The collected works of ClassicHasClass - Page 15

That's actually a really good idea ...

The CoCla is a soft power on system, assuming that the rear switch is on. I would look at the PRAM battery. If the PRAM battery is dead, it will not generate video. There is no CUDA switch on the CoCla.

I concur with robespierre: don't mess with the motherboard like that. You will eventually kill it, if you haven't already.

I note, FWIW, that when my CC board's caps failed, there was no visible issue.

But the PRAM battery will be easier to fix and I'd do that first. The CC boards last longer if you leave the rear switch on and use soft-power to start it and shut it down.

The CoCla never had such a switch on any production run.

The weird oddball of the Workgroup Server line, a 601 class Power Macintosh crammed into a Quadra 950/Apple Workgroup Server 95-type case. It has quite a strange history -- a prototype of it ran Apple's brief, sorrowful flirtation with NetWare, which went over like a lead balloon with testers and was repurposed to run MacOS instead.

There aren't many of them out there; this was a chance eBay find. It was incredibly grimy, but cleaned up nice, and boots into Mac OS 7.6. I'm thinking about slapping MkLinux on it for yuks.

http://www.floodgap.com/retrobits/ans/9150/

The IIgs keyboard works fine with Macs. As I mentioned above, my CC motherboard with bad caps had no visible failed ones either. It's to the point where just about every 68030-based Mac's capacitors have blown by now -- my IIsi, IIci and SE/30 all needed recaps (the CC was replaced with a Mystic LC575 motherboard instead).

Tried MkLinux on it over the weekend. It's, uh, incredible that it runs at all, I suppose.

Screenshots added to the link above.

I can't imagine it's substantially un like AIX on RS/6000 and PPC, though.

Admittedly I have plenty of AIX iron around here to play with.

Second vote for Charles. My SE/30 and IIsi were recapped by him, and he's working on two IIcis for me. Considering the fiddly work involved, his rates are very reasonable.

Smoke signals.

Or, maybe this: http://www.theregister.co.uk/2014/09/15 ... ing_world/

No, I know all that; my SR-7100 is set up that way. This one I just wanted to see how the stock hardware performed.

You're more cooked if you're on a system where /bin/sh == /bin/bash. OS X is such a system. I quickly built a standalone bash for 10.4+ PPC/Intel if you want one of those.

http://tenfourfox.blogspot.com/2014/09/ ... -bash.html

But I think IRIX is very low risk.

Vulnerability not fully fixed.
https://bugzilla.redhat.com/show_bug.cgi?id=1141597#c23

Second patch seems to pass muster:

http://seclists.org/oss-sec/2014/q3/734

I updated the OS X universal bash already (10.4-10.9, PPC and x86).
http://tenfourfox.blogspot.com/2014/09/ ... dated.html

What are you saying, VP? That the floppy makes your internal drives hard?

Thank you, I'll be here all week.

Hey, I'm just staying classy.

Remember that the CoCla needs System Enabler 401 as well. See this handy chart: http://support.apple.com/kb/TA47341

Oh, FFS.

http://tenfourfox.blogspot.com/2014/09/ ... -4327.html

4.3.28 is out, and the 10.4+ universal binary is updated, which should fix all five CVEs finally.

http://tenfourfox.blogspot.com/2014/09/ ... dated.html

They were called MPTasks but they were really a very primitive type of thread. They worked very well for a specific purpose, but they could not interact with the GUI, and essentially were a function of the PowerPC nanokernel. I used a limited version of them in Classilla for timers.

Right, but in classic Mac OS, all UI stuff was cooperatively multitasked, so any component of an application in the "blue task" (as it was called) that didn't yield could still hang up the system. You can't do that in OS X.

TeamBlackFox wrote: Has Classic given up on Classilla? Or is TenFourFox just taking up all of your time?

Mostly the latter. Classilla's layout engine is probably hopeless, but I am trying to get it to accept SHA-2 certificates and brush up its SSL a bit.

gxemul does use a lot of CPU. It's just a very heavyweight emulator with no dynamic code compilation that I know of.

That's exactly what I mean. That's exactly what it's doing on my G5 right now (it's spreading it over four cores, but it's clearly pegging one).

The Commodore 1802 (and its close relative the CM-141) is a piece of crap. It's actually a substandard composite monitor that cheats by taking the monochrome signal off the DB-9 RGB port a 128 would have. You don't have a 128, so the pathetic monochrome RGB mode it does offer is worthless to you, and the monitor will fail if you look at it wrong. Find a Commodore 1702 instead, which is a superior and indestructible composite monitor suitable for any classic computer.

If you do end up getting a 128, then I'd recommend either a 1084 or a 1902, not the 1802, both of which have a "real" digital RGB mode.

sgifanatic wrote:

ClassicHasClass wrote: The Commodore 1802 (and its close relative the CM-141) is a piece of crap. It's actually a substandard composite monitor that cheats by taking the monochrome signal off the DB-9 RGB port a 128 would have. You don't have a 128, so the pathetic monochrome RGB mode it does offer is worthless to you, and the monitor will fail if you look at it wrong. Find a Commodore 1702 instead, which is a superior and indestructible composite monitor suitable for any classic computer.

If you do end up getting a 128, then I'd recommend either a 1084 or a 1902, not the 1802, both of which have a "real" digital RGB mode.

