26 Nov
2010
26 Nov
'10
8:48 a.m.
Ok, so I have been doing research, and this affects everyone here who
collects SGI, apple, and anything using scsi. in the near future, the supply
of smaller SCSI drives will eventually dry up, due to failure from age and
lack of supply, as I believe they are no longer manufactured, unless for the
industrial/server market. I could be wrong on that part, but I take it as
such because my searching for them brings up nothing new. what does that
mean for collectors? it means that your neat little apple or SGI or etc will
run, until the scsi disk dies, then we will be left cannibalizing machines
for disks, paying ridiculous prices for leftover scsi disks, which will
skyrocket because of the small supply.
So what is our option? I have seen SCSI to IDE adapters around, but they go
up into the $100's to $200's, say you have a minimum 10 machines, that
leaves you somewhere between $1,000 and $2,000 total, before the disk or
devices. If we could design a simple SCSI to IDE interface, we could be set
for a good long while on storage for these devices for a decent amount of
time. There are all sorts of adapters to IDE, there is CF to IDE, SD to IDE,
SD to CF which could be placed in a CF to IDE adapter if need be, i'm sure
one could also rig up a USB drive to SCSI if one tried, but I could be
wrong.
Having said that, I know nothing about programming microcontrollers or
simple processors, but I have seen people on the board who do. I'm sure if
we can find a chip capable, and lay out a schematic, source code, and
possibly a board layout, members could source their own boards and
components to build their own devices at home (given they have a chip
programmer), so no one person has to take on the responsibility of supplying
parts, kits, finished devices unless they so choose to. This is just an
idea, if nothing becomes of it, no big deal, just putting it out there....
-Joe
26 Nov
26 Nov
11:52 a.m.
On 11/26/2010 11:48 AM, joe lobocki wrote:
Ok, so I have been doing research, and this affects
everyone here who
collects SGI, apple, and anything using scsi. in the near future, the supply
of smaller SCSI drives will eventually dry up, due to failure from age and
lack of supply, as I believe they are no longer manufactured, unless for the
industrial/server market. I could be wrong on that part, but I take it as
such because my searching for them brings up nothing new. what does that
mean for collectors? it means that your neat little apple or SGI or etc will
run, until the scsi disk dies, then we will be left cannibalizing machines
for disks, paying ridiculous prices for leftover scsi disks, which will
skyrocket because of the small supply.
First don't go to IDE as that is also going the way of the dodo.
Current servers are using hot swap SATA.
The other alternative is SCSI to USB as small USB 1gb to 16GB
thumb drives are dirt cheap and easy to use. The smaller end
of those are more in line with what those systems required.
So what is our option? I have seen SCSI to IDE
adapters around, but they go
up into the $100's to $200's, say you have a minimum 10 machines, that
leaves you somewhere between $1,000 and $2,000 total, before the disk or
devices. If we could design a simple SCSI to IDE interface, we could be set
for a good long while on storage for these devices for a decent amount of
time. There are all sorts of adapters to IDE, there is CF to IDE, SD to IDE,
SD to CF which could be placed in a CF to IDE adapter if need be, i'm sure
one could also rig up a USB drive to SCSI if one tried, but I could be
wrong.
At $100 I'd consider SCSI to IDE cheap.
Having said that, I know nothing about programming
microcontrollers or
simple processors, but I have seen people on the board who do. I'm sure if
we can find a chip capable, and lay out a schematic, source code, and
possibly a board layout, members could source their own boards and
components to build their own devices at home (given they have a chip
programmer), so no one person has to take on the responsibility of supplying
parts, kits, finished devices unless they so choose to. This is just an
idea, if nothing becomes of it, no big deal, just putting it out there....
It would be very hard to kit up a SCSI to IDE converter for less than
$60-100
and that assumes the people doing it are contributing labor, time and
cash for free. then there is the issue of surface mount, people building
incorrectly assembled kits to be supported.
The other headache is that modern drives are far larger than most of
systems referred to can address (without mods/patches) so you run into
something I encountered recently while building a small dedicated system
using an ITX board, the whole system needed maybe 2GB of storage and
the smallest drive I could buy was 80gb! Fortunately the system also
supported USB thumb drives so it was easy and cheap to go that way.
