Alesis HD24
Jim Gilliland
My HD24 arrived yesterday, and I've been putting it through its paces.
So far I'm delighted with it. I'm going to point out several of its
weaknesses in this note, but don't take that as too much criticism - so
far, this is a great little unit.
I've done all the basic stuff with it. I had a recording on a DA38 tape
that I needed to get into the DAW, so I used the HD24 as the
go-between. Everything went smoothly.
Everything I've done so far has come in via the digital inputs, so I
can't comment on the quality of the converters.
I took the bit of money that I saved by buying this from Musician's
Advocate and invested in a 60GB IDE drive. That adds about 5 hours of
24 track recording time.
I don't know what to make of the guy who was complaining about caddy
availability. At some point, I'm sure that I'll want to have an extra
caddy or two, but it's clear that I could get buy with just these two
for a long time to come. It's unbelieveably easy to install drives into
these caddies. You just slide the top off, connect the power and data
cables, install a couple of screws to hold the drive in place, slide the
top back one. If you're really in a hurry, you could even operate it
without the screws and/or without the top. So it'll be no problem to
get along with two caddies.
After I finished my recording, I transferred it to the PC via FTP. It's
slow - even slower than I expected. The HD24 has only a 10Mbit ethernet
port, and I can see why they didn't invest in a faster one - the HD24
can't even keep this one full. It runs at about 5Mbits/sec. So
transfer over the LAN takes a while. But it still has the big advantage
of being an unattended process. If it's a long session, just start up
the transfer at the end of the day and it'll be done by morning. For
shorter sessions, you can transfer it over lunch.
Just to compare, 24 bit recording in real time transfers at about 1Mbit
per second per track. So if you're transferring fewer than five tracks,
FTP will be faster than real-time. But if you're transferring a full 24
track recording, it'll take about five times longer to transfer via FTP
it than it did to record it.
I can already see a couple of things that I'm hoping Alesis will address
in a software update someday. For one thing, this machine is ONLY a 24
bit recorder. I think it would be useful to have a 16 bit option. But
disk space is cheap enough that this isn't a big deal.
A bigger deal, though, is the fact that the HD24 can only be set to
record entirely from its digital inputs or entirely from its analog
inputs. It would be VERY useful to be able to, say, set tracks 1-16 to
come from digital sources, and use the analog inputs for tracks 17-24.
You can't do that with the HD24 at present.
I'm going to try to hook it up in a closed-loop MMC configuration today
and see how that works. If all goes well, I'll use it for its first
real show tomorrow night.
I've taken the liberty of setting up a discussion group for the HD24 on
Yahoo Groups. It's simply called HD24. I would encourage new HD24
owners to join in - we'll definitely need a place to discuss this unit.
http://groups.yahoo.com/group/hd24
jsp...@smumn.edu
>slow - even slower than I expected. The HD24 has only a 10Mbit ethernet
>port, and I can see why they didn't invest in a faster one - the HD24
>can't even keep this one full. It runs at about 5Mbits/sec. So
>transfer over the LAN takes a while.
If that is a 10Mbit full duplex connection than the reality is that you would
see as much as 5Mbit upstream. That would be as good as it gets. I am
surprised that they did not go with a 10/100 interface. I don't think that the
cost difference would be that big of a deal.
jason
Jason Spartz
Multimedia & IT Support
Saint Mary's University
Winona MN 55987
jsp...@smumn.edu
Chris Moylan
drives is limited to a speed for 24 tracks that falls below the 10Mbit
speed limit. Also, most 10Mbit connections run at half-duplex by
default, which will give you the full 10Mbit's on download from the
unit, but as you said the machine can't keep this up.
The algorithm used for the filesystem is still the right way to go for
this type of device. Professional video and audio hard drives have
been using similar systems for years for A/V data, and FAT32/HFS is a
non-ideal storage system for data of this type. The only reason
people are using this in standalone HD recorders is because EIDE
drives are now (almost) fast enough to not require a special drive
architecture/driver model. It is unclear whether or not Alesis will
release the specs for their filesystem so that people can write native
drivers for Linux/Windows/whatever to read the drives on a local
system, but it seems as though this would be a good idea. Perhaps
integrating such a thing into ADAT/Edit would behoove them as well.
