13 Feb
2008
13 Feb
'08
3:12 a.m.
I was wondering what the current status is of using multiple Delta 1010s
with jackd. I have seen several references to an .asoundrc for this
kind of a setup, but I have not seen anyone say how this is working for
them. I currently have a single 1010 installed in a 3 GHz P4. While
running at 5 msec latency it is running at about 6% CPU utilization with
jack and Ardour2 running. How much does a second card add to the CPU
load? Also, can xruns be avoided under these circumstances?
I'm planning on using the machine for auxiliary processing during live
performance. I know that RME cards would be the best option, but I
really can't afford the card and the A/Ds.
BTW: I had a significant amount of problems eliminating xruns on the
machine. I started by moving from KDE to IceWM, but that did not
improve things. After that I went from a normal kernel to the 2.6.22
real time kernel in the Gentoo pro audio overlay. That also did not
seem to help very much. Even at 100 msec, I was still getting lots of
xruns. I then turned off hyperthreading and added rtirq. These changes
helped, but I could still generate xruns by just typing in a terminal
window or moving the mouse. The thing that actually solved the problem
was changing from the ATI Mach 64 on the motherboard to a NVIDIA 5200.
That seemed to be the magic. I can now run Ardour with no xruns at 5
msec latency.
--
Terry Hull - President
Network Resource Group, Inc.
13 Feb
13 Feb
7:48 a.m.
On Wed, 13 Feb 2008 02:12:37 -0600
"Terry Hull" <tah(a)nrg-inc.com> wrote:
I was wondering what the current status is of using
multiple Delta 1010s
with jackd. I have seen several references to an .asoundrc for this
kind of a setup, but I have not seen anyone say how this is working for
them. I currently have a single 1010 installed in a 3 GHz P4.
I have a second 1010 on the way to me now. If it's repairable (I think
it's just a cap replacement, which I've done successfully before), I'll be
doing a big writeup on running 2 1010s in a similar system. Stay tuned.
--
======================================================================
Joe Hartley - UNIX/network Consultant - jh(a)brainiac.com
Without deviation from the norm, "progress" is not possible. - FZappa
12:19 p.m.
On 13 February 2008 at 7:47, Joe Hartley <jh(a)brainiac.com> wrote:
I have a second 1010 on the way to me now. If
it's repairable (I think
it's just a cap replacement, which I've done successfully before), I'll be
doing a big writeup on running 2 1010s in a similar system. Stay tuned.
Could you share with us the symptom and specifics of the solution
involving replacing a cap? I'm wondering if the noisy channels
problem mentioned here a couple of times is related somehow.
Thanks....
--
Kevin
12:46 p.m.
On Wed, 13 Feb 2008 09:19:04 -0800
Kevin Cosgrove <kevinc(a)cosgroves.us> wrote:
On 13 February 2008 at 7:47, Joe Hartley
<jh(a)brainiac.com> wrote:
Heh - I just Googled to see if I could find a good link to describe
the process and the first link I come up with is a post of mine to the
Planet CCRMA list about the problem!
My symptom was a horrendous buzz on all 8 output channels. When I took
apart the breakout box, I saw a quadrupling circuit on the left hand side
of the bottom board (toward the front of the case, underneath the small
board that's on top) that had a couple of oozing caps.
I replaced them with 2200uF 35v caps from DigiKey:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=493-1879-ND
If you've ever done electronic soldering before, it's a trivial procedure:
remove the 2 small 2200uF 25v caps and squeeze the larger caps in there.
The 35v caps are recommended, it was the underspec'ing of the caps that was
causing the problems in the first place.
--
======================================================================
Joe Hartley - UNIX/network Consultant - jh(a)brainiac.com
Without deviation from the norm, "progress" is not possible. - FZappa
I have a second 1010 on the way to me now. If
it's repairable (I think
it's just a cap replacement, which I've done successfully before), I'll be
doing a big writeup on running 2 1010s in a similar system. Stay tuned.
Could you share with us the symptom and specifics of the solution
involving replacing a cap? I'm wondering if the noisy channels
problem mentioned here a couple of times is related somehow.
1:03 p.m.
On 13 February 2008 at 12:45, Joe Hartley <jh(a)brainiac.com> wrote:
kc> Could you share with us the symptom and specifics of the solution
kc> involving replacing a cap? I'm wondering if the noisy channels
kc> problem mentioned here a couple of times is related somehow.
Heh - I just Googled to see if I could find a good link to describe
the process and the first link I come up with is a post of mine to the
Planet CCRMA list about the problem!
My symptom was a horrendous buzz on all 8 output channels.
One unit I had started having troubles with buzz on the output
channels a few months after I bought it. At first power cycling the
unit would get it to behave, and at first it was only a few channels.
Eventually it was all channels, and I couldn't get it to behave.
I returned the unit under warranty and got a replacement.
When I took
apart the breakout box, I saw a quadrupling circuit on the left hand side
of the bottom board (toward the front of the case, underneath the small
board that's on top) that had a couple of oozing caps.
That happens more often in electronic equipment than the public generally knows.
Remember the laptop battery recall from months ago? I've seen quite
a few recalls for capacitor problems over the last 3 decades of
working in the electronics industry, first as a part-time soldering
and assembly person, and now as an analog IC designer.
I replaced them with 2200uF 35v caps from DigiKey:
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail?name=493-1879-ND
If you've ever done electronic soldering before, it's a trivial procedure:
remove the 2 small 2200uF 25v caps and squeeze the larger caps in there.
