[Mad Scientist Jr[_2_]]

I thought I had a few good USB2 choices for 8 or 16 simultaneous ins
to the computer (my system = Win XP Pro w/Cubase SX3):

* Alesis Multimix 16
* TASCAM US-1641
* Edirol UA-101

However a recent discussion about USB2 vs Firewire (see below) has me
wondering if the firewire route is better. Can anyone confirm this?
Has anyone had any throughput issues with any of the above USB2
interfaces when recording 8-16 tracks simultaneously? Can anyone
recommend a good firewire equivalent of the above, in a comparable
price range?

Thanks...

Newsgroups: rec.video.production, rec.video.desktop
From: "Richard Crowley"
Date: Fri, 5 Sep 2008 10:47:37 -0700
Local: Fri, Sep 5 2008 1:47 pm
Subject: transferring VHS to digital video on computer?

"Mad Scientist Jr" wrote ...

USB2 is 480 mbps, whereas Firewire is only 400 mbps.
How is Firewire faster? Please explain...

Firewire is...
1) Full-duplex. It has an independent and concurrent data path
in EACH direction. This means that continuous stream of video
data can be flowing from the video gadget into the computer
without having to be interrupted to send commands, handshaking
protocol, etc back the other direction.
http://en.wikipedia.org/wiki/Duplex_communication

OTOH, USB has only one data path and it must be interrupted
and turned around to send data, commands, etc. the other way.
http://en.wikipedia.org/wiki/Simplex_communication

2) IEEE1394 (aka "Firewire") was designed from the ground up
to be *isosynchronous*. This means it is capable of an uninterrupted,
sustained stream of data for an indefinite period of time.
http://en.wiktionary.org/wiki/isosynchronous
http://www.allwords.com/word-isosynchronous.html

OTOH, USB is designed for burst-mode signaling which is
quite appropriate for most kinds of data transfer. Except for
high-speed, real-time data such as DV video.

The actual throughput (in 2006) attained with real devices is
about two thirds of the maximum theoretical bulk data transfer
rate of 53.248 MB/s. Typical hi-speed USB devices operate
at lower speeds, often about 3 MB/s overall, sometimes up
to 10-20 MB/s
http://en.wikipedia.org/wiki/Univers...#USB_signaling

[[email protected]]

On Sep 6, 2:33 pm, Mad Scientist Jr
wrote:
I thought I had a few good USB2 choices for 8 or 16 simultaneous ins
to the computer (my system = Win XP Pro w/Cubase SX3):

* Alesis Multimix 16
* TASCAM US-1641
* Edirol UA-101

However a recent discussion about USB2 vs Firewire (see below) has me
wondering if the firewire route is better. Can anyone confirm this?
Has anyone had any throughput issues with any of the above USB2
interfaces when recording 8-16 tracks simultaneously? Can anyone
recommend a good firewire equivalent of the above, in a comparable
price range?

Thanks...

Newsgroups: rec.video.production, rec.video.desktop
From: "Richard Crowley"
Date: Fri, 5 Sep 2008 10:47:37 -0700
Local: Fri, Sep 5 2008 1:47 pm
Subject: transferring VHS to digital video on computer?

"Mad Scientist Jr" wrote ...

USB2 is 480 mbps, whereas Firewire is only 400 mbps.
How is Firewire faster? Please explain...

Firewire is...
1) Full-duplex. It has an independent and concurrent data path
in EACH direction. This means that continuous stream of video
data can be flowing from the video gadget into the computer
without having to be interrupted to send commands, handshaking
protocol, etc back the other direction.http://en.wikipedia.org/wiki/Duplex_communication

OTOH, USB has only one data path and it must be interrupted
and turned around to send data, commands, etc. the other way.http://en.wikipedia.org/wiki/Simplex_communication

2) IEEE1394 (aka "Firewire") was designed from the ground up
to be *isosynchronous*. This means it is capable of an uninterrupted,
sustained stream of data for an indefinite period of time.http://en.wiktionary.org/wiki/isosyn...nchronous.html

OTOH, USB is designed for burst-mode signaling which is
quite appropriate for most kinds of data transfer. Except for
high-speed, real-time data such as DV video.

The actual throughput (in 2006) attained with real devices is
about two thirds of the maximum theoretical bulk data transfer
rate of 53.248 MB/s. Typical hi-speed USB devices operate
at lower speeds, often about 3 MB/s overall, sometimes up
to 10-20 MB/shttp://en.wikipedia.org/wiki/Universal_Serial_Bus#USB_signaling

there are issues with firewire and windows machines.
the quality/type of firewire card/port in your window's machine can
make a difference!
have you found a firewire printer or keyboard??

[Ken[_8_]]

On Sat, 6 Sep 2008 11:53:10 -0700 (PDT), wrote:

there are issues with firewire and windows machines.
the quality/type of firewire card/port in your window's machine
can make a difference!
have you found a firewire printer or keyboard??

http://www.wikio.com/product/reviews...ge1-sort0.html

http://musicthing.blogspot.com/2004/...-keyboard.html

[Richard Crowley]

wrote from Goooooooogle Groups ...
there are issues with firewire and windows machines.

Examples?

the quality/type of firewire card/port in your window's machine can
make a difference!

The quality/type of most any hardware "can make a difference!"
If you want to play some silly Mac vs. PC game, you'll have to
find some other children to play with.

have you found a firewire printer or keyboard??

What's the point? Low-speed devices generally have no need of
high-speed interfaces like Firewire. Besides Apple guaranteed
by their excessive licensing fees that Firewire would be used
only in higher-end products (like camcorders, etc) that really
needed high-bandwidth .

Furthermo Firewire may be on its way out, replaced by USB3
http://en.wikipedia.org/wiki/Usb#USB_3.0

Apple has since dropped the licensing fees for Firewire in an
attempt to keep it alive, but it may be too little too late.

