On 06/02/2014 05:57 PM, Chris Elmquist wrote: It's been a long time, but we had a Stearns in Santa Clara (courtesy of Neil). ISTR that it had one (1) ISA slot as well. Or am I thinking of another sort-of PC? Oh, don't forget the almost-PC early 16-bit Columbia and Eagle systems. Could one really call the Sanyo MBC-500 series "PC Compatible"? --Chuck

On Monday (06/02/2014 at 07:49PM -0700), Chuck Guzis wrote: No--you're right. There was an ISA-like slot in it. Same connector anyway but not the same signaling. There were few PC plugins that worked in the slot. Only simple stuff like serial and parallel ports but nothing with memory (such as a video card) or that required DMA (like a hard drive) would work in the slot. Nicely done... but, in those days, we were all still learning and so it took a few years and several hundred machines before we knew what hit us. -- Chris Elmquist

There is very little that's typical about the FTS-88 IMHO. For one thing it has an optional GPIB por and an optional network port (their own standard) But the really odd thing is the video display. The video RAM does not appear in the memory map. To access it you use one of the channels of the DMA controller. To write data to video RAM you use the DMA chip to read the data from normal memory. This data is then written to locations in video memory by hardawre on the video board. Very odd. The normal ekybaord is bmassive with a large number of function keys. It's cotnroleld by a COP800 IRIC. WIth external ROM. I think this is the only time I've seen a 2758 EPROM (half a 2716, basically). The basic configuration has 3 main PCBS in the cardcage. The CPU board (which also cotnaisn the keyboard itnerfce, network interface, DMA controller, interrupt controller, RAM, ROM, etc), the video board, and the disk controller (which also carries the standard serial ports (7201 based IRIC), GPIB port, and 3 optional serial ports (8251 based). The drives were standard Mitsubishi units, standard Shugart interface. Perhaps I've bene lucky, but they don't seem to be particularly unreliable. The main problem with these machines seems to be the internal video monitor (whcih is spproximately MDA rates, an MDA monitor can be hooked up exernally). The oroginal monitors (there are various different manufacturers/models) all seem to eat flyback transformers. -tony

Err, IBM PC? CP/M 86 was a supported OS. But actually, given that few users chose that, there may well ahve bene mroe of another CP/M 86 box out there... Not quite as strange as the Sirius character genretor, the Apricot has mcu hthe same idea. These machines had 2K*16 of video RAM, which stored, IRIC, 11 bit characters and 5 bits of atributes (intensified, underlines, inverse, etc). The 11 bits were fed as an address to part of the main DRAM array, the output of that went into the shfit registers, then to the CRT. So the character generator was actually part of the normal DRAM user memory. Note I said 11 bit chracters, There was a reason for that. The display was 80*24 or 80*25 (I forget which). In other words, <2^11 chracters on the screen. So you could actually have a different chracter displayed at every location Given that the character patters were in RAM, you could therefore have bitmapped graphics (load a differnet chracter into each lcoation of the 2**16 video RAM, then set/reset bits of the 'character generator) area). [FTS monitor] I manaaged to keep one going using bits of 2 different oens(flyback form one, horizotnal oscillatro from another, a somehat fatter output transisotr on a large heatsink... Didn't you end up with one of those Zentih montoros wtih the PSU that combines the efficiency of a linear wit hthe reliability of a switcher? I made an interface to get compositie video from the FTS. Of course at a strange scan rate. But amazingly my old TRS-870 Model 1 monitor could lock to it if you twidled the osicllaotr coil. The montioro didn't like it much, but it was enough for tesitng to prove the vidoe PCB was OK, etc. -tony

On Jun 1, 2014, at 12:56 AM, Fred Cisin <cisin at xenosoft.com> wrote: I dunno, I think the 3.5" disks on the HP 150 that came out in 1983 were less compatible, since everything else was still doing 5.25. :-)

On 06/01/2014 06:48 AM, Chris Osborn wrote: Fred said "format"--and if you don't get the significance of this, consider the NEC 9800-series "PC". Certainly not IBM 5150+ compatible, but holding at one point 70% of the Japanese PC market. 9801-series PCs used differently-sized (8" 5.25" and 3.5"), but the *format* did not change--each disk held about 1.23MB, regardless of size. The Victor/Sirius, on the other hand, uses variable-bit-rate, "zoned" (i.e., the disk is divided up into zones, each zone having a different number of sectors per track) GCR-encoded data. I can't think of a more different floppy *format*. --Chuck

