Finally, Death of the 3.5 inch floppy disk
in [Embedded]
|
From: David Brown on 29 Apr 2010 02:15 Mark Borgerson wrote: > In article <OOednZWhLLmMEkXWnZ2dnUVZ8nCdnZ2d(a)lyse.net>, > david.brown(a)hesbynett.removethisbit.no says... >> Doug McIntyre wrote: >>> John Tserkezis <jt(a)techniciansyndrome.org.invalid> writes: >>>> Spehro Pefhany wrote: >>>>>> The only way to load device drivers (drive interfaces, SCSI drivers >>>>>> etc) when installing windows is via the drive at A:. And that's your >>>>>> only option. >>>>> My latest machine lacks floppy support on the motherboard (Asus P6T >>>>> WS). They suggest using a USB flash drive or USB floppy for RAID >>>>> drivers. >>>> That's nice, but USB flash drives won't ever map to A: or B:. This is >>>> done intentionally, and it makes perfect sense. But it doesn't help the >>>> fact that Windows will not look at *any* other drive than A:. >>> >>> I have a Flash Drive that mimics part of its space as a USB Floppy >>> that *does* map to drive A: or B:. Unfortunately it doesn't work very >>> well with most systems. :( >>> >>> >>>> So, that leaves USB interfaced FDDs, or, as already suggested, creating >>>> an alternative boot disk with the drivers included. >>> Of course this is all only just for WinXP (ie. that Windows release >>> from 8 years ago), or Server 2003 from 7 years ago.. >>> >> Unfortunately, XP is still the best version of windows for many uses. >> Lots of companies feel they don't have time to waste testing for >> compatibility with Win 7, or finding drivers for it, or re-training >> staff, or handling the support. It's better with the devil they know. >> Besides, Win 7 has no advantages over XP if you are actually /using/ the >> computer, rather than admiring the pretty clock on the desktop. > > That's not exactly true if you're running the 64-bit version of Win 7. > It does allow you to use more memory effectively. The downside is > that it does require signed drivers---some of which weren't immediately > available. About the only application I use that needs that much memory > is Matlab. There's a 64-bit version of XP - I'm running it now. But to be fair, 64-bit XP is an oddity that few people have and few developers test for, while the 64-bit support for Win7 is much more mature (it's now almost as well supported as 64-bit in Linux ten years ago). The application I see that needs lots of memory is for virtual machines - it's good to have a 64-bit host and lots of GB's if you want to run several VMs at the same time. Requiring signed drivers, however, makes a system pretty much useless for embedded development work - you don't get signed drivers for the dozens of hardware debuggers, cards, and other bits and pieces that you need. I seem to remember there being ways around the driver signing requirements, however. >>> Vista & Win7/Server 2008 either release have methods to read in >>> RAID/HBA drivers off flash or USB devices during installation while >>> booted into WinPE. And its easy to make a new WinPE boot environment >>> with said drivers if needed. >> I had to install windows (XP and Win 7) on a couple of computers >> recently - it is often faster to install Windows from scratch than to >> start using a typical "pre-installed" system (after it takes ages to >> install windows from a hidden partition, you then have to waste more >> time removing all traces of the "demo" and time-limited junk that comes >> with system). While Win 7 installation is mildly improved over XP, it's >> still seriously inefficient. And once you have the basic system >> installed, you then have to find and install the drivers - which are >> often totally absurd (I had to download a 100 MB file for an Ethernet >> driver, including it's useless utilities - and it wouldn't even install >> until I'd added dotnet runtimes!). > > I'd like to give the originator of .net a piece of my mind----for about > as much time and memory as it has cost me! >> The Windows developers really should get hold of a few Linux >> distributions to see how OS installers /should/ be made - they have a >> decade or so catching up to do. >> > Good point. I've installed Ubuntu several times---and it has always > been pretty straightforward. > Linux distros have worked hard to learn from the best aspects of Windows - it would be good if MS tried to emulate some of the best /important/ features of Linux (they copied a lot of KDE's appearance when making Aero - but that is only skin-deep and totally irrelevant when you are actually using the machine).
