Sorry, can't resist an opportunity for a plug. You can see a Mathatronics, Programma 101 and an HP 9845 all in one gallery at The National Museum of Computing (UK). I'll go you one further! You can also see contemporary Freiden, Wang plus assorted other systems nearby. If you sign up to volunteer with us, and have the requisite skill set, etc., etc., you might get to run and maintain the little beggars as well. Go on, you know you want to! -- Colin Rick Bensene <rickb at bensene.com> wrote: Liam Proven wrote: Warning - long rambling response.... It was actually called the Programma 101, or "P101" for short. It was a stored-program desktop programmable calculator. Its development began in 1962. It was not introduced to the market in Europe until 1964, and wasn't available formally in the US market until the fall of 1965 (though some did make it into the US prior to then via informal methods). There were a number of other electronic calculators had come on the scene prior to the P101, including one machine that was also a stored program calculator called the Mathatronics Mathatron. Though the Mathatron did not have offline storage for its programs, it did use core memory which was non-volatile, so programs were not lost when the machine was powered off, while the P101 did lose its memory when turned off (it used magnetostrictive delay lines), making offline storage of programs and data more of a requirement. The Mathatron could do everything that the P101 could with the exception of magnetic card functionality, and it was introduced in the US almost two years prior to the P101. Was the Mathatron a personal computer? Not by my definition. The claim that the P101 was the first-ever "Personal Computer" is debatable, though the father of the machine, Pier Girogio Perotto, insisted that it was the first personal computer until the day he died. Before he died, I debated this with him via Email (though it was interesting, as I had to use translators to/from Italian). He was insistent despite my arguments to the contrary. In my opinion, the machine should be classed as one of the earliest desktop learn-mode programmable calculators (though it wasn't the first..the Mathatron was), not a personal computer. It had a lot of the characteristics of a computer, in that it was programmable, it had a magnetic card for writing out programs and data, and reading them back in. It had a serial printer that could both record output of calculations and programs, and could also list out programs. It was a numeric-only printer, with special characters for annotation (math function, memory register identification). It couldn't print general text. It could add, subtract, multiply and divide, along with calculating square root and absolute value as built-in functions. Programs had to be written for other functions such as logarithms and trig. It wasn't terribly fast, especially when executing programs, because each program step had to be "waited for" as it circulated through the delay line, and memory register access also was serial in nature. It was capable of simple conditional branching. It had a number of memory registers (quite limited) that could store variables and constants. The machine was an all transistor design (no ICs), and used a magnetostrictive delay line for register and program step storage. The keyboard and printer were heavy mechanical devices more reminiscent of a mechanical desktop calculator. Programs were "learned" into the machine by entering the steps from the keyboard a step at a time. It did not have peripheral interfacing capabilities, so the machine was what it was with no real means for expansion or additional peripheral devices. Given that it had very limited memory capacity, and variable storage registers were traded off for program steps, it could only do so much, though through clever use of the mag cards, programs could be "chained" and larger capacity for programs and data created. However, it required manual operation to load/unload cards, and slowed down the process. The machine was capable of some pretty amazing calculations, including its use by NASA for calculating antenna aiming information for the US Apollo spacecraft program. Personal computer? There were machines that significantly predated it that were more capable, and could be considered "personal". The Stanley Frankel-designed Royal McBee/General Precision LGP-30 (1956) was a true computer, that could easily be used at a personal level. It was fully programmable, had peripheral interfacing capabilities, had a versatile I/O typewriter that could do both text and numbers, and plugged into a standard electrical outlet. It had much more memory (magnetic drum-based), and much larger programs could be written, though in order to do math like could be done simply on the P101, math routines had to be written, along with I/O routines to accept input and print output. Monroe, in the early to mid-1950's had "desk-sized" (not desktop) machines called the Monrobot series that could do more than the P101 and were more computer-like in their later models. Clary also built the DE-60, another desk-sized programmable calculator that was computer-like. IBM's 610 "Auto-Point" calculator was a very capable stored program calculator/computer that could out-perform the P101, but it was quite large and expensive. All of these machines came about in the 1950's, quite a long time before the Programma 101 came out. These machines could potentially be considered "Personal Computers" before the P101 was even a concept. The Programma 101 was pretty simple to program, probably less complicated than these earlier machines, but that also limited it capabilities. The key things that the P101 had going for it was that it was a desktop machine (though rather large and heavy - 80 pounds), and it was comparatively inexpensive, selling for around $3900. These predecessors didn't fit on a desk (though the Mathatron was a desktop machine, though it took up quite a bit of square footage on an average desk), and were significantly more expensive. Was the P101 the first personal computer? My opinion is that it isn't, but then some might argue. Some consider the Berkeley SIMON computer, developed in the late 1940s, with articles on how to build the relay-based machine published in Radio Electronics in the early 1950's, to be the first personal computer. Some also consider HP's 9100A calculator (1968) the first personal computer. This calculator made everything developed prior to it look like a toy. It had much larger memory capacity, had logs and trig built-in, was an all-transistor microcoded design, and could interface to a significant number of peripherals including plotters, printers, extended memory, and even custom interfaces made for instrumentation control systems. But in my mind, it was still a calculator, though an extremely powerful one. My personal definition of the first real desktop personal computer would be the HP 9830. It had a full alphanumeric keyboard and display, was programmed in BASIC, had large memory capacity, could interface to a wide range of peripherals (plotters, printers, storage, instrumentation interfaces, telecommunications). It could be used as a calculator by simply typing in math expressions and pressing a key, and the answer would be displayed. It was introduced in 1972. It sat on a desk, and was easily usable in a productive fashion by anyone after spending a little time reading through the well-written manual. All that said, the P101 was an amazing machine for its time. It was immensely popular, and spawned some follow-on machines from Olivetti that improved on its capabilities. However, HP's domination of the high-end programmable calculator market beginning with the 9100A eventually knocked Olivetti out of that line of business, though Olivetti did stick with producing office calculators for some time. Rick Bensene The Old Calculator Museum http://oldcalculatormuseum.com

