• Registered: 2011-07-20
  • Posts: 5

Topic: Minimum Buffersize: What has an influence

Dear Community,

I have two PCs operating Windows 7 64 Bit. Both are multicore CPUs with the Hammerfall DSP Multiface II installed
involving the same driver version (3.08.5).

I tried to find out the minimum latency achievable for both systems. On system I, I reached down to 32 samples without
any problems. On system II, a buffersize of 64 samples does lead to a completely distorted signal, the minimum size is 128.
Completely distorted means that every buffer seems to be corrupt, not just a single dropout or so. Testing the bad system
with dpclat shows no problem.

What can be the reason for this bad behavior?

Thank you in advance.

Best regards,

Hauke

  • Registered: 2011-07-20
  • Posts: 5

Re: Minimum Buffersize: What has an influence

Hi again,
I solved this issue myself. If I boot the PC with the Hammerfall PCI card with a Multiface connected, the driver seems to be in a specific mode (bad mode) for the Multiface I.
If I then disconnect the MF I and plug a Multiface II to the PCI card, the Multiface II can be operated only down to a buffersize of 128.
If I boot the PC with the Multiface II connected, the driver seems to be in the Multiface II mode (good mode) and the delay is fine down to 32 samples.

By the way, I made some latency measurements and I am impressed. Compared to what I saw with other soundcards, the timing
reached with the Multiface II is really outstanding!!

Best regards,
Hauke

  • From: Germany
  • Registered: 2007-04-10
  • Posts: 9,857

Re: Minimum Buffersize: What has an influence

There is no such thing as a good or bad, MF I or MF II mode for the driver...


Regards
Daniel Fuchs
RME

Regards
Daniel Fuchs
RME

RME Support's Website

  • Registered: 2011-07-20
  • Posts: 5

Re: Minimum Buffersize: What has an influence

So, I wonder why I observed this different behavior then.. But you will definitely know this better than me, it was just my guess :-)