3.3 (There ain't no sech animal; it is actually 3.30, three point THIRTY) was limited to 32M (due to DOS using "relative sector number", and being unable to comprehend more than 65536 sectors per drive letter. 3.31 (three point thirty one) was the version that supported larger drives.

It will display the registers. AL has the major version, AH has the minor version. If AH is other than 1F (on a 3.xx system that supports >32M), then I am wrong. (AX: 1F03

How about 6.00, 6.10, 6.20, 6.21, 6.22, 6.23, and 6.30? Well, THOSE are OT for thisa list.

On Fri, 22 Jun 2001, Sellam Ismail wrote: Ooops. 6.23 was a typo. I don't think that there was ever such. Sorry about that. 6.22 is the last release from MICROS~1 that I'm aware of, although internally, WIN95 is 7.00, WIN98 is 7.10, ... PC-DOS 6.30 was IBM's version of MS-DOS 6.2x (PC-DOS 6.10 was IBM's version of MS-DOS 6.00) -- Grumpy Ol' Fred cisin(a)xenosoft.com

Actually, didn't IBM come out with PC-DOS 7.0, then 7.1, then 7.3, then 8.0, then Y2K patch it to make PC-DOS 2000?

On Sun, 24 Jun 2001, Jeff Hellige wrote: I don't have a copy of that one. If it would be convenient, could you check what funtion 30 returns? DEBUG A MOV AH, 30 INT 21 INT 3 G does AX show 0A07 ? -- Grumpy Ol' Fred cisin(a)xenosoft.com <A HREF= "http://www.xenosoft.com/dogears" >DogEars</A>

Do the same rules apply to collecting and attempting to fix computer equipment?

Today I posted some questions regarding an IBM S/23 I am trying to get working (VCF 5.0 is coming up fast).

Should I simply not even try to do anything, lest I risk criticism for doing it wrong?