Drive do not generally die from old age, I have many drives from the early
80s that still work though see infrequent use. So it would be better to
buy
the drives you need as you can now and stockpile them for later. Also you
can then do drive to drive copies as both backup and failure recovery.
Allison
2:57 p.m.
On 27/11/2010, at 6:52 AM, allison wrote:
The other alternative is SCSI to USB as small USB 1gb to 16GB
thumb drives are dirt cheap and easy to use. The smaller end
of those are more in line with what those systems required.
Anyone know of such a device? There are plenty of devices to allow you to access a SCSI
disk via USB but I haven't found one that works the other way....
Huw Davies | e-mail: Huw.Davies at kerberos.davies.net.au
Melbourne | "If soccer was meant to be played in the
Australia | air, the sky would be painted green"
27 Nov
27 Nov
2:02 a.m.
I have not seen much in the way of parallel SCSI target hardware
anywhere, other than efforts mentioned earlier here to use various chips
for target emulation on PC's.
I would think that some effort with an FPGA could produce a fairly nice
10mhz scsi target, and if a PIC was attached to it with USB support, one
could write a program that could communicate with the PIC and host
emulated storage on the USB host.
I don't think the concept of USB storage does much for you, unless you
were to also include the storage on the USB /SCSI hardware device and
make the SCSI access the storage which the SCSI target would also be
accessing.
That would be interesting, but not so useful either as that storage
would still not make sense to the host, unless you were using fairly
modern disk mappings and they could be read via current USB drivers from
a host.
If you put say an SDram on the small dongle with the USB device and SCSI
target hardware, it would run fairly well though.
On 11/26/2010 2:57 PM, Huw Davies wrote:
On 27/11/2010, at 6:52 AM, allison wrote:
The other alternative is SCSI to USB as small USB
1gb to 16GB
thumb drives are dirt cheap and easy to use. The smaller end
of those are more in line with what those systems required.
Anyone know of such a
device? There are plenty of devices to allow you to access a SCSI disk via USB but I
haven't found one that works the other way....
Huw Davies | e-mail: Huw.Davies at kerberos.davies.net.au
Melbourne | "If soccer was meant to be played in the
Australia | air, the sky would be painted green"
9:09 a.m.
I see, this wouldn't be for high power servers, this is more of a way to
keep older machines like classic macs from being unusable in the future,
instead allowing an SD card or usb drive for example to be used in place of
a drive and given whatever special formats are needed. basically this would
just be a way of emulating a hard drive using flash memory.
On Sat, Nov 27, 2010 at 4:02 AM, jim s <jws at jwsss.com> wrote:
I have not seen much in the way of parallel SCSI
target hardware anywhere,
other than efforts mentioned earlier here to use various chips for target
emulation on PC's.
I would think that some effort with an FPGA could produce a fairly nice
10mhz scsi target, and if a PIC was attached to it with USB support, one
could write a program that could communicate with the PIC and host emulated
storage on the USB host.
I don't think the concept of USB storage does much for you, unless you
were to also include the storage on the USB /SCSI hardware device and make
the SCSI access the storage which the SCSI target would also be accessing.
That would be interesting, but not so useful either as that storage would
still not make sense to the host, unless you were using fairly modern disk
mappings and they could be read via current USB drivers from a host.
If you put say an SDram on the small dongle with the USB device and SCSI
target hardware, it would run fairly well though.
On 11/26/2010 2:57 PM, Huw Davies wrote:
> On 27/11/2010, at 6:52 AM, allison wrote:
>
> The other alternative is SCSI to USB as small USB 1gb to 16GB
>> thumb drives are dirt cheap and easy to use. The smaller end
>> of those are more in line with what those systems required.
>>
> Anyone know of such a device? There are plenty of devices to allow you to
> access a SCSI disk via USB but I haven't found one that works the other
> way....
>
> Huw Davies | e-mail: Huw.Davies at kerberos.davies.net.au
> Melbourne | "If soccer was meant to be played in the
> Australia | air, the sky would be painted green"
>
>
>
>
7:07 a.m.
