[LAU] RME Multiface and an extra 8 output channels

Ralf Mardorf ralf.mardorf at alice-dsl.net
Thu May 16 08:16:22 UTC 2013


On Wed, 2013-05-15 at 21:04 +0200, Jörn Nettingsmeier wrote:
> iain, are you absolutely sure you need sample rates above 48k? i'd be 
> curious to hear more about your usecase.
> 
> if you really do, e.g. for creating ultrasonics, then the only solution 
> that comes to mind is an rme hdspm plus a pricey converter for 96 or 
> even 192 khz, such as two adi-qs or an m16.

Perhaps the noise floor at higher sample rates could be an issue for the
OP.

Sometimes higher sample rates allow to use a lower latency, for my RME
card, another model, not the MF, it's not like that, if I set it to
higher sample rates, the frames/period automagically becomes higher too.

IMO 48 KHz is high enough, not only for my home studio, also for
professional usage.

FWIW, since my card still is within the warranty period I'm testing if
everything does work on a Windows I installed to do the tests. I noticed
that for sample rates from 32 KHz to 96 KHz nothing connected to the
analog inputs of the card, the noise floor shown by TotalMix is at -110
[1] and for sample rates from 128 KHz to 192 KHz it's at -54. If I
connect an elCheapo no-name device, the noise floor for sample rates
from 32 KHz to 96 KHz does raise to -87.

Even if I would use sample rates higher than 48 KHz, it would be ok for
my taste, but it's not what you want for professional usage.
A Yamaha MT 44D, a 4-Track cassette recorder already has a better SNR
[2].

[1] From the user manual: "(SNR): 113 dB RMS unweighted, 116 dBA @ 44.1
KHz", no information about the SNR for other sample rates.
[2] From the service manual: "Signal-to-Noise Ratio (EIAJ) (Dolby OFF)
55 dB", "(Dolby C ON) 67 dB".

So at least the card I'm using has got a noise floor for sample rates
> 96 KHz, that perhaps isn't ok for the OPs needs.

Perhaps it's not like that for more expensive RME cards.

Regards,
Ralf



More information about the Linux-audio-user mailing list