If this is aimed at me, and I think it most certainly, is, then you might say so... There's something that I find offensive too. And that is being flamed for offering what I consider to be correct advice. Please remember that I give my time, knowledge, and experience freely. If people here don't want such advice, then tell me and I will leave. I've thought about leaving several times due to being flamed for things like this. And I can assure you that _I_ will not be the loser.... Oh, for %deity's sake. Nobody is going to sue anybody, surely. OK, let me make a few points. You started by asking for methods for undoing stuck screws. I provided a number of such methods that have worked for me in the past (note : I have never needed to use a hammer or similar tool to dismantle _any_ part of a classic computer, but I digress). I also suggested that there might be an alternative method to do what you wanted to do (namely to remove the parts from the monitor chassis). I have learnt many times over that if somebody is asking how to do something, it's best to find out what the _real_ problem is as there may be an alternative solution that's easier, better, or whatever. I know some people don't like being told they're going about it the wrong way, but.... Also, you now say that the 2 monitor chassis assemblies have the same part number. In this case, I think it's safe to assume they _are_ compatible (in fact they might well be identical). So the height problem is certainly not an incompatability. It's most likely a fault. In the circumstances, the first thing to try when the height is too low is, most liklely, a tweak of the height control. To be honest, I'd probably have tried that too (knowing that a height controller, as opposed, say, to a disk controller PLL adjustment, is easy to set correctly). But when you find that you can't get enough height out of the monitor that way, then the next thing to do is not to look for more adjustments, but to look for the fault. The fact that the height control was sealed with silicone rubber should have indicated that most likely this is not a common field adjustment. In other words you should be able to drop any monitor chassis in place of any other without having to adjust things -- they would have come correctly set up from the factory. In which case the lack of height would suggest a component fault. Now, as to what I would have done differently, had it been my machine.... My first thought would have been to swap the CRT only. Not the entire monitor. For at least 2 reasons it's a good idea to replace as small a part as possible. Those 2 reasons are : 1) Historical accuracy -- keep as much of the machine original as possible. There may have been changes in design that happened during the production life of the machine (I know plenty of examples where this did happen). You don't want to confuse people by having the 'wrong' parts fitted. 2) Lack of incompatibilites. Often seemingly identical subassemblies can have subtle differences. I've been caught with this too often. I prefer to keep with parts that I _know_ have once worked together and therefore can be made to work together again. My second thought would be to look for faults when tweaking an adjustment doesn't have the desired effect. IMHO classic computers are repaired by very different procedures now to how they were repaired when in current production. For one thing, there's no source now of known-good FRUs for most classic computers. Even New-In-Box parts can have failed in storage. Therefore, it's necessary to do a lot more tests and a lot less module-swapping. But since this is a hobby for most of us, we don't count the time we spend looking for the fault. The time doesn't matter, the cost of parts might. It's cheaper to replace a small component rather than an FRU, too. For this reason I often ignore much of what is printed in the official service manual for a classic computer. I will make use of the 'reference data' -- the schematics, pinouts, voltage charts, waveforms, source listings and so on. But I almost always ignore the fault-finding charts. I work from the schematics, making measurements, checking waveforms, and figuring out what is really going on. There is an attidude here that I find hard to accept. I have no problem with the fact that you are probably relatively inexperienced in fixing monitors (or analogue stuff, or...). Everybody has to start out somewhere. I can remember making a lot of mistakes when I was starting out. The thing is, I was prepared to learn from my mistakes, from my problems, and so on. I don't consider that learning should stop just when you leave school (or university, or...). I want to stop learning the day that I die, and not before. There are many things (related to computers, electronics, engineering) I don't know. And many things I've learnt due to working with classic computers. My attidude if I come up against something that I don't know (like 'How does the vertical deflection system of a monitor work', or 'How do I write a device driver') is to go away and learn about it. Read books, experiment, ask questions, etc. And then I have got a solution if I ever come against the same problem in the future. Repairing and using classic computers is a hobby for most of us here. A hobby is something we do because we want to do it, it's fun. This doesn't mean that everything I do related to classic computers has to be enjoyable. But equally, if I found that everything was unenjoyable then I'd find some other hobby. But I consider learning to be an integral part of this hobby. OK, so you have a monitor that's not working right, how do you learn about it? Well, I am going to assume that the manufacturer never provided a schematic or component level repair manual. But a mono monitor is like the last few stages of a monochrome TV. So we look at books on repairing TVs. (There's a good reason why I have 30 years of tha annual 'Radio and Television Servicing' books on the shelf). We look at circuit diagrams for TVs. We look at the monitor and try to identify the main components -- at least the ICs. If the numbers are unreadable, then we trace enough pins to make a sensible guess as to what the chip must be (this is something that, I will admit, comes with expeirence). We try to find circuits in TV repair books that use similar components. And then try to match up the other parts -- the resistors and capacitors round the chip. Most likely they'll not tbe the same value, but the arrangement -- the circuit topology -- is often very similar. There are only so many ways you can use a particular chip. For example I had a problem with a Hitachi colour monitor on my DEC Rainbow. It was badged DEC VR241, but it was certainly a Hitachi chassis. The vertical output stage was part of a ceramic hybrid module. A look at some early 1980s Hitachi TV schematics producted one using a hybrid with almost identical circuitry from what I could deduce about the one in my monitor. Needless to say that schematic was a great help. If 'ME' = Mechanical Engineer, then I am even more suprised by the use of a hammer. I've never used a hammer for any mechanical engineering that I can remember. I'm not an engineer at all. I am also not a 'doctor' in this sense. You don't need an electronics lab. FWIW, the tools I most often need to fix computers (including test gear) will all fit into one normal-sized 'holdall' bag. In fact I take them along to HP calculator club meetings. [That, BTW, is more than can be said for my 'mechanical' toolkit which is most definitely not portable] But, more important that tools or test equipment, IMHO, is the ability to think about what you are doing and to stop and think if there's a better way to do it. Nobody is going to flame you for making a mistake. I've debugged several computers to component level using nothing more than an analogue VOM (multimeter) and a bit of experience. Test equipment is not a magic box that will find the faults for you. I keep on about experience. Yes, some of us have been working on such machines for rather a long time. And _all_ of us (if we are honest) will admit to making mistakes. Sometimes very silly mistakes. But experience comes from mistakes, provided you are prepared to realise that you have made a mistake and what uyou did wrong. No, please carry on. The only way to learn is by doing things. And keep on asking questions -- that's another good way to learn. But bear in mind that if you ask for advice it may not be what you want to hear. It may help you in the end, though -tony

I have no idea if all System/23 monitors are

totally different. And obviously the decision on what

indeed where to get parts from) depends on that.