Thanks for the info. The monitor is working in color with my 64, though. I will try to look for a 1702 also. And yes, I am looking for an Amiga 500/600, after which a 128 will probably be on the list of stuff to get.

Right, the 1802 displays colour in composite mode; it's just monochrome for RGB (its RGB "support" was its selling point back in the day and suckered a lot of people who really should have bought a 1902). But it's just a sucky monitor overall.

If you're going to get an Amiga, then you probably want to just get a 1084S (the stereo version). That'll connect to pretty much any Commodore Amiga or 8-bit, including the 500, 128 and 64. The problem with the 500 is that they sprawl across desks like the flat 128 does. I like my tower Amigas much more than the 500 (and the 500 had a GVP A530 sidecar, so it wasn't for lack of power, per se).

During the Spindler/Amelio years? No doubt the latter. Everybody knew that Copland was doomed, but no one knew what could replace it, and this new thing called Linux seemed like it was worth a shot.

Very nice!

Did your environment chips ever work in this machine? Were they intentionally disabled?

I can't get my damn cat OUT of my computer room.

Proud to be a tcsh bigot, but I updated the TenFourFox bash to .30 anyway.

In my case, tcsh works, is easy to get, doesn't change much, and I'm used to it. It hasn't cheesed me off enough to look at another shell.

Plus, as a product of the University of California, csh syntax is now wired into my brain.

My bet is disk rot of some sort.

She knows where she likes to be. In winter she sits behind the G5 exhaust fans and soaks up the heat.

I've got a DEC PRO-380 with a sticker on the front saying "VENIX." I've got to find a video cable for that thing.

That's exactly what it is; only the WGS 9150 was any different, and all the other PowerPC WGSes are just rebadges. The 9600 is a great beige Mac, and the last six slot one (I have one from johnklos). That said, being a beige Mac, it has a limited bus, and you really need a G3 or G4 card to make it practical. Those aren't hard to find, but add that to the cost. It can run OS X through XPostFacto, though I would not exceed 10.3 or 10.4 on this machine.

The server software that came with it was, at best, perfunctory. Even regular 9600s can do basic file server duty over AFP. It will run regular Mac OS 9.1 with no problems.

I have awesome personality disorder.

So I have an HP 9000/350 running HP-sUX 08.00. It came with a sticker on the 670H hard disk saying "Homer" and a Homer Simpson squeeze toy. So that's its name. This is homer: http://www.floodgap.com/iv/1572

Although the 670H hard disk is a freaking tank and will probably never die short of a head-on missile attack (gawd, HP sure doesn't make computers like they used to), I wanted a bootable tape as a recovery method since I don't have HP-UX media (except for the C8000). After a lot of watching on eBay, I finally landed a 16-track HP 9144A tape drive.

The 9144A and 9145A drives are interesting in that they use QIC form factor tapes, but they have to be pre-formatted as IOTAMAT at the factory (DC 600HC tapes are the 3M equivalent). They can do error correction and slow random access; in fact, HP-UX treats them as very slow CS/80 disk drives. The 350 will boot from it, which is ideal, and I even have some HC tapes lying around. Perfect!

The first problem is that HP used a polyurethane-isocyanate composite for the pinch roller on the capstan that degenerates into black gooey muck over time. I had expected this, but I hadn't expected it to have already leaked over parts of the drive. That was an extensive cleanup (I discovered 91% isopropanol dissolves it well, which was good because that's what I was using to clean the head and the rotation sensor). Then I had to rebuild the capstan roller; for that, I selected a 9/16" ID - 3/32" wall rubber O-ring and glued it to the capstan with Loctite GO2 gel, which I picked since it has a space-filling vinyl-silicone formulation that seemed able to take more compression than regular epoxy. It cured for 24 hours. I thought that came out rather well: http://www.floodgap.com/iv/2354

I powered it on; the rear electronics said "P 2" (self-test passes, HP-IB device 2, which is correct). I inserted a brand-new 600HC tape as a positive control. The new drive wheel gripped perfectly and moved the tape properly; the BUSY light came on and it spun through the tape and found its format keys, and turned the BUSY light off. I pressed Unload, it rewound the spools, and I ejected the tape successfully. No residue on the tape, so I concluded the positive control was a success. The negative control was a regular DC 6150 I had in a box. I put that in, the drive put its BUSY light on, spun the tape, buzzed a few times, and illuminated FAULT . That was as expected. I ejected the tape, confident that I had proven both the electronics and the mechanism were now working, since it could read and certify its own tapes, and properly rejected unformatted QIC tapes.

The second problem I haven't been able to solve yet. SAM did not see the tape drive when I tried to install it, so I manually did so with mknod (major 4 for character and major 0 for block; minor 0x070200 for slow HP-IB bus and device 2). When I order the drive to accept data with tcio, it stops with a write error. When I try to have tcio certify the tape with tcio -uv, it says it fails. The drive does light BUSY and tries to write but invariably is unable to do so.

Any suggestions? Clearly it can read, or it would have rejected the IOTAMAT cassette. I see only a unified read-write head on the mech.

I hadn't thought of that, though it has a Protect light that should have illuminated. But that's a good suggestion to verify. I'll do that this week, thanks.

Ugh. Heartbreaking on a classic C128D.