I decided to go with one of the Alesis units because the competing
price-level unit from Mackie is literally a PC-in-a-Mackie-box which
is extremely scary to me. I have vast experience with the PC
platform, and it is the last place I want to handle my lovingly
crafted audio tracks. Hopefully the Alesis unit will suffice as an
interim solution until I can afford, and have enough room for a real
24 track solution from Studer/Otari/Sony/MCI/etc. Ahh yes, analog
tape. Right now I track everything 8 tracks at a time on a reel
machine and transfer that to my PC, which will soon be
dismantled/destroyed and replaced with the HD24.
Currently my analog tape machine is the only thing I can count on for
tracking. It is absolutely unacceptable that PC-based recorders have
ridiculous latency issues, crashes, random behavior (I SWEAR I armed
track 4, etc). I can't wait to sink the PC to the bottom of the
Charles River. ;-) Ok venting complete...
Chris
jsp...@smumn.edu wrote in message news:<9stvj...@drn.newsguy.com>...
Jim Gilliland
>
> I suspect that Alesis's proprietary file-streaming driver for IDE
> drives is limited to a speed for 24 tracks that falls below the 10Mbit
> speed limit.
That shouldn't be an issue - the ATA100 IDE interface that is used in
the HD24 is theoretically capable of 100MBytes/sec. In the real world,
it only operates at some fraction of that spec, but the spec implies
data transfers that are _80_ times faster than the 10Mbits/sec ceiling
of the ethernet port. Even if you assume that the drive system operates
at 10% of its capacity (and I submit that 30-40% is more likely), that
would mean that the drive can deliver data 8 times faster than the
ethernet port could accept it.
So the bottleneck is elsewhere.
BTW, ftp transfers one file at a time, so the number of tracks isn't an
issue either.
> Also, most 10Mbit connections run at half-duplex by
> default, which will give you the full 10Mbit's on download from the
> unit, but as you said the machine can't keep this up.
Agreed.
> The algorithm used for the filesystem is still the right way to go for
> this type of device. Professional video and audio hard drives have
> been using similar systems for years for A/V data, and FAT32/HFS is a
> non-ideal storage system for data of this type. The only reason
> people are using this in standalone HD recorders is because EIDE
> drives are now (almost) fast enough to not require a special drive
> architecture/driver model. It is unclear whether or not Alesis will
> release the specs for their filesystem so that people can write native
> drivers for Linux/Windows/whatever to read the drives on a local
> system, but it seems as though this would be a good idea. Perhaps
> integrating such a thing into ADAT/Edit would behoove them as well.
Agreed. A Windows IFS for this file system is something that someone
should put together. Alesis should either do it themselves or release
the spec so that someone else can do it. It appears that they do intend
to release the spec - the FAQ on their website suggests that they expect
others to use this file system over time. They call it ADAT/FST (for
File Streaming Technology).
> I decided to go with one of the Alesis units because the competing
> price-level unit from Mackie is literally a PC-in-a-Mackie-box which
> is extremely scary to me. I have vast experience with the PC
> platform, and it is the last place I want to handle my lovingly
> crafted audio tracks. Hopefully the Alesis unit will suffice as an
> interim solution until I can afford, and have enough room for a real
> 24 track solution from Studer/Otari/Sony/MCI/etc.
If I'm not mistaken, the Alesis is also basically just a PC with
firmware. The Mackie MDR2496 and the Alesis HD24 are really quite
similar in many ways.
Chris Moylan
> Chris Moylan wrote:
> >
> > I suspect that Alesis's proprietary file-streaming driver for IDE
> > drives is limited to a speed for 24 tracks that falls below the 10Mbit
> > speed limit.
>
> That shouldn't be an issue - the ATA100 IDE interface that is used in
> the HD24 is theoretically capable of 100MBytes/sec. In the real world,
> it only operates at some fraction of that spec, but the spec implies
> data transfers that are _80_ times faster than the 10Mbits/sec ceiling
> of the ethernet port. Even if you assume that the drive system operates
> at 10% of its capacity (and I submit that 30-40% is more likely), that
> would mean that the drive can deliver data 8 times faster than the
> ethernet port could accept it.
>
> So the bottleneck is elsewhere.