So, OK, I assume now that these monitors are

but how similar are the CRTs?

from lesser systems, and scrap the remains, but I suspect like many on this

Tony's actions in the UK are perfectly reasonable for him, but here in SoCal I run across so much stuff that more than 95% of it goes to scrap. Of the 5% I bring home and mess with, about 75% still goes to scrap after a few months or years of storage. My problems are; Space. I am currently at a good 200% of reasonable capacity, and forced to pay an extra $30 month to park my new car on the street instead of my packed garage. The worst to me about this is that in NO WAY is the extra stuff blocking my cars parking spot worth more than a couple months of this extra $30. If I wouldn't pay that much to buy, why the heck am I keeping and pay that much? Stupid projects that bog down and tie up my workbench areas. I am frequently tempted to do what Tony does, and get myself into a real mess while I wait for a part, tool, or knowledge on how to continue. (see stupid 486 minitower below) A very reasonable course of activity for me is to take the parts I need from lesser systems, and scrap the remains, but I suspect like many on this list I find that VERY hard to do. Instead of what odd bit did you add to your collection, what was the last thing you tossed? My list for June is a yellow 13" mac RGB monitor that I couldn't get to work properly, tried to fix and broke something, plus found out it was already working and I hooked it up wrong. Item two are a couple of 486/66 minitowers that had just too many things wrong with them(bad power supplies, missing memory, bad hard drives, funky motherboards (vlb), and finally nasty old case I cut my finger on.

Not everyone in the house needs a PC; I think I'm more happy with one PC in a central room where I can

I feel bad...Only 20 functionals in the house, but I do have over 150 varying PC's in my storage building, mostly PS/2's....

I feel bad...Only 20 functionals in the house, but I do have over 150 varying PC's in my storage building, mostly PS/2's.... -----Original Message----- From: owner-classiccmp(a)classiccmp.org [mailto:owner-classiccmp@classiccmp.org]On Behalf Of Mike Ford Sent: Tuesday, June 26, 2001 12:58 PM To: classiccmp(a)classiccmp.org Subject: Whats a reasonable collection? Hmmm, I was thinking of cutting down to about 50 computers, about half networked and the other half as curiousities or spares. I have six areas that each need a working mac and pc, (5) main servers each running something different (Linux, NT, OS2, AppleShare (one 3.0.4 to boot IIgs systems, and one running a newer IP version)), fun servers (HP/ux on some Apollo's, various on older macs and PCs), dedicated systems (scanner, couple phone things, (2) CDR burners), various old systems running old software. I want to have as many different kinds of segments on my network as I can manage, which means running a model 95 OS/2 system to bridge between 10bt and tokenring on twinax.

BTW, I'm also very fond of MCA. Just glad I haven't got on my own desktop. =)

Responses inline: On Wed, 27 Jun 2001, Mike Ford wrote:

> >BTW, I'm also very fond of MCA. Just glad I haven't got on my own desktop. > >=) > > A model 9595 was the first computer I was really impressed with > mechanically when I opened it up. Its built like a Porsche, and actually > LOOKS good internally.

The 8595/9595 is an excellent machine. Another one that I like very much is the MCA/PCI IBM PC Server 320 (8640).

> I have a couple dozen of various models, with about a dozen 9595 servers. > One of my big scores in the auction last month was a box of 3Com MCA 10bt > cards I had been trying to buy for a year or two.

I am trying to find some of those 100Base-T Olicom MCA cards.