[[email protected]]

On Sep 6, 3:28 pm, Ken wrote:
On Sat, 6 Sep 2008 11:53:10 -0700 (PDT), wrote:
there are issues with firewire and windows machines.
the quality/type of firewire card/port in your window's machine
can make a difference!
have you found a firewire printer or keyboard??

http://www.wikio.com/product/reviews...wire-page1-sor...

http://musicthing.blogspot.com/2004/...ewire-keyboard....

opps,
well there are printers and I should have specified typing keyboard.
Firewire is better!!!!

[Mike Rivers]

Mad Scientist Jr wrote:

a recent discussion about USB2 vs Firewire (see below) has me
wondering if the firewire route is better. Can anyone confirm this?
Has anyone had any throughput issues with any of the above USB2
interfaces when recording 8-16 tracks simultaneously?

You can read about troubles with anything on the Internet. USB2 audio
interfaces are still relatively new and were pretty slow getting off the
block. I guess there are some now that are greater than 8 channels, and
I haven't heard any complaints. The advantage of Firewire is not in the
throughput of the interface (the numbers are a bit faster for USB2 than
Firewire) but that it takes less computer resources to move data. This
has pretty much been overcome, though, because everyone has a CPU that's
faster than it needs to be - until they start piling on more programs.
So if you have a marginal system then Firewire might be better.

The thing about USB, at least when comparing them on a PC, is that it
can use standard Windows drivers, which means that Microsoft can break
it at any time, but that the hardware manufacturer doesn't have to worry
about differences in the computer side of the interface. Firewire
devices require their own driver, and if someone comes up with a
Firewire interface that wasn't accommodated in the audio interface
manufacturer's driver, it probably won't work until he updates his
driver. Some do, some do slowly, some never seem to.

I'd say you shouldn't worry about it. Pick an interface that you like
for other reasons than how it connects to the computer.


--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)

[Mike Rivers]

Soundhaspriority wrote:

Firewire is better, period.

Unless you have to fuss around to make it work.

--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)

[Mike Rivers]

Soundhaspriority wrote:

Use TI based host card. No muss, no fuss.

That may be good advice for certain devices today, but what about the
audio interface that you buy three years from now?

Another problem with that piece of device is that you usually can't tell
what chip set is in the host interface until you install it and even
then, you may not be able to tell. It doesn't always help, either, to
know a specific make and model interface card. Manufacturers change
parts that they believe are essentially interchangeable (and they are,
for things like disk drives and video cameras that look to the computer
like a disk drive). So if I tell you to get an Adaptec card because mine
has a TI chip in it, yours, two years later, might have a VIA chip set.

You can probably make a good living by buying up a stock of cards that
you've found that have the TI Firewire chip, tell everyone that this is
the only one that works for sure (as far as you know, of course) and
selling them at a profit.

What I'm getting at is that when dealing with computers, there's no sure
thing other than that there's no sure thing.


--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)

[[email protected]]

On Sep 6, 4:28 pm, "Richard Crowley" wrote:
wrote from Goooooooogle Groups ...

there are issues with firewire and windows machines.

Examples?

the quality/type of firewire card/port in your window's machine can
make a difference!

The quality/type of most any hardware "can make a difference!"
If you want to play some silly Mac vs. PC game, you'll have to
find some other children to play with.

have you found a firewire printer or keyboard??

What's the point? Low-speed devices generally have no need of
high-speed interfaces like Firewire. Besides Apple guaranteed
by their excessive licensing fees that Firewire would be used
only in higher-end products (like camcorders, etc) that really
needed high-bandwidth .

Furthermo Firewire may be on its way out, replaced by USB3http://en.wikipedia.org/wiki/Usb#USB_3.0

Apple has since dropped the licensing fees for Firewire in an
attempt to keep it alive, but it may be too little too late.

boy are you trolling!!! no mention of Apple until you !!
firewire has a new 3.2 gig standard now ...blows usb out....

In December 2007, the 1394 Trade Association announced that products
will be available before the end of 2008 using the S1600 and S3200
modes which were already (mostly) defined in 1394b. The 1.6 Gbit/s and
3.2 Gbit/s devices will use the same 9-pin connectors as the existing
FireWire 800 and will be fully compatible with existing S400 and S800
devices. It will compete with the forthcoming USB 3.0..

FireWire S800T (IEEE 1394c-2006)
FireWire is enhanced to share gigabit Category 5e cable
FireWire is enhanced to share gigabit Category 5e cable

IEEE 1394c-2006 was published on June 8, 2007.

It provides the following improvements

* A new port specification which provides 800 Mbit/s over the same
RJ45 connectors with Category 5e cable which is specified in IEEE
802.3 clause 40 (gigabit Ethernet over copper twisted pair)
* An automatic negotiation that allows the same port to connect to
either IEEE Std 1394 or IEEE 802.3 (Ethernet) devices.
* Various minor updates to IEEE 1394b


http://en.wikipedia.org/wiki/FireWire

[Mike Rivers]

Soundhaspriority wrote:

Mike, I can't help anyone prepare for the future.

Maybe you can't, but people who put their money into this stuff would
like to know that they'll be able to continue to use what they have when
they upgrade something.

The fact is, Firewire is
an isochronous interface, the equivalent of FEDEX "guaranteed service." USB
is not. That's why all the better interfaces use Firewire.

I'm not equipped to debate this issue. I can tell you that Dan Lavry
won't build a Firewire interface because he's measured random and
inconsistent timing differences between channels. He probalby won't make
a USB interface for the same reason. But perhaps the fact that it
determines its own timing makes it less suitable for a multichannel
audio interface. If the channels are asynchronous, then the errors are
probably insignificant. But if they're supposed to be phase-coherent
(like a stereo pair, or a significant amount of leakage spanning two or
more tracks) then the phase errors will not be consistent.

As far as the
problems you mention, yes, computers suck. Always have, always will.

Don't use a computer. Then you won't need Firewire.


--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)

[[email protected]]

On Sep 6, 4:28*pm, "Richard Crowley" wrote:


What's the point? Low-speed devices generally have no need of
high-speed interfaces like Firewire.