Ahh, but that doesn't take into account what is involved in getting
the data off the drive into a useable format. As far as I can figure,
the system contains an FTP server hybrid that abstracts the FST file
system into "WAV" and "AIFF" files with certain/exact filenames that
cannot be changed. I believe these "files" don't actually exist and
are just pointers that direct the unit to begin to extract and convert
the FST-interleaved data into WAV or AIFF files for export in
real-time. Pretty neat. However, I believe this is the cause of the
mediocre performance. Perhaps a BIOS upgrade in the future will
improve performance. In my experience so far, it really isn't that
bad, and is certainly convenient for offline editing of individual
tracks. It certainly works perfectly, I had no network issues,
crashes, or weirdness with the Ethernet and FTP capabilities in the
past few days of experimentation.
> BTW, ftp transfers one file at a time, so the number of tracks isn't an
> issue either.
I discovered this as well, although I'm not certain if it can accept
more than one connection at once. I expect not.
...
> > system, but it seems as though this would be a good idea. Perhaps
> > integrating such a thing into ADAT/Edit would behoove them as well.
>
> Agreed. A Windows IFS for this file system is something that someone
> should put together. Alesis should either do it themselves or release
> the spec so that someone else can do it. It appears that they do intend
> to release the spec - the FAQ on their website suggests that they expect
> others to use this file system over time. They call it ADAT/FST (for
> File Streaming Technology).
All in all the sytem and it's FST file layout seems to be very well
designed and operates smoothly. I have only gone up to 16 tracks thus
far, but all my "Songs" were created as 24 track songs, and it
apparently writes 0's in place of unrecorded tracks as evidenced by 24
full size tracks appearing in each FTP folder for 24 track songs. And
this thing is fast on the play/scrub/etc. latency tip, you can zip
around the disk by setting up elaborate locates and it literally seems
instantaneous (compared to my PC...or to analog tape ;-) )
> If I'm not mistaken, the Alesis is also basically just a PC with
> firmware. The Mackie MDR2496 and the Alesis HD24 are really quite
> similar in many ways.
I dunno man, the Mackie, if you look at the pictures of them being
constructed that they have on their web site is literally an ATX
motherboard, even the back of the box has the metal slats like the
back of a PC. the way the back of the Alesis is laid out, I think it
would be impossible for a standard PC to fit. However, there are lots
of mini-board PC products out there, which may also account for the
use of a 10Mbit network port, as many of the single-board PC type
deals have only 10Mbit Ethernet. I'm gonna open it up and take a look
one of these days. The unit itself is also much smaller than the
Mackie, assuming the MDR and HDR are the same size (I have only seen
the HDR in person).
You know what else? The ADAT sync works really well syncing to my
analog tape machine, far better than my previous MTC/word clock combo
through the Echo Layla (sucks). I am using a MOTU MTPAV as the
translation station. As far as features and technical functionality
goes, I couldn't be happier with this box. I haven't really given it
a super-critical listen, but it seems to sound quite nice as well.
Certainly better than older ADAT's. ;-) The LRC remote is quite
cheesy, I'll have to get a better remote later on, there is supposedly
a dedicated remote on the way...
Chris
Jim Gilliland
>
> Jim Gilliland <usemyl...@altavista.net> wrote in message news:<3BF41955...@altavista.net>...
> > Chris Moylan wrote:
> > >
> > > I suspect that Alesis's proprietary file-streaming driver for IDE
> > > drives is limited to a speed for 24 tracks that falls below the 10Mbit
> > > speed limit.
> >
> > That shouldn't be an issue - the ATA100 IDE interface that is used in
> > the HD24 is theoretically capable of 100MBytes/sec. In the real world,
> > it only operates at some fraction of that spec, but the spec implies
> > data transfers that are _80_ times faster than the 10Mbits/sec ceiling
> > of the ethernet port. Even if you assume that the drive system operates
> > at 10% of its capacity (and I submit that 30-40% is more likely), that
> > would mean that the drive can deliver data 8 times faster than the
> > ethernet port could accept it.
> >
> > So the bottleneck is elsewhere.
>
> Ahh, but that doesn't take into account what is involved in getting
> the data off the drive into a useable format. As far as I can figure,
> the system contains an FTP server hybrid that abstracts the FST file
> system into "WAV" and "AIFF" files with certain/exact filenames that
> cannot be changed. I believe these "files" don't actually exist and
> are just pointers that direct the unit to begin to extract and convert
> the FST-interleaved data into WAV or AIFF files for export in
> real-time. Pretty neat.