Thanks, but I volunteered about 15 years ago. There were so many hassles, that I decided my own collectio nwas enough. Anyway, woking on one 9845B is quite sufficient. Any machine with over 1000 ICs in it, where even the PSU contains 17 ICs spread over 5 PCBs arranged as a sort-of 3 dimensional sculpture is not something to take on lightly. -tony (who traced out the schematics of the 9845B and 98780 monitor).

Evan wrote: then 101 I guess I didn't really explain my thoughts about the differences in my diatribe. I am referencing "programmable calculators", not calculators in general. A plain old calculator is not a computer in any way, as it has no way to program it or store a program. I limit the scope of my definition during the timeframe from 1963 (when the Mathatron was introduced) through around 1972, when the HP 9830 came about. That machine really blurred the line between programmable calculator and computer. Even though the machine was (in my opinion) a computer, it was conceived by HP's calculator division, and was marketed as an advanced programmable calculator. An interesting factor that calculator companies took into account when classifying their machines as "programmable calculators" versus "computers" is that from a marketing standpoint in the early 1970's, computers were intimidating to people, while calculators were much more familiar to people in business and government. Thus, even more advanced machines made after this timeframe, such as HP's 9825, were marketed as programmable calculators rather than computers, though the machine was definitely more of a computer than a programmable calculator. The term programmable calculator allowed these machines to sneak through the bean counters who generally went into panic when they saw the term "computer" on a purchase requisition. I am also considering just the base device itself, without add on peripherals in cases where such were available. * Primarily has input, and is operated through, a keyboard that has a numeric keypad (0-9), and discrete function keys for mathematical & programming operations. Lacks an alphanumeric keyboard. * Operates primarily as a decimal math machine, meaning its registers store numbers in BCD (Binary-Coded Decimal), Excess-3, or other 4-bit representations of decimal digits, and the ALU is generally not a pure binary device (most ALUs were 4-bit parallel BCD adders, or bit-serial adders that had correction circuitry to handle inter-bit and inter-digit carry). * Don't have the ability to perform Boolean logic operations (bitwise AND, OR, NOT and derivatives) as built-in functions. * Typically do not have the ability to manipulate text. * Typically does not have much in the way of main storage. In terms of K-bytes of memory, most programmable calculators from this period max out at around 2K. Many machines had much less than this. * Output is in the form of numeric display (CRT, numeric indicator tubes(Nixie, Panaplex, VF, LED); limited function output printer (numeric and function annotation); or both. * Programs were initially entered via learn-mode, where the operations provided for by the machine's keyboard were stored into program memory as "steps". No "language" per-se, just keyboard operations. * In most cases, programmable calculators can't do self-modifying code (though there are a few exceptions where program step and register memory are overlapped) * Could not support any high level language...all programming operations were machine-level built-in functions. No language compilers or interpreters were possible. * Limited instruction sets, mostly confined to the basic keyboard functionality of the machine. Conditionals based on simple conditions (X=Y, X=0, X<>0, X<0, error condition exits, sense switch set/not set). * Limited subroutine capabilities (nesting depth, mainly) * Lack of any kind of stack architecture (for programming, not register architecture). Recursion is difficult if not impossible due to lack of ability to stack working registers. On some machines a stack architecture could be programmed, but I'm speaking more about a built-in type of functionality. * No interrupt handling architecture. I/O is typically polled or handled by pausing execution until an I/O is completed. Obviously, there are exceptions to everything...some machines stretch these rules a bit, but overall, most machines that have these characteristics would be classified by me as a programmable calculator rather than a computer. Here are some examples of these criteria applied to a good selection of machines -- * The Olivetti Programma 101 matches all of these criteria as being a programmable calculator rather than a computer. * The HP 9100A/B and 9810/9820/9821 match most of these, though self-modifying code was possible on the 9100's, although rather tedious to do. These machines could be significantly increased in capabilities with add-ons, but again, I'm only considering the base unit and not add-ons. * The HP 9830 isn't a programmable calculator by these criteria, though bitwise Boolean has to be "programmed" rather than being built-in functions. I consider this machine to be a computer. * The Wang 700/500/600-series calculators match these criteria, and are thus programmable calculators and not computers. They had some amazing computer-like capabilities when paired with peripheral add-ons, but as standalone machines, definitely were high-end programmable calculators. * Wang's 2200-series machines were definitely computers by these criteria. * The Computer Design Corp. (Compucorp, also sold under Monroe and Sumlock brands) programmable calculators stretch some of these criteria, but only when used with external devices. Additional programming capabilities/instructions were accessible through punched-card programming, which required an external punched card reader. These machines did have some really amazing programming capabilities that blurred the line between programmable calculator and computer, but only when used with external add-ons. * The Tektronix Model 31 stretches some of these (can print text, but can't really input or process it, and does have interrupt and DMA capabilities), but I still consider it a programmable calculator rather than a computer. * Programmable machines by Sharp and Casio certainly match these criteria. * Monroe EPIC 2000/3000 programmable calculators match all of these criteria. * Friden programmable calculators match all of the criteria, though the early machines (1151, 1152, 1154) had a really weird ALU and numeric representation that wasn't BCD or other four-bit decimal representation. * Mathatronics' machines match all of these criteria...at least running standalone. With add-on devices, the capabilities could be dramatically expanded, but I'd still consider these programmable calculators rather than computers even when maxed out with add-ons. I hope that this serves to clarify my statement as to why I consider the Olivetti Programma 101 a programmable calculator rather than a "personal computer". Rick Bensene The Old Calculator Museum http://oldcalculatormuseum.com