On Fri, Nov 26, 2010 at 11:48, joe lobocki <jlobocki at gmail.com> wrote:
So what is our option? I have seen SCSI to IDE
adapters around, but they go
up into the $100's to $200's, say you have a minimum 10 machines, that
leaves you somewhere between $1,000 and $2,000 total, before the disk or
devices. If we could design a simple SCSI to IDE interface, we could be set
for a good long while on storage for these devices for a decent amount of
time. There are all sorts of adapters to IDE, there is CF to IDE, SD to
IDE,
SD to CF which could be placed in a CF to IDE adapter if need be, i'm sure
one could also rig up a USB drive to SCSI if one tried, but I could be
wrong.
I have previously proposed here a SCSI-to-SD interface; I've only done
preliminary brainstorming on it, but think it should not be too hard. The
SD card interface is very simple, both from a logic and electrical point of
view: if you don't care much about speed it can be done with SPI. (the speed
should be sufficient to keep up with 68k Mac, Amiga and Atari) All logic
can probably be done in a single AVR or even PIC chip, if you run that at
3.3V, the only additional hardware is the SCSI level shifters.
I do not know enough about low-level SCSI to start designing this yet, and
since I gave away my Amiga 3000 and classic Macs my interest in this project
has all but disappeared... still it'd be kinda fun, I think.
Joe.
--
Joachim Thiemann :: http://www.tsp.ece.mcgill.ca/~jthiem
7:26 a.m.
I have previously proposed here a SCSI-to-SD
interface; I've only done
preliminary brainstorming on it, but think it should not be too hard. The
SD card interface is very simple, both from a logic and electrical point
of
view: if you don't care much about speed it can be done with SPI. (the
speed
should be sufficient to keep up with 68k Mac, Amiga and Atari) All logic
can probably be done in a single AVR or even PIC chip, if you run that at
3.3V, the only additional hardware is the SCSI level shifters.
After seeing this project http://micha.freeshell.org/ramdisk/index.php
I'm very inclined to at least TRY something on this area. Seems simple
enough for trying.
8:35 a.m.
On Sat, Nov 27, 2010 at 10:26, Alexandre Souza - Listas <
pu1bzz.listas at gmail.com> wrote:
I have previously proposed here a SCSI-to-SD
interface; I've only done
Yes, I have been looking at that before I got busy with "real life" (is
academia real? discuss...) I figured that chopping off the second (RAM
controller) AVR and add SD-card smarts to the SCSI control AVR might be
possible.
Joe.
--
Joachim Thiemann :: http://www.tsp.ece.mcgill.ca/~jthiem
preliminary brainstorming on it, but think it
should not be too hard. The
SD card interface is very simple, both from a logic and electrical point
of
view: if you don't care much about speed it can be done with SPI. (the
speed
should be sufficient to keep up with 68k Mac, Amiga and Atari) All logic
can probably be done in a single AVR or even PIC chip, if you run that at
3.3V, the only additional hardware is the SCSI level shifters.
After seeing this project http://micha.freeshell.org/ramdisk/index.phpI'm very
inclined to at least TRY something on this area. Seems simple
enough for trying.
28 Nov
28 Nov
11 p.m.
I have a row of Micro PDP11's waiting on RD52/53 hard disks.
I'd call them ST506 interface. I presume MFM means much the same thing.
Two cables one for control and one for data.
Regards
?
Rod Smallwood
-----Original Message-----
From: cctech-bounces at classiccmp.org [mailto:cctech-bounces at classiccmp.org]
On Behalf Of Joachim Thiemann
Sent: 27 November 2010 15:07
To: General Discussion: On-Topic and Off-Topic Posts
Cc: cctech at classiccmp.org
Subject: Re: SCSI to IDE
On Fri, Nov 26, 2010 at 11:48, joe lobocki <jlobocki at gmail.com> wrote:
So what is our option? I have seen SCSI to IDE
adapters around, but they
go
up into the $100's to $200's, say you have a
minimum 10 machines, that
leaves you somewhere between $1,000 and $2,000 total, before the disk or
devices. If we could design a simple SCSI to IDE interface, we could be
set
for a good long while on storage for these devices for
a decent amount of
time. There are all sorts of adapters to IDE, there is CF to IDE, SD to
IDE,
SD to CF which could be placed in a CF to IDE adapter if need be, i'm sure
one could also rig up a USB drive to SCSI if one tried, but I could be
wrong.