High speed devices can't use a low end interface like USB. If they
could you'd see them on them.


Furthermo Firewire may be on its way out, replaced by USB3http://en.wikipedia.org/wiki/Usb#USB_3.0


Jeez, Richard, that's wrong. Firewire is being updated to improve
firewire. Nothing to do with USB 3.


Apple has since dropped the licensing fees for Firewire in an
attempt to keep it alive, but it may be too little too late.

Also incorrect.


The two biggies in the debate will still be true in the yet to be
released versions, which a USB will still rely on the CPU while FW
won't, so any best case, non-real world scenario specs of speed are
out the window, as they always have been for FW/USB comparisons. And
for real time audio and video, isosynchronous/full duplex beats out
burst mode/single data path every time, no matter what the speed specs
are. They don't matter.

[Richard Crowley]

"Mike Rivers" wrote ...
I'm not equipped to debate this issue. I can tell you that Dan Lavry won't
build a Firewire interface because he's measured random and inconsistent
timing differences between channels.

Firewire, USB, AES/EBU, MADI and even SPDIF are merely streams
of digital samples. Any "timing differences between channels" can only
happen back when the samples were made (or if/when they are converted
back to analog) and these functions are far removed from the method of
transmitting, storing, even processing the digital datastream, however
many channels it contains.

"Inconsistent timing differences" between WHAT "channels"?
Firewire (and USB, etc, etc.) is a single stream of ones and zeroes.
Multiple channel sample data is interleaved and travels together.

He probalby won't make a USB interface for the same reason.

Perhaps he doesn't understand how modern digital interfaces work.

But perhaps the fact that it determines its own timing makes it less
suitable for a multichannel audio interface.

Neither Firewire nor USB "determine their own timing". They send
and recieve data asynchronously at the pleasure ot the controlling
equipment at each end.

If the channels are asynchronous, then the errors are probably
insignificant. But if they're supposed to be phase-coherent (like a stereo
pair, or a significant amount of leakage spanning two or more tracks) then
the phase errors will not be consistent.

Since the channel data are interleaved, they travel together in pairs
in exactly the same manner that concurrent stereo sample pairs are
stored on a CD or DAT. Exactly how do we propose that these
*interleaved* samples of multiple channels can become "asynchronous"
relative to each other?

By BS meter is well into the "absurd gobbledygook" range.
Does Mr. Lavry publish any coherent support for this theory?

[Richard Crowley]

wrote from Gooooooooooogle Groups...
"Richard Crowley" wrote:

What's the point? Low-speed devices generally have no need of
high-speed interfaces like Firewire.

High speed devices can't use a low end interface like USB. If they
could you'd see them on them.

Jeez, dwgriffi, that's wrong. USB is beig updated to improve USB.
Nothing to do with Firewire.

You mean high speed devices like hard drives and optical drives?
You mean high speed devices like digital TV tuners and video
AD converters. Do you mean high speed devices like wireless
networking modems? etc. etc. etc.

Furthermo Firewire may be on its way out, replaced by USB3
http://en.wikipedia.org/wiki/Usb#USB_3.0

Jeez, Richard, that's wrong.
Firewire is being updated to improve firewire. Nothing to do with USB 3.

USB is beig updated to improve USB. Nothing to do with Firewire.

Apple has since dropped the licensing fees for Firewire in an
attempt to keep it alive, but it may be too little too late.

Also incorrect.

"Firewire not dead, but its on life-support:"
http://blogs.zdnet.com/Apple/?p=57

"IEEE has approved the new FireWire 2008 specification, which
will include the S1600 and S3200 standards, running at 1.6Gbps
and 3.2Gbps each. Firewire and i.Link desperately need the upgrade
in order to remain competitive with USB, which will receive an
upgrade to 4.8 Gb/s in version 3.0."
http://blogs.zdnet.com/gadgetreviews/?p=312

"New 3.2Gbps FireWire spec approved, not as fast as USB 3.0"
"USB dominated Firewire primarily because of apple's initially
outrageous royalties attached to Firewire. This despite the fact
that it is an IEEE standard (IEEE 1394). IEEE allows this but it
is still pathetic of apple."
http://digg.com/hardware/New_3_2Gbps...t_as_ USB_3_0

"Apple iPod Officially USB 2.0 Only - No More Firewire!"
http://digg.com/apple/Apple_iPod_Off...ire_?FC=serrs1

The two biggies in the debate will still be true in the yet to be
released versions, which a USB will still rely on the CPU
while FW won't,

That is completely implementation-dependent. There is nothing about
USB that prevents anyone from implementing it with a standalone
controller. In fast, as more high-speed devices like video gadgets
migrate to USB, I fully expect that we will see a new generation of
independent USB DMA controllers.

OTOH Firewire *requires* a standalone controller because it has
too much complexity and overhead to be implemented as a CPU-
dependent I/O driver.

so any best case, non-real world scenario specs of speed are
out the window, as they always have been for FW/USB comparisons.
And for real time audio and video, isosynchronous/full duplex
beats out burst mode/single data path every time, no matter what
the speed specs are. They don't matter.

Hey, I don't want to see FW disappear. All my video equipment
still uses it. But that doesn't keep me from believeing that Apple
is desparately trying to revive an antique, has-been interface
back to life in the face fo stiff competition from USB3, etc.

And apparently the USB3 people agree that full-duplex is better.

[Richard Crowley]

"Soundhaspriority" wrote ...
The real problem with USB is that it has a much larger software
component than Firewire.

That is entirely implementation-dependent. As more high-speed
applications migrate away form Firewire to USB, I fully expect
to see USB controllers doing independent DMA just as Firewire
does now.

[Richard Crowley]

"Soundhaspriority" wrote ...
USB 2.0 host adapters do DMA. The NEC chip uses a 4K slice of ram. But in
the USB protocol, as currently defined, only the host knows the shape of
the tree. The complexity of mediating the tree is what slows it down. If
the generality of the USB protocol is to be preserved, there's no way
around it. Could a special processor be assigned to tree mediation? I
guess it could.