On 06/03/2014 11:44 AM, Tony Duell wrote: It's a matter of perspective, I suppose. I was looking at it from a conventional viewpoint--with the disk turning at a constant speed, which is the way most people view them today. Saying "variable speed spindle" doesn't convey quite the same level of arcaneness to me. At least one "1.2M" PC drive (Weltech) used a 250Kbps bitrate and spun the media at 180 RPM for high denisty and 300 RPM for double-density. The drives did not work very well. We still have a holdover of this on 3.5" USB drives--if you insert a 1.3MB (Japanese DOS) floppy, the drive will automatically change from 300 to 360 RPM. Let's not forget the Lisa "Twiggy" disks in this category, also. Listening to the drive speed change as the heads moved in and out was pretty annoying. I recall my reaction when I read about Victor's rather arcane scheme for their recording in one of the electronics trade mags. Plain GCR would have gotten them over 1MB on a double-sided 96 tpi DSDD floppy. We figured it must have been hangover from the Commodore days. I was surprised that they didn't try to fit a 6502 in as a disk controller. Another memory of Victor is that they sued anyone they could who used the Victor name, including an electronics dealer who'd been in business with the same name for 30 years. A rather ugly business. --Chuck

On 2014-06-03 8:35 PM, Chuck Guzis wrote: What Victor did with their diskettes is something that that has been routinely done with disk drives for some time now, what is different is the way they did it. For rotating magnetic media the tracks near the outer edge are physically longer than those on the inner tracks so in a normal diskette drive the bits on the other track are further apart. What techniques like what Victor did do is alter the bit rate as you move across the surface so that on the outer tracks you had approximately the same linear bit density as on the inner tracks. It is not practical to make this adjustment for every track so the surface is divided into zones and when you move from one zone to the next the amount of data per track changes. Now there are two ways that you can do this, one is to shift the clocking rate of the data, the other is to vary the spindle speed. The later is the technique used in the Victor 9000. The down side of this technique is when you move from zone to zone there will be a delay while you wait for you spindle speed to settle. Any hard disks drives that I am aware of, run at a constant spindle speed, but alter the clock rate, which has a much lower penalty when moving from zone to zone, if any at all. There was a comment made earlier about them being sued over the Victor name, it was actually the Sirius name they where sued over. After leaving Commodore Chuck Peddle started the Sirius Systems and they later merged with the Victor Company, an established calculator manufacturer who not too many years later went bankrupt.

How about the east german Robotron A7150 aka CM1910? That machine has am Siemens-AMS Bus compatible Card Cage (Multibus I) called MMS16 using DIN41462 connectors and two Z80A Based Subsystems, one for the Floppy and Winchester Disk, the other one for the (color)graphics using an uPD7220. The bus is multimaster capable and there are even dual port RAM Cards for use with multible CPU Cards. There is one connector for the system bus and another one for the local Z80 Busses for the two subsystems. The Keyboard is connecetd serial using 20mA Teletype interface and has a Mode for CP/M86 that will pu out ASCII Characters and an additional mode where is delivers Scan codes. The CPU is an I8086 (or russian K1810WM86) and mostly there is a 8087 on a doughterboard installed. The DMA is an 8257, the Centronics Interface is using an 8255 and serial is an 8251 on the CPU board but there are Z80 SIOs on other Boards too (ASP). Clock is 4.9Mhz. It is a bit slower than a average XT because of the multimaster bus arbitation. Besides of programs using the serial interfaces, most other DOS things will work unsing a double scan CGA Mode. There are SCP (CP/M86), Mutos (some Unix) and BROS 1810 (IRTS?) Os'es available too. DOS was Called DCP on that machine (Disk Control Program) The MMS16 Bus was first developed for use in machine controls, there where Z8000 CPU Cards for that Bus too.... http://de.wikipedia.org/wiki/A_7150 http://www.robotrontechnik.de/index.htm?/html/computer/a7150.htm CPU: http://img-fotki.yandex.ru/get/5635/119497768.1/0_9e411_2106ad6a_XL Here are more pictures from the internals: http://habrahabr.ru/post/162043/ (russian site) The computer had two names, A7150 in the GDR and then it was listed in the eastern countries SKR Standard System as CM1910 where CM are cyrillic letters, so it is pronounced SM1910.. Regards, Holm -- Technik Service u. Handel Tiffe, www.tsht.de, Holm Tiffe, Freiberger Stra?e 42, 09600 Obersch?na, USt-Id: DE253710583 www.tsht.de, info at tsht.de, Fax +49 3731 74200, Mobil: 0172 8790 741