On 3/7/2012 7:58 PM, Rick Bensene wrote: I t could do register compares and other logic related to calculation steps It was fully equipped in the arithmetic department with operations and branching. In 1963 a 1403 4k memory module cost $250,000 to a friend who was implementing a fore-runner to Ticketmaster, automating a group of manual answering services to have a central location. The cost of memory was nowhere near what it would be a the end of the decade, and would blow the budget for any device that was "personal" if it had much. output was mostly on adding machine paper for me. No the operations were more complex than just hitting keys. There were a number of mode to take into account when doing programming. I recall doing exactly this, modifying a program. The register and programming memory was overlapped and could be modified while the program was running. I did just that as mentioned just to do so. On a device with so few resources there was not much one could do. Recall that there was 256 bytes (if memory serves) of ram on the Atari 2600. It had lots of program storage, none easily modified, but only a couple of orders of magnitude more storage than the 101 did. Admittedly there is way more to the story, but I don't agree with the subjective bits of size, etc. Architecturally this was a computer, not just a calculator. I just wish mine worked. The one I have locks up and doesn't run, sad to say, gets the red warning light and stops. Someday hopefully I'll get the time and bravery to fix it. Jim

On 8 March 2012 21:31, Tony Duell <ard at p850ug1.demon.co.uk> wrote: Perhaps Alan Sugar, sorry, /Lord/ Sugar was right in his description. Speaking of the CPC464, he said: ?Our ?rst computer was a very typical Amstrad concept. We sat down and observed all this computer stuff. And we saw what people were actually buying with computers. They needed lots of cables and cassette decks, and then they had to plug it into a television. ?I decided that the Amstrad philosophy is an all-in-one piece, so we would present our product as complete with a keyboard, cassette mechanism and monitor ? It looked like a mug?s eyeful for the old man when he walked into Dixons. He looks at this thing, with its whacking great big keyboard and a monitor, and he has visions of a girl at Gatwick airport where he checks himself in for his holidays. And he thinks, ?That?s a real computer, not this pregnant calculator thing over there called a Sinclair?? [p123 David Thomas. Alan Sugar: The Amstrad Story. Century, 1990.] -- 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 Tel: +44 20-8685-0498 ? Cell: +44 7939-087884

