Der Mouse wrote:
>> A windowed environment is a GUI, yes?
>
> Not necessarily. NetBSD - and presumably others - ships with
> window(1), which provides a text-mode windowing environment. Nothing
> graphical about it any more than any text "terminal" necessarily is.
Fair enough.
>> What's meant here is that the Chris M has a point: you *do* cheat
>> yourself of some functionality when using a non-windowed environment
>> "these days".
>
> I'm not sure "cheat" is really an appropriate word, but yes, you lose
> some functionality when you drop windowing.
Just going with the way things were phrased in the original post.
> Sometimes that functionality is irrelevant, or worth losing to get some
> other benefit. Sometimes it's not.
Exactly my point.
>> However, you also cheat yourself of functionality exactly by using a
>> windows environment. There are things that are much more efficient
>> if you don't have to wrestle the point-and-click interface.
>
> Windowing environments don't necessarily mean point-and-click. In my
> own X-based environment, for example, I can work productively for hours
> - and not just all in the same window, either - without touching the
> mouse. As you yourself said,
>
>> Look at *real* power users, even on windowed systems. They hardly
>> touch the mouse.
>
> It's not GUI environments that lose the functionality you're talking
> about; it's about a particular subclass of GUI environments that are
> designed - misdesigned, arguably - so as to compel their users to
> switch between keyboard and mouse comparatively frequently (on a
> timescale of seconds to minutes).
As I assumed was meant by the OP.
>> It's all keyboard shortcuts, and it's *way* faster. The downside is
>> having to master all those cryptic gestures and key combinations.
>
> Sounds to me as though you're talking about primarily point-and-click
> windowing environments with keyboard "shortcuts" grafted on, rather
> than environments designed from the ground up to be keyboard-driven.
> (That one particularly dominant windowing environment is an especially
> egregious example of this doesn't help....)
Yes, I was going by the *popular* definition of GUI/Windowed environment, not the strict one.
,xtG
tsooJ
>
>Subject: Re: ST506 WTB:Micropolis 1325
> From: "Steven N. Hirsch" <shirsch at adelphia.net>
> Date: Sat, 07 Apr 2007 08:42:34 -0400 (EDT)
> To: "General Discussion: On-Topic Posts Only" <cctech at classiccmp.org>
>
>On Thu, 5 Apr 2007, Sridhar Ayengar wrote:
>
>> John Kourafas wrote:
>>> Also looking for a Micropolis 1325 MFM Drive, 71/80MB , I've seen both the
>>> ST506 and Mic. 1325 on eBay for like 600.00 which I think is crazy...
>>
>> What's the largest capacity 3.5" MFM HDD available?
>
>I have an extensive collection of MFM/RLL drives and have _never_ run
>across one with that interface in a 3.5" form-factor. Not sure that
>anything of this sort existed. 5.25" 1/2-ht. was about as sophisticated
>as they got.
I have several miniscribe 20mb in all flavors of interface st412, SCSI
and IDE. I have some really old 3.5" WD 10mb. to name a few.
They show up around the beginning of the IDE era but by time IDE hit 40mb
they seemed to have disappeared.
Allison
Dave McGuire wrote:
> For the rest...*Someone* has to make the ruling, preferably
> leaning to the "not part of the OS" side more often than not, and the
> people who want everything handed to them on a platter will just need
> to bite the bullet and learn how to type "./configure;make;make
> install"...or go run another OS.
Hear, hear.
,xtG
tsooJ
Chuck Guzis wrote:
The "half good" memory dates back long before 256K chips--it goes all
the way back to 16K ones. I've got a batch of Intel 2109s that
differ from each other by a suffix -1 or -2. The idea is that the
suffix indicates which half is good; otherwise the chips are
electrically the same. I remember when the local Intel sales guy
dropped off a bag of the things on my desk and I sorted through them
and found a batch where all 16K worked in my application.
Cheers,
Chuck
------
Billy:
Actually it goes back further than that. Intersil offered the 6002 (a 2K
Dynamic RAM) with options of 1K good. I never saw it on smaller chips like
the 256 byte or the 1K. But it would not surprise me. This was the era
when a new state of the art memory chip was $15-20 each and in very short
supply. A half good part could be very useful in a terminal or small
system. By small, I mean 4K or 8K bytes, typical of many memory boards
available in the early 1970's.
Billy
On Wed Apr 11 00:44:13 CDT 2007 Ethan Dicks ethan.dicks at gmail.com wrote
>Are there any 'gotchas' with the ARS-2000IU?
> Max drive size?
> Strange SCSI packet issues?
> Will it work with really old SCSI implementations?
>Really new ones?
1. I haven't seen any gotchas yet, I've tried it with smaller 1gb-10 gb IDE drives no prob.
as I said before I needed the new bios to ecognize the CF card to IDE converter.
2. I haven't tried a big drive >128mb as I don't have any spare ones that big, but I'm getting
one in this weekend and I'll try it out and report.
3. No glitches with the SCSI at all (as far as I can tell)
4. I've tried it with several Adaptec cards right throiugh to some newer PCI ones and no glitches
so far.
I just got in an Acard AEC-7720U that I'm going to replace the bigger ARS-2000. I had one before
and it worked OK until I reversed the power supply and fried it. Yea, I know that shouldn't be possible
as it uses a standard floppy power connector, but I broke off the plastic tab and guessed wrong the next time.
It seems to have the same chipset so it should work the same as the ARS-2000. I'll be putting it into the
5155 in the next few days. If anything goes wrong, I'll post the results.