It is no bigger a computing task than an optical mouse or a fingerprint
reader or a gigabit network card. Any of which you can buy for $15
at retail.

OTOH, someone, presumably Intel, could define a protocol that uses the
same connectors, but different behavior for the interface chips to define
a simple nonnetworking protocol, like a UWB cable replacement. Non
networking protocols are always the fastest, because they are minimal.

I don't even have lunch with the guys that do the I/O protocol
design. and if I did, I couldn't talk about it here.

[Scott Dorsey]

Mike Rivers wrote:
Soundhaspriority wrote:
The fact is, Firewire is
an isochronous interface, the equivalent of FEDEX "guaranteed service." USB
is not. That's why all the better interfaces use Firewire.

I'm not equipped to debate this issue. I can tell you that Dan Lavry
won't build a Firewire interface because he's measured random and
inconsistent timing differences between channels. He probalby won't make
a USB interface for the same reason. But perhaps the fact that it
determines its own timing makes it less suitable for a multichannel
audio interface. If the channels are asynchronous, then the errors are
probably insignificant. But if they're supposed to be phase-coherent
(like a stereo pair, or a significant amount of leakage spanning two or
more tracks) then the phase errors will not be consistent.

Firewire has stuff in it to allow it to operate in realtime, with
guaranteed deadlines.

But for recording and playing back, you don't really need the interface
to operate in realtime. You just have to have lots of buffering, and
an interface that is generally faster than the datastream you're sending.
For the most part, data integrity is more important than making deadlines
because buffering has become remarkably cheap.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Mike Rivers]

On Sep 6, 8:51 pm, "Soundhaspriority" wrote:

The other point that I must regrettably contradict Mike about is on drivers.
Audio devices CANNOT use Windows standard drivers, because the only "Windows
Audio Device" that is supported is 16 bits, two channels. All USB audio
devices with greater than two channels or 16 bits have proprietary drivers.

Then why is it that I can use my cEntrance MicPort Pro at 24-bit 96
kHz without loading a driver? Connect it for the first time and
Windows finds a USB audio device.

Again, let me reiterate that I'm not an expert when it come to the
internal workings of operating systems or data protocols, but I do
pick up a few tidbits here and there. As cEntrance explained it to me,
the device identifies itself to Windows and the driver handles it at
its highest sample rate. If I want to run it at 44.1 kHz or have the
ability to set the sample rate from a program, then I have to use the
cEntrance driver.


The real issue for Firewire is that it is highly advisable to use a host
card with a chipset recommended by the audio device manufacturer.

But
(a) they don't all do that and
(b) as I said before, it's not always possible for the end user to
identify the chipset

Good theory, good intention, but not always possible in today's
marketplace. If there were only a couple of Firewire host interfaces
(he card, not the chip) then it might be manageable, but that's not
the situation.

If we all used ProTools, there would be no problems either. But we
don't.

[Richard Crowley]

"Mike Rivers" wrote ...
"Soundhaspriority" wrote:

The other point that I must regrettably contradict Mike about is on
drivers.
Audio devices CANNOT use Windows standard drivers, because the only
"Windows
Audio Device" that is supported is 16 bits, two channels. All USB audio
devices with greater than two channels or 16 bits have proprietary
drivers.

Then why is it that I can use my cEntrance MicPort Pro at 24-bit 96
kHz without loading a driver? Connect it for the first time and
Windows finds a USB audio device.

It isn't more than 2 channels, which is the limit of the generic drivers.

[hank alrich]

Buzzard Breath wrote:



snort

--
ha
Iraq is Arabic for Vietnam

[Mike Rivers]

Richard Crowley wrote:

As more high-speed
applications migrate away form Firewire to USB, I fully expect
to see USB controllers doing independent DMA just as Firewire
does now.

Only if "mass" market products require it. They aren't going to do any
favors for the little corner of the industry that makes multi-channel
high quality audio interfaces.


--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)

[Richard Crowley]

"Mike Rivers" wrote ...
Richard Crowley wrote:

As more high-speed
applications migrate away form Firewire to USB, I fully expect
to see USB controllers doing independent DMA just as Firewire
does now.

Only if "mass" market products require it. They aren't going to do any
favors for the little corner of the industry that makes multi-channel high
quality audio interfaces.

Fortunately, at that level, there's no difference between a digital
stream of video and a digital stream of a bunch of audio channels.
I was assuming that digital video would be the mass-market 'killer-
app" that would drive the market.

[nickm]

"Mad Scientist Jr" wrote in message
...
I thought I had a few good USB2 choices for 8 or 16 simultaneous ins
to the computer (my system = Win XP Pro w/Cubase SX3):

* Alesis Multimix 16
* TASCAM US-1641
* Edirol UA-101

However a recent discussion about USB2 vs Firewire (see below) has me
wondering if the firewire route is better. Can anyone confirm this?
Has anyone had any throughput issues with any of the above USB2
interfaces when recording 8-16 tracks simultaneously? Can anyone
recommend a good firewire equivalent of the above, in a comparable
price range?

Thanks...

Newsgroups: rec.video.production, rec.video.desktop
From: "Richard Crowley"
Date: Fri, 5 Sep 2008 10:47:37 -0700
Local: Fri, Sep 5 2008 1:47 pm
Subject: transferring VHS to digital video on computer?

"Mad Scientist Jr" wrote ...

USB2 is 480 mbps, whereas Firewire is only 400 mbps.
How is Firewire faster? Please explain...

Firewire is...
1) Full-duplex. It has an independent and concurrent data path
in EACH direction. This means that continuous stream of video
data can be flowing from the video gadget into the computer
without having to be interrupted to send commands, handshaking
protocol, etc back the other direction.
http://en.wikipedia.org/wiki/Duplex_communication

OTOH, USB has only one data path and it must be interrupted
and turned around to send data, commands, etc. the other way.
http://en.wikipedia.org/wiki/Simplex_communication

2) IEEE1394 (aka "Firewire") was designed from the ground up
to be *isosynchronous*. This means it is capable of an uninterrupted,
sustained stream of data for an indefinite period of time.
http://en.wiktionary.org/wiki/isosynchronous
http://www.allwords.com/word-isosynchronous.html

OTOH, USB is designed for burst-mode signaling which is
quite appropriate for most kinds of data transfer. Except for
high-speed, real-time data such as DV video.