I have previously proposed here a SCSI-to-SD interface; I've only done
preliminary brainstorming on it, but think it should not be too hard. The
SD card interface is very simple, both from a logic and electrical point of
view: if you don't care much about speed it can be done with SPI. (the speed
should be sufficient to keep up with 68k Mac, Amiga and Atari) All logic
can probably be done in a single AVR or even PIC chip, if you run that at
3.3V, the only additional hardware is the SCSI level shifters.
I do not know enough about low-level SCSI to start designing this yet, and
since I gave away my Amiga 3000 and classic Macs my interest in this project
has all but disappeared... still it'd be kinda fun, I think.
Joe.
--
Joachim Thiemann :: http://www.tsp.ece.mcgill.ca/~jthiem
29 Nov
29 Nov
4:34 a.m.
On 11/29/2010 02:00 AM, Rod Smallwood wrote:
I have a row of Micro PDP11's waiting on RD52/53
hard disks.
I'd call them ST506 interface. I presume MFM means much the same thing.
Two cables one for control and one for data.
RD52 is the quantum D540 and the 53 is the Micropolos 1325.
You can also use the ST225 (20mb) as a RD31 as well as ST512 (10mb).
Yes, that St506 is the general MFM pattern drive interface.
The alternate for Qbus PDP-11 or microVAX is one of the SCSI boards.
Any board that does MSCP to media(any media) will work.
Allison
5:24 a.m.
On Nov 29, 2010, at 7:34 AM, allison <ajp166 at verizon.net> wrote:
Q540 actually
I have a row
of Micro PDP11's waiting on RD52/53 hard disks.
I'd call them ST506 interface. I presume MFM means much the same thing.
Two cables one for control and one for data.
RD52 is the quantum D540 and the 53 is the Micropolos 1325.
You can also use the ST225 (20mb) as a RD31 as well as ST512 (10mb).
The latter (RD32) is an ST-251.
-Dave
--
Dave McGuire
Port Charlotte, FL
27 Nov
27 Nov
9:27 a.m.
On 26 Nov 2010 at 10:48, joe lobocki wrote:
Ok, so I have been doing research, and this affects
everyone here who
collects SGI, apple, and anything using scsi. in the near future, the
supply of smaller SCSI drives will eventually dry up, due to failure
from age and lack of supply, as I believe they are no longer
manufactured, unless for the industrial/server market. I could be
wrong on that part, but I take it as such because my searching for
them brings up nothing new. what does that mean for collectors?
In my experience, SCSI drives are still plentiful and show no
imminent sign of disappearing. With a bit of looking, even MFM
drives aren't hard to locate and they've been obsolete for more than
20 years. Aside from PATA drives, I have more SCSI hard drives than
any other type.
So perhaps your reaction is a bit premature...
--Chuck
9:53 a.m.
On 11/27/2010 09:27 AM, Chuck Guzis wrote:
mIn my experience, SCSI drives are still plentiful and
show no
imminent sign of disappearing. With a bit of looking, even MFM
drives aren't hard to locate and they've been obsolete for more than
20 years. Aside from PATA drives, I have more SCSI hard drives than
any other type.
PATA drives, and many but not all SCSI drives, are out at the far end of
the bathtub curve. This means that their reliability is now very
questionable. Contrary to popular belief, the MTBF figures in the drive
specifications do not give any information about expected longevity of a
single drive, so an MTBF of 200,000 hours does not mean that a single
drive is expected to last 22.8 years. The MTBF only applies during the
design life of the drive, which is usually five years. Beyond that,
you're living on borrowed time.
10:36 a.m.
On 27 Nov 2010 at 9:53, Eric Smith wrote:
PATA drives, and many but not all SCSI drives, are out
at the far end
of the bathtub curve. This means that their reliability is now very
questionable. Contrary to popular belief, the MTBF figures in the
drive specifications do not give any information about expected
longevity of a single drive, so an MTBF of 200,000 hours does not mean
that a single drive is expected to last 22.8 years. The MTBF only
applies during the design life of the drive, which is usually five
years. Beyond that, you're living on borrowed time.
I fail to see what you're driving at. Most of my SCSI drives were
purchased new between, oh, 15-20 years ago and all have fewer than
1000 hours on them.
My situation can scarcely be unique.
--Chuck
12:18 p.m.
