ProSoundWeb Community
Sound Reinforcement - Forums for Live Sound Professionals - Your Displayed Name Must Be Your Real Full Name To Post In The Live Sound Forums => Audio Measurement and Testing => Topic started by: Janne Paul on March 01, 2017, 05:39:12 PM
-
So I bought SysTune prior any measurement experience. Now I don't get how to configure it for dual FFT measurements. The phase plot looks chaotic, line graph is just bouncing around.
What's wrong with my config?
I'm using my iPhone as pink noise generator. The left iPhone channel goes to Roland Octacapture channel 2 ("Reference channel").
The right iPhone channel goes to amp and left speaker.
ECM8000 mic inputs to channel 2 on Octacapture ("Signal channel".
Octacapture connected with USB to PC.
-
So I bought SysTune prior any measurement experience. Now I don't get how to configure it for dual FFT measurements. The phase plot looks chaotic, line graph is just bouncing around.
What's wrong with my config?
I'm using my iPhone as pink noise generator. The left iPhone channel goes to Roland Octacapture channel 2 ("Reference channel").
The right iPhone channel goes to amp and left speaker.
ECM8000 mic inputs to channel 2 on Octacapture ("Signal channel".
Octacapture connected with USB to PC.
I don't know how the iThingy generates pink noise, but I'd use only 1 output (say, Left) and y-cord that to the reference input and DUT input. That way you're assured both are receiving the exact same signal.
Also you need to do an initial loop-back test. If that doesn't give flat magnitude response and a phase response that doesn't wrap (wrap is what you're seeing as a big saw-tooth in your phase window), you've got problems to address before attempting measurement.
While the examples at Rational Acoustics are based on their Smaart product, the concepts are universal so you might want to check out their tutorials and white papers. I'd be really surprised if AFMG doesn't have similar resources.
Forum moderator Doug Fowler is a SysTune instructor and has some YouTube videos. Start with this one:
https://www.youtube.com/watch?v=zFGKAwHQTtg
Finally, the PSW forums include one specific to measurement and testing:
http://forums.prosoundweb.com/index.php/board,78.0.html
I will ask the mods to move your post to that forum.
Have fun, good luck.
Tim Mc
-
So I bought SysTune prior any measurement experience. Now I don't get how to configure it for dual FFT measurements. The phase plot looks chaotic, line graph is just bouncing around.
What's wrong with my config?
I'm using my iPhone as pink noise generator. The left iPhone channel goes to Roland Octacapture channel 2 ("Reference channel").
The right iPhone channel goes to amp and left speaker.
ECM8000 mic inputs to channel 2 on Octacapture ("Signal channel".
Octacapture connected with USB to PC.
You say that you have both the reference and the mic on channel 2. That would be a starting point.
You also don't have any delay set.
That will screw up the phase trace A LOT
-
I downloaded Smaart trial version and installed it on my MacBook and got it working.
It should prove that my measurement hardware and wiring is correct. So the problem should be with the configuration of Systune. Or my Windows PC (where Systune is installed)
"You also don't have any delay set."
I have now set the delay to the same value (5.08ms) as in Smaart and also tried the peak and auto function. Makes no difference.
"You say that you have both the reference and the mic on channel 2."
It's a typo. My misstake. The setup is now:
iPhone as pink noise generator. The left iPhone channel goes trough an y-cord to the reference input and DUT input.
-
I downloaded Smaart trial version and installed it on my MacBook and got it working.
It should prove that my measurement hardware and wiring is correct. So the problem should be with the configuration of Systune. Or my Windows PC (where Systune is installed)
"You also don't have any delay set."
I have now set the delay to the same value (5.08ms) as in Smaart and also tried the peak and auto function. Makes no difference.
"You say that you have both the reference and the mic on channel 2."
It's a typo. My misstake. The setup is now:
iPhone as pink noise generator. The left iPhone channel goes trough an y-cord to the reference input and DUT input.