The actual throughput (in 2006) attained with real devices is
about two thirds of the maximum theoretical bulk data transfer
rate of 53.248 MB/s. Typical hi-speed USB devices operate
at lower speeds, often about 3 MB/s overall, sometimes up
to 10-20 MB/s
http://en.wikipedia.org/wiki/Univers...#USB_signaling

There can be issues with both firewire and USB 2.0 interfaces/devices.
However, that should be the exception to the rule in both cases - and
usually is :-)

I have two multi i/o Firewire interfaces: M-Audio FW1814 which has always
been reliable with both laptop and desktop machines - provided I've used it
with Windows XP or Vista 32 bit OS's. However there is no Vista 64bit driver
as yet, which caused me a bit of a problem about a year or so ago :-/ hence
my reason for getting a second interface: MOTU Traveler which does have 64
bit Vista support. Both the M-Audio and MOTU interfaces support 8 analogue
channels in plus another 8 if used with an ADAT compatible mic pre-amp like
the Behringer Ultragain ADA8000. The M-Audio FW1814 has 2 XLR mic inputs
with phantom power and 6 line inputs, whilst the MOTU traveler has 4 XLR mic
inputs with phantom power and 4 line inputs. Of the two interfaces the
M-Audio is the cheaper one. Both are more expensive than the Tascam US-1641
but I think the M-Audio FW1814 is roughly in the same ball park as the
Alesis Multimix 16. Edirol make a FW-101 interface. I've never had audio
throughput issues with either of my two firewire interfaces. M-Audio has
dropped the ball quite badly with respect to Vista 64bit support for its
firewire interfaces in my opinion, whereas both MOTU and Edirol have had 64
bit vista drivers available from January 2007. MOTU offers a number of
different firewire audio interfaces if you take a look at their product
range.

In terms of the USB devices I've tried, I have used several 2 channel
devices and one 8 channel device briefly and they've been OK too.

Some computers offer more USB sockets and this is perhaps where you need to
check whether the sockets you are using are directly connected to the USB
controller on the motherboard, or whether they are connected via an internal
hub as you can find on some notebooks. Some USB devices don't perform to
their best ability if connected via a hub. Additionally USB interfaces like
IEEE1394 interfaces are made by a number of manufacturers, and not all are
quite equal in the way they work with external devices. In my experience,
I've found that you can minimise the chances of running into problems by
sticking with Intel CPU's and Intel chipsets where USB is concerned, and
Intel CPUs, Intel chipsets and Texas Instruments IEEE1394 chipsets for
firewire.

I want to give either or both the Tascam US1641 and Alesis Multimix 16
interfaces a try-out.myself soon.

I wouldn't like to say definitively that firewire is better than USB 2.0,
but my experiences with firewire have shown firewire devices to be good
performers. Firewire has always been my personal preference over USB 2.0
for portable audio interfaces simply because with firewire I usually only
have a single device attached and with USB there are likely to be a number
of devices attached if you include things like dongles, printers, a mouse
etc...

[Bill Ruys]

"Soundhaspriority" wrote in message
...


Use TI based host card. No muss, no fuss.

Bob Morein
(310) 237-6511
Indeed, the TI chipset for FireWire is a good place to start, but it is no
cure-all. I spend enough time on DAW forums to know that FireWire on the PC
can be a crap-shoot, even with the "recommended" card.

Bill.


** Posted from http://www.teranews.com **

[Bill Ruys]

"Richard Crowley" wrote in message
...
wrote from Goooooooogle Groups ...
there are issues with firewire and windows machines.

Examples?

Take a look at the Mac and Windows hardware user forums at
www.motunation.com. You will find that the Windows hardware forum is
completely loaded with Windows/FireWire issues that simply don't exist on
the Mac. It's one of the reasons I went with a PCIe audio interface (Motu
2408 Mk3 & 24i/o).


the quality/type of firewire card/port in your window's machine can
make a difference!

The quality/type of most any hardware "can make a difference!"
If you want to play some silly Mac vs. PC game, you'll have to
find some other children to play with.


The fact of the matter is that often F/Wire audio interfaces have real
problems on the PC platform. I hang out on a number of DAW forums and
F/Wire issues on the PC are far more common than F/Wire issues on the Mac.
And I'm not playing "some silly Mac vs PC game". I'm a PC user with no
particular interest in switching to the Mac. They are both just tools and
we pick the platform we're most comfortable with.

Bill.


** Posted from http://www.teranews.com **

[Bill Ruys]

Soundhaspriority" wrote in message
...

"Bill Ruys" wrote in message
...
"Richard Crowley" wrote in message
...
wrote from Goooooooogle Groups ...
there are issues with firewire and windows machines.

Examples?

Take a look at the Mac and Windows hardware user forums at
www.motunation.com. You will find that the Windows hardware forum is
completely loaded with Windows/FireWire issues that simply don't exist on
the Mac. It's one of the reasons I went with a PCIe audio interface
(Motu 2408 Mk3 & 24i/o).

Bill, the newer Mac notebooks have dysfunctional Firewire, because they
switched to an Agere chip. See
http://www.google.com/search?q=apple...e7&rlz=1I7ADBF

Bob Morein
(310) 237-6511

Just did some reading from you link... Wow, seems like Apple really dropped
the ball on what was their point of difference. I guess the affected
Macbooks are yet another platform to avoid for audio work. The minefield
broadens!

This weakens the argument that F/Wire is the best thing since sliced bread
for audio, doesn't it? I've yet to find a PC that doesn't work with my
UA-101, which I use for mobile work. Any PC made in the last 5 years with
USB2 has just worked. I can't say the same for my band-mate's F/Wire
micro-lite; it is very fussy about the chipset.