----- Original Message -----
From: "Chuck Guzis" <cclist at sydex.com>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Saturday, November 27, 2010 1:36 PM
Subject: Re: SCSI to IDE
On 27 Nov 2010 at 9:53, Eric Smith wrote:
I purchased a case of 20 pcs of 2GB Quantum 50 pin SCSI drives that were
never used from a guy on craigslist. Some tech school had purchased a bunch
many years ago to show the students how to set up a SCSI RAID and this box
was never opened and sat in storage until it was sold for the massive price
of $15 total. Yea, I get lucky once in a while.
Decent working 50 pin SCSI drives in the 40-500MB range are probably getting
hard to find these days. Quite a few were in old macs that got recycled by
now, so if you happen to have one it is probably in a machine. The older
5.25" drives are harder to find. 500-2GB can still be found if you look hard
enough. Anything over 2gb should be common, especially in 68 pin. Last few
years there has been a flood of cheap SCA drives around from decommissioned
servers in the 9-72GB range.
IDE drives under 500 MB are getting harder to find, but 2GB and larger
(which is what you would probably use to replace a SCSI drive if you had an
adapter) will be common for a while. Snooping around a recycler for any
length of time will show you how common IDE drives are (and how quickly they
are getting stripped for recycling). I was lucky enough to be able to take a
huge box of IDE drives home from a scrapper (all under 2GB, mostly under
200MB), test them for bad sectors and keep the ones I wanted then return the
discards for a very cheap price (about 50% made the cut out of 50).
What I find to be rare are older laptop hard drives (they seem to not last
very long), especially the early PS/2 ESDI laptop drives and early Apple
SCSI laptop drives. Kind of hard making an adapter that will fit in a
laptop. Even laptop drives under 10GB are not as common as you would think
(outside of a machine).
MFM drives can still be found, but they are not that cheap (many people are
using SCSI or IDE cards in machines that were built with MFM). ESDI not so
common, same with HVD drives.
PATA drives, and many but not all SCSI drives,
are out at the far end
of the bathtub curve. This means that their reliability is now very
questionable. Contrary to popular belief, the MTBF figures in the
drive specifications do not give any information about expected
longevity of a single drive, so an MTBF of 200,000 hours does not mean
that a single drive is expected to last 22.8 years. The MTBF only
applies during the design life of the drive, which is usually five
years. Beyond that, you're living on borrowed time.
I fail to see what you're driving at. Most of my SCSI drives were
purchased new between, oh, 15-20 years ago and all have fewer than
1000 hours on them.
My situation can scarcely be unique.
--Chuck
3:42 p.m.
Chuck wrote:
I fail to see what you're driving at. Most of my
SCSI drives were
purchased new between, oh, 15-20 years ago and all have fewer than
1000 hours on them.
My situation can scarcely be unique.
The MTBF is a statistical measure or prediction of the behavior of a
population of drives (not a single drive!) during the design life
(typically five years). The MTBF rating of drives you purchased 15-20
years ago means absolutely NOTHING at this late date. If the drives
keep working, that's great, but you don't have any particular basis for
expecting them to do so for any length of time.
Eric
4:21 p.m.
On 27 Nov 2010 at 15:42, Eric Smith wrote:
The MTBF is a statistical measure or prediction of the
behavior of a
population of drives (not a single drive!) during the design life
(typically five years). The MTBF rating of drives you purchased 15-20
years ago means absolutely NOTHING at this late date. If the drives
keep working, that's great, but you don't have any particular basis
for expecting them to do so for any length of time.
For those wondering about "bathtub curves" see the following very
good paper:
http://db.usenix.org/events/fast07/tech/schroeder/schroeder_html/index
.html
Which essentially sheds a lot of light on the MTTF issue (which
appears to be part snake oil). The far end of the bathtub curve is
characterized as the "wearout" end, just as it would be with, say, an
automobile. It makes a lot of difference if you drive a 1957 Chevy
Belair with 5,000 miles on the clock and one with 750,000 miles.
True, components such as tires will fail due to their nature, but
which would you rather have?
So time spent spinning is quite germane in a disk drive.
But as I mentioned, MTTF on hard disks largely seems to be a
statistical trick not particularly observed by reality--and the
Schroeder-Gibson paper.
New drive or old, you pays your money and takes your chances. But my
money will be on the low-mileage units.
