I use MATAA, which runs on GNU Octave or Matlab. It works on Linux, MacOS and Windows. MATAA is free, open-source software. Yes, I wrote (most of) the code. See here to learn more about MATAA.
I am looking into ways to make MATAA automatically determine the settings of the RTX level switches. If at all, I want this to work on Linux, MacOS and Windows. However, this is my first time working with USB HID, and I am trying to read up on this. The USB HID terminology is confusing me big time...
+1
mbrennwa
My dream would be a MATAA function to measure automatically distorsion versus DUT output level. This would be great to plot preamp harmonics versus output level, or THD+N of power amplifiers versus output power, for example.
I've done this manually with REW and an Excel spreadsheet where I log the RTX settings, but it is tedious and it takes time. The RTX is good enough to warrant this software functionality and MATAA would be a great fit IMO.
I think greedy boyz like me would appreciate 😀.
You must have thought of this already. Haven't you 😉?
Pierre
Edit: I see you responded to mkc already. Good luck with USB HID!
Hi Pierre
The cool thing with MATAA is that you can easily do this on your own in Octave or Matlab. Just use an Octave script (or Matlab) to do whatever you need. I spent a few minutes to come up with a quick-n-dirty example that measures the distortion at ten different voltages and plots the results. See attachment (you'll have to delete the ".asc" extension from the file name).
But I guess these kinds of questions are not so much related to the RTX and should be discussed in a "MATAA channel" (try the MATAA mailing list or GitHub, or maybe open up a new thread here on diyAudio).
In audio-Tester there are also 3 windows "Rife-Vinc" 1, 2, 3. (sorry "vince" was a typo)
Hi mbrennwa,
Unfortunately i do not have much time at the moment to look at it. However, a quick Google search suggests that Octave supports USBTMC. Perhaps we should see if Jens could be persuaded to add this natively to the RTX6001.
Or else, we should look into how to add the use of native USB to octave.
Mogens
Unfortunately i do not have much time at the moment to look at it. However, a quick Google search suggests that Octave supports USBTMC. Perhaps we should see if Jens could be persuaded to add this natively to the RTX6001.
Or else, we should look into how to add the use of native USB to octave.
Mogens
Hi JensH
we are missing the damn SPDIF output behind the RTX close to the USB port...
I was thinking to do my own addon around an HC04 or so linked to the xmos spdif pin but realize we need also to pick up power supply and xmos recomend to latch the spdif with mclock .
Any chance you could design a small PCB with a pulse transformer, bnc, coax, optical, AES and provide the assembly PCB with a new back panel and the few wires to be hand soldered .. ?
I m sure a GB of 80unit at 80€ ish would be successful 🙂
if not, would you be ready to publish the information required to elaborate this ourselves?
many thanks
we are missing the damn SPDIF output behind the RTX close to the USB port...
I was thinking to do my own addon around an HC04 or so linked to the xmos spdif pin but realize we need also to pick up power supply and xmos recomend to latch the spdif with mclock .
Any chance you could design a small PCB with a pulse transformer, bnc, coax, optical, AES and provide the assembly PCB with a new back panel and the few wires to be hand soldered .. ?
I m sure a GB of 80unit at 80€ ish would be successful 🙂
if not, would you be ready to publish the information required to elaborate this ourselves?
many thanks
Thank you, got it, appreciated. However this is not what I had in mind.
I guess what i wrote wasn't clear... my query was to seek how MATAA could provide automatic changes of the output and input gain settings of the RTX so that the stimulus and measurements can span several decades and provide meaningful results even when the distorsion is low. I am familiar with Matlab but developing such a function (with automatic RTX setting changes) is not a 10 minute job for me. Would it be feasible to code an exemplar function to measure, for instance, harmonic distortions versus output level for a DUT that has unity gain from 0.1 mV up to the output level that yields 1% distortions? This would be quite useful to derive other standard test functions afterwards by lazy me for example 😀.
And on a related matter, could MATAA provide an auto-ranging function for the RTX? Could be as simple conceptually as a function that finds and returns the max input gain setting that does not overload the input for a continuous signal.
Apologies if this is out of place, but I feel these would augment the functionality of the RTX and be of interest to its users. This is why I posted this here. I agree discussion of the coding itself should be taken elsewhere. And this is not meant to be an endorsement for MATAA. It just seems to be a natural tool to try and share how to do such things.
Really hope you get that USBHID interface sorted out. Good luck and thank you!
I have already asked this to Jens ....
It would be very useful and relatively easy to realize
You should do some loopback testing before trusting the results.
Hello,
Auto-ranging is a different creature. The RTX6001 does not have this feature. The user is forced to select the output and input voltages range best suited to the DUT at hand. Pick the incorrect output and or input selector switch position and you will get compromised results at best.
The input selector switch position is fairly straight forward. Each switch position is labeled with the maximum voltage without overload. Select the switch position that you want to use.