So far I have seen no reference in this thread to the Seattle Gazelle, the target machine for the original version of DOS - before Microsoft. The Gazelle is an S100 machine that uses a serial terminal for its console. There is nothing PC-compatible about it yet Microsoft released a version of DOS 2 for the machine. My example used 8? full-height floppies in a separate cabinet. I was told the floppy format had 1024-byte sectors. There were no BIOS, just a ROM monitor that would boot the machine on a good day. Jim Sent from Windows Mail From: Holm Tiffe Sent: ?Tuesday?, ?June? ?10?, ?2014 ?2?:?27? ?AM To: General Discussion: On-Topic and Off-Topic Posts How about the east german Robotron A7150 aka CM1910? That machine has am Siemens-AMS Bus compatible Card Cage (Multibus I) called MMS16 using DIN41462 connectors and two Z80A Based Subsystems, one for the Floppy and Winchester Disk, the other one for the (color)graphics using an uPD7220. The bus is multimaster capable and there are even dual port RAM Cards for use with multible CPU Cards. There is one connector for the system bus and another one for the local Z80 Busses for the two subsystems. The Keyboard is connecetd serial using 20mA Teletype interface and has a Mode for CP/M86 that will pu out ASCII Characters and an additional mode where is delivers Scan codes. The CPU is an I8086 (or russian K1810WM86) and mostly there is a 8087 on a doughterboard installed. The DMA is an 8257, the Centronics Interface is using an 8255 and serial is an 8251 on the CPU board but there are Z80 SIOs on other Boards too (ASP). Clock is 4.9Mhz. It is a bit slower than a average XT because of the multimaster bus arbitation. Besides of programs using the serial interfaces, most other DOS things will work unsing a double scan CGA Mode. There are SCP (CP/M86), Mutos (some Unix) and BROS 1810 (IRTS?) Os'es available too. DOS was Called DCP on that machine (Disk Control Program) The MMS16 Bus was first developed for use in machine controls, there where Z8000 CPU Cards for that Bus too.... http://de.wikipedia.org/wiki/A_7150 http://www.robotrontechnik.de/index.htm?/html/computer/a7150.htm CPU: http://img-fotki.yandex.ru/get/5635/119497768.1/0_9e411_2106ad6a_XL Here are more pictures from the internals: http://habrahabr.ru/post/162043/ (russian site) The computer had two names, A7150 in the GDR and then it was listed in the eastern countries SKR Standard System as CM1910 where CM are cyrillic letters, so it is pronounced SM1910.. Regards, Holm -- Technik Service u. Handel Tiffe, www.tsht.de, Holm Tiffe, Freiberger Stra?e 42, 09600 Obersch?na, USt-Id: DE253710583 www.tsht.de, info at tsht.de, Fax +49 3731 74200, Mobil: 0172 8790 741

I think a disk turnign ant differnet speeds to read/write different tracks is somewhaat more arcane than a changing data rate. However , the origianl Sirius drives were varaible rate. I don't have the scheamtic to hand, but IIRC there's a microcontroller (8048?) that takes in a 4 bit speed value for each drive, takes in the tacho signals and outputs a cotnrol signall to the motor driver. And yes, you can hear the speed change ans youformat a disk. -tony

On 6/5/14 7:25 AM, Toby Thain wrote: 400K 3.5 and Mac Twiggy used PWM derived from the sound circuit. Lisa Twiggy and all other variable-speed drives had dedicated hardware in the drives for motor speed control.

On 6/5/14 7:35 AM, Simon Claessen wrote: Lisa 400K had dedicated hardware with no connection to sound

Because Apple! Duh! At least they belatedly realized how mind-bogglingly stupid it had been to make their own floppy drives, and switched to Sony. In 1983 it was still a lot less expensive for them to build a Woz controller than to buy a 1793 or 765/8272. The entire Woz controller cost less than just the support components needed for a 1793 or 765/8272. It became even more cost-effective when they started using the IWM ASIC in 1984.

On 06/05/2014 12:17 PM, Eric Smith wrote: Even cheaper than a simple WD1770? Almost no external support needed. Atari saw the wisdom of this in the ST. Since the 1770/1772 is in my 1984 databook and is not labeled "Preliminary" I'd assume it would have been available in 1982 at least in early steppings to the likes of Apple. I think your first explanation of "duh!" probably fits the picture better. --Chuck

On 06/05/2014 05:57 PM, Eric Smith wrote: But they do, just fine. Probably because the signals aren't really that fast. At any rate, I can find absolutely no mention of the output drive capability of the 1770/1772/1773--not even in the application notes. Very strange. Outputs to drives are low-true, however, so the usual 7438 wouldn't be appropriate without an inverter beforehand. I can't recall, but I don't believe that the ST used any output buffering--but it's been too long to be certain. A possibility, but at the expense of compatibility with anything but another Mac. --Chuck

