Metering the IMD of EMU0404 with Multi-Instrument Pro-3

[fotios]

Originally posted by janneman



I think 0404 can do 192kHz but not at 24bit

Jan Didden

Hi Jan, glad to meet you again after a long time.
As for bit depth / sample rate, all models of EMU from 0202 up to 0404 (included also the unsurpassed PCI 1212M-V2) can do run meassurements in extreme settings like 24bit/192KHz in record-playback mode simultaneously. This depends only in the O.S. of PC used, and it can be obtained only under x64bit OS which can manage up to 128GB SDRAM. I have mentioned in the first posts the PC that i use ( C2D 3GHz, 8GB SDRAM, 1333MHz FSB). Under 32bit OS these settings of sound card cause hang-ups continuously, instead under 64bit with 4GB SDRAM -at least- the measurements executed on the fly.
In the first post, the plot has obtained at 192Ks/s with 24bit resolution.
The issue it is that, for commercial type measurements (narrowed between 20 to 20000Hz) there is no reason of using a so high sampling rate, instead we need all the available bit depth of sound card. A sample rate of 48Ks/s is more than enough by giving a bandwidth of 48/2 = 24KHz in measurement. Instead, for research and experiments for our purpose, there are the problems which stated very correctly by Conrad Hoffman.
In the followig link, there is a thread with very interesting discussions about those issues:

Issues with Emu tracker and THD measurements

Regs
Fotios

[fotios]

Originally posted by Conrad Hoffman
I have an 0202 and am very happy with the measurement performance, though the output is quite noisy. This can disappear in a system measurement, but it bothers me.

My real problem with these sound card/software combinations is how to verify the results. IMHO you need access to conventional high quality ($$) instrumentation to compare readings, and then must always use the exact software setup.

I've seen too many cases where changes in buffers, sample rates, hidden scope settings and PC settings affect the results. It doesn't inspire confidence that any given measurement is truth.

A thought- we need a reproducible calibration circuit that gives a known spectral output. Something anybody could build and confirm that their soundcard measurement system agreed with everybody else's. Ideas?

Hi Conrad. Glad to meet you again after 3 months when we had this great and helpfull discussion in the thread that i mentioned above in Jan Didden. In the concrette thread we had the helpfull participation of members like "BFNY" which was busy in the measuring instrument industry for many years. Me, as well, i had -in the background- the support and the advices of the very polite CEO of Virtins Tech who is a real expert in everything about PC based measuring software.
Well, the only that i did from then was comparisson of measurement plots that i obtained with my PC with those included in the handbook of Virtins M.I. Pro, by placing ready setups (e.g. for CCIF IMD meas) included in the concrette software. I am in place to say now, that the ready setups offered from software -for all kind of audio measurements- it is the only way to get real (at least by 90%) and trustable results. I must to point out, that the ready setups does not make any trick, simply these offer ready and tested - for infinite times, a thing that we aren't in place to make by alone- setups for concrette measurements. To do the same by alone, i must to spend out about 3 to 4 months occupied every day exclusively with this software... that thing is impossimple for a working man. Maybe i will do that, when and IF will be retired after 17 years at least!

Regs
Fotios

[fotios]

Originally posted by richwalters
My experience of the picosoft 4224 is short of a disaster. The downloaded pre order sample, behaves quite differently to the real thing. Nunu.....be careful.Each time a software download comes the unit runs riot.....

richy

Hi richy

I can understand your anger, because this 4224 as i have seen it is a quite expensive instrument (arround 620 euro) if you had the intention to use it mainly as DSO, because with the same ammount you can by a stand alone DSO (bench top) which includes also PC communication interface and software. Usually people purchase such type instruments like pico, because those offer some better ADCs of 12 to 16 bit compared to the standard 8 bit flash ADC offered from the bench top DSOs, as well some most sophisticated software, like the FFT analyzer which includes some ready audio measurements like THD and SNR.
I had tried in the past the pico software, and i was disapointed (fortunatelly!) for its FFT purpose. Moreover i had already a good Hameg DSO. I have mentioned above the reasons of superiority of sound cards in audio measurements.
But the problems that you reffer are very curious. I have the sense that something wrong exists in the communication of picoscope with your PC. For example, pico 4224 has a 2.0 USB interface. If you are using it with an old PC that includes a main board of 2 years old, you may know that the front USB ports that offers are usually USB 1.1. Only the back ports are compatible with USB 2.0 interface. Those instruments, can run correctly only with USB 2.0.
The above mentioned it is only one of the possible reasons for the bad behaviour of pico. I don't think that the instrument has any operating problem. Do a search in your PC settings please.

