On Mon, 29 Mar 2010 00:39:46 +0000 Jacob Meuser
<jake...@sdf.lonestar.org> wrote:
> > With this card:
> > FCC ID: IBACT-SB32PNP49
> > Model#: CT3620
> >
> > You patch works perfectly. `aucat -l -m play` runs without a fuss
> > and both `aucat -i track01.wav` and `cdio cdplay` run very well.
>
> excellent.
>
> > As for not having a working system with an ISA bus for testing the
> > ancient stuff, I'll quite happily send you one, but I'm afraid you'd
> > quite begrudgingly receive it. ;)
>
> yeah, I don't really need/want an old i386 ... maybe if ISA DMA
> worked on alpha or the hp* platforms ...
I figured you'd hold a grudge for sending a "gift" like that. ;)
I think the fastest things here with an ISA bus are Pentium II 400MHz
to 600MHz dell boxes. I know we have ISA for alpha, hp300, loongson, and
possibly others, but I've got no clue about the status of DMA on them.
> > I'll start testing the other four SB cards soonish.
>
> if you (or anyone else) have any other working ISA audio devices, it
> would probably be a good time to make similar changes in other
> drivers.
>
I tested everything in my box of ISA audio cards. There are probably
other ISA audio cards around here sitting in systems, but digging the
systems out would be painful, very painful.
All of the following tests were done with your sb(4) patch, as well as
the ISA related patches from oga@ and ariane@ from t...@. The testing
was fairly simple and used a good quality recording:
# aucat -l -m play
# aucat -i track01.wav
It's the same ancient P54C Pentium 133MHz system which I posted the
dmesg for previously. It has has a noisy 50-pin SCSI disk, and the
expected ancient (dirty) power supply. --Perfect for abusive testing.
***********************************************************************
MODEL_: Creative SoundBlaster 32 PnP-ISA
MODEL#: CT3620
FCC_ID: IBACT-SB32PNP49
CHIPS_: CT1745A
CHIPS_: CT1749-DAQ
CHIPS_: CT1971-TDQ
Like many older PnP-ISA devices, it shows up a bit wonky, namely as sb1
rather than sb0. It has an IDE interface (shows as wdc0) and two SIMM
slots (no mem installed).
It works fine.
As expected, `aucat -l` failed but `aucat -l -m play` worked fine.
The default mixerctl output volume is a bit low.
outputs.output=128,128
outputs.master=128,128
Bumping these up to 192 works well enough, although there's the typical
light hiss (with nothing playing) when volume is increased. The light
hiss is more quiet than the hard drive or power supply, so I doubt
anyone will complain. ;)
The IDE interface on this card (wdc(4)) is a pain in the ass and likes
to conflict with anything it can find. IDE interface was not tested.
***********************************************************************
MODEL_: Creative SoundBlaster 16 ISA (jumper config)
MODEL#: CT1750
FCC_ID: IBACT-SB16MCD
CHIPS_: CT1746B-6
CHIPS_: CT1748A
CHIPS_: CT1745A
It shows up as expected as sb0. It has an IDE interface but it is not
detected. This card has a manual volume control on the card itself
(e.g. external wheel).
It works fine.
As expected, `aucat -l` failed but `aucat -l -m play` worked fine.
Again the defualt mixerctl output volume is a bit low, even when the
manual volume wheel is maxed.
outputs.output=128,128
outputs.master=128,128
Bumping these up 192 results in minimal (but expected) light hiss. This
older card actually sounds better than the newer SB32 card above.
IDE interface was not tested.
***********************************************************************
MODEL_: Creative SoundBlaster Vibra 16XV PnP-ISA
MODEL#: CT4170
FCC_ID: ?
CHIPS_: CT2511-SBT
It works fine. Sound quality is much better than the previous two cards
(more headroom).
As expected, `aucat -l` failed but `aucat -l -m play` worked fine.
Again, the default mixerctl output volume is a bit low. Bumbing it up
from 128,128 to 192,192 results a very tiny amount of light hiss.
What's that little thumping noise? -Oh, it's a helicopter outside. ;)
Flase alarm. Sorry. But it actually happened and I thought it was funny.
***********************************************************************
MODEL_: Creative Soundblaster AWE64 PnP-ISA
MODEL#: CT4520
FCC_ID: ?
CHIPS_: CT1972-NAS
CHIPS_: CT8920-NBQ
Whooo! This even has a subwoofer output.
It works fine. Again the sound quality is better than the older cards.
And again, `aucat -l` failed but `aucat -l -m play` worked fine.
Again, the default mixerctl output volume is a bit low. Bumbing it up
from 128,128 to 192,192 results an almost unnoticeable amount of light
hiss.
***********************************************************************
MODEL_: ESS AudioDrive PnP-ISA
MODEL#: ?
FCC_ID: E5X1835 1833/5
CHIPS_: ES1688F VBA2393A
CHIPS_: ES968F U6213
By the chip numbers, this card is not supported by ess(4).
The ESS cards have a soundblaster compatibility mode and can show up as
sb(4) devices. When this card comes up as sb0, it does not work. If I
disable sb(4) in the kernel, the card doesn't show up at all.
***********************************************************************
MODEL_: ADLIB Gold 1000 8-bit ISA
CHIPS_: GOLD CTRL L4A0662
FCC_ID: GKKGD1000
NOTES_: Made In Canada. Circa 1991
No hope.
***********************************************************************
MODEL_: CRYSTAL Audio MJ-021-R00 CD32-3DI PnP-ISA
FCC_ID: M6RCS32A3D
CHIPS_: CS4232-KQ UTAYNP9611
This is a Windows Sound System card, wss(4), and like other PnP cards,
it inconveniently shows up as wss1 rather than wss0.
It works perfect and sounds gorgeous (by comparison to the above).
Unlike the various soundblaster cards above, this card does not give
the expected error when using `aucat -l` namely:
aucat: <default>: can't open device
No error is given, but it doesn't work. If `aucat -l -m play` is used,
the card works extremely well.
The default volume in mixerctl is great (220,220), but mixerctl shows
no "outputs" at all (save "monitor.monitor")?
***********************************************************************
If you want the dmesg for any of the above configurations, or want a
"not_too_painfully_slow" ISA system, let me know.
jcr
