USB 2.0 or Firewire 400??

I'm cleaning up the storage bins.
I found a few firewire 400 hard drive enclosures.
My primary PC has firewire 400 and USB 3.0 ports...no contest.

But the other machines have USB 2.0 and firewire 400 ports.
I did a bunch of googling, but most of the data presented pertains
to Apple Machines. It's suggested that, for Apple, the sustained transfer
rate of firewire 400 exceeds USB 2.0 by a significant margin.

What's the real-world throughput of firewire 400 vs USB 2.0
external hard drive on a windows 7 PC (say 2.6GHz. dual core Intel
7200 RPM drives) when copying files?

I'd just test it, but I can't find the cable.

If I can't expect significant increase in throughput, I'll quit looking
and put the enclosures in the recycle pile.

There's also the question of power available from a firewire 400 port.
Specs vary considerably by device.
My expectation is that I could use a firewire powered 2.5" hard drive
for storing backups and get more speed than a USB 2.0 thumb drive.

2.5" USB hard drives powered by USB have been iffy with insufficient
available current to start the drive. Firewire should fix that???

mike wrote:
I'm cleaning up the storage bins.
I found a few firewire 400 hard drive enclosures.
My primary PC has firewire 400 and USB 3.0 ports...no contest.

But the other machines have USB 2.0 and firewire 400 ports.
I did a bunch of googling, but most of the data presented pertains
to Apple Machines. It's suggested that, for Apple, the sustained transfer
rate of firewire 400 exceeds USB 2.0 by a significant margin.

What's the real-world throughput of firewire 400 vs USB 2.0
external hard drive on a windows 7 PC (say 2.6GHz. dual core Intel
7200 RPM drives) when copying files?

I'd just test it, but I can't find the cable.

If I can't expect significant increase in throughput, I'll quit looking
and put the enclosures in the recycle pile.

There's also the question of power available from a firewire 400 port.
Specs vary considerably by device.
My expectation is that I could use a firewire powered 2.5" hard drive
for storing backups and get more speed than a USB 2.0 thumb drive.

2.5" USB hard drives powered by USB have been iffy with insufficient
available current to start the drive. Firewire should fix that???

Firewire performance can be a function of the controller
chip. These are my numbers for a pair of Firewire 400 enclosures.

Host ----------- Firewire 400
30MB/sec

The second configuration is daisy-chain mode. The enclosures happen
to have two Firewire 6 pin connectors. The second connector is
via controller chip "regeneration". The performance is bus-limited
by the passthru transfer of the chip in the first enclosure.

Host ----------- Firewire 400 ----------- Firewire 400
20MB/sec

So if they were supposed to do 50MB/sec, nobody
told my enclosures that. The devices were never connected
in daisy chain mode again, after the initial test was
carried out. I would hook them up directly, or not
at all.

*******

The bus power is wired-OR. The "strongest" device sources
the power. Apple computers had an extra output at
25V, for powering the monitor right from the computer.
This mean the monitor did not need its own power
supply (also a means of lock-in). The other benefit of
the 25V output option on the power supply, is it could
be used as a bus power source for Firewire.

The total watts available, is bus_voltage times max_allowed_current.

On IBM PC compatible Firewire ports, 12VDC is a typical
choice for bus power. As that is the most-positive voltage
available. It means a bus-powered enclosure, has a lower
limit on power caused by the lower bus voltage used. A
Firewire enclosure "always has to deal with Apple",
so power conversion from VBUS to other voltages,
is required.

My Firewire enclosures were AC powered, and the power supply
was inside the enclosure. This meant fewer wall warts, but
also meant that the device was not looking for bus power to
run (didn't matter what it was connected to, and would
work with a 4 pin Firewire computer). There was sufficient power
(40W footprint), that the internal power supply did the work.
There were a few "phantom power" issues, with the LED on the
front of the unit remaining lit, even when the AC power on the
enclosure was switched off. That means the bus power has "leaked"
into the enclosure when it shouldn't have, and caused the
front LED to be dimly lit.

I don't see a strong reason for using that stuff now,
and I'd just retire it. At times, it can be just as
goofy as USB2, with no obvious benefit.

A USB3 peripheral connected to a USB2 port, can
do on the order of 35MB/sec, which is good enough
if that's all you've got. I wouldn't waste the extra
time digging out my Firewire crap, to get roughly
the same mediocre performance.

My USB3 enclosures have wall warts, so I never
have to worry about power. And the enclosures
were relatively cheap. I run them with the
aluminum cover off, because they have
no cooling to speak of. These were an impulse buy
at my local computer store, without researching them.

NST-366S3-SV $25

http://www.newegg.com/Product/Produc...82E16817392076

I seem to remember there were some Macintosh web sites
with benchmarks for Firewire 400, so if you want
numbers, they'll still be sitting there. It's just my
enclosures that apparently weren't the best.