Regs
Fotios

[richwalters]

fotios;
Got the message. Notebook is a Toshiba satellite A100 and Pico mentions this model is not on their problem list. It does work fast with MathCad and many other complicated programs and I like it. But with the 4224 pico6 soft, nup. Experts are stuck like Apollo 13 ground crew looking for answers and I'd bet it's a silly one.The updates are specifically for USB bug problems and it seems I'm just unlucky and xxx off. I've un-installed/installed so many times and delected stagnant files and still doesn't work. They understand and will refund.

I have a suspicion that the current demand of the 4224 when working hard could be dragging down the 5V USB. The psu does get quite warm to touch. Will check specs.

Naturally I'm on the hunt and open to any advice who is using FFT and mixed signal stuff.
Thanks
richy

[fotios]

Originally posted by richwalters
fotios;
Got the message. Notebook is a Toshiba satellite A100 and Pico mentions this model is not on their problem list. It does work fast with MathCad and many other complicated programs and I like it. But with the 4224 pico6 soft, nup. Experts are stuck like Apollo 13 ground crew looking for answers and I'd bet it's a silly one.The updates are specifically for USB bug problems and it seems I'm just unlucky and xxx off. I've un-installed/installed so many times and delected stagnant files and still doesn't work. They understand and will refund.

I have a suspicion that the current demand of the 4224 when working hard could be dragging down the 5V USB. The psu does get quite warm to touch. Will check specs.

Naturally I'm on the hunt and open to any advice who is using FFT and mixed signal stuff.
Thanks
richy

Hi again richy
Please give me a little time to check the specs of your Notebook and -again- the specs of pico 4224.
Stay tuned.
Fotios

[fotios]

Hi richy
From a first view, your notebook it looks very good. I have seen as well, the power consumption of pico 4224 which is 2,5Watt. I don't know how many power can sustain your notebook (via its USB ports) but from experience, i know that the power supplies of notebooks are worsen year by year. For example, me i have an old notebook (Uniwill) which i bought before 5 to 6 years. Its power supply, it is a real monster (not in dimensions but in power handling) compared with the current offered with modern notebooks. First of all, it is plenty of ventilation holes and includes as well a mini blower for forced air cooling. Today's power suplies are sealed (by some way it is worthy because the lower consumption of C2D CPUs compared with the P4 that includes my old notebook) and this is to my technician's eyes a horrowing sight. I remember well, that i had loaded in its 2 (only) USB ports, an optical mouse and a 4 port USB hub in which i had connected one external DVD writter and one HDD to copy moovies! During this time, i don't appreciated its powerfull supply (when i plug it on the mains, allways i can see a momentary sparcle!) but today i know its value.
Anyway, it seems like you have found the problem, i agree with you in the voltage drop of USB ports caused from pico. There is another one way to check it, by using a powered USB hub between your notebook and pico, but there is again the problem of reduction of communication speed! What a confusion!
Before few days, i did a walk in Virtins Tech. site, and i found new hardware stuff. They offer for example, a series of 3 PC based Oscilloscopes (like pico scopes exactly) from 250 up to 350 USD, included of course their nice software which of cost it is from 50 up to 200 USD. When i asked sales manager for their purposes and because i have paid the software before 1,5 year, he, very sincerelly replied to me that for FFT analysis it is better the use of a sound card with its 24bit depth and its sampling rate of 192KHz, instead the 8bit/150MHz PC based oscilloscopes! This was an amazing proof of honesty (Virtins technology, it is a new company found in Singapore) that i have seen few times in my 30 years career.
Summarizing, i think that a PC based oscilloscope does not offers nothing except portability. OTOH, there are stand alone DSOs very light operating as well with batteries, if this is the problem. Their connectivity to PCs the only that offers in reality, it is the storage of plots etc. A bench top DSO it is prefferable i think. Now, for FFT analysis as i told (not only me, but many others as well) a good external USB 2.0 sound card (Conrad Hoffman said that he uses the most economic EMU0202 of 100 Euro with success) with ITS OWN power supply it is the better solution.
Before 1,5 year i remember that i was confused and i did uselless expences from my ignorance. Now, not!

Regs
Fotios