For audio Tester: go to options, select calibrate and enter the selected voltage scale in mv. (1.0V = 1000mv)
The output selector and output voltage is not as straight forward.
The left hand output selector position is 0.100Vrms full scale. Set the output at 0dBFS the output voltage is 0.100Vrms. Set the output at -20dBFS the output voltage is 0.010Vrms.
The center output selector position is 1Vrms full scale. Set the output at 0dBFS the output voltage is 1Vrms. Set the output at -20dBFS the output voltage is 0.100Vrms.
The right hand output selector position is 10.0Vrms full scale. Set the output at 0dBFS the output voltage is 10.0Vrms. Set the output at -20dBFS the output voltage is 1.00Vrms.
You will need a calculator or spread sheet handy to calculate between the particular dBFS value and desired output voltage. I am developing a little spreadsheet for my use. I will share it here.
If you look at the early system I and system II AP machines you can write scripts to execute programs you are talking about.
If you look at the AP APx500 SW many of the procedures you are speaking of are preprogramed in Sequence Mode.
Speaking of manual switch positions to pick switch position that you think should be okay. Practically speaking they may not work out so well. As you get closer to the 0dBFS setting the generator output distortion will dominate, it will be greater than the DUT contribution. You should do some loopback testing before trusting the results.
The only difference between the plot in post 1836 and the one below is 10dB greater output. The generator output distortion increases to a large degree.
DT
Hello,
Auto-ranging is a different creature. The RTX6001 does not have this feature. The user is forced to select the output and input voltages range best suited to the DUT at hand. Pick the incorrect output and or input selector switch position and you will get compromised results at best.
The input selector switch position is fairly straight forward. Each switch position is labeled with the maximum voltage without overload. Select the switch position that you want to use.
For audio Tester: go to options, select calibrate and enter the selected voltage scale in mv. (1.0V = 1000mv)
The output selector and output voltage is not as straight forward.
The left hand output selector position is 0.100Vrms full scale. Set the output at 0dBFS the output voltage is 0.100Vrms. Set the output at -20dBFS the output voltage is 0.010Vrms.
The center output selector position is 1Vrms full scale. Set the output at 0dBFS the output voltage is 1Vrms. Set the output at -20dBFS the output voltage is 0.100Vrms.
The right hand output selector position is 10.0Vrms full scale. Set the output at 0dBFS the output voltage is 10.0Vrms. Set the output at -20dBFS the output voltage is 1.00Vrms.
You will need a calculator or spread sheet handy to calculate between the particular dBFS value and desired output voltage. I am developing a little spreadsheet for my use. I will share it here.
If you look at the early system I and system II AP machines you can write scripts to execute programs you are talking about.
If you look at the AP APx500 SW many of the procedures you are speaking of are preprogramed in Sequence Mode.
Speaking of manual switch positions to pick switch position that you think should be okay. Practically speaking they may not work out so well. As you get closer to the 0dBFS setting the generator output distortion will dominate, it will be greater than the DUT contribution. You should do some loopback testing before trusting the results.
The only difference between the plot in post 1836 and the one below is 10dB greater output. The generator output distortion increases to a large degree.
DT
Attachments
Ah, ok, I got this wrong. I am trying to wrap my head around the USB HID stuff. I a completely lost... I someone out there could recommend some reading for USB HID dummies, that would be great.
This is possible, and I don't think it needs automatic level switching at the RTX. Just set the levels at the RTX such that the DUT in signal will be in the right range, and the DUT output will not clip the RTX input and leave it there for the test. Even if "autoranging" might reduce the residual distortion introduced by the RTX, the 1% threshold is far and beyond above what the RTX will contribute at the fixed / pre-set switch positions.
Note that the latest MATAA version (as of today...) has a new function that will automatically analyse the harmonic distortion stuff 😉
If / when I manage to get the USB HID communication working, that should be technically possible. I don't have much experience with autoranging methods, but I would not be surprised if this was tricky to get right. What would/should be the criteria to switch levels? Would it be okay to switch levels within a measurement sequence? I personally like to keep this kind of stuff under manual control, but I can see how others may prefer to have this handled automatically by the software.
Ok -- I just don't want to hijack Jens' RTX thread to discuss my software.
Thank you! Although I am no expert, after a quick search I think I would personally start by having a look at GitHub - signal11/hidapi: A Simple library for communicating with USB and Bluetooth HID devices on Linux, Mac, and Windows. (HIDAPI). Happy to follow up off line.
@mbrennwa
If you have not seen it yet, the PsychHID from Psychtoolbox (interfaces between Matlab or Octave and the computer hardware) Psychtoolbox-3 - Overview could be useful.
If you have not seen it yet, the PsychHID from Psychtoolbox (interfaces between Matlab or Octave and the computer hardware) Psychtoolbox-3 - Overview could be useful.
I have successfully ran Soundeasy through separate sound cards for input and output. Looks like some interesting things can be done.