Responses inline:

On 27 Jun 2001, Iggy Drougge wrote:

> Vance Dereksen skrev: > > >I love microchannel equipment. 8-) > > I did an MCA pick-up today, entirely free. Got a set of sealed little TR > adaptors, various IBM hard drives, different protocol adaptors, two 8514/A > adaptors, a Future Domain SCSI adaptor, a RAM expansion card, XGA2 ID > plugs, as well as a huge load of ISA junk and a Xircom parallel Ethernet > adaptor.

That Ethernet should come in handy. If I were you, I would lend that FD SCSI to Jaromir Dolecek. He's writing MCA card drivers for NetBSD.

> Question 2: I got a little black plastic box by IBM with one DB25m, one > DB25f and one DA9f. I suppose this is some kind of test equipment, but > what?

It's a serial/parallel wrap plug. The advanced tests on PS/2 reference diskettes need them to do port testing. It's a handy thing to keep around.

> BTW, I'm also very fond of MCA. Just glad I haven't got on my own desktop. > =)

Why not?

Well most of the cards aren't jumpered because of the, "silly ADF files". They are all software setup via the Reference disk.

I hate to tell you this, but other buses such as Zorro, PCI or Turbochannel aren't jumpered either, yet without the need for ADF files. ADF files are just jumpers on a disk.

Zorro was around in 87 when the MCA machines appeared. It was strictly Amiga though. It was always 'Plug & Play' with boards plugged into Zorro II and III slots supporting Autoconfig. I'm not sure about the Zorro slots on the A1000 and A500.

Iggy Drougge wrote:

> I hate to tell you this, but other buses such as Zorro, PCI or Turbochannel > aren't jumpered either, yet without the need for ADF files. > ADF files are just jumpers on a disk.

Well, I never heard of Zorro, and I wouldn't know Turbochannel if I saw it. I suspect PCI doesn't use jumpers because most cards are, "Plug and Play". Besides, PCI wasn't around when Microchannel was king of the corporate world.

on 6/28/01 1:12 PM, Brian Chase at bdc(a)world.std.com wrote:

Didn't EISA also suffer from still being slow? I seem to remember that it didn't gain much/any speed over ISA, just that the data path went from 16 to 32bits, allowing it to move more data at the same bus speed? To remain compatible with ISA, it would seem they'd have to retain the original bus speed as well.

Oh no, EISA cranks out speed, if you have the right boards. EISA bus mastering boards can blast 32 bit data directly into RAM using a seperate data path in the chipset.

On Thu, 28 Jun 2001, Chad Fernandez wrote: > Iggy Drougge wrote:

> > I hate to tell you this, but other buses such as Zorro, PCI or > > Turbochannel aren't jumpered either, yet without the need for ADF files. > > ADF files are just jumpers on a disk.

> Well, I never heard of Zorro, and I wouldn't know Turbochannel if I saw > it. I suspect PCI doesn't use jumpers because most cards are, "Plug and > Play". Besides, PCI wasn't around when Microchannel was king of the > corporate world.

There's EISA. I thought it was an okay bus while maintaining backwards compatibility with ISA. It'd have been more successful if it were cheaper. As for other non-jumpered busses, Sun's Sbus is nice.

On 28 Jun 2001, Iggy Drougge wrote: > Brian Chase skrev:

> > There's EISA. I thought it was an okay bus while maintaining backwards > > compatibility with ISA. It'd have been more successful if it were > > cheaper. As for other non-jumpered busses, Sun's Sbus is nice. > > EISA is similar to MCA, it's got the same descriptor files, only they're > called INF instead of ADF.

Do all EISA and MCA based systems require these? I don't seem to remember having to deal with them on MCA based RS/6000 AIX systems, or say under NetBSD/i386 for an old 80386 EISA box I have.

Is it possible that maybe the existence of .ADF and .INF files is just a brain-damaged Microsoft OS requirement?

EISA is similar to MCA, it's got the same descriptor files, only they're called INF instead of ADF. -- En ligne avec Thor 2.6a.

Well, I never heard of Zorro, and I wouldn't know Turbochannel if I saw it. I suspect PCI doesn't use jumpers because most cards are, "Plug and Play". Besides, PCI wasn't around when Microchannel was king of the corporate world.