But the OP speculated that SCSI drives will become unobtainium. My
point was that the possibility is very remote, at least during my
lifetime--and that not all drives have been run to death.
Now, wanting a cheap substitute is a different matter--and we should
be candid about this. I paid about $1000 for 330MB CDC 5.25" SCSI
drive sometime around 1989 and I thought I was getting a heckuva
deal. If there were a strong demand for the same drive at the same
price today (adjusted for inflation), I'd imagine that an
enterprising soul could find a ready supply of working drives, even
if they had to be remanufactured.
--Chuck
4:34 p.m.
----- Original Message -----
From: "Chuck Guzis" <cclist at sydex.com>
To: "General Discussion: On-Topic and Off-Topic Posts"
<cctalk at classiccmp.org>
Sent: Saturday, November 27, 2010 7:21 PM
Subject: Re: SCSI to IDE
Now, wanting a cheap substitute is a different
matter--and we should
be candid about this. I paid about $1000 for 330MB CDC 5.25" SCSI
drive sometime around 1989 and I thought I was getting a heckuva
deal. If there were a strong demand for the same drive at the same
price today (adjusted for inflation), I'd imagine that an
enterprising soul could find a ready supply of working drives, even
if they had to be remanufactured.
--Chuck
Not to be a party pooper, but technology and materials have changed quite a
bit in the last 30 years of hard disk manufacturing. Exactly how easy would
it be for somebody to remanufacture that 330MB 5.25" SCSI drive (assuming
something is wrong with it like bad motor, crashed heads, etc) with current
methods?
4:46 p.m.
Not to be a party pooper, but technology and materials
have changed quite a
bit in the last 30 years of hard disk manufacturing. Exactly how easy would
it be for somebody to remanufacture that 330MB 5.25" SCSI drive (assuming
something is wrong with it like bad motor, crashed heads, etc) with current
methods?
An excellent point. I think it is completely possible to do basement
hard disk surgery with drives of this density (although semimodern
devices might throw up roadblocks, since there is magic in them).
Sure, their is precision machined parts and tight tolerances, but
these days you can get extremely nice machine tools for a fraction of
their original cost. Likewise, surplus cleanroom equipment pops up
from time to time.
Come to think of it, Ian and I did hard disk surgery on just such a
disk - a Zenith PC built up and badged as a CDC Cyber console. It had
stickion problems, if I recall properly, and the drive was popped open
on his kitchen table and unstuck. And the drive still works, even
after a full surface scan.
--
Will
5:24 p.m.
Chuck wrote:
So time spent spinning is quite germane in a disk
drive.
Obviously.
However, there are failure mechanisms that are not related to spinning,
so having an old drive with few operating hours is NOT a guarantee that
it will work reliably now.
Anecdotal evidence suggests that leaving a drive non-operational for a
long time is bad, and that spinning it for a few hours a month may
actually extend its life.
Eric
6:17 p.m.
On 27 Nov 2010 at 17:24, Eric Smith wrote:
Anecdotal evidence suggests that leaving a drive
non-operational for a
long time is bad, and that spinning it for a few hours a month may
actually extend its life.
Aw, and I was about to propose a modest wager. I have a system with
a 20-year old hard drive that gets powered up for a short time every
few months. Over the next, say, 3 years, I'm willing to pay you $20
for every time the drive doesn't spin up when powered on once each
month if you'd pay me $20 for every time it did.
That's probably not a bet you'd like to take, I'm guessing.
I have a lot of comparatively old equipment. It's not used a lot,
but it is used every so often.
It seems that many hard drives (care to buy a JTS hard drive?)
exhibit early mortality; survivors tend to keep plugging away if
given the juice every once in awhile.
--Chuck
28 Nov
28 Nov
6:41 p.m.
On 11/27/2010 09:17 PM, Chuck Guzis wrote:
On 27 Nov 2010 at 17:24, Eric Smith wrote:
Well, lessee I have that system with a pair of Quatum D540s aka DEC RD52s
The system runs about once every few weeks and has done that since 1990
with those drives. It has not failed yet. My microVAXs have a bunch of
the
seagate 1.07gb drives that were revered for their longevity that are
sill operable
15 years since I got them and they were NOT new then.
