Soundcard performance - realistic?

Hi,

I recently splurged for a hi-end soundcard (RME Fireface USB). Not cheap but I insisted on 192k sampling AND USB interface.

Downloaded the Virtins measurenments software, and after a lot of problems got the software talk to the hardware in MS XP.

To my surprise I got the loop-bacjk results sown in the attachements. To say it looks nice is an understatement - it is extremely good, so good I am doubting these results. Even at 100mV it is still incredible good.
Has anyone reading this seen similar results for this or any other soundcard?

jan
 

Attachments

  • RME1Vloop.pdf
    40.5 KB · Views: 118
  • RME100mVloop.pdf
    45.6 KB · Views: 78
Also the top price... ;)

I did some more testing with my RME and it looks like there is an issue with the ASIO driver.
RME has a test app called Digicheck which is like a background level meter on steroids. Digicheck sees my input in 'hardware level' mode but not in 'asio mode'.
I'm trying to get something out of the RME users forum.

jan
 
Jan, which OS are you using? For me, it was a lot easier to get my soundcard to do what I want in 7 than XP or Vista, but "experts" assure me that XP really is optimum IF things are configured correctly. Of course, one apparently needs to be an expert to get XP configured correctly... :D When I was forced to switch out my lab computer to my old XP box (I hate you, HP!), I could never get ASIO to work properly.
 
Also the top price... ;)

I did some more testing with my RME and it looks like there is an issue with the ASIO driver.
RME has a test app called Digicheck which is like a background level meter on steroids. Digicheck sees my input in 'hardware level' mode but not in 'asio mode'.
I'm trying to get something out of the RME users forum.

jan

Matthias (of RME) ...is one of the best ...;)
I have one Fireface 800 and babe gets her duty very well.
The Hadware Level get levels directly from the internal DSP on the Fireface.
Try to install a old ( < 5.51 ) Digicheck and look if still fail.
The major functions of Digicheck get info from the internal DSP, this is the best way to check any audio functions.
 
Jan, which OS are you using? For me, it was a lot easier to get my soundcard to do what I want in 7 than XP or Vista, but "experts" assure me that XP really is optimum IF things are configured correctly. Of course, one apparently needs to be an expert to get XP configured correctly... :D When I was forced to switch out my lab computer to my old XP box (I hate you, HP!), I could never get ASIO to work properly.

Hi SY :)

i think that the numeric comparision is the mother of all test.

I did... and when i have compared individually (bit @ bit with C++) the audio stream this is my results :

(Same audio source for all the S.O. (rip from commercial CD) )

incoming data stream -> (XP) -> outcoming data stream
incoming data stream = 100 % @ outcoming data stream

incoming data stream -> (Win7) -> outcoming data stream
incoming data stream = 100 % @ outcoming data stream

these are the mathematical results.
For example, if the number 1258 passes through the engine of Windows XP and its output is still the number 1258...all the differences you hear are ...personal :)

EDIT
I did the test with external loopback, but here the result are biased because the audio stream passes from the internal DSP on sound card, so it's a inefficient test.

PS
since i have wrote the real-time (sound card input) FFT to steal data...and...just for fun ... http://www.bonisaudio.com/Bonis_Free/ago1.zip
 
Last edited:
Jan, which OS are you using? For me, it was a lot easier to get my soundcard to do what I want in 7 than XP or Vista, but "experts" assure me that XP really is optimum IF things are configured correctly. Of course, one apparently needs to be an expert to get XP configured correctly... :D When I was forced to switch out my lab computer to my old XP box (I hate you, HP!), I could never get ASIO to work properly.

WIN XP SP3 on one of those mini-boxes. I suspect indeed that there's an ASIO issue, but RME installs it's own ASIO drivers so that *should* be correct.

/OT I organised for tonight an AES evening about industrial IP with a patent lawyer who specialises in audio, is also an audio-nut. Should be interesting.

jan
 
Yeah, I don't think that, properly set up, there's going to be any sonic or measurement differences, the key is being able to get XP to give me that "1258" instead of something else that it *thinks* I want. :D What's probably simple and intuitive to you is totally arcane to me! 7 (and to a lesser extent Vista) made it easy for me to get the 1258 to come out, rather than MXXLVIII.
 
Yeah, I don't think that, properly set up, there's going to be any sonic or measurement differences, the key is being able to get XP to give me that "1258" instead of something else that it *thinks* I want. :D What's probably simple and intuitive to you is totally arcane to me! 7 (and to a lesser extent Vista) made it easy for me to get the 1258 to come out, rather than MXXLVIII.

:D:D:D
 

Bksabath

Banned
2007-12-01 11:21 pm
Hi Jan that sound card looks really good
The best I have seen around
I have the Audiophile 192 PCI and no mather what 50 Hz noise floor is at around -80 dB and the res barely goes to -100


Price is out my budget for the use I have for it (mostly building circuits out of Linear Audio) but as I said I have never seen one so good and I have been looking for a while
 
Well I found out (nudget by jcx) that I was not really testing the hardware but only the software.

In fact, I've been fighting with it for several days now, and talking to one of the designers, to get it work with Virtins or ARTA.
Just today I finally had the RME ASIO test tool working, but the measurement software still doesn't see it...

jan