Bill.


** Posted from http://www.teranews.com **

[Laurence Payne[_2_]]

On Mon, 8 Sep 2008 11:19:15 +1200, "Bill Ruys"
wrote:

Take a look at the Mac and Windows hardware user forums at
www.motunation.com. You will find that the Windows hardware forum is
completely loaded with Windows/FireWire issues that simply don't exist on
the Mac. It's one of the reasons I went with a PCIe audio interface (Motu
2408 Mk3 & 24i/o).

Yeah, but is it a Firewire thing or a MOTU thing?

[Mike Rivers]

Bill Ruys wrote:

The fact of the matter is that often F/Wire audio interfaces have real
problems on the PC platform. I hang out on a number of DAW forums and
F/Wire issues on the PC are far more common than F/Wire issues on the Mac.

Go over to the Mackie Onyx forum and see how many happy Mac users who
own 400F, 1200F, and Satellite interfaces are happy. Mackie says it's an
Apple problem, the users say it's a Mackie problem. You just can't count
on Macs being trouble-free any more.


--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)

[Bill Ruys]

"Soundhaspriority" wrote in message
...

"Bill Ruys" wrote in message
...
"Soundhaspriority" wrote in message
...


Use TI based host card. No muss, no fuss.

Bob Morein
(310) 237-6511
Indeed, the TI chipset for FireWire is a good place to start, but it is
no cure-all. I spend enough time on DAW forums to know that FireWire on
the PC can be a crap-shoot, even with the "recommended" card.

Bill.

** Posted from http://www.teranews.com **

It has nothing to do with Firewire. Two problems occur:

1. When a Firewire host adapter shares a bus with a storage device, say a
RAID adapter. When the computer is capturing audio, the Firewire adapter
becomes bus master to accomplish DMA transfer. It must relinquish bus
master to the storage device. The RAID adapter will not relinquish the bus
until the write is complete. This constant switching between bus masters
creates dead time when no audio can be received.

Bus contention isn't a factor with PCIe and I have seen problems with audio
interfaces on PCIe F/Wire adapters. I would also guess that not too many
people running into this problem have RAID controllers sitting on the PCI
bus. Most motherboards these days host the ATA/SATA adapters directly from
the chipset and are natively capable of RAID levels 0 & 1 - some will even
do RAID 5. They don't share the PCI bus.

If, as you say, it had "nothing to do with FireWire", it wouldn't matter
whose F/Wire chipset you used. Clearly, chosing the right FireWire chipset
can have a significant impact on audio. All bus contention issues aside,
surely this means it has *everything* to do with FireWire?

The same problem occurs with USB, or any other bus adapter, but it's
masked because USB audio interfaces are smaller and subject to less
demanding use. You're hearing about Firewire because these users are more
intense and yell louder.

First generation USB often used to generate bus contention issues. But it's
been a long time since USB was hosted on the PCI bus. Again, it's not on
any PC built in the last 5 years. Back then, we used to disable USB on the
motherboard to eliminate audio glitching problems. Recently I have actually
fixed glitching audio problems by removing PCI based WiFi cards and
replacing them with USB Wifi adapters. That would have been a big no-no
when USB was hosted on the PCI bus.

2. In laptops, the OS will sometimes refuse to give a Cardbus adapter a
dedicated interrupt. This would occur with ANY Cardbus adapter, be it USB,
SATA, Firewire, SPDIF, etc. The problem is most often found in laptops
from 2003 and older. In some of these older laptops, the USB
implementation could be OK. I'll give you a specific example: my trusty
Compal CL-50. The workaround in this case is to use the built-in
Ilink/Firewire port instead, because the driver is written well, and it
has the dedicated interrupt.

Newer laptops have an extended interrupt protocol that gives the Cardbus
adapter a dedicated interrupt. A 1GHZ Asus subnotebook, weighing only 2.75
pounds, can capture 12 channels at 96/24, the highest I've tested. This is
beyond the channel capability of any USB interface on the market.

We've had plenty of IRQs to play with since the IOAPIC was intruduced quite
a number of years ago. PCIe gets around this further by packetizing
interrupt requests; there are no hard IRQ lines anymore with PCIe, although
it emulates them for backward compatibility with legacy PCI drivers.
ExpressCard is basically PCIe, so should be better than Cardbus.

A modern, well designed DAW shouldn't suffer bus contention. I've seen too
many examples of people following the manufacurers recommendations regarding
F/Wire chipset, motherboard, etc and still having unresolved issues. Some
of the responcibility must also lie with the audio hardware/driver
manufacturer. I say this because I have seen people make the switch from a
given brand of F/Wire audio hardware to RME and have all their problems go
away, with no change to the underlying DAW hardware (agere chipset
excluded).

I currently run a MOTU PCIe-424 card with a 2408 Mk3 and a 24i/o. I
couldn't easily match that setup with anything in the F/Wire domain, but if
I needed to go F/Wire, I would go RME, purely because they seem to have less
problems with PC based F/Wire than anyone else.

There is still a lot of variability with FireWire based audio on the PC.
There was always less variability on the Mac platform, but as you point out,
the new MacBooks seem to have taken that a step backwards.

Bill.


** Posted from http://www.teranews.com **

[Mike Rivers]

Soundhaspriority wrote:

. . . But for laptops, Firewire is the way to go. The compatibility
problems are solved by knowledge. When you buy a Firewire audio interface,
you buy the host adapter at the same time, whatever is recommended by the
hardware vendor. These cards are cheap, around $20 from Newegg, Firewire
Depot, etc.

You can buy a relatively inexpensive PCI host adapter card for a desktop
PC, but PCI slots have threatened to become extinct for a few years now
and probably will, soon. PCIe Firewire cards are still kind of
expensive. You don't buy a host adapter for a laptop, other than perhaps
a CardBus or whatever the new external card is called that I can't
remember. Those aren't down to $20 yet.