Doesn't sound too bad.
The 35v caps are recommended, it was the
underspec'ing of the caps that was
causing the problems in the first place.
Did you measure the voltage somehow, or did you get documentation
from M-Audio somehow to figure out the new recommended voltage rating?
Excellent write-up. Thanks a bunch!
--
Kevin
1:19 p.m.
On Wed, 13 Feb 2008 10:03:25 -0800
Kevin Cosgrove <kevinc(a)cosgroves.us> wrote:
There was a brand of motherboards that famously had capacitor issues
a couple of years ago as well. I'm glad that having worked throughout
my high school years as an electronic assembler, I know which end of the
soldering iron to hold. I've been able to revive a lot of "dead" gear
over the years!
I remembered a discussion on the Planet CCRMA list about the caps when I had
my problem, and there was a great post from someone who knew more theory
than I did about why 25v was too small for the circuit, and that 35v was
the way to go.
When I took
apart the breakout box, I saw a quadrupling circuit on the left hand side
of the bottom board (toward the front of the case, underneath the small
board that's on top) that had a couple of oozing caps.
That happens more often in electronic equipment than the public generally knows.
Remember the laptop battery recall from months ago? I've seen quite
a few recalls for capacitor problems over the last 3 decades of
working in the electronics industry, first as a part-time soldering
and assembly person, and now as an analog IC designer. The 35v caps
are recommended, it was the underspec'ing of the caps that was
causing the problems in the first place.
Did you measure the voltage somehow, or did you get documentation
from M-Audio somehow to figure out the new recommended voltage rating? Excellent write-up. Thanks a bunch!
No problem; we all learn something when we post our successes here.
--
======================================================================
Joe Hartley - UNIX/network Consultant - jh(a)brainiac.com
129 Petta. Lake Rd, Saunderstown, RI 02874 - vox 401.338.9214
Without deviation from the norm, "progress" is not possible. - FZappa
2:54 p.m.
On Wednesday 13 February 2008 19:18, Joe Hartley wrote:
On Wed, 13 Feb 2008 10:03:25 -0800
Kevin Cosgrove <kevinc(a)cosgroves.us> wrote:
There was a brand of motherboards that famously had capacitor issues
a couple of years ago as well. I'm glad that having worked throughout
my high school years as an electronic assembler, I know which end of the
soldering iron to hold. I've been able to revive a lot of "dead" gear
over the years!
I remembered a discussion on the Planet CCRMA list about the caps when I
had my problem, and there was a great post from someone who knew more
theory than I did about why 25v was too small for the circuit, and that 35v
was the way to go.
IIRC the caps problem on some mobo's was due to someone having stolen the
formula for the electrolyte from someone or other, but the formula was not
complete, and the caps that were manufactured were obviously faulty. They
ended up being used in various mobo's, and consequently failed.
This takes me back to the 60's. We had an Alba TV, a console set. like a piece
of furniture, and our first TV. I seem to remember, and being very young at
the time, the first program I saw was horse racing. Not too interesting for a
youngster, to say the least. Anyway this TV worked ok at first, and there was
only one TV channel available in Jersey CI at the time, which was BBC1. Then
came Channel Television, a local commercial tv station, ads , and all. Now
you had to change channels, and the Alba had a turret tuner, a rotary switch
that had the coils tuned to the various tv frequencies slotted in around it.
Switching tv channels using physical contacts didn't seem to be a good idea,
and there were always problems trying to get some sort of reception, and
having to waggle the turret tuner back and forth to get to see TV became
unbearable.
Back to the caps thing, we decided (that is my dad) to buy a new TV, so down
we went to a local store (Fortuna), and Brian the owner explained to us the
virtues of the new polyester caps that Sobell used in their TV's, and if the
TV went down the tubes we could take it back and wrap it around his neck.
Well we bought the Sobell TV, and it worked ok. Only black and white, but
that's all that there was at the time, and we didn't have to take it back and
wrap it around his neck because of failed caps.
Sorry for this being totally OT, but just some stuff from my past.
Nigel.
When I
took
apart the breakout box, I saw a quadrupling circuit on the left hand
side of the bottom board (toward the front of the case, underneath the
small board that's on top) that had a couple of oozing caps.
That happens more often in electronic equipment than the public generally
knows. Remember the laptop battery recall from months ago? I've seen
quite a few recalls for capacitor problems over the last 3 decades of
working in the electronics industry, first as a part-time soldering
and assembly person, and now as an analog IC designer. The 35v
caps are recommended, it was the underspec'ing of the caps that
was causing the problems in the first place.
Did you measure the voltage somehow, or did you get documentation
from M-Audio somehow to figure out the new recommended voltage rating? Excellent write-up. Thanks a bunch!
No problem; we all learn something when we post our successes here. 
6:32 p.m.
On Wednesday 13 February 2008, Joe Hartley wrote:
There was a brand of motherboards that famously had
capacitor issues
a couple of years ago as well.
Please elaborate. My PC has started to exhibit random lock-ups recently
(shortly after starting X), and although there's no evidence of dodgy caps,
there is significant folklore suggesting my eagerly awaited power supply
replacement may not fix the problem.
I have an MSI K8T-Neo FIS2R from 2004.
--
David Haggett
6197
days inactive
6197
days old
linux-audio-user@lists.linuxaudio.org
7 comments
5 participants
participants (5)
-
David Haggett -
Joe Hartley -
Kevin Cosgrove -
Nigel Henry -
Terry Hull