Paul

On 12/17/2015 2:47 PM, Paul wrote:
mike wrote:
I'm cleaning up the storage bins.
I found a few firewire 400 hard drive enclosures.
My primary PC has firewire 400 and USB 3.0 ports...no contest.

But the other machines have USB 2.0 and firewire 400 ports.
I did a bunch of googling, but most of the data presented pertains
to Apple Machines. It's suggested that, for Apple, the sustained
transfer
rate of firewire 400 exceeds USB 2.0 by a significant margin.

What's the real-world throughput of firewire 400 vs USB 2.0
external hard drive on a windows 7 PC (say 2.6GHz. dual core Intel
7200 RPM drives) when copying files?

I'd just test it, but I can't find the cable.

If I can't expect significant increase in throughput, I'll quit looking
and put the enclosures in the recycle pile.

There's also the question of power available from a firewire 400 port.
Specs vary considerably by device.
My expectation is that I could use a firewire powered 2.5" hard drive
for storing backups and get more speed than a USB 2.0 thumb drive.

2.5" USB hard drives powered by USB have been iffy with insufficient
available current to start the drive. Firewire should fix that???

Firewire performance can be a function of the controller
chip. These are my numbers for a pair of Firewire 400 enclosures.

Host ----------- Firewire 400
30MB/sec

The second configuration is daisy-chain mode. The enclosures happen
to have two Firewire 6 pin connectors. The second connector is
via controller chip "regeneration". The performance is bus-limited
by the passthru transfer of the chip in the first enclosure.

Host ----------- Firewire 400 ----------- Firewire 400
20MB/sec

So if they were supposed to do 50MB/sec, nobody
told my enclosures that. The devices were never connected
in daisy chain mode again, after the initial test was
carried out. I would hook them up directly, or not
at all.

*******

The bus power is wired-OR. The "strongest" device sources
the power. Apple computers had an extra output at
25V, for powering the monitor right from the computer.
This mean the monitor did not need its own power
supply (also a means of lock-in). The other benefit of
the 25V output option on the power supply, is it could
be used as a bus power source for Firewire.

The total watts available, is bus_voltage times max_allowed_current.

On IBM PC compatible Firewire ports, 12VDC is a typical
choice for bus power. As that is the most-positive voltage
available. It means a bus-powered enclosure, has a lower
limit on power caused by the lower bus voltage used. A
Firewire enclosure "always has to deal with Apple",
so power conversion from VBUS to other voltages,
is required.

My Firewire enclosures were AC powered, and the power supply
was inside the enclosure. This meant fewer wall warts, but
also meant that the device was not looking for bus power to
run (didn't matter what it was connected to, and would
work with a 4 pin Firewire computer). There was sufficient power
(40W footprint), that the internal power supply did the work.
There were a few "phantom power" issues, with the LED on the
front of the unit remaining lit, even when the AC power on the
enclosure was switched off. That means the bus power has "leaked"
into the enclosure when it shouldn't have, and caused the
front LED to be dimly lit.

I don't see a strong reason for using that stuff now,
and I'd just retire it. At times, it can be just as
goofy as USB2, with no obvious benefit.

A USB3 peripheral connected to a USB2 port, can
do on the order of 35MB/sec, which is good enough
if that's all you've got. I wouldn't waste the extra
time digging out my Firewire crap, to get roughly
the same mediocre performance.

My USB3 enclosures have wall warts, so I never
have to worry about power. And the enclosures
were relatively cheap. I run them with the
aluminum cover off, because they have
no cooling to speak of. These were an impulse buy
at my local computer store, without researching them.

NST-366S3-SV $25

http://www.newegg.com/Product/Produc...82E16817392076

I seem to remember there were some Macintosh web sites
with benchmarks for Firewire 400, so if you want
numbers, they'll still be sitting there. It's just my
enclosures that apparently weren't the best.

Paul
Thanks for the input.
I read many of the Apple reports. Some said 50% faster than
USB2. Others said about the same in real world. Pick any number
and you can find a review that says that number. Sigh.
I liked that the firewire enclosures had power switches.
Since I'd use it for archival storage it would be off most of the time.
I've got a bunch of AC power switches at the wall warts for the USB
drives. PITA.
I think I'll take your advice and retire them.

mike wrote:

Thanks for the input.
I read many of the Apple reports. Some said 50% faster than
USB2. Others said about the same in real world. Pick any number
and you can find a review that says that number. Sigh.
I liked that the firewire enclosures had power switches.
Since I'd use it for archival storage it would be off most of the time.
I've got a bunch of AC power switches at the wall warts for the USB
drives. PITA.
I think I'll take your advice and retire them.

My most modern Mac, has USB1.1 and Firewire 400.
And it has GbE. For serious work, I do stuff
over the network. I've done file transfer over
the USB1.1 port, but even small transfers (a
few Photoshop files), might take a fraction of
an hour. If I wanted, I could fit a USB2 card
into the machine - and Apple could easily have
done that too, except they wanted to emphasize
Firewire by "making USB suck". That's why it
doesn't have a NEC USB2 chip. They could easily
have fitted USB2 at the time.

