM   #12
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
Sorry for the unintentional double posting (posts # 9 and # 10) . It would be nice if a moderator could delete one of them.

Quote:
Originally Posted by curryman View Post
The meters might be calibrated to random noise signals and not for sinusoids
Are you able to repeat the test with random noise? (Maybe I can do so this evening)
Curryman
You can try if you wish but you’ll get more confused. See for example here:
http://www.meyersound.com/pdf/cinema...ech_report.pdf
What is to keep from this?
A. There is 14db to 15.7db difference between the Peak and RMS values on all these test signals.
B. Their RMS value are to be held 21 db to 28db below the Digital Full Scale.
C. Their Peak value are to be held 6.8 db to 15.8db below the Digital Full Scale (the 6db overhead for even the True professional PPMs was a known factor from the analog past)
As you can see from the screenshot I have posted, ( http://www.diyaudio.com/forums/attac...screenshot.jpg ) the relationship is not much different there (shown: Bit Depth Meter, Vu Meter, Peak Meter and the miniDSP RMS Meter, all registering their indications while monitoring in parallel one channel from the same FM radio station program)

Quote:
Also not sure where actual rms calculation takes place? In the DSP, PIC or PC?
Take PC out of this.

Quote:
Being a bit slow might be a problem with data-transfer between miniDSP and plugin.
I would say no.

Quote:
I agree it would be nice to have a peak indicator.
It would be useful

George
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]
  Reply With Quote
Old 25th February 2013, 09:48 PM   #13
diyAudio Member
 
Join Date: Apr 2005
Location: Melbourne
I think the problem here George as Curryman has alluded to is that the data transfer between the board and the software isn't fast enough to keep up with the peaks. And the peak hold would have to be done in the software.

The meters themselves whilst being more or less useless for music signals are, apart from reading 2db for every 1db change to the attenuators reasonably accurate for steady state signals.

So I just run the input attenuators at 0db and have adjusted the analog input level to the board so the meters read -9.3db for a 1khz 0db down (ie. flat out) signal off a test CD. As CD is my loudest analog source (only by a couple of db) this works out fine.

I don't apply any eq boost at all (just cut if needed) so this works fine and the 3.3db gives just a bit of extra headroom for safety.

For a typical loud "pop" track the input meters read just into the high 20's.

Obviously if your were applying large amounts of boost via eq you would have to take this into account and reduce the signal to the board by a similiar amount.
  Reply With Quote
Old 25th February 2013, 10:30 PM   #14
diyAudio Member
 
vacuphile's Avatar
 
Join Date: Apr 2011
Location: Seaside
George, I am very thankful for your work on this. I use MiniDSP for first prototyping speakers, before creating an analog analog. This is very useful information.

The question I would like to throw at you and others is the following: your measurements show how important it is to get the gain structure right. Too much input level and it starts to clip, and too little ... what then? The lower the signal, the lesser bits will be used in the AD-DA conversions. Since I run these prototypes from an analog signal with volume control, I must have listened to music coded with 8 bits or even less. The remarkeable thing is that this does not seem to degrade the sound quality perceptually that much. In other words, I would argue to stay on the safe side, and allow for more headroom than you instinctively would. Am I out on a limb here?

Anyways, the reason I translate the MiniDSP settings into an analog active filter is so that gain structure falls out of the equation, but I am almost tempted not to bother.
  Reply With Quote
Old 26th February 2013, 07:18 AM   #15
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
Quote:
Originally Posted by vacuphile View Post
The question I would like to throw at you and others is the following: your measurements show how important it is to get the gain structure right. Too much input level and it starts to clip, and too little ... what then? The lower the signal, the lesser bits will be used in the AD-DA conversions. Since I run these prototypes from an analog signal with volume control, I must have listened to music coded with 8 bits or even less. The remarkeable thing is that this does not seem to degrade the sound quality perceptually that much. In other words, I would argue to stay on the safe side, and allow for more headroom than you instinctively would. Am I out on a limb here?
Vacuphile thank you for your kind words.
We have to be as much out on a limb here as much out have opted to be in the professional fields.
See Meyer Sound link
Consider how much more $$$ they have to invest for “throwing away” (reserve for) 21 db to 28db on the huge audio installations in theatres, auditoriums, planetariums ect. It would be instructive if Ed Simon and Tom Danley chime in here to give us first hand information on the topic (I bet they will admit that compressors have saved their pockets).
In broadcasting (big money there too) , 28db headroom is not enough.