Yes, that's exactly what's going on in their "FTP Server" - it is making
WAV or AIFF files on the fly out of the data on the disk. Still, that
shouldn't require many cycles. These formats aren't that complex.
> However, I believe this is the cause of the
> mediocre performance. Perhaps a BIOS upgrade in the future will
> improve performance. In my experience so far, it really isn't that
> bad, and is certainly convenient for offline editing of individual
> tracks. It certainly works perfectly, I had no network issues,
> crashes, or weirdness with the Ethernet and FTP capabilities in the
> past few days of experimentation.
That might be the reason, but I really can't see why it would take that
many CPU cycles to create the proper format. There is really very
little in a WAV file other than the raw data. On the other hand, I
can't come up with a better explanation <g>.
You're right that it works well, and I transferred two complete
multitrack concert recordings from the HD24 to my PC in about four hours
today. The recordings were about 60 minutes and 90 minutes in length,
so that transfer time was quite reasonable.
> > BTW, ftp transfers one file at a time, so the number of tracks isn't an
> > issue either.
>
> I discovered this as well, although I'm not certain if it can accept
> more than one connection at once. I expect not.
Apparently not. When I tried to start a second concurrent transfer, it
just sat waiting for the first one to complete.
> ...
> > > system, but it seems as though this would be a good idea. Perhaps
> > > integrating such a thing into ADAT/Edit would behoove them as well.
> >
> > Agreed. A Windows IFS for this file system is something that someone
> > should put together. Alesis should either do it themselves or release
> > the spec so that someone else can do it. It appears that they do intend
> > to release the spec - the FAQ on their website suggests that they expect
> > others to use this file system over time. They call it ADAT/FST (for
> > File Streaming Technology).
>
> All in all the sytem and it's FST file layout seems to be very well
> designed and operates smoothly. I have only gone up to 16 tracks thus
> far, but all my "Songs" were created as 24 track songs, and it
> apparently writes 0's in place of unrecorded tracks as evidenced by 24
> full size tracks appearing in each FTP folder for 24 track songs. And
> this thing is fast on the play/scrub/etc. latency tip, you can zip
> around the disk by setting up elaborate locates and it literally seems
> instantaneous (compared to my PC...or to analog tape ;-) )
I did one twenty-four track recording just to see it work at that data
rate. I'll be doing one for real tomorrow night, and I wanted to see it
work in practice mode first. My 24-track test worked perfectly.
> > If I'm not mistaken, the Alesis is also basically just a PC with
> > firmware. The Mackie MDR2496 and the Alesis HD24 are really quite
> > similar in many ways.
>
> I dunno man, the Mackie, if you look at the pictures of them being
> constructed that they have on their web site is literally an ATX
> motherboard, even the back of the box has the metal slats like the
> back of a PC. the way the back of the Alesis is laid out, I think it
> would be impossible for a standard PC to fit. However, there are lots
> of mini-board PC products out there, which may also account for the
> use of a 10Mbit network port, as many of the single-board PC type
> deals have only 10Mbit Ethernet. I'm gonna open it up and take a look
> one of these days. The unit itself is also much smaller than the
> Mackie, assuming the MDR and HDR are the same size (I have only seen
> the HDR in person).
True. I didn't really think that the HD24 used a PC motherboard, but I
think it uses an Intel CPU and support chips. I haven't opened mine up
either.
> You know what else? The ADAT sync works really well syncing to my
> analog tape machine, far better than my previous MTC/word clock combo
> through the Echo Layla (sucks). I am using a MOTU MTPAV as the
> translation station. As far as features and technical functionality
> goes, I couldn't be happier with this box. I haven't really given it
> a super-critical listen, but it seems to sound quite nice as well.
> Certainly better than older ADAT's. ;-) The LRC remote is quite
> cheesy, I'll have to get a better remote later on, there is supposedly
> a dedicated remote on the way...
I haven't even unwrapped the LRC yet. I did try to get a MIDI-MMC
closed loop configuration to work with my Spirit 328 mixer, but with
limited success. It worked, but it was somewhat quirky. I'll probably
experiment with that again when I have more time. BTW, the clearance on
the MIDI connectors on the HD24 is pretty tight. I found that typical
MIDI cables with molded plastic plugs wouldn't fit. I had to buy a pair
of cables with all metal plugs to get a solid connection.