On 06/05/2014 08:01 PM, Fred Cisin wrote: It's really surprising that the attitude didn't hurt Apple badly. Other companies before them attempted to "lock in" their customer base (i.e. create a captive customer community) and some fared rather badly. It's just very strange in a period where everyone was trying to be compatible with the IBM PC. I guess it really helps to have a user community of devoted acolytes. --Chuck

On Thu, Jun 5, 2014 at 10:43 PM, David Riley <fraveydank at gmail.com> wrote: When the Twiggy and 3.5" drive decisions were made, Apple most definitely did already have legions of fanatical acolytes, although obviously not as many as today. However, said acolytes were already accustomed to Apple using disk formats that were not interchangeable with anything else, and for the most part didn't care. There was a third party opportunity for add on products for compatibility. The Twiggy was being designed for use across Apple's entire product line well before the plans for the in-development Lisa started looking even vaguely similar to the Lisa that actually shipped. Apple knew that they needed higher-capacity floppies, and thought (incorrectly, as it turns out) that they could engineer and manufacture a better drive than they could buy, at lower cost. As such, there was no need for the disk format to match anyone else's. It was also not yet obvious to everyone that the PC would end up with >90% market share. Since Apple wasn't trying to build machines with capabilities comparable to the PC, but rather to far surpass it, they wouldn't have seen any value to spending a single additional cent of production cost on adding any form of PC compatibility. It's possible that if a PC-compatible floppy format that existed in 1982 met Apple's capacity and performance requirements, AND was less expensive to produce than their home-grown stuff, they might have been willing to use it. Or maybe not; they did have a significant amount of NIH mentality. Apple switched to the Sony 3.5 inch drive well before there was any solid evidence that it would become an industry standard, so there was still no particular need for Apple's 3.5 inch format to match anyone else's. IIRC the only contemporary (1983-ish) almost-PC-compatible machines using the Sony 3.5 inch drives did not actually use the later 720K standard format, so even if Apple had used MFM rather than Woz GCR, it still wouldn't likely have been compatible with anything significant, unless IBM had later deliberately chosen to be compatible with Apple, which would have been unlikely. While the Twiggy turned out to be a fiasco, I don't think that the general idea of using an incompatible disk format was actually a mistake.at the time.

On 06/06/2014 03:08 PM, Tony Duell wrote: There was such a half-length ISA PCB from Micro Solutions called the "Match Point PC". It was interposed in the normal floppy cable and, using software provided, enabled one to read and write Apple II floppies and even read and write the floppies generated by the Microsoft (?) Z80 card for the Apple. I have one such card. A few PALs and some SSI TTL. I also have an 8" floppy generated for an Apple II setup--the recording is plain IBM 3740-style FM, but the data is in the form of RWTS codes within each sector. Again, very strange. --Chuck

On Fri, Jun 6, 2014 at 4:08 PM, Tony Duell <ard at p850ug1.demon.co.uk> wrote: However, if Apple had sued a standard MFM controlelr at the normal data No, because the IBM PC didn't have 3.5 inch floppy drives at that time. If it had, it's possible that Apple might have considered doing it, but I suspect that they still wouldn't have. IBM PC compatibility wasn't something that they cared much about. They were trying to build something far better than a PC. Whether they succeeded is obviously a matter of opinion.

On 06/06/14 12:14 AM, Chuck Guzis wrote: ...Which they didn't exactly have in January, 1984. :) If they had, the Lisa may not have failed. Many Macs were sold to people who had not previously owned an Apple product - it was also sold as a business machine. The Mac created that pool of loyal users. As we all know, they soon walked back on the Mac's incompatible interfaces and added SCSI, Ethernet, later USB and PCI, etc. (Third parties provided other interoperability solutions.) Being PC compatible would not have made any sense for Apple from a marketing perspective; their angle was "being the alternative". And the competition appears to have been, ah, fruitful. Apple was able to do things differently (e.g. picking 68000 over 808x; building an entirely new, carefully considered UI Toolbox and integrated applications) and users had a meaningful choice. --Toby

