Subject: Re: [linux-audio-user] Delta 66/new mobo question
From: Mark Knecht (mknecht_AT_controlnet.com)
Date: Thu Sep 02 2004 - 21:05:29 EEST
Dave Phillips wrote:
> Mark Knecht wrote:
>
>> Mark Knecht wrote:
>> <SNIP>
>>
>>> Well, obviously an 800MHz machine will at first glance seem
>>> underpowered compared to all the newer stuff out there, but my
>>> thought is that for just recording/playback you'll probably be fine
>>> if you pay close attention to the rest of your hardware.
>>
>>
>>
>> I meant to include the comment that with an 800MHz machine you should
>> not plan on using many, if any, plugins.
>
>
> Ja, I kinda figured that... :(
>
> I can get a good deal on something not-state-of-the-art that would still
> be much faster than my current box, so it's still a consideration.
> Thanks especially for your recommendation re: hard-disk, I'll follow up
> on that too. Btw, is there support for that disk in the 2.4 kernel
> series ? I'm not planning a kernel upgrade at this time and would rather
> avoid it right now.
>
> Best,
>
> dp
>
1394 hard drives work pretty well for me under both the 2.4 Planet
kernels and 2.6 Gentoo kernel. I have more trouble with CDRW/DVD drives
which do not work well under either in my experience. This is probably
not an issue for you. I do 1394 chips and software for a living so I try
a lot of this out on my company's nickle.
1394 performance under Linux is not what it should/could be. The Linux
1394 stack doesn't optimize gap count automatically so throughput is
slow. (Maybe 5-11MB/S?) I think there may be some little stand alone
apps that will allow you to set the gap count by hand which would help.
Humm...I just remembered that I had my main Pro Tools 1394 audio drive
in my laptop bag so I plugged it in and ran hdparm to get some results.
Better than I thought:
root_AT_flash ~ # hdparm -tT /dev/sda1
/dev/sda1:
Timing buffer-cache reads: 1640 MB in 2.00 seconds = 818.49 MB/sec
Timing buffered disk reads: 56 MB in 3.15 seconds = 17.75 MB/sec
Speeds are higher under Windows, but this isn't bad at all.
root_AT_flash ~ # uname -a
Linux flash 2.6.8-gentoo-r2 #3 Fri Aug 27 11:05:03 PDT 2004 i686 Mobile
Intel(R) Pentium(R) 4 CPU 3.06GHz GenuineIntel GNU/Linux
root_AT_flash ~ #
I don't have a 2.4 series kernel here to test with. My laptop's chipset
doesn't work well under 2.4. Too new and many important things not
supported.
I don't remember how to extract the actual IDE drive parameters out of
this 1394 drive case. I think the drive is a 7200RPM but it could be
5400. I don't remember and I'm sure anything you bought today would be
faster. I've used this one for 18-24 months at least. The drive is set
up with larger than normal (32K) cluster sizes and is a FAT drive so
that I can use with both Pro Tools and other OS's as required. (Heck - I
just did!) ;-)
From dmesg:
<SNIP>
ieee1394: Host added: ID:BUS[0-00:1023] GUID[413f0200b723013d]
ip1394: $Rev: 1224 $ Ben Collins <bcollins_AT_debian.org>
ip1394: eth1: IEEE-1394 IPv4 over 1394 Ethernet (fw-host0)
eth0: link up, 10Mbps, half-duplex, lpa 0x0000
ieee1394: Node added: ID:BUS[0-01:1023] GUID[0001d20000030fc1]
ieee1394: The root node is not cycle master capable; selecting a new
root node and resetting...
ieee1394: Node changed: 0-01:1023 -> 0-00:1023
ieee1394: Node changed: 0-00:1023 -> 0-01:1023
sbp2: $Rev: 1219 $ Ben Collins <bcollins_AT_debian.org>
scsi0 : SCSI emulation for IEEE-1394 SBP-2 Devices
ieee1394: sbp2: Logged into SBP-2 device
ieee1394: Node 0-00:1023: Max speed [S400] - Max payload [2048]
Vendor: WDC WD40 Model: 0BB-32CXA0 Rev:
Type: Direct-Access ANSI SCSI revision: 06
Attached scsi generic sg0 at scsi0, channel 0, id 0, lun 0, type 0
SCSI device sda: 78165360 512-byte hdwr sectors (40021 MB)
sda: asking for cache data failed
sda: assuming drive cache: write through
/dev/scsi/host0/bus0/target0/lun0: p1
Attached scsi disk sda at scsi0, channel 0, id 0, lun 0
root_AT_flash ~ #
<SNIP>
HTH,
Mark
This archive was generated by hypermail 2b28 : Thu Sep 02 2004 - 21:08:05 EEST