Chris Moylan
> > I believe these "files" don't actually exist and
> > are just pointers that direct the unit to begin to extract and convert
> > the FST-interleaved data into WAV or AIFF files for export in
> > real-time. Pretty neat.
>
> Yes, that's exactly what's going on in their "FTP Server" - it is making
> WAV or AIFF files on the fly out of the data on the disk. Still, that
> shouldn't require many cycles. These formats aren't that complex.
You are correct, PCM is hardly rocket science, and is most likely very
similar to how the data is encoded on the drive, although all the
channels are interleaved I would expect. How do you feel about the
drive-to-drive copy time though, that seemed a tad on the slow side to
me as well, should we benchmark some of the file operation times on
the HD24 for fun? I would be interested in knowing what kind of wait
I am in for when I am thinking through ways to move data around, so an
FTP/drive-to-drive standard throughput rate might be nice to know. I
wonder if this changes much for faster drives, I am using the drive
that came with the box and some 5400 RPM 40gb Maxtor I had lying
around.
> > However, I believe this is the cause of the
> > mediocre performance. Perhaps a BIOS upgrade in the future will
...
> That might be the reason, but I really can't see why it would take that
> many CPU cycles to create the proper format. There is really very
> little in a WAV file other than the raw data. On the other hand, I
> can't come up with a better explanation <g>.
Maybe I'll ask Alesis, I sure am curious. I'm dying to open this
thing up, but something about having the all-in-one recording solution
makes me think it'll lose some of its magic if I can see how it's
made.
Did you notice that the insides of the drive bays are powder-coated
aluminum? They apparently had to scratch off some of the powder
coating so the little ground contacts on the side of the caddy can
make solid contact. I'll bet the powder-coaters were supposed to
mask off that area, and didn't, so they had to manually sand them
down. I thought it was kinda funny, and probably typical of problems
seen in the manufacture of extremely complex products. They got the
job done, though.
Is the input-normalling awesome or what? I know other units have this
feature, but it is totally perfect for my application, which is
tracking everything on my 8 track reel machine, sync and load, sync
and load, etc.
...
> True. I didn't really think that the HD24 used a PC motherboard, but I
> think it uses an Intel CPU and support chips. I haven't opened mine up
> either.
It may well use Intel processing, as it is extremely powerful and
affordable these days. I am quite confident in Intel's hardware for
the most part, but looking inside that Mackie and seeing a durn ATX
motherboard with about a zillion cards jabbed into it scared me away
pretty quick. The Alesis seemed like a leaner, cleaner solution.
Yeah, no virtual tracks, but that means you are going to hit
hard-drive throughput limits, fragmentation resulting in dropped
audio, etc. when the track number gets up there. The Alesis box is
like a tape machine, it does what it does, does it well, and doesn't
pretend to do everything. I hate products that try to do everying,
being overly complicated and prone to breakdowns (are you listening
Logic Audio???).
...
> I haven't even unwrapped the LRC yet. I did try to get a MIDI-MMC
> closed loop configuration to work with my Spirit 328 mixer, but with
> limited success. It worked, but it was somewhat quirky. I'll probably
> experiment with that again when I have more time. BTW, the clearance on
> the MIDI connectors on the HD24 is pretty tight. I found that typical
> MIDI cables with molded plastic plugs wouldn't fit. I had to buy a pair
> of cables with all metal plugs to get a solid connection.
I have never been able to get MMC to work right, albeit my only
attempt was with Logic and my reel-to-reel recorder through its MMC
interface. The freakin thing would seek all over the place, randomly
not-play, etc. I might give it a whirl some day with the Alesis, but
its not a priority. You really had a problem with the Midi
connectors? I used a couple Hosa molded-junk Midi cables for MTC and
didn't have a problem. We musn't forget there are 48 TRS connectors
back there, so something may end up squished. ;-)
This has been an interesting discussion, hopefully more people will
share their experiences with the HD24 as well, as they become widely
available.
Chris
Mike Rivers
> I am quite confident in Intel's hardware for
> the most part, but looking inside that Mackie and seeing a durn ATX
> motherboard with about a zillion cards jabbed into it scared me away
> pretty quick. The Alesis seemed like a leaner, cleaner solution.
Possibly, but each of those zillions cards (it's not that many) is
replaceable if there's a failure. If Mackie finally got their ribbon
cable problems resolved, there shouldn't be a problem.