On Jun 6, 2014, at 8:20 AM, Al Kossow <aek at bitsavers.org> wrote: That's interesting. One of the first projects that I was looking at once I moved into the PC group (ESD...I was in GSD) at IBM was to look at a 68K board for the original IBM PC (at that point in time we weren't really welded to Intel CPUs and were investigating alternatives that would give us more "headroom" in terms of CPU capabilities). It was admitted internally that the 8088 was a TTM (time-to-market) ploy and we wanted something that would give us more capabilities going forward. The problem (if you want to call it that) was that by the time we actually started looking at alternatives the PC was already wildly successful and (IBM being IBM) S/W compatibility took over all future discussions. As a side note, my 2nd project was what was to become the IBM PC Convertible. We were really pushing technology with that one (no really). We were almost at the prototype stage (we had physical mockups and complete schematics and board layouts) when corporate politics intervened and the project was moved from Boca Raton to Austin where it took 4(!) years to finally ship at which point it wasn't nearly so ground breaking. TTFN - Guy

On 6/6/14 8:49 AM, Guy Sotomayor wrote: Which is exactly what screwed Apple until Steve came back. Any new OS had to support all the (important) old Mac OS apps, trap-patching and all. The notion about caring about old versions of our (or third party) compatibility slowly disappeared, as you know, until now, where anything older than x86-64 isn't 'interesting'. Who knows when the split to Apple-unique ARM will happen, locked to the net and the mothership. You KNOW they want to.

On 6/6/14 8:54 AM, Sean Caron wrote: "TNT" (The code name for the first PCI Mac) stood for "The New Tessaract" Tessaract was the machine with "BLT" (Bus-like thing) which was a crossbar switch for expansion. Someone had a prototype in there garage, and I have it here now at the museum. Some of the I/O blocks from "Grand Central" the I/O controller ASIC made it into TNT. Tessaract was the PPC version of Hurricane, which was the 88110 Mac. Hurricane was contemporary with Cyclone (the 840AV). You forgot Spindler. That's when the market expansion strategy happened. Apple wanted to get out of the low-end low-profit Mac business, so they licensed clone makers who proceeded to make high-end machines instead. Apple did sell one machine based on the Tanzania motherboard that was licensed, the 4400 http://www.everymac.com/systems/apple/powermac/specs/powermac_4400_200.html which very few people have ever heard of. That is pretty much identical to the Motorola StarMax product. Companies like Power Computing and Supermac had their own ASIC designers. Power Computing's team was mostly ex first-generation PPC engineers.

On Jun 6, 2014, at 11:54 AM, Sean Caron <scaron at umich.edu> wrote: Al has hashed that over several times here, and there are plenty of other accounts, but the short answer is "politics happened". My understanding too many cooks, and other projects kept glomming onto it so they wouldn't get cancelled. They eventually took it out back and shot it, skinned it and clad NextSTEP in its trappings, which is how we got OS X. Do I recall that they almost went with Solaris as the kernel? That would have been interesting. - Dave

On 6 June 2014 17:54, Sean Caron <scaron at umich.edu> wrote: Weird, 2nd time in 2 days when I Google for something & find one of my own old ClassicCmp messages. Anyway, yes, that did happen. As I said in 2011: << NT 4.0 supported PowerPC, but the only Macs it would run on were based on PReP, the PowerPC Reference Platform. http://en.wikipedia.org/wiki/PowerPC_Reference_Platform There were some Mac clones built to this, but they were due to appear right about the time that Steve Jobs returned to Apple and killed off the MacOS licensing programme. The best of the PReP clones were the Motorola ones; as soon as Motorola learned of Jobs' return, they killed the development programme. This infuriated many - there were finished review machines circulating. MacUser ruffled a lot of people's feathers by making the Motorola Mac clone the magazine's Mac of the Year for, IIRC, 1996 or 1997. I saw the hardware. Of course, under NT4 the ADB ports didn't work, and under MacOS, the PS/2 and parallel ones didn't. But it was real. The Jobs killed the licensing programme and it was all over. Motorola was furious. I couldn't agree more. Haiku is making progress but it's got a long way to go. Absolutely agree there, too. AU/X for PowerPC would have given them a contender, but then again, without NeXT's world-class development tools and Jobs' eye for design, Apple's comeback would have failed. Same if they'd bought Be. Sadly. NeXT was the right thing to do. -- Liam Proven * Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk * GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com * Skype/AIM/Yahoo/LinkedIn: liamproven Cell: +44 7939-087884

On Fri, 6 Jun 2014, Toby Thain wrote: You never met any of the Apple][ acolytes??!? Price might have also been a factor. The Mac brought that down to a point where it could be considered, albeit unfavorably. If you bought "COMPLETE SYSTEM", then PC and Mac were close to even. However, if you bought the minimum and bought aftermarket expansion, the same could be done MUCH cheaper on PC. (5150 from IBM cost me $1280, total fully loaded was <$1800) There are advantages and disadvantages to "starting fresh" V "upgrading what's already available". (Such as not having to add kludges on top of kludges, V software already available) Documentation revisions took much longer than the PORT for Wordstar, SuperCalc, etc. -- Grumpy Ol' Fred cisin at xenosoft.com