I was in Micro Center yesterday and did a little survey. I found three
PCI and one PCIe Firewire host adapters on the shelf. None said anything
on the box about the chipset used. I found one of the companies on the
web, found the card that I saw at Micro Center on the web page, and
looked for the chipset. Nope.

Perhaps you could post a list of cards that you know and the chipsets
used, or which cards you know use the TI chipset (what ever happened to
Oxford, that used to be the one to have?). But then it probably wouldn't
be of much use because manufacturers change parts at will, without
changing packaging or model numbers, so what you got last month might
not be what I get this month unless you sell me yours.


--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)

[Mike Rivers]

Soundhaspriority wrote:

I reject the assertion that Firewire interfaces used with the recommended
adapter chipset are especially problematic.

It's pretty rare these days to find an audio hardware manufacturer
recommending a specific Firewire host adapter or chipset. First off,
they don't want to make it look like their hardware is fussy. They want
it to be the one that anyone can use with any computer. Second, they
just don't know. They don't have the resources to test every Firewire
adapter on the market, and they especially don't have the resources to
keep up with how fast that end of the seesaw is changing.

It does seem possible for a company to build hardware (and drivers) that
works successfully on a large range of computers, and some companies
don't do such a good job at that. It depends on what the company is
selling and for how much. RME is the name that comes up most often when
someone manages to solve all his problems when giving up with one audio
interface and buying a different one. M-Audio had a reputation for fast
response with driver updates when a generic (not individual system
specific) problem is uncovered. Mackie has been saying for well over a
year that they're working on updated drivers for the "F" series of audio
interface hardware but has just recently announced a call for beta
testers among users of a specific combination of hardware and OS.

There are simply no guarantees. Today I'd say that you can hedge your
bets by buying an RME interface if you can find one that suits your
needs, and that it won't be too fussy about what's on the computer end.
But there are plenty of good reasons (not just cost) for buying a
Mackie, or a MOTU, or an M-Audio, or a Presonus or whatever because it
might be a better match for your audio interfacing, and that's the swamp
you're really trying to drain.




--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)

[Arny Krueger]

"Mike Rivers" wrote in message
news:a6%wk.788$jE1.502@trnddc03
Bill Ruys wrote:

The fact of the matter is that often F/Wire audio
interfaces have real problems on the PC platform. I
hang out on a number of DAW forums and F/Wire issues on
the PC are far more common than F/Wire issues on the
Mac.

Go over to the Mackie Onyx forum and see how many happy
Mac users who own 400F, 1200F, and Satellite interfaces
are happy. Mackie says it's an Apple problem, the users
say it's a Mackie problem. You just can't count on Macs
being trouble-free any more.

I guess the recent rise in Mac sales have put enough product in user hands,
for the usual expected weaknesses to finally be reported.

The world is full of people who confuse a relatively small number of samples
in user hands with a trouble-free product.

An absence of trouble reports does not prove the absence of design and
production errors.

[Arny Krueger]

"Soundhaspriority" wrote in message


Mike, I can't help anyone prepare for the future. The
fact is, Firewire is an isochronous interface, the
equivalent of FEDEX "guaranteed service." USB is not.
That's why all the better interfaces use Firewire. As far
as the problems you mention, yes, computers suck. Always
have, always will.

USB has an isosynchronous mode which provides the desired guaranteed
service, but it extracts the inherent cost in terms of inability to have
multiple nominally concurrently active connections.

[hank alrich]

Arny Krueger wrote:

"Mike Rivers" wrote in message
news:a6%wk.788$jE1.502@trnddc03
Bill Ruys wrote:

The fact of the matter is that often F/Wire audio
interfaces have real problems on the PC platform. I
hang out on a number of DAW forums and F/Wire issues on
the PC are far more common than F/Wire issues on the
Mac.

Go over to the Mackie Onyx forum and see how many happy
Mac users who own 400F, 1200F, and Satellite interfaces
are happy. Mackie says it's an Apple problem, the users
say it's a Mackie problem. You just can't count on Macs
being trouble-free any more.

I guess the recent rise in Mac sales have put enough product in user hands,
for the usual expected weaknesses to finally be reported.

The world is full of people who confuse a relatively small number of samples
in user hands with a trouble-free product.

An absence of trouble reports does not prove the absence of design and
production errors.

Actually that is compounded by Apple ****ing up lately in some areas
they had previously had together. I guess that's "progress".

--
ha
Iraq is Arabic for Vietnam

[[email protected]]

On Sep 7, 8:40*pm, "Bill Ruys"
wrote:


This weakens the argument that F/Wire is the best thing since sliced bread
for audio, doesn't it? *I've yet to find a PC that doesn't work with my
UA-101, which I use for mobile work. *Any PC made in the last 5 years with
USB2 has just worked. *I can't say the same for my band-mate's F/Wire
micro-lite; it is very fussy about the chipset.


Even though FW is less surefire on a PC, I'd still put my money on
MOTU providing a not ready for primetime interface, as L. Payne opens
the door to. After having several different manufacturer's FW
interfaces in and out of my Mac studio in the past 10 years, MOTU's
were the only ones that weren't rock solid FW-wise. Close, but I
don't rate the programming in their hardware highly.


Soundhaspriority wrote:
I reject the assertion that Firewire interfaces used with the recommended
adapter chipset are especially problematic.
It's pretty rare these days to find an audio hardware manufacturer
recommending a specific Firewire host adapter or chipset. First off, they
don't want to make it look like their hardware is fussy.


Actually for the past few years it's been pretty common. Digi, Apogee
and Presonus all do, and I'm sure if you looked you'd find more. RME
particularly has a lot to say about Firewire chips.

[Mike Rivers]

On Sep 6, 11:41 pm, "Soundhaspriority" wrote:

"Mike Rivers" wrote in message
Then why is it that I can use my cEntrance MicPort Pro at 24-bit 96
kHz without loading a driver? Connect it for the first time and
Windows finds a USB audio device.

The specific driver is contained in a flash memory in the MicPort. When you
plug it in, Windows first sees a small "flash disk" with an autorun file on
it that installs the driver. Cleverly, once the driver is automatically
installed, you, the user, won't see the "disk" listed under My Computer,
because the drive letter is de-assigned.