Now in another place I had a set of servers one was populated with
a pair of MAXTOR 9GB UWscsi drives.. in actual use the mttf was 11months
(24x7) and in three years all four of those drives died. They were
replaced
with Seagate barcudas, two years later all four drives going strong.
Some drives were failure prone (anyone remember the Bigfoot drive?)
and some tended to had odd problems (micropolos 1325 with heads sticking
to the return bumper). And some seemingly similar drives Segate ST225
and ST250 had amazingly different reliability. I Have all my ST225s and
they
all work and none are less than 20 years old, most of the st250s never made
it to 5 years with intermittent use.
The point if not registering yet, is some drives were terrible (JTS and
bigfoot),
and some were not. A few like the seagate ST3660 were junk but the 3660A
was unbreakable, Seagate got a black eye on the early ones. I still have 4
low time 3660As in storage and two in use still.
Over the years I sought out used drives that were known to be better than
average and those that weren't known were used to their death. Some drives
will meet or exceed their MTBF and some don't. Of those that don't my
experience
is as a family they didn't. My collection of spares include those I
know from
experience have a better than average life.
However none of the standing drives I keep as spares have failed me and I do
cycle through them on occasion for archival reasons. They are spares and
also backups.
Oh, and speaking of drives said to be total crap I have a Syquest 270mb
removable cartridge drive (in an syquest parallel to IDE adaptor case)
and thirteen carts for it that is still in occasional use. Story is
those had
a failure rate of 99% in less than 6months. I also have a functional
500mb JTS
and a 2GB bigfoot that I've been waiting on to die forever. Must be lucky.
Allison
Anecdotal evidence suggests that leaving a drive
non-operational for a
long time is bad, and that spinning it for a few hours a month may
actually extend its life.
Aw, and I was about to propose a modest wager. I have a system with
a 20-year old hard drive that gets powered up for a short time every
few months. Over the next, say, 3 years, I'm willing to pay you $20
for every time the drive doesn't spin up when powered on once each
month if you'd pay me $20 for every time it did.
7:33 p.m.
On 28 Nov 2010 at 21:41, allison wrote:
Oh, and speaking of drives said to be total crap I
have a Syquest
270mb removable cartridge drive (in an syquest parallel to IDE adaptor
case) and thirteen carts for it that is still in occasional use.
Story is those had a failure rate of 99% in less than 6months. I also
have a functional 500mb JTS and a 2GB bigfoot that I've been waiting
on to die forever. Must be lucky.
I've got 4 of the Quantum Bigfoots (of varying capacities) and only
one doesn't work--I think it's the 2GB CY version. I suspect a
problem in the electronics and not the HDA itself.
I've still got a Syquest Sparq removable cartridge drive that works
just fine as well as a couple of functioning Iomega Jaz drives. My
1GB JTS drive, on the other hand, worked fine for a couple of months,
then started a slow downhill trend of bad sectors. I've not checked
it for functionality in years.
It's the infant mortality that'll getcha.
--Chuck
27 Nov
27 Nov
10:59 p.m.
On Sat, Nov 27, 2010 at 7:24 PM, Eric Smith <eric at brouhaha.com> wrote:
Chuck wrote:
I can echo this, as I have ran servers for years straight, without a hard
drive failing, but I have pulled drives off the shelf and have them fail on
power up. one must also account for manufacturing defects that can manifest
themselves for any amount of time before causing a failure. environment also
plays a role, as simple events such as seasons changing between extreme hot
and extreme cold changes the structures of metal bearings with very tight
clearances, causing friction easily, and humidity changes along with
condensation also have an effect on drives. I am sure, even in a slight
vacuum at 72 degrees constant and being spun up every month or so a drive
would still fail.
So time spent spinning is quite germane in a disk
drive.
Obviously.
However, there are failure mechanisms that are not related to spinning, so
having an old drive with few operating hours is NOT a guarantee that it will
work reliably now.
Anecdotal evidence suggests that leaving a drive non-operational for a long
time is bad, and that spinning it for a few hours a month may actually
extend its life.
Eric
28 Nov
28 Nov
2:39 p.m.
On Sat, Nov 27, 2010 at 7:21 PM, Chuck Guzis <cclist at sydex.com> wrote:
?It makes a lot of difference if you drive a 1957
Chevy
Belair with 5,000 miles on the clock and one with 750,000 miles.