Don't be afraid to take a screwdriver to it and look under the hood.
Hopefully you won't let out any smoke. I'd be curious as to how
integrated it is. While you might be scared of "a zillion cards
jabbed into it", I think back to the Alesis 1622 mixer, which was all
on one board and was a horror show when something failed. It was
essentially unrepairable.
> The Alesis box is
> like a tape machine, it does what it does, does it well, and doesn't
> pretend to do everything.
It's important to recognize that.
Have you done any testing yet to determine monitor throughput delay -
analog input to analog monitor output? Or response to punching in and
out?
--
I'm really Mike Rivers (mri...@d-and-d.com)
Jim Gilliland
>
> How do you feel about the
> drive-to-drive copy time though, that seemed a tad on the slow side to
> me as well, should we benchmark some of the file operation times on
> the HD24 for fun? I would be interested in knowing what kind of wait
> I am in for when I am thinking through ways to move data around, so an
> FTP/drive-to-drive standard throughput rate might be nice to know.
I did one drive to drive copy, but I didn't really pay attention to the
performance. I just started it and walked away. It seemed to go fast
enough <g>.
> Did you notice that the insides of the drive bays are powder-coated
> aluminum? They apparently had to scratch off some of the powder
> coating so the little ground contacts on the side of the caddy can
> make solid contact. I'll bet the powder-coaters were supposed to
> mask off that area, and didn't, so they had to manually sand them
> down. I thought it was kinda funny, and probably typical of problems
> seen in the manufacture of extremely complex products. They got the
> job done, though.
No, I didn't notice that. I didn't do anything special with the Western
Digital drive that I added, and it seems to work fine.
> Is the input-normalling awesome or what? I know other units have this
> feature, but it is totally perfect for my application, which is
> tracking everything on my 8 track reel machine, sync and load, sync
> and load, etc.
It is pretty cool, but not that useful to me. I would like to have the
ability to route any input to any channel, but that's probably asking a
bit much. It sure would have come in handy last night - we had four
acts to record, and we had them spread all over the 32 channel snake.
Fortunately, the manager of the event coordinated it well and kept me
well informed, so I was able to get everything tracked that I needed.
Still, I had to swap a lot of XLRs around between sets.
Jim Gilliland
>
> > I'm gonna open it up and take a look one of these days.
>
> think it uses an Intel CPU and support chips. I haven't opened mine up
> either.
Well, I took a look inside and found that I was wrong about the
processor. It's not an Intel at all - the HD24 uses a Motorola ColdFire
CPU. It also uses an Altera ACEX PLD chip and an Analog Devices DSP
chip.
The only connector on the mainboard that isn't already in use is the BDM
connector. That's typically used to troubleshoot a system. So there
aren't any obvious paths for upgrading or adding features to this unit.
The inside is cleanly laid out, with a power supply at the rear right,
and a mainboard (about 5x13 inches) across the back of the bottom
panel. Above that are two boards containing all of the TRS ports, one
for A/D, the other for D/A.
The connections to the drive ports use 40-conductor IDE cables, so it's
clear that they aren't worried about drive performance.
All the connectors across the bottom back panel are integral to the
mainboard - this includes ethernet, wordclock, three Lightpipes in and
out, MIDI and ADAT Sync.
The ethernet is a CrystalLAN chip on the main board. Too bad, it can't
be readily upgraded. (And worse, the latest ColdFire chips from
Motorola include an integrated Ethernet 10/100 port, so Alesis could
have done faster Ethernet for less money if the newer chip had been
available when they designed the HD24.)
The only other components are the meter panel and the control panel at
the front.
Dave
alumnium base? I have now 4 caddies from Alesis and all of them are
all plastic?
Has anyone found a replacement caddy elsewhere? I tried but could not
find one that fits. I hear Alesis should have a stock of caddies come
in this week, anyone hear different?
By the way, I have now 2 sets of IBM drives I am using, 2ea 40 GB and
2 ea. 60 gb 7200 rpm drives and they are working fantastic....
I also have been recording using the A/D on the Alesis and yes, they
blow away the ADAT converters.....
If you guys know more about the caddy issue let me
know.....davidk@eaglecanyon.com
Jim Gilliland <usemyl...@altavista.net> wrote in message news:<3BF803D5...@altavista.net>...