From: Paul Carpenter on 29 Apr 2010 03:53 In article <83rugtFvq4U1(a)mid.individual.net>, invalid(a)invalid.invalid says... > Spehro Pefhany wrote: > > On Wed, 28 Apr 2010 13:15:21 -0700, Joerg <invalid(a)invalid.invalid> > > wrote: > > > >> Spehro Pefhany wrote: > >>> On Wed, 28 Apr 2010 11:49:10 -0700, Joerg <invalid(a)invalid.invalid> > >>> wrote: > >>> > >>> > >>>> Yes, and that wear is clearly visible. However, the typical disk is used > >>>> as file storage and only once in a while read back, and then only small > >>>> parts of it. > >>> I have an old HP logic analyzer that boots off of a floppy. > >>> > >> AFAIK there's also plenty of scopes from Tek and others where that's the > >> only way to get screen shots over to your PC. Unless you bought the now > >> pretty much unobtanium GPIB interface for beaucoup $$$. But mostly I see > >> that with production machines. One floppy slot and absolutely zilch in > >> terms of other interfaces. CNC gear become almost useless without being > >> able to feed data into it. > > > > Many, maybe most, of them have an old-fashioned serial interface too, > > for which people have cobbed together interfaces so that they can be > > controlled from a central point. There are half a dozen, from several > > different suppliers, in a college machine shop that I'm familiar with- > > used for teaching CNC machining. > > > > Occacionally I have been asked to take a look at a machine shop. Mainly > because it gets messy in there and they'd rather not carry disks around > and worst case get a splotch of gunk or metal chafings into a drive > (happened to me once). But usually there was only one or two of the > machines that had RS232, sometimes none. Some machine shops have very old CNC equipment, last year I had an enquiry to find a spare PDP 11/73 card for one, that MIGHT have had a serial but I dread to think what format of media it might have had. -- Paul Carpenter | paul(a)pcserviceselectronics.co.uk <http://www.pcserviceselectronics.co.uk/> PC Services <http://www.pcserviceselectronics.co.uk/fonts/> Timing Diagram Font <http://www.gnuh8.org.uk/> GNU H8 - compiler & Renesas H8/H8S/H8 Tiny <http://www.badweb.org.uk/> For those web sites you hate
From: Albert van der Horst on 29 Apr 2010 06:40 In article <7dtgt5d4gen2icp8qql138b9mmf6t81m0k(a)4ax.com>, George Neuner <gneuner2(a)comcast.net> wrote: >On Tue, 27 Apr 2010 18:55:51 -0700, Joerg <invalid(a)invalid.invalid> >wrote: > >>Stuart Longland wrote: >>some of the [diskettes] I have at home are slowly decaying with age. >> >>Interesting. What's decaying about them? I've got Fuji MF2HD from the >>90's and they still work fine. > >The media does not have high enough coercivity to retain magnetic >alignment indefinitely - given enough time it loses orientation and >your data simply fades away. >And unlike hard disks, diskette R/W heads actually touch the recording >surface and gradually wear away the media. This doesn't explain why virtually pristine disks (written only once and a visibly impeccable surface) have difficulties reading and are almost impossible to reformat. I go with those saying that the plastic in the surface deteriorates. And it seems to me that double density, and especially single density disk are more reliable. I managed to recover most from Osborne CP/M disks with visibly damaged surfaces, and used very intensively. (Remember those CP/M machines had no hard disk. Floppies were even used for -- small -- databases. ) > >George Groetjes Albert -- -- Albert van der Horst, UTRECHT,THE NETHERLANDS Economic growth -- being exponential -- ultimately falters. albert(a)spe&ar&c.xs4all.nl &=n http://home.hccnet.nl/a.w.m.van.der.horst
From: GregS on 29 Apr 2010 09:50 In article <83re9eFvhoU1(a)mid.individual.net>, news(a)analogconsultants.com wrote: >George Neuner wrote: >> On Tue, 27 Apr 2010 18:55:51 -0700, Joerg <invalid(a)invalid.invalid> >> wrote: >> >>> Stuart Longland wrote: >>> some of the [diskettes] I have at home are slowly decaying with age. >>> >>> Interesting. What's decaying about them? I've got Fuji MF2HD from the >>> 90's and they still work fine. >> >> The media does not have high enough coercivity to retain magnetic >> alignment indefinitely - given enough time it loses orientation and >> your data simply fades away. > > >Hmm, I have disks dating back to 1990 and none of them has ever lost >data or caused read errors. But some posters said that they still can >have lost writeability. No idea why. > >I did always make sure to never buy disks from dubious sources but >always the good stuff, name brands. > > >> And unlike hard disks, diskette R/W heads actually touch the recording >> surface and gradually wear away the media. >> > >Yes, and that wear is clearly visible. However, the typical disk is used >as file storage and only once in a while read back, and then only small >parts of it. > I have tried using some old 3.5's and had some pretty bad luck. I still have people that want to read the larger floppies, and try finding adaptors for that. Speaking of machine shops, one CF card got currupt, and to fix that, required replacing everything. greg
From: Joerg on 29 Apr 2010 12:07
Paul Carpenter wrote: > In article <83rugtFvq4U1(a)mid.individual.net>, invalid(a)invalid.invalid > says... >> Spehro Pefhany wrote: >>> On Wed, 28 Apr 2010 13:15:21 -0700, Joerg <invalid(a)invalid.invalid> >>> wrote: >>> >>>> Spehro Pefhany wrote: >>>>> On Wed, 28 Apr 2010 11:49:10 -0700, Joerg <invalid(a)invalid.invalid> >>>>> wrote: >>>>> >>>>> >>>>>> Yes, and that wear is clearly visible. However, the typical disk is used >>>>>> as file storage and only once in a while read back, and then only small >>>>>> parts of it. >>>>> I have an old HP logic analyzer that boots off of a floppy. >>>>> >>>> AFAIK there's also plenty of scopes from Tek and others where that's the >>>> only way to get screen shots over to your PC. Unless you bought the now >>>> pretty much unobtanium GPIB interface for beaucoup $$$. But mostly I see >>>> that with production machines. One floppy slot and absolutely zilch in >>>> terms of other interfaces. CNC gear become almost useless without being >>>> able to feed data into it. >>> Many, maybe most, of them have an old-fashioned serial interface too, >>> for which people have cobbed together interfaces so that they can be >>> controlled from a central point. There are half a dozen, from several >>> different suppliers, in a college machine shop that I'm familiar with- >>> used for teaching CNC machining. >>> >> Occacionally I have been asked to take a look at a machine shop. Mainly >> because it gets messy in there and they'd rather not carry disks around >> and worst case get a splotch of gunk or metal chafings into a drive >> (happened to me once). But usually there was only one or two of the >> machines that had RS232, sometimes none. > > Some machine shops have very old CNC equipment, last year I had an > enquiry to find a spare PDP 11/73 card for one, that MIGHT have had > a serial but I dread to think what format of media it might have had. > I had to coach someone through repair and calibration of a circuit board test bed from the 80's. All nicely DOS-based so it worked right off the bat :-) -- Regards, Joerg http://www.analogconsultants.com/ "gmail" domain blocked because of excessive spam. Use another domain or send PM. |