Turbochannel is the bus I've seen most often with DEC RISC workstations, although I think they were not the only vendor to use it. The MIPS boxes and the early Alpha boxes were Turbochannel. Mostly, it was used for frame buffers, but I think there were one or two non-graphic Turbochannel cards.

On Thu, Jun 28, 2001 at 12:25:21PM -0700, Ethan Dicks wrote:

Of course there are: I have an Ethernet adapter (10 Mbit) in my DECstation 5000/125 which is a TurboChannel card and AFAIK there also exist some SCSI cards for the TurboChannel.

Well, I never heard of Zorro, and I wouldn't know Turbochannel if I saw it. I suspect PCI doesn't use jumpers because most cards are, "Plug and Play". Besides, PCI wasn't around when Microchannel was king of the corporate world.

Chad Fernandez skrev: Zorro was. =? BTW, is ruining people's quoting a hobby of yours?

Chad Fernandez skrev: Yes, since you put your comments about the quoted material. It look like this: 4 -- En ligne avec Thor 2.6a. Computers are not intelligent. They only think they are.

And _my_ preference is to keep things in chronological order and with the

On Fri, 29 Jun 2001, Ethan Dicks wrote the following words of wisdom: > And _my_ preference is to keep things in chronological order and with the > new comments interspersed for context. Oh, yeah, and with the irrelevant > stuff *not* quoted.

Agreed. I remember when there was a time when careless, clueless, bandwidth and time wasters on Usenet were flamed to a crisp for not doing this. The lines of text that were irrelevant were replaced with a single "[...]". Also, severe flamage would also result when those blasted careless, clueless, people would fail to properly quote the names of those whose words they were replying to.

What do you think, do we need to go back to using severe flamage in order to correct this problem? The flames weren't attempts at being nasty; they were merely a very fast method of education. :-)

> Unfortunately, this is the style of yester-year, of 1200-baud UUCP feeds

However. it still works well, and produces messages that make sense.

> and old Usenet. None of this is easy to do with MicroSloth products.

Nothing useful is easy to do with Micro$loth products.

Only thing I don't like is a lack of fast SCSI cards. I have all kinds of good scsi cards for my EISA computer, but not my Microchannel. I

I have an EISA server built by Unisys, and I have some very nice SCSI cards for it built by Adaptec and Buslogic. The only Adaptec card for Microchannel is based in the 1540 series (ISA) and is only good for 5megs/sec, just like the ISA version. My EISA SCSI cards are good for 10 or 20megs/sec depending on the card. I wish Adaptec would have made a 32-bit Microchannel card based on the 7800 series Adaptec chip, like the EISA cards. Microchannel SCSI cards from Buslogic must be very rare.... I haven't seen a single one on Ebay since I have been looking. I haven't really looked into the IBM Fast/Wide because I really want a differential card, and the Diff version seems to be RS6000 only.

Besides it uses funky connectors and the cables seem to be quite rare.

I want to have as many different kinds of segments on my network as I can manage, which means running a model 95 OS/2 system to bridge between 10bt and tokenring on twinax.

Mike Ford skrev: Is it possible to run TR on twin-ax? I thought it was only run on STP or those big jaw-like IBM connectors.

Instead of what odd bit did you add to your collection, what was the last thing you tossed? My list for June is a yellow 13" mac RGB monitor that I couldn't get to work properly, tried to fix and broke something, plus found out it was already working and I hooked it up wrong. Item two are a couple of 486/66 minitowers that had just too many things wrong with them(bad power supplies, missing memory, bad hard drives, funky motherboards (vlb), and finally nasty old case I cut my finger on.

The PC-DOS 2000 that I have shows 'PC-DOS Version 7.0 Revision 1' if you do a 'VER' command. Jeff -- Collector of Classic Microcomputers and Video Game Systems: Home of the TRS-80 Model 2000 FAQ File http://www.geocities.com/siliconvalley/lakes/6757

What are some of the symptoms of y2K incompliancy? I have been setting old machines (1994 -> before) to current dates and have been getting no trouble, at least as far as I could tell. John A.