Latency on emu10k1 sound?

Matt Dawson matt at mattsnetwork.co.uk
Tue Sep 27 08:52:52 PDT 2005


Hi all,
Anyone else noticed severe latency on the Soundblaster Live! (5.1 player)? 
Playback of anything, including timidity and MP3 seems stuttery and hesitant. 
It's hard to put into words, but it's similar to the effect you would get 
when playing back a stream and the input can't keep up with playback. I've 
removed all the debugging entries from my kernel config, but nothing seems to 
help.

System:
Gigabyte GA-K8VT800 Pro socket 754, Via KT800 chipset, onboard sound disabled.
AMD Athlon 64 Newcastle 3000+ CPU
1GB DDR 400
Radeon 9200SE AGP 8X
SB Live! Player 5.1

dmesg (relevant bits):
pcm0: <Creative EMU10K1> port 0x9000-0x901f irq 18 at device 10.0 on pci0
pcm0: <eMicro EM28028 AC97 Codec>

cat /dev/sndstat:
FreeBSD Audio Driver (newpcm)
Installed devices:
pcm0: <Creative EMU10K1> at io 0x9000 irq 18  (4p/3r/4v channels duplex 
default)

uname -a:
FreeBSD workstation1 6.0-BETA5 FreeBSD 6.0-BETA5 #2: Tue Sep 20 00:18:46 BST 
2005     md001 at workstation1:/usr/obj/usr/src/sys/WORKSTATION1  i386

sound section of my kernel config:
device          sound           # FreeBSD newpcm driver
device          "snd_emu10k1"   # Creative EMU10K1 processor

pciconf -lv:
pcm0 at pci0:10:0: class=0x040100 card=0x80641102 chip=0x00021102 rev=0x07 
hdr=0x00
    vendor   = 'Creative Labs'
    device   = 'EMU10000 Sound Blaster Live! (Also Live! 5.1) - OEM from DELL 
- CT4780'

*Note* No, it isn't a Dell OEM. It was purchased a couple of years ago from a 
local PC box-shifter. I don't buy prefab peecees.

    class    = multimedia
    subclass = audio
emujoy0 at pci0:10:1:      class=0x098000 card=0x00201102 chip=0x70021102 
rev=0x07 hdr=0x00
    vendor   = 'Creative Labs'
    device   = 'EMU10000 Game Port'
    class    = input device

Using the 4BSD scheduler. Although this is an amd64 machine, I'm running i386 
as some ports I need do not work on amd64. The card itself is on an IRQ of 
its own in a slot which does not share a PCI interrupt with anything else. It 
worked fine on 5.4-RELEASE.

I'm simply trying to decide whether to file a PR or not and suggestions on 
gathering relevant information to do so would be appreciated.
-- 
Matt Dawson.

matt at mattsnetwork.co.uk
MD2657-RIPE OpenNIC M_D9


More information about the freebsd-current mailing list