See "Requirements" on this page, and also page 6 of the user manual.
https://artalabs.hr/download.htm
https://artalabs.hr/download.htm
Hello,
While I own a legit copy of ARTA, I have never even looked at the manual, I use the instructions by Kimmo for use with VCad.
Guess I should have gone looking for them.
After reading the requirements, it looks like I could run ARTA on a toaster.
Thank you,
David.
Any used laptop with Win XP or Win7 will do. The battery should be ok so you could measure without AC-Adapter if required. ARTA runs on Linux/Wine as well with some minor restrictions: The control panel is not available and never was. I have a licensed ARTA for years now but just discovered that REW offers more nice options when measuring THD+N.
I an trying to make FRD files in ARTA for the first time but i get an error?
I follow this instruction:
https://pearl-hifi.com/06_Lit_Archive/15_Mfrs_Publications/ARTA 1.7/FR _Measurement_Using_ARTA.pdf
This is how i do it:
I follow this instruction:
https://pearl-hifi.com/06_Lit_Archive/15_Mfrs_Publications/ARTA 1.7/FR _Measurement_Using_ARTA.pdf
This is how i do it:
Is anyone aware of a currently available sound card with high S/N and low distortion which can have it's brick-wall filter moved with the sample rate for the input?
I am running ARTA and AUD3 with a ASUS Essence STX ii which has the filter fixed for 44.1KHz sample rate.
My old Xonar SE moved it's filter and could plot up to 48Khz but had poor S/N and distortion.
It is a PIA to swap sound cards for different functions.
I have not found any references to switching the filter in the Essence STX ii.
I need the wider BW to look for resonences in audio output transformers used in tube amplifiers.
Thanks.
Sven
I am running ARTA and AUD3 with a ASUS Essence STX ii which has the filter fixed for 44.1KHz sample rate.
My old Xonar SE moved it's filter and could plot up to 48Khz but had poor S/N and distortion.
It is a PIA to swap sound cards for different functions.
I have not found any references to switching the filter in the Essence STX ii.
I need the wider BW to look for resonences in audio output transformers used in tube amplifiers.
Thanks.
Sven
Last edited:
Member
Joined 2003
Member
Joined 2003
Other feature requests:
ARTA:
-addition of left gate for impulse response -> FR processing. Include left window option for spatial response export.
-add tukey 0.25 and tukey 0.5 window functions.
-window cursor and gate positions stay in place when loading a PIR file.
-window cursors stay in place when changing to FR1, FR2, SPA then back to IMP
-ability to have more than one impulse response open at a time would be great as well, so one could take multiple measurement and toggle between them.
-ability to leave FR window open and do other things like adjust window positions would be nice.
-generator for setting levels in the impulse response "sweep" section would be a nice addition. Generators exist in the noise and MLS sections, but not for sine sweep.
LIMP:
-fast sine sweep that isn't slow stepped would be welcome by every user I am sure
-some automated cable compensation process would be great, similar to the basic channel calibration. It is a bit clunky to short the leads, run measurement, calculate RLC, record numbers, enter manually. It could be as simple as a dialog that says "short the leads and push this button"
-requiring re-calibration when changing sampling rate or FFT size seems a bit odd to me, I'm sure there is a way around this to avoid re-calibration when changing these settings.
Many years ago, I messaged the developer about the first point.
The answer was that it would be (a lot) less precise to do a fast sweep method?
Which to me doesn't make a lot of sense, except maybe seen from the mechanical side of things.
The cable compensation thing also always gets me.
I have to really force myself into the flow of always enabling it and always going through the same calibration steps again.
Which is rather annoying in a dedicated setup (calibration is also never ever off)
Also, the standard recommended way of measuring speakers is sub-optimal.
One should actually use a constant voltage method (also recommended by AES), instead of a voltage divider method.
The last one won't give you an accurate reading, since the voltage heavily drops at the Fs for example
Member
Joined 2003
If all you want is basic impedance, the difference in precision would be negligible, as comparisons to REW or DATS will show. The comparison should really be to the PN noise signal, where fast sine sweep will be much better at low freq than the noise signal, and take the same amount of time. As well, the option for stepped sine would remain for those that have all day to wait around and want that extra precision.
I think the voltage divider is fine, it seems fairly common across ARTA, REW, and Soundeasy. However, for "voltage source" I would think you can simply change from small value of sense resistor to large value. If the sense resistor is say 400 ohms instead of 10-30ohm, the measurement will be closer to constant voltage since the 400 ohm resistor will be the dominating factor.
Last edited:
Yeah those are exactly my thoughts as well.
Pink noise doesn't always work very well at the lower frequencies.
Member
Joined 2003
A constant voltage source can simply be created by adding a resistor inside the feedback loop, right after the amplifier's output.
I know the divider method is (very) common, but it actually create a lot of implications (plus it's wrong according to AES standards)
Especially when one tries to measure T/S parameters for example.
That all has to do with how the speaker (read Cms mostly) behaves at certain voltages.
SB Acoustics for example (but there are MANY more brands) use a lot higher voltage than 0.1V
I haven't done all the error calculations just yet, but I am willing to bet that using a divider network actually creates a lot more errors than a fast sine-wave sweep.
For a more constant voltage across the driver you need the lowest possible sense resistor and a source with very low output impedance (i.e. an amplifier).
No you don't if you add this sense resistor inside the voltage feedback of the amplifier.
The downside is that the amplifier itself has to have quite some voltage overhead to compensate.
But that's practically still very doable
How would that help? Under normal situation, a driver is connected to the amplifier via a crossover circuit which has a resistance and on top some reactive impedance (inductor, capacitor). Just the resistance is -including the cable - easily 1 Ohm or more.