I beleive the Compucorp 322 was the first ever handheld programmable calculator, it beat the HP65 bya few weeks. Of course the latter was a much more capable mahcine (and I seem to remembner some of the HP advertising material for the 65 called it 'your personal computer' :-)). How much is known about the internals of the Compucorp machines? I've read the service manual on Katie Wasserman's site (I think that's where I saw it), but it doesn't have much detained information on the processor. I don't think there are even scheamtics of some of the boards, only those you could resonably repair, like the power converter. How do you classiby the single-board evaluation sstems (like the intel SDKs)? Those often ahd just a he keybod with function keys and a numeric-only display. Are those computers? Playing Devil's Advocate for a moment, I believe the String Varioable ROM was avialkable as a factor-installed internal option board (most of the ROMs were). And thys you could order an HP9830 from HP with string variable apability. Is that a 'stock' 9830 :-) I cetianly regard the 9830 as a computer. In fact I think it is a strong contentor for the title of 'first personal comptuer' Let me add at this point that the fact we can't agree on an exact dividing line between 'programmable calcualtor' and 'computer' doesn't make the terms worthless. Nobody can agree o nthe precise wavelength where 'orange' becomes 'yellow' but that doesn't mean it's silly to say 'connect the orange wire to take 3 and the yellow wire to tag 15' Realisitcally, some, if not many, prograambble calculators are close to being computers, even if they don't meet the strict definitions. I don't think anyone is goign to object to a discussion of the HP9810, 9820 or 9830 here (even if you all them calculators, they are clearly things that run programs, they are clearly over 10 years old, they are clearly interesting machines to some of us here(). And I don;t think anyone would object if you gave a presenation o nthem at a vintage computer event. My point is not that there can only be one fucntion on a key, but that to get a particular function you press are key with that name on it (maybe after pressing a shift key). You don;t type the name out. Or to be somewhat silly about it, to enter the SIN or COS functions on a comptuer you press 3 keys in sequence. The first key of 'SIN' is the same as the last key of 'COS'. I don;t know of any claculator where that owuld be the case. Corret, AFAIK. I don;t think even the advanced I/O ROM for the 9830 allowed user trapping of iterrupts. It was strickly something that the firmware could do at the machine code level. Some periperals used interrupts (the cassette drive, the last of the RS232 intefaces, etc), but they ahd their own firmware to cotnrol them, it was not something the suser could do. IU find it strange that a machien can change form being a 'programmable calculaotr' into a 'computer' if you add exxtensiuons to it. They were indeed. I think they're one of the landmark products. -tony

Can someone translate the advertisement: http://www.devidts.com/be-calc/adv_18199.html

Ummm .... can someone who * actually speaks the language * translate it please ....

On 03/08/2012 04:32 AM, Evan Koblentz wrote: So lets see : The P101 : a computer on your desk. With its small size and the ease of use of an office appliance, the P101 offers you all essential advantages of electronic calculations that up to now needed bigger machines. To know : speed, logic (?), programming- and reading capabilities. It is also the first desktopcomputer that can register programs on magnetic cards. Where the normal calculator is insufficient, and a bigger machine is undesired, the P101 will find its niche. For scientific investigations and technical calculations. For administrative and statistical purposes. Jos

This was about 10 years after my father bought our first house for $4000 in 1953, and inflation had not run up that much. 3900 was a lot of loot. I'm sure it was way less than the IBM 610 you cite, as almost everything then IBM was quite pricy, but I'd not attach the adjective inexpensive to it. Jim On 3/6/2012 11:07 PM, Rick Bensene wrote:

_first_personal_computer-0/ As I understand things the usual differentiator between a "programmable calculator" and "<stored programme> Computer" is that its possible for the program in a computer to change itself. So can the P101 store the result of a calculation into programme memory? <Note. US readers may not agree with this definition because it makes ENIAC into a stored program calculator whilst they may like to claim it's a computer>

On 3/6/2012 11:08 PM, dwight elvey wrote: Dwight, the feature that separated it from just a programmable calculator was that it could compute and store program steps on the fly. I wrote programs specifically to do that. One entered the programs as a series of steps, and you lost storage registers as you added program steps. However you could calculate the location of a program step and carefully modify it. It isn't much, but I think that tips it over into a really small computer with stored program. The cards were for backup, and reusable which was quite a feature in the days of punched cards. the Wangs that came later had only the card reader, and you typing to program it. Anyone recall whether it could generate program steps on the fly or not? I don't recall ever seeing anyone do it. I don't defend it as useful on a machine with effectively 64 locations total (48 years dulls the memory) and only used it to prove the capability. Jim