True, components such as tires will fail due to their nature, but
which would you rather have?
I can attest to this one... My '67 Chevy Impala
(http://www.chdickman.com/impala/67_Chevrolet_1.jpg) got a new set of
bias ply tires about 15 years ago. After only 5,000 miles I had a
tread separation last summer. Not so good on the interstate at
65MPH... It is a long way between exits in some parts of Ohio. Now I
am looking for some age appropriate whitewalls that are not from the
same batch as the ones that came apart.
I agree with Eric's science and I hope that Chuck's experience, which
I share as far as computer drives go, continues. I figure that things
like stick head syndrome will catch up with most old drives and they
will be done.
So where are these MFM drive sources that people keep talking about?
Most I have seen have incredible prices for working drives.
-chuck
10:57 a.m.
On 27 Nov 2010 at 15:58, Alexandre Souza - Listas wrote:
They are common in certain parts of world. Here in
Brazil I'm
looking for AGES for a SCSI zip drive. I can find, at $200+ prices,
since all of them are sold at premiun prices for
keyboard/synthesizer/sampler peop;e.
So the issue really is one of cost, not availability, correct?
--Chuck
29 Nov
29 Nov
12:16 a.m.
<<<< in the near future, the supply of smaller SCSI drives will eventually
dry up
That's why when a batch of 151 new ones popped up on eBay I bought them :-)
If anyone is trying to restore a machine I may part with one or two (I have
a few machines in the restoration list myself inc some servers that need
about 8 each) but I reside in Australia so postage should be factored in.
Contact me off list if you need some help restoring a genuine vintage
machine that needs a small SCSI drive but please note that I can't do much
until maybe March 2011 as its all packed up for a house move (again :-(
++++++++++
Kevin Parker
m: 0418 815 527
++++++++++
http://au.linkedin.com/in/krparker
-----Original Message-----
From: cctalk-bounces at classiccmp.org [mailto:cctalk-bounces at classiccmp.org]
On Behalf Of joe lobocki
Sent: Saturday, 27 November 2010 03:48
To: cctech at classiccmp.org
Subject: SCSI to IDE
Ok, so I have been doing research, and this affects everyone here who
collects SGI, apple, and anything using scsi. in the near future, the supply
of smaller SCSI drives will eventually dry up, due to failure from age and
lack of supply, as I believe they are no longer manufactured, unless for the
industrial/server market. I could be wrong on that part, but I take it as
such because my searching for them brings up nothing new. what does that
mean for collectors? it means that your neat little apple or SGI or etc will
run, until the scsi disk dies, then we will be left cannibalizing machines
for disks, paying ridiculous prices for leftover scsi disks, which will
skyrocket because of the small supply.
So what is our option? I have seen SCSI to IDE adapters around, but they go
up into the $100's to $200's, say you have a minimum 10 machines, that
leaves you somewhere between $1,000 and $2,000 total, before the disk or
devices. If we could design a simple SCSI to IDE interface, we could be set
for a good long while on storage for these devices for a decent amount of
time. There are all sorts of adapters to IDE, there is CF to IDE, SD to IDE,
SD to CF which could be placed in a CF to IDE adapter if need be, i'm sure
one could also rig up a USB drive to SCSI if one tried, but I could be
wrong.
Having said that, I know nothing about programming microcontrollers or
simple processors, but I have seen people on the board who do. I'm sure if
we can find a chip capable, and lay out a schematic, source code, and
possibly a board layout, members could source their own boards and
components to build their own devices at home (given they have a chip
programmer), so no one person has to take on the responsibility of supplying
parts, kits, finished devices unless they so choose to. This is just an
idea, if nothing becomes of it, no big deal, just putting it out there....
-Joe
4911
days inactive
4914
days old
27 comments
14 participants
participants (14)
-
ajp166＠verizon.net -
cclist＠sydex.com -
chd＠chdickman.com -
eric＠brouhaha.com -
huw.davies＠mail.vsm.com.au -
jlobocki＠gmail.com -
joachim.thiemann＠gmail.com -
jws＠jwsss.com -
mcguire＠neurotica.com -
pu1bzz.listas＠gmail.com -
rodsmallwood＠btconnect.com -
teoz＠neo.rr.com -
trash80＠internode.on.net -
wdonzelli＠gmail.com