Quote:
Originally Posted by KenTripp View Post
For a typical loud "pop" track the input meters read just into the high 20's.
For a typical compressed material , you are marginally OK.

Quote:
Originally Posted by KenTripp View Post
I think the problem here George as Curryman has alluded to is that the data transfer between the board and the software isn't fast enough to keep up with the peaks.
I have said that I doubt it. I have to think over for a way to test it. (*)

Quote:
Originally Posted by KenTripp View Post
And the peak hold would have to be done in the software.
Apart from input level scaling, input anti-aliasing filtering and output low pass filtering, everything else is done in software.

George

(*) Now see this quote from ADAU1701 data sheet:
Quote:
ADAU1701 programs can be loaded on power-up either from a serial EEPROM through its own self-boot mechanism or from an external microcontroller. On power-down, the current state of the parameters can be written back to the EEPROM from the ADAU1701 to be recalled the next time the program is run.
My interpretation is that the sw resides inside 1701 and executed from within. But I may be wrong.
>Edit (Serial port, SPI port and Multipurpose Pins tmax is in the tens of ns. It is only the I2C Port that has tmax in the hundreds of ns) .
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]

Last edited by gpapag; 26th February 2013 at 07:34 AM.
  Reply With Quote
Old 26th February 2013, 07:23 AM   #16
minidsp is offline minidsp  Hong Kong
diyAudio Member
 
minidsp's Avatar
 
Join Date: Dec 2009
Dear All,

First off, Thanks to George for bringing up this issue along with some good investigative work. Few weeks ago we did get a warning that we should have a look at it. We already started into looking into it to review what was the issue. It was a 2 fold issue it seems:
- A bug in the software (easy the fix)
- An incorrect type of DSP block used for the metering (harder to fix since it would mean "breaking the configuration")

Since Monday as we learned of the issue and found a fix by modifying the DSP structure. We are now going ahead and started to update all DSP structure for the metering to perform as it should. Problem is that it will unfortunately break the DSP configuration as previous .xml file will not be compatible anymore. (Shifted address)

In the future, we might see how we can build a script to "update" them. A second step though. In the next few days, we'll first post a new set of plug-ins with updated DSP block for monitoring and make an announcement. Hoping this info helps and thanks for your patience as we work through this issue.

Feel free to email us if you have any questions.

DevTeam
__________________
www.minidsp.com - Low cost & modular audio DSP kits for DIYers - Follow our tweets @ minidsp
  Reply With Quote
Old 26th February 2013, 07:57 AM   #17
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
Dev Team

You have responded here in less than 24 hours from the time you received my e-mail. You have my respect

George
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]
  Reply With Quote
Old 26th February 2013, 08:46 AM   #18
diyAudio Member
 
Join Date: Aug 2010
Thank you very much DevTeam for this Info! This saves me some time as I don't "need" to investigate it further this evening
  Reply With Quote
Old 26th February 2013, 01:32 PM   #19
gpapag is offline gpapag  Greece
diyAudio Member
 
gpapag's Avatar
 
Join Date: Nov 2002
Location: Athens-Greece
In the meantime, you can have a look and download these peak meter programs (configurable and calibreable)

Audio Level Meter

George
__________________
["Second Law is a bitch." - SY] ["The Road To Heaven:Specify the performance & accept the design. The Road To Hell:Specify the design & accept the performance"-Bruno Putzeys]
  Reply With Quote
Old 28th February 2013, 12:41 AM   #20
diyAudio Member
 
Join Date: Sep 2009
Location: Australia
Hello George and Devteam.

Do you know if these issues also apply to the "2x8" minidsp platform?

(I also echo the thanks from George to Devteam, for quickly looking at the issue)
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
signal level sakshi Analog Line Level 2 31st December 2012 11:40 AM
Signal levels shown in MiniDSP software look too low - is it working? DrNick miniDSP 2 10th December 2012 08:12 AM
Signal level 39tomcat Analog Line Level 2 23rd March 2012 03:43 PM
How do I turn an high current signal into a line level signal?? sardonx Tubes / Valves 8 27th August 2005 11:16 AM
how to bring down line level signal to preamplifier level deji Solid State 15 15th April 2004 04:13 PM


New To Site? Need Help?

All times are GMT. The time now is 03:50 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2