I have other hardware for my retired Mac family.

PTP 250 had two Adaptec SCSI cards, and external cabling
to SCSI enclosures. Two disks used to sit on top of the
computer case. The inside of the computer was full
of (low density) disk drives.

Mac G4 has two SCSI cards too, a narrow card for the scanner
and a wide card for the enclosures (so enclosures could be
moved between machines). The G4 also got an IDE card of
some sort, ACard brand. But I don't remember exactly why
I needed that. So I actually had plenty of ways to
connect stuff, and owned more cards than slots on the
Macs. Even my original Quadra had excess hardware - it
got the must expensive video card I've ever owned :-)
Silly, really. But that's the Apple ecosystem for you.
Stuff like that just seems to happen.

Paul

On Thu, 17 Dec 2015 15:47:22 -0800, mike wrote:

I liked that the firewire enclosures had power switches.
Since I'd use it for archival storage it would be off most of the time.
I've got a bunch of AC power switches at the wall warts for the USB
drives. PITA.

I have a vision of a medium-sized plywood box with a set of about 10 outlets
mounted along one face and a corresponding set of switches mounted along the
adjoining face, allowing you to turn individual outlets on or off at will.
Am I close? :-)

--

Char Jackson

On 12/18/2015 6:48 AM, Char Jackson wrote:
On Thu, 17 Dec 2015 15:47:22 -0800, mike wrote:

I liked that the firewire enclosures had power switches.
Since I'd use it for archival storage it would be off most of the time.
I've got a bunch of AC power switches at the wall warts for the USB
drives. PITA.

I have a vision of a medium-sized plywood box with a set of about 10 outlets
mounted along one face and a corresponding set of switches mounted along the
adjoining face, allowing you to turn individual outlets on or off at will.
Am I close? :-)

ONlY TEN??? I've got 36 and counting, although not all are individually
switched and several are occluded by wall-warts.

It started as one of the wide, thin UPS boxes with switched outlets and
the guts removed sitting under the printer.
It's devolved to a tree-structure of power strips with individual outlet
switches plugged into those. My Anthro workstation looks like a combination
of spider web and rat's nest.

I ordered one of these without thinking it thru.
http://www.outletpc.com/al2647-rosew...FUZcfgodTNoJNA

Sockets are too close together and rotated wrong direction for most of
the wall-warts. And the warts cover the switches. Bummer.

What I'd like is a 6-foot long strip with switched outlets every 4
inches or so. That would dramatically decrease the rat's nest.

I considered one big honkin' 12V/5V supply to replace a bunch of wall warts,
but, I don't like the idea of a single point failure destroying everything.

On Fri, 18 Dec 2015 10:09:38 -0800, mike wrote:

On 12/18/2015 6:48 AM, Char Jackson wrote:
On Thu, 17 Dec 2015 15:47:22 -0800, mike wrote:

I liked that the firewire enclosures had power switches.
Since I'd use it for archival storage it would be off most of the time.
I've got a bunch of AC power switches at the wall warts for the USB
drives. PITA.

I have a vision of a medium-sized plywood box with a set of about 10 outlets
mounted along one face and a corresponding set of switches mounted along the
adjoining face, allowing you to turn individual outlets on or off at will.
Am I close? :-)

ONlY TEN??? I've got 36 and counting, although not all are individually
switched and several are occluded by wall-warts.

It started as one of the wide, thin UPS boxes with switched outlets and
the guts removed sitting under the printer.
It's devolved to a tree-structure of power strips with individual outlet
switches plugged into those. My Anthro workstation looks like a combination
of spider web and rat's nest.

I ordered one of these without thinking it thru.
http://www.outletpc.com/al2647-rosew...FUZcfgodTNoJNA

Sockets are too close together and rotated wrong direction for most of
the wall-warts. And the warts cover the switches. Bummer.

What I'd like is a 6-foot long strip with switched outlets every 4
inches or so. That would dramatically decrease the rat's nest.

I considered one big honkin' 12V/5V supply to replace a bunch of wall warts,
but, I don't like the idea of a single point failure destroying everything.

I use a bunch of those little 6" extension cords that are made just for
getting the wall warts away from the outlets.

--

Char Jackson

On 12/20/2015 02:11 AM, Char Jackson wrote:

[snip]

I use a bunch of those little 6" extension cords that are made just for
getting the wall warts away from the outlets.


I have a few. Occasionally, I find one useful. They're really expensive
for what you get. A normal 6-foot cord works too, and you can plug in 2
(or is some cases 3) wall warts.

Anything with properly-spaced and oriented outlets would either be a big
waste of space, or custom-built for what you have now, and not adaptable
to what you have next year.

--
5 days until the winter celebration (Friday December 25, 2015 12:00:00
AM for 1 day).

Mark Lloyd
http://notstupid.us/

"In fact, when you get right down to it, almost every explanation Man
came up with for *anything* until about 1926 was stupid." [Dave Barry]