The product was pretty mature.
One factoid that colors this market is not that PCI-E is going away forever but that most people (including myself) are pretty well taken with the convenience of plugging a USB card into a laptop.
Another situation is that not a lot new of substance has happened with A-D converter chips since 2008 (e.g. PCM 4222).
Jan Didden brought over a Scarlett 2i2 USB which seemed to work very well, albeit without the 192kHz sample rate that I'd prefer.
I bought one based on this recommendation, and now that I have it, I can't in good conscience recommend it to anybody else.
Basically, its only a 16/96 device with the corresponding -93 dB SNR limitation you'd expect.
Focusrite claims that in ASIO mode it has 24 bit performance, but its best SNR and DR are still limited to about 93 dB, not the > 100 dB you'd reasonably expect from a true 24 bit device.
Good enough for speaker measurements, but that is about it.
In most of the non-ASIO modes that are used by most Windows analytical and editing programs, it does not even try to expose a 24 bit interface. The lists of supported formats are limited to 16 bits.
These are well-known limitations, documented on the Focusrite web site, and eBay is covered with people trying to sell their Scarlett 2i2 off.
Some are available on the web: http://html.midifan.com/review/2i2/[MME]%20Scarlett%202i2%20USB%202448.htm
Sure here are Rightmark .sav files, one from a 24/96 test and another from a 24/44 test:
View attachment SCARLETT SAV files.zip
Last edited:
It works now, thx 
Strange, it works here (the whole link, you can't just click on it, tags are breaking it). Anyway, if you type "Scarlett 2i2 rmaa" in google, it should be the first link.
Strange, it works here (the whole link, you can't just click on it, tags are breaking it). Anyway, if you type "Scarlett 2i2 rmaa" in google, it should be the first link.
Thanks Arny, I'm beginning to think USB audio is going the way of everything else. My Edirol UA-5 which died easily did better than many units
I have auditioned, it worked at real 24/48 in full duplex with two instances of an application (one in and one out) with no glitches. Folks like the Audacity team don't help by continuously claiming you don't need this so they don't even flag a warning. I wasted 2 or 3 hours of my time figuring this out.
Under XP my Maudio 0204 does record 24/96 BTW, these things are trivial to verify.
I've found that simply using a 24/96 field recorder offline eliminates all this OS/application nonsense.
Last edited:
As far as I can tell, the EMU0404 remains king of the hill:
With a low distortion 1kHz sine generator, distortion gets down to .00014%.
I also have a Fuckusrite 2i4, which does something like this:
Perhaps the 2i2 is better, but for me this was quite a disappointment, because it was produced after the EMU0404.
Recently I also bought an Asus Xonar U7:
Which isn't too bad. I think the opamps may be responsible for the upward sloping distortion, so I am going to replace the mysterious opamps marked 43K5 2A3 with OPA 1642. Opened a thread just today on this, but it will be a couple of weeks before I can report on any possible results.
May I suggest the following: most will have ARTA, or know how to download it. It would be very enlightening if folks with other soundcards could make similar measurements of theirs in loop-back, 96kHz 24 bits, so that we get some comparable data.
With a low distortion 1kHz sine generator, distortion gets down to .00014%.
I also have a Fuckusrite 2i4, which does something like this:
Perhaps the 2i2 is better, but for me this was quite a disappointment, because it was produced after the EMU0404.
Recently I also bought an Asus Xonar U7:
Which isn't too bad. I think the opamps may be responsible for the upward sloping distortion, so I am going to replace the mysterious opamps marked 43K5 2A3 with OPA 1642. Opened a thread just today on this, but it will be a couple of weeks before I can report on any possible results.
May I suggest the following: most will have ARTA, or know how to download it. It would be very enlightening if folks with other soundcards could make similar measurements of theirs in loop-back, 96kHz 24 bits, so that we get some comparable data.
Last edited:
Yes, emu0404 is the king of the hill but it's discontinued.
There are quite big differences in between the rmaa results in the link I posted and arnyk's, especially thd+n and imd graphs. The other measurements posted here agree more with midifan than arnyk so I wonder if his unit is very bad or if there is a setup problem ?
There are quite big differences in between the rmaa results in the link I posted and arnyk's, especially thd+n and imd graphs. The other measurements posted here agree more with midifan than arnyk so I wonder if his unit is very bad or if there is a setup problem ?
Rightmark test results here: ProSound: Ïðîôåññèîíàëüíûé çâóê
Frequency response (in the range of 40 Hz - 15 kHz), dB
+0.08, -0.70
The noise level in dB (A)
-92.7
Dynamic range, dB (A)
92.7
Harmonic distortion%
0.011
Harmonic distortions + noise, dB (A)
-76.2
Intermodulation Distortion + Noise,%
0.015
The saeparation of channels, dB
-63.9
Intermodulation at 10 kHz,%
0.034
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.