• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

E-mu 0202 + audiotester

I use an 0202 with Visual Analyzer for various tests at home and at work. No doubt you can get something better, but it does a nice job for the money. You can run different programs for input and output, say the generator from Visual Analyzer and record with Audacity. Even though there's no explicit option for it, you can also play wav files and analyze them using Visual Analyzer.


2002-09-19 3:52 pm
EMU 0404-USB and ARTA gets my vote but obviously it depends on what exactly you'r going to measure. The phantom powered mic inputs is nice but if you never do acoustic stuff obviously those are superfluous.

Also the slightly better performance of the 0404 maybe is not needed if you measure circuits that are typically higher in noise and distortion.

ARTA is a good starting point though IMO. You can use it with the built in generator or with some other software/hardware based generator.


I was going by this info. I downloaded the specs and the A/d's are the same but the D/A's are indeed different. The 0202 also had a lower listed jitter.

Take a peek at the test data from the 0202 and it looks pretty good to me. The dip in response above 20K can be corrected for with Audiotester, and the other thing audiotester does is asynchronous measurements so external sources can be used for automated capture. One of the things i want to measure is RIAA accuracy and frequency response using test records as a source.

Of course the 0404 will do everything the 0202 will do and then some, but given my propensity to blow things up, I'd rather two 0202's than a single 0404 :)

I have seen ARTA but it does not mention 192K sampling possibilities. (audiotester at least alludes to it stating maximum frequency measurement in an FFT is 96K (SF/2)

I currently have and love the stingray but it is tough to integrate with other RTA's. Winaudio MLS supports it and i have the basic version which gets me to 48K, but the 1meg sample rate suggests possibilities much higher. Sure it is only 8 bit, but that is plenty and a reasonably automated FR measurement out to 500Khz would be perfect for me. My initial goal is not to look into the shadows which requires bit depth but to look into behavior well out beyond the audio band which requires sample rate.

My initial thinking was that if audiotester supported the stingray (they provide the dll's) then an asynchronous mode measurement would give a nice plot by simply sweeping a signal generator. I have pretty much given up on this dream which has inspired my acceptance of a 96Khz top end for an easy automated measurement. Then i can break out the stingray and excel for the higher bandwidth needs.

I have actually been starting to think i can just use the data logger function of the stingray to create a file and then use something like LTspice to interpolate it out but the learning curve for that is a bit steeper than I have time for since i am essentially a luddite. Converting the datalogger output to a WAV file seems like another interesting option (is that possible?) since i do not need real time results!



2002-09-19 3:52 pm
Isn't Stingray a 12bit device? Website says so.

However I have a similar aproach/set up as you. I went for a Picoscope 3224 (12bit 20MS/s for a 10MHz BW) for the fast stuff and a audiointerface + ARTA and other sowtware as a basis.

I think you could use the Scope as a logger and if you can convert that data to wav it should be possible to analyse in ARTA. Or you could use a recorder software as Audacity or similar to capture a waveform/sweep etc. with your audiointerface and export that as a wav into ARTA.

I have a Echo Audiofire4 but I'm thinking on a EMU 0404-USB for the measurement rig because the Audiofire4 has interupt problems which I have not been able to get rid off. Other firewire interfaces have given good results in the same app as the Audiofire4.

EMU 0404-USB seems very good to me. It has plenty of I/O and very good measured performance.

Hopefully the USB connection is less buggy than the firewire port. Firewire seems to be problematic even though some say it's better and more stable than USB. That does not reflect my experiences though and from surfing the net it seems like klicks and interupts are very common with firewire. Sometimes it can be solved with the settings in Windows but it seems sometimes and cen not be made to work. Crappy firewire chipset on the motherboard and/or crappy drivers can be the explanation.

A couple of years ago I used ISA based Clio lite and I have been looking out for a new set up for years. Right now it looks like EMU 0404-USB with ARTA is the obvious way to go.

For about $200 you get performance that you had to pay at least ten times as much for ten years ago, even outperforming AudioPrecision1

dsavitsk said:

One other thing to watch with the 0404 USB is that the SPDIF outputs require you to use ASIO drivers. Lots of software, including RMAA, can't do this. This only matters, of course, if you are testing a DAC.

Audiotester does support asio and I use the spdif output on my media server with my normal external dac for system measurements and the analog inputs of the M-Audio Audiophile 2496 (PCI based) for the signal coming from back from the mic pre-amplifier.

I received the "alternate audio dll" from the developer via email - you have to ask for it.

An interesting side note is that I have not been successful in getting asio to work on my old laptop with an M-Audio Transit USB. It causes Audiotester to crash every time.. :hot: Otherwise it works fine to 48kHz in full duplex.
I picked up the 0202 and it works with audiotester with the occasional crash. Arta keeps giving me this error unless i go to the 48K sample rate.

any ideas?


  • picture 4.png
    picture 4.png
    13.9 KB · Views: 544
I still get crashes with audiotester (running parallels on a mac) but when it is working the sweep function is sweet. I was unable to get ARTA to work at anything other than the 48K sample rate and the sweep i ran was painfully slow.

there are a number of other issues with the setup that have me scrambling and i'm sure a dedicated PC would solve all of them. For now the two tedious tasks at hand are measuring outputs which generally show their colors below 96K and the measurement and tuning of riaa EQ accuracy.

I haven't fired it up with winaudio MLS or rightmark but they are on my list to do.

I do agree the usb power makes it perfect for a portable measuring system.



2008-02-18 1:32 am
I use an 0202 with Visual Analyzer for various tests at home and at work. No doubt you can get something better, but it does a nice job for the money. You can run different programs for input and output, say the generator from Visual Analyzer and record with Audacity. Even though there's no explicit option for it, you can also play wav files and analyze them using Visual Analyzer.

Hi Conrad, could you please be a bit more specific about your setup and configuration? I have an EMU 0404 and I have some problems to make it up and running with Visual Analyzer (when started it always give me an error about the input mixer not being available). I can generate signals and hear them at 0404 but it seems the input side is not working.

I have also Audacity and here I can record from mic using the 0404 so, apparently, the input side in 0404 is working. You can send me a PM in case you would prefer that way.

Thanks in advance

The mixer error seems to be normal for the E-mu and I haven't spent much time investigating it. Be sure you're running Visual Analyzer 10.0.5 and have the most recent drivers from Creative for the E-mu. Go to your PC sound setup under control panel and select the input you're using at full volume. I never use software gain control, always the E-mu gain control. More on that at the end. You can't afford to throw away bits! Typical settings for me are FFT size=16384, sampling=44100, stereo, 16 bits, I/O=E-MU 0202. For the waveform gen, buffer samp=4096, sampling=96000, buffers=4, 16 bits. There are many combinations of settings that will give false results or crash the program. IMO, these are just limitations of what a given machine can deal with in terms of sample rates and bits, and I'm not sure how the program could foresee this without limiting the performance to below what's possible. Thus, it doesn't bother me too much. I also have the advantage of having used a lot of traditional bench type test equipment, so it's really obvious to me when the Visual Analyzer results are bogus. This is common when making THD measurements with inadequate resolution, having an alias condition, or if the vertical spectrum scale is too good to be true.

The E-mu front end and gain control is a bit odd. I think mine insists on a low value DC return path, as if I were to ac couple the input, the front end amp would drop out at a bit over half volume setting, or even an unpredictable level depending on the charge of the input coupling cap. Depending on what you're connecting, it may not hurt to shunt the input with 10k or so.