On 06/06/14 11:58 AM, Fred Cisin wrote: Apple II acolytes weren't the Mac market and didn't become Mac loyalists, on the whole. (That was what I was trying to imply, but I didn't spell it out.) The Mac was pitched to a new market, and I saw first hand that this worked. It could actually live up to the "not what you thought a computer was" promise (or, "the computer for the rest of us"). Yes. I'm not saying the Mac was perfect, but it doesn't take too much imagination (of the alternative) to see that we are probably better off for having had the alternative branch of development. --Toby

On 06/06/2014 09:40 PM, Eric Smith wrote: Let's not forget all of the Apple IIs that were sent to schools--very many an early student's exposure to computers started with an Apple II in school. My reaction to the first Mac was that the screen was way too small for me. It was Osborne all over again. --Chuck

On Sat, 7 Jun 2014, Toby Thain wrote: Where else would a ][ user want to go? EXCEPT: ][ users who were into the hardware kinda liked the PC! HHold the two motherboards side by side! The overall design and layout were appealingly similar, although the PC was a little bigger, and they hadn't learned from the "shortage of slots" issue of the ][! It was more like Kaypro; too small, but not as ridiculous as 5" But no external video connector. no slots, . . .

FWIW, not in my case. We had Beebs at achool, we also had a few RML 380Z machines. I preferes the latter (if only because I knew the Z80 processor). Much later on I got my own BBC micro and Advanced User Guide and relaisd just what a well-designed machine it was (IMHO it is the best of all the 8-bitters, had they used a 6809 CPU it would have bene close to perfect!). I have not had the same feeling abotu any Mac (or IBM PC compatible) machine, though. -tony

* Eric Smith <spacewar at gmail.com> [140607 00:41]: Well, I was big into the Apple II but never purchased a Mac (I started using a second hand Compaq luggable and adding my own hard drive before moving on to assembling PC clones from parts.) On the other hand I knew someone who was also very into the Apple II and then moved on to Macs. So in my limited experience it was pretty split. Todd

Did it? Do you ahve the figures fdor the cost of the 2 devices (in 1-offs and in 'production' quantities)? If you're going to allow Apple (or whoever) to make their own drives, there is no reason why they couldn't have used a 1770 to directly drive a bit of ribbon cable to a special drive (presumably with inverted signal lines). That would at least have given a disk which was redable on other machines. Interestingly Amstrad, a British manufacutrer who were know nfor cheap machines and cutting every corner possible used the 1700 seires and provided '38 line drivers, etc. I do not believe tht Apple couldn't have done the same (maybe for a tirival price increase o nthe Mac, but if you're spending that sort of money o na machine that increase is lost in the noise...) In other words I do not beleive that cost was the primary motivation for using the IWM. -tony

On Thu, 5 Jun 2014, Eric Smith wrote: But none of the others had that extra thumb slot like the Lisa Twiggy, that made it so much easier to fingerprint your disks. Besides, using off-the-shelf "standardized" components would open the way for after-market peripherals!! Jobs wan't a SEALED machine, not an open one. The velcro lid, and ease of making cards and peripherals, of the ][ had been Woz' idea.

On 06/06/14 5:53 PM, Tony Duell wrote: The only issue I recall here is that the Apple SCSI formatter looked for Apple OEM drive firmware. If you used third party formatters (e.g. FWB or several others) then drive interoperability with Mac SCSI was excellent. --Toby

Everything I have read (books and web pages) says that RS423 is single-ended. The Macc had differental data lines. The Mac seems therefore to be RS422 data lines and RS423 handshake lines. FWIW, 'Designing Cards nad Drivers for the Macintosh Family' (I may have slightly mis-rememebred that title) calls them RS422 ports. Which they also aren't Again siad books sais 'Based on TI's Nubus' (or somehtign simialr) which implies to me it is not pure Nubus. The book, alas doesn';t give the differences other htan the trivial ones (limitation on the size of the PCB, no -5.2V supply). And yes, it is certainyl close enough to the TI spec that hte standard ICs will work. I thought, but can't find the details, I read somewhere tht there was one fairly obscure thing that Apple did differently and which meant it wasn;t strictly Nubus. If I come acorss the detials I'll post them. -tony