This smelled kind of fishy to me, so I asked my friend at CEntrance,
who said it is not correct. There is no "flash disk" in the MicPort
Pro, no autorun, and no driver installation. Firmware in the MicPort
Pro developed by CEntrance allows them to run 3 channels of 24-bit 96
kHz audio through USB1.1. This firmware code is one of the company's
products and they've licensed it for use in products from other
manufacturers. See http://centrance.com/consulting/usb/ for an
overview.

[Geoff]

Mike Rivers wrote:
Soundhaspriority wrote:

I reject the assertion that Firewire interfaces used with the
recommended adapter chipset are especially problematic.

It's pretty rare these days to find an audio hardware manufacturer
recommending a specific Firewire host adapter or chipset. First off,
they don't want to make it look like their hardware is fussy. They
want it to be the one that anyone can use with any computer. Second,
they just don't know.

But they find out pretty soon in the support forums...

geoff

[david correia]

In article ,
"Bill Ruys" wrote:

"Soundhaspriority" wrote in message
...


Use TI based host card. No muss, no fuss.

Bob Morein
(310) 237-6511
Indeed, the TI chipset for FireWire is a good place to start, but it is no
cure-all. I spend enough time on DAW forums to know that FireWire on the PC
can be a crap-shoot, even with the "recommended" card.

Bill.


** Posted from http://www.teranews.com **



The recommended chipset inside FW hard drives in the Mac world has
always been Oxford.

I have about 6 of them and they've always been reliable. We use them for
backing up data, a very important task for a busy Protools studio. They
saved by ass big time once.




David Correia
www.Celebrationsound.com

[Mike Rivers]

wrote:

It's pretty rare these days to find an audio hardware manufacturer
recommending a specific Firewire host adapter or chipset.

Actually for the past few years it's been pretty common. Digi, Apogee
and Presonus all do, and I'm sure if you looked you'd find more. RME
particularly has a lot to say about Firewire chips.

Well, maybe they've decided that it's better to have fewer, but happier
customers than to no **** off the computer hardware manufacturers. Or
perhaps this is a marketing ploy to lock a user into the equipment that
he bought. Say you chose an interface from a company that recommends a
TI Firewire chipset and you searched far and wide and stretched your
budget to near the breaking point to find a laptop computer with a TI
Firewire chipset. Then you decide that the audio interface isn't really
what you want, you return it, and exchange it for one for which the
manufacturer recommends an NEC chipset. Then what do you do? Exchange
your computer as well? Take your chances with the IT Firewire chip?
Start searching for an external adapter card with an NEC chipset? Shoot
the messenger?

I still contend, though, that it's difficult for "the average user" to
determine what chipset is used in his Firewire host adapter, and to, for
sure get what he thinks he's getting. Once you have one you may be able
to determine what you have, but some people have a problem with going
through a lot of hardware before finding what they really need. When I
first got a Mackie Onyx mixer Firewire card, I had a click'n'pop problem
with the laptop computer that I wanted to use it with, using the
cheapest Cardbus Firewire adapter that my friendly local computer store
had. I forget what chipset it had - it didn't say on the box.

Thinking that the wrong Firewire chipset might be the problem, I
exchanged it for another, and another, and another, and finally
eliminated the problem when I updated the network interface card driver
for the computer. Now I have two Cardbus Firewire adapters, one with an
NEC chipset the other which I've not been able to identify. Both work
fine with the Onxy mixer card, the Mackie Satellite, the Mackie 1200F, a
Mad Dog brand and a completely nameless Firewire hard disk enclosure,
and my Nomad Jukebox 3. That's all the Firewire stuff I have.

I'll acknowledge that there are optimum combinations, and I'll accept
that there may be some combinations that are simply incompatible. But to
require a certain combination, unless you're selling a turnkey system, I
think is bad business all around. IBM worked on that model for a long
time, but see where it got them? g

What ever happened to standards? They must have gone away with the
influx of mass market computers.



--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)

[Mike Rivers]

Soundhaspriority wrote:

If anybody is making a buying decision, I'll be happy to help them out.

OK, let's play. I have a Dell 2650 laptop, an IBM T40 laptop, and a Dell
Precision 370 without a Firewire card, all running Windows XP. The 370
has an available PCI and PCI-Ex1 slot, neither laptop has a built-in
Firewire port, both have a CardBus slot available. The IBM also
accommodates an ExpressCard but I don't know which type. I want to use a
Mackie 1200F and Onxy mixer. I'd prefer not to have to change host cards
when I change interfaces. After all, they're both made by Mackie. What
should I buy?

Show your work.

How did you determine the optimum chipset for the audio hardware? What
is it? How did you locate host adapter cards with that chipset? Can you
assure me that when I buy what you recommend, it WILL have the chipset
that you think it has?

(Hint: The mixer Firewire card uses a BridgeCo chip, the 1200F uses a TI
chip - but I only know that because I took them apart, and I'm a friend
of the company. Forget that I told you. Do your own research and tell me
how far you got and how reliable your source is)



--
If you e-mail me and it bounces, use your secret decoder ring and reach
me he
double-m-eleven-double-zero at yahoo -- I'm really Mike Rivers
)

[Mike Rivers]

On Sep 9, 3:26 pm, "Soundhaspriority" wrote:

Mike, are you seriously interested in buying? I'll be happy to work with you
provided it's not an academic exercise. In other words, please don't waste
my time. First step: please supply links to the product support pages for
the two laptops. I need to check out the motherboard chipsets.

It's an academic exercise. I want to know just how much trouble it
really is. My point is that I believe that it's too much trouble for
most users. If you can make it look easy, I'll concede.

You've already made it seem like too much trouble, though, by asking
me to dig up support links for (obsolete) laptop computers. I gave you
the makes and model numbers. The only other information I can offer is
that the IBM specific model number is 1875DLU.

Take it from there or take it elsewhere.