I have found the phase trace in smaart to be much smother than Systune. I almost got the impression that the phase trace in Systune was unaffected by the averages. If i recall i could set the averaging to infinity, and the phase would still jump around like crazy.
-
I have found the phase trace in smaart to be much smother than Systune. I almost got the impression that the phase trace in Systune was unaffected by the averages. If i recall i could set the averaging to infinity, and the phase would still jump around like crazy.
I used Systune quite a lot when it first came out.
But, like you, I could not get the phase trace to display anything useful. I was using Smaart 5 as my standard for system alignment before that, and the phase trace was fine.
When Smaart 7 came out with live multiple traces-I stopped using Systune.
It may have different now, but since I have changed computers, I cannot seem to be able to get my license back. That seems to be problem with AFMG. But that is a different story.
-
So I bought SysTune prior any measurement experience. Now I don't get how to configure it for dual FFT measurements. The phase plot looks chaotic, line graph is just bouncing around.
What's wrong with my config?
Why are you using such a short FFT size. It is limiting your frequency resolution to ~190Hz. A more appropriate value would be a size of 32k or the default of 64k, giving you ~1Hz frequency Resolution. You also probably do not need nearly as many averages. I find 4 usually gives me a stable display for both Magnitude and Phase.
If you continue to have problems, please contact us at [email protected] .
Bruce
-
It may have different now, but since I have changed computers, I cannot seem to be able to get my license back. That seems to be problem with AFMG. But that is a different story.
You might want to contact us at [email protected] so we can sort that out for you. It looks like this is the first we've heard about this from you.
Bruce
-
I have found the phase trace in smaart to be much smother than Systune. I almost got the impression that the phase trace in Systune was unaffected by the averages. If i recall i could set the averaging to infinity, and the phase would still jump around like crazy.
The phase trace is affected by the number of averages in SysTune. If the phase is jumping around as much as you say, I would suspect environmental factors are at play to cause that.
Bruce
-
You might want to contact us at [email protected] so we can sort that out for you. It looks like this is the first we've heard about this from you.
Bruce
I went back and forth about Systune and my EASE license with our IT guy and AFMG a number of years ago.
It could never be resolved so I just gave up and went another direction for my usages.
We spent a lot of money for the licenses that were not usable.
-
I used Systune quite a lot when it first came out.
But, like you, I could not get the phase trace to display anything useful. I was using Smaart 5 as my standard for system alignment before that, and the phase trace was fine.
When Smaart 7 came out with live multiple traces-I stopped using Systune.
It may have different now, but since I have changed computers, I cannot seem to be able to get my license back. That seems to be problem with AFMG. But that is a different story.
The reason I chose Systune over Smaart was that i visited a Danley stand at a tradeshow. In the same stand was also Linea Research. The LR guy showed me how you can control their DSP within Systune from an iPad. Did get excited over that you could invert the measurement curve and overlay phase linear EQ correction. I later bought that DSP and obviously now wanted Systune.
Just a story for you that people actually but stuff after seeing them at trade shows, but also to consider the systune integration with your DNA SC48. Can complement each other beautiful if it can be made to work.
-
Yes! ??? I got it working, thanks to input from both Ivan and Bruce. It was the infamous "error 60" (error 60=problem is 60cm behind the screen, if you what I mean :).
The delay was set incorrectly and also the FFT size. I also had to swith display from FULL to WINDOWED phase response.
Big thanks guys!
-
Yes! ??? I got it working, thanks to input from both Ivan and Bruce. It was the infamous "error 60" (error 60=problem is 60cm behind the screen, if you what I mean :).
The delay was set incorrectly and also the FFT size. I also had to swith display from FULL to WINDOWED phase response.
Big thanks guys!
That looks much better :)
-
Yes! ??? I got it working, thanks to input from both Ivan and Bruce. It was the infamous "error 60" (error 60=problem is 60cm behind the screen, if you what I mean :) .
The delay was set incorrectly and also the FFT size. I also had to swith display from FULL to WINDOWED phase response.
Big thanks guys!
Excellent! You are welcome.