At 21:08 Uhr +0100 7.6.2014, Tony Duell wrote: Well, what is "pure Nubus" here? The MIT implementation? Texas Instruments' 1983 specification? IEEE 1196-1987? [3] Assuming you are talking about Apple's "Designing Cards and Drivers" (3rd ed., 1992) aka DC&D, while they don't summarize in one place Nubus features the Macintosh does not support, they do talk about them. (1) Cache coherency is supported, but does not use the Nubus method (DC&D, ch. 3) (2) Interrupts - Nubus offers interrupts as write transactions on shared memory, a shared interrupt line for all slots, or a per-slot interrupt line. The Macintosh only implements the latter, using a 6522 VIA as interrupt controller (DC&D ch. 3). The shared memory pattern for communicating interrupts in multi-processor systems re-appeared with PCI-X and PCI-E [1]. (3) Block transfers - while later (68040 based) CPUs allow block transfers between Nubus masters, the mainboard does not support them. Generally spoken, the Nubus is targeted towards a multi-master system, as opposed to contemporary buses like ISA, ECB or S100 which are glorified extensions of CPU buses. Apple tried to go that way with its A/ROSE based intelligent extension cards, and there were master cards from i?86 to LISP machines to I/O processors to the Radius Rocket, but the main Nubus appplication were graphics cards, which did not really fit the Nubus model, and soon required PDS slots for adequate performance. So, both the "PC" idea and Apple's view of its Macintosh operating system [2] were at odds with the underlying concept of Nubus, intelligent subsystems. Using Nubus for a serial card was just too expensive, compared to an ISA card. hauke [1] <http://mail-index.NetBSD.org/tech-kern/2014/05/30/msg017180.html> [2] Macintosh New Technical Notes, HW 16 - NuBus Interrupt Latency (I Was a Teenage DMA Junkie) [3] Apple DC&D refers to ANSI IEEE 1196-1990 ("Nubus90"), which neither a web search nor IEEExplore search have surfaced. I assume once Apple decided to move towards PCI, the update was shelved. Does anyone here have (ever seen) a 1196-1990 draft? -- "It's never straight up and down" (DEVO)

?? sorry?? Apple Nubus uses eurocard connectors. maybe you are confused with apple ][ edgeconnectors? On 09-06-14 20:06, Jecel Assumpcao Jr. wrote: -- Met vriendelijke Groet, Simon Claessen drukknop.nl

On 6/9/14 4:57 PM, Jecel Assumpcao Jr. wrote: BFD, they had other incompatibilities, 12.5MHZ and double-clocked burst mode All of this closer to the Nubus 90 spec TI Nubus form factor and was only ever used by TI and LMI

Rougthly, yes. The sound and disk speed data were stored in unused locaitons in the vdieo memory area (I think one was in odd bytes, the otehr in even bytes)(. At the end of every scan line (IIRC), a pair of byters were loaded into counters that produced a pair of PWM signals. One was used fro the sound port (via a custom Apple IC which appears to be a volume cotnrol and nto a lot else, maybe some filtering), the other was the disk speed PWM signal. The 800K drive does nto use this PWM signal, it's still variable speed (IIRC), but changes the speed automatically based on the current head position. -tony

On 06/04/2014 12:33 PM, Tony Duell wrote: 2316 ROM. Probably overkill for the task, but a commodity part. There's a copy of the schematic online, if you're interested. It appears to be just photographs of the factory schematics, but it's quite readable. But then again, it was used all over the place. It's not *too* broken if you avoid changing modes... --Chuck

On 06/04/2014 08:33 PM, Eric Smith wrote: If you don't change mode after initialization, there's nothing wrong. On the other hand, if you change direction of of a port (say, from input to output), the 8255 "forgets" whatever was last written to the output port. To quote the 8255 data sheet: "All of the output registers, including the status flip-flops, will be reset when the mode is changed." So, if using the 8255, set it up and don't touch the mode register ever again. This can trip you up if you're not aware of it. --Chuck

The minor annoyuance is that you can't individually control the dirction of the port lines. You can set port A, port B, and each half of port C to be input or output (at least in the simplest no-handshake mode). The major problem is that any write ot the mode cotnrol register clears all output lines to zero. This is ridiculous. Fristly it prevents you cases. But worst than that it ven makes the intial setup of the system hard. At the time it was made, the most likely thing to drive from a noutptu port line was a TTL input. And as is well-known, TTL input float high, anf ae much easier to pul lhig htan to pull low. After a reset, all ports o nthe 8255 wwre set to be inputs, so were effectively flating. They could be trivially puleld up with a resistor if necessary. So after a reset, the TTL device driven by the 825 sees a l=high elve on its input. As soon as you write ot the mode register, that becomes a low. You can then set the lien how you like, but that low state has to extist, maybe only for a brief period. But your hardware hs to be able to handle it. On a more resonaly parallel prot IC, one where writign to a mode register does nto clear the outptu data registers, you can write 1's to output data registers (even when the lines are set to be inputs), then set the lines to be outputs. The external device will ways see ahigh level (firstly from the floating 'input port' and/or a pullup resistor, then the extrnal hardware cna be designed to have active low inputs, and there will be no glitches to worry it as the parallel port IC is set up. Btu alas that doesn't work with the 8255. The 8255 is proablay OK fror reading switches/keyboards and driving LEDs/displays, but a right pain to use for anything remotely complicated. What I want to know is why clearing the data registers upon writing to the mode register was regarded as a desirable thign to do. I cna think of no useful applciaiton for this -tony