The AEC-7720U's are going for about $30 on eBay right now (two vendors seem to have a big
stock of them), they are the best deal IMO.
This was a message from Gary Fisher
I picked up one of the terminal units for the Computer Museum
and just took at look at whats inside. It is a repackaged Burroughs
B25, which is, as you noted, a Convergent 80186 system.
I'd be interested in seeing pics of the insides of the main
server unit.
Convergent OEMed this stuff to lots of people. I think the
Microdata 1000 may be one as well.
I found one of a five disc set today for what appears to be
CTOS rel 9.1-D for the 1000.
On 4/11/07, Jerome H. Fine <jhfinedp3k at compsys.to> wrote:
> But I once had a project that
> used a real DEC TU-58. Not the fastest "random"
> access device!!!!!!!!!!!!!!!!!!
They work better as "sequential" access devices - being long and thin
and travelling in one dimension, go figure. We used to optimize file
order on our console TU58s to speed up the boot times on our 11/725s
and 11/730s. Since the file order doesn't change, one just builds a
TU58 with EXCHANGE with each file following the other. The console's
8-bit-micro must cache the directory block, since the tape didn't whip
back to the start between each file.
Using unaltered console tapes from DEC resulted in, IIRC, about 15
minutes from turning the key to booting the hard disk. Replacing that
tape with one of our own devising shortened that pre-boot time to well
under 3 minutes.
I'd hate to rely on a TU-58 and no other block-addressable media on a
PDP-11, though. I survived a PDP-8 with a TD8E and TU56, but it was
somewhat tedious (cool to watch, though). TU-58s weren't as cool,
IMHO.
-ethan
I have no connection to the seller other than living in the same city
- and I don't collect Apple ][ computers, but I figured someone here
might appreciate the heads-up.
Looks like a IIc:
http://montreal.craigslist.org/sys/309660731.html
Though I am tempted to ask how much he wants just to play with it for
a few days and then pass it on to someone else... but I have other
things to keep me occupied (1581! Yay!) and I know nothing about Apple
8-bits...
Joe.
This email is short on questions; just a vintage computer putzing
report. I do have one question of opinion at the end.
I've had a northstar horizon 8/16 system for a year and a half now, but
I was able to finally get everything together and a bit of time to get
it running.
The 8/16 is a normal horizon (in the aluminum cabinet, not wooden), with
a beefier power supply. There may be some other mechanical differences,
such as many more punchouts in the back. The purpose of the system is
to host multiple CPUs running in a S-100 backplane, each with its own
local memory, using the Z80 down on the motherboard to act as a server
for the shared resources. As the 8/16 name implies, you can have 8b
(z80) or 16b (8086) CPUs, or a mix of them. Mine has two Z80 cards,
each with 64KB DRAM, in addition to the Z80 on the motherboard and 64KB
DRAM that is uses on the S-100 BUS.
The box has a single 5.25" floppy and a 30 MB Rodime hard drive (ST506
type interface). The way things are set up in the horizon, you can't
boot directly off the hard drive; the usual procedure is to boot the
floppy, and the floppy contains a bootstrap to load the OS from the hard
disk.
I first booted into HDOS from floppy, the hard disk version of NSDOS.
It has both non-destructive and destructive disk tests. "LI" shows that
there is no meaningful HDOS file system on the drive.
Next, I ran only the non-destructive test since one goal is to see what
is on the hard disk. HDOS, like NSDOS, has a command for reading
arbitrary absolute sectors from the floppy, but it isn't supported on
the hard drive -- instead you can load sectors relative to a named file,
which doesn't help me here.
Next I booted turbodos from floppy. "DIR" shows no meaningful file
system on either of the two partitions on the drive.
Finally, I'm at an impasse. I assume that this system had *something*
on the drive, although I suppose a previous owner did a FORMAT on the
drive before passing the system on. It would be easiest for me to just
format the drive and install either HDOS, or more likely, TurboDos and
get on with it. If I had more time I'd look into finding a mechanism to
read the hard drive sector by sector and make a copy, but the reality is
I have more projects than I have time for, so this seems unlikely.
Does anybody who has read this far have an idea what to do next? Format
and reinstall? Write my own driver to dump the disk first? Find an old
PC with a controller card that could interface to the drive?
>
>Subject: Re: Some progress with my PDP-11/73 system
> From: David Betz <dbetz at xlisper.com>
> Date: Tue, 10 Apr 2007 08:58:01 -0400
> To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at classiccmp.org>
>
>>> I soldered it back together but both drives still fail. At least
>>> one of the drives spins up but then back down again.
>>
>> That can happen if the heads are stuck against the end stop inside
>> the drive. There's a rubber bumper that goes sticky. Alison
>> posted more information about that just a few days ago.
>
>Yes, I saw those messages. At this point I'm not up for opening up
>the drive to attempt a fix though. I guess if one of my good ones
>stops working I may change my mind. Instead, I'm starting to look
>into what is necessary for a minimal RT-11 bootable RX50 diskette.
>Unfortunately, that's the only medium I have for moving things to the
>hard drive on the PDP-11.
To boot a floppy under RT-11 is fairly easy. You will need a bootable
disk, That is a copy of RT-11 with the bootdevice you plan to use configured.
It also must have the hard disk driver as well and basic utilities. RT-11
fits on small devices (even tu58 256kB).
FYI: an option is to use TU-58 emulator on a PC and a serial line to
a PDP-11 as a fairly straight forward way to get stuff on a an -11
with blank media. At the extreme the DD boot can be hand entered
into uODT as its something like 30 words long. I know I used to boot
a romless 11/23 that way.
Allison
