Hi.
I have a FW410 with Cubase SX2 on 2ghz DP G5.
does anybody has similar setup?
If you do have similar setup, what's your latency?
I put mine on 256sample and I get
input latency: 12.698ms
output latency: 14.014ms
but it takes up some CPU so
while I mix, I put it on 512sample or 1024sample
and I get latency like 18ms~30ms.
is it normal?
I had layla24 with PC and when I set to 512 sample I've got like
13ms.
M-audio FW410 users with SX2.
-
- KVRist
- 90 posts since 20 Jul, 2003
I don't know if this is normal or not but i'm a bit scared because i expect my FW Audiophile to arrive and i don't want any latency problems. With my SBLive i work on 5.33 ms @ 44.1 (kx drivers) and works just fine, i dont want bigger latency on m-audio.. i donno how this can be possible 
Last edited by moldofunker on Fri Jul 30, 2004 10:52 am, edited 1 time in total.
-
twolegstoneworks twolegstoneworks https://www.kvraudio.com/forum/memberlist.php?mode=viewprofile&u=7585 - KVRian
- 1426 posts since 12 Jun, 2003 from Denmark
I'm getting 8ms latency with the FW410 and a Dell Inspiron 8600. 8ms is not much and there's absolutely no glitches whatsoever.
-
- KVRAF
- 6937 posts since 4 Jun, 2004 from Utrecht, Holland
Assuming the samplerate is 44.1kHz, calculation shows with a 512 byte buffer the minimal latency to expect is 512 / 44.1 = 11.6 ms.
I read somewhere else someone with a FW410 used a buffersize of 64 bytes in CuBase and experienced a latency of 8ms. Theoretical minimum is then 64 / 44.1 = 1.45 ms, so 6 to 7ms are lost somewhere on the path...
If you get 13ms latency with a 256 byte buffer the loss is in the same range (around 7ms) so I wouldn't worry too much. Although 30ms is kinda high, it's perfectly explainable: also 7ms away from the 23ms you'd expect with a 1024 byte buffersize.
I read somewhere else someone with a FW410 used a buffersize of 64 bytes in CuBase and experienced a latency of 8ms. Theoretical minimum is then 64 / 44.1 = 1.45 ms, so 6 to 7ms are lost somewhere on the path...
If you get 13ms latency with a 256 byte buffer the loss is in the same range (around 7ms) so I wouldn't worry too much. Although 30ms is kinda high, it's perfectly explainable: also 7ms away from the 23ms you'd expect with a 1024 byte buffersize.
