Lookin for a simple measurement system

Hi,

I'm new here and lookin for a suggestion pc/mac base simple and not expensive software to do measurement of:

1. compression ratio/slope (I attach the picture from DAOC thread)
2. phase in eq (I attach the picture from Studer 169 thread)

Right now I use old True RTA from True Audio for frequency response.
Any reply will be very appreciated.
Thank You

Nescafe said:

Hi,

I'm new here and lookin for a suggestion pc/mac base simple and not expensive software to do measurement of:

1. compression ratio/slope (I attach the picture from DAOC thread)

Click to expand...

I can't think of anything off-hand that will plot output level vs input level to see a compressor's operation as a function of ratio. I suppose you could use an RTA or transfer-function measurement program and send tones into the compressor at various input levels and enter the output level into a spreadsheet. For each sweep you vary the ratio.

2. phase in eq (I attach the picture from Studer 169 thread)

Right now I use old True RTA from True Audio for frequency response.
Any reply will be very appreciated.
Thank You

Click to expand...

To measure phase, or more comprehensively, to measure a filter's transfer function (which gives magnitude and phase response), you need a program which can do that. I'm a shill for Rational Acoustics' Smaart, but there are other options which may be less expensive.

-a

Hi Andy Peters,

Thank You for replying, right now I'm using protools with 1kHz test tone and automation level (like sweep) and capture back then check the wave, but it cannot tell the transfer function and not ideal for checking and matching left right response.

And for phase check, yes I use smaart before, borrowing from my fiend along with his laptop, but I cannot do it again and again  ;D lookin for the cheaper substitution for my own

Thank You

In terms of flexibility the ultimate measurement system might be any old DAW plus Matlab (or the open source alternative Octave). The function tfestimate() alone is a workhorse (for estimating transfer functions using the Welch periodogram method.. Which is stats-speak for Fourier analysis in the presence of noise and other errors).
You will have to write code, and it is by no means instant gratification. You will have to get the audio files into Octave (easy if your DAW project folder and current Octave folder coincide). You will have to study the theory to correctly choose parameters and interpret results.

The upshot is, that you can perform any analysis you can think of if you understand it we'll enough to write it ( or use some of the many built in functions ).

Feel free to ask for pointers if you go down this road. It is very doable.

VictorQ said:

In terms of flexibility the ultimate measurement system might be any old DAW plus Matlab (or the open source alternative Octave).

Click to expand...

Definitely a +1 to Octave!

-a

Another alternative, in the area of 'writing your own', is 'Labview'

For those fortunate enough to have access, it has windows audio drivers, including asio, with no additional hardware required.

A whole realm of library functional blocks is available for real time data input and visualisation as well as dsp and such.

It uses the 'visual' defining of the system block diagram and auto-generates code with the capability of inserting scripts.
The scripting can be done in some variety of languages, c++ and variants being the most refined.

The end result can be compiled into an executable+runtime installer which can be distributed.

It's pretty interesting stuff.

Funny, back in the day we used to roll our own plots with pencil and paper... We used Log or log-log paper referring to the scales for X and Y axis.

These days there so much free stuff... Back in the late '70s I had to program my then pretty crude computer to plot out response charts for filter designs using the tab function on the dot matrix printer... Crude but better than paper.

JR

I remember using the mainframes with teletype and fortran for calculating filter coefficients in the early 80s - that was my earliest memory into into clunky computers.

Write the app, submit it to the queue and a day or so later, get the printouts.

Fortunately I missed punch-cards!

---

Also, just using plain old Visual Basic or so is a pretty easy path into writing an audio transfer function tool.

There's plenty of code about for recording wavefiles from Windows audio devices and also a fair bit of basic graphing libraries.

The 'sync-ing' of record/playback doesn't need to be all that tight for this app.

I think there is a free VB version from MS.

Nescafe said:

I'm new here and lookin for a suggestion pc/mac base simple and not expensive software to do measurement of:

Click to expand...

What about this?
http://www.virtins.com/multi-instrument.shtml

alexc said:

I remember using the mainframes with teletype and fortran for calculating filter coefficients in the early 80s - that was my earliest memory into into clunky computers.

Write the app, submit it to the queue and a day or so later, get the printouts.

Fortunately I missed punch-cards!

Click to expand...

Yup, I recall using Hollerith punch cards to submit my Fortran homework back in the '60s to run on the university IBM 360 overnight. By the late '70s I had my own Heathkit/DEC LSI 11 so I had some computing on demand. 

---

Also, just using plain old Visual Basic or so is a pretty easy path into writing an audio transfer function tool.

Click to expand...

Yup, my early crude filter design software was written in Basic. I found it very powerful to plug in actual capacitor values I had in stock so I could tweak the cheaper to change resistor values to dial in complex multi-pole filters for my analog delay designs and such.

JR

There's plenty of code about for recording wavefiles from Windows audio devices and also a fair bit of basic graphing libraries.

The 'sync-ing' of record/playback doesn't need to be all that tight for this app.

I think there is a free VB version from MS.

Click to expand...