The Olivetti M20 ( a native Z8000 machine ) had a 8086 board one could use run DOS. The machine still wasn't PC hardware compatible. The graphics was superior if you had a color machine. Eight colors. PC applications quickly became too hardware specific. Dwight

The 80286 and 80186 were actually used in IBM PCs of various generations, were they not? If so, the basic instruction set would be a PC-compatible aspect, would it not? /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

John Dvorak wrote about the interrupt conflict in the Nov 7, 1983 issue of InfoWorld. http://books.google.com/books?id=0C8EAAAAMBAJ&pg=PA112 Michael Holley -----Original Message----- From: cctalk-bounces at classiccmp.org [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Michael Holley Sent: Sunday, June 01, 2014 12:21 PM To: 'General Discussion: On-Topic and Off-Topic Posts' Subject: RE: Least PC-Compatible MS-DOS machines? I recall that the 80186 had an interrupt conflict with MS-DOS. Michael Holley -----Original Message----- From: cctalk-bounces at classiccmp.org [mailto:cctalk-bounces at classiccmp.org] On Behalf Of Mouse Sent: Sunday, June 01, 2014 10:01 AM To: cctalk at classiccmp.org Subject: Re: Least PC-Compatible MS-DOS machines? The 80286 and 80186 were actually used in IBM PCs of various generations, were they not? If so, the basic instruction set would be a PC-compatible aspect, would it not? /~\ The ASCII Mouse \ / Ribbon Campaign X Against HTML mouse at rodents-montreal.org / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B

On Sun, 1 Jun 2014, Mouse wrote: I can't recall any model from IBM that used an 80186. Yes, 80186 and 80286 were completely PC-Compatible. There were plenty of PC-Compatible machines that used them. But, there was a LOT more than processor needed to be compatible with PC software. For example, although MS-DOS had OS calls to write characters to the screen, virtually all commercial software ignored them, and wrote directly to video memory. That made it pretty easy to build an MS-DOS machine that couldn't run software sold for the PC. -- Grumpy Ol' Fred cisin at xenosoft.com

On 06/01/2014 10:01 AM, Mouse wrote: That's a straw man. If you're going to run MS-DOS, you're going to have to embody the x86 instruction set somehow, either by hardware implementation or by emulation. I'm not aware that IBM ever used an 80186 in a PC. The integrated peripherals are utterly different from the XT-AT standard. Tandy did try to use one in the 2000, but the incompatibilities were considerable. Curiously, the USR Courier modems used an 80c188, as did other peripherals. --Chuck

On 06/01/2014 04:17 PM, David Riley wrote: Embedded environments certainly seem to be the place to find "interesting" CPUs. I was surprised to see an 80C188EB used on a SCSI interface for a Cipher tape drive, along with a pair of Z8002s--all on the same board. I've seen a 6502 used in a QIC tape drive. At least one model of Diablo dot-matrix printer used a couple of Rockwell PPS-8s. Laser printers certainly used to have some interesting CPUs buried in them. I've seen 68K family CPUs as well as the occasional NS32016 chip. If you just bin old laser printers, you don't know what you're missing by not taking a peek inside. --Chuck

On 1 June 2014 23:20, Chuck Guzis <cclist at sydex.com> wrote: I believe the original RM Nimbus used an 80186. Research Machines is a UK company that focuses solely on the educational market, so business/home software compatibility was not much an issue for it. Or so it thought. In the end, of course, it had to go compatible to survive. The BBC Master 512 - a 6502-based BBC Micro with a 2nd processor - also came with a '186, along with DR DOS-Plus and GEM. -- Liam Proven * Profile: http://lproven.livejournal.com/profile Email: lproven at cix.co.uk * GMail/G+/Twitter/Flickr/Facebook: lproven MSN: lproven at hotmail.com * Skype/AIM/Yahoo/LinkedIn: liamproven Cell: +44 7939-087884