6502.org

Sorry for the title, I didn't know what to call it.

I was looking at a prototype cash register schematic. Looking at the address lines it is hooked up to some 74244 and then the address lines go to the rest of the system. The enable lines are tied to ground. There is the same circuit on the data bus also.
Is their any purpose to this?

Edit: forgot to put the picture

Some kind of electrical/static protection, or perhaps to provide a known level of drive?

The 74LS244 is an octal buffer/driver. In this application, it apparently is being used to boost the relatively weak output of the 6502's address outputs. As the address bus is always "on," so must also be the buffer, which is why the enables are continuously grounded.

Very neat, could be useful later on. When would you need to boost the signal though? Would that need to be done if the address lines are very long?

In the early days when the 6502 was NMOS only and the glue logic was mostly LSTTL, the NMOS could not drive that many LSTTL inputs. Current-production CMOS 65c02's however have much, much stronger pin drivers, and loads are usually CMOS anyway, giving you a benefit both coming and going, so to speak.

More likely is when there are many devices on the bus. Expansion busses often do this, but if the main PCB has a lot of devices connected it could also need buffering. A possible case in older systems could be when 1-bit RAM chips are connected to the bus, there'd be 8 of them all driven by the address bus (or at least a number of the lower pins).

I think this is especially important with the NMOS 6502's (and not CMOS 65C02's) as they can't drive as many loads on the bus, but you could look at the datasheets for that information.

WDC's data sheets have been notoriously deficient over the years, but fortunately the actual parts are usually much better than the data sheet lets on. In the case of the pin drivers, I suspect they improved them many years ago and just never updated the data sheet to reflect that. I have done some brief tests on WDC's W65C816S's pin drivers which I suspect are the same ones used they used on the W65C02S. Their behavior was pretty much symmetrical, able to pull up just as hard as they can pull down, unlike TTL which cannot pull up as hard as down. If you had to boil my test results down to approximations and treat the circuits as just a resistance, the data pin drivers acted very roughly like a SPDT switch with 50Ω in series with the common terminal (ie, the output); and the address bus pins, as a SPDT switch with 60Ω in series. The time constant of 60Ω times the capacitive load of 10 CMOS loads is around 3ns, which is less added delay than you'll get from a bus transceiver IC.

In a separate test on WDC's W65C22S VIA (not the W65C22N) I/O pins years earlier, I found they were each able to pull to within 0.8V of either rail with a 220-ohm resistor to the opposite rail, meaning a 19mA load, even pulling up, and give 50mA into a dead short. Rockwell's R65C22 could pull down with 100mA into a dead short, but could not pull up as hard, not being symmetrical like WDC's.

(This is from my page on the many differences between the NMOS and CMOS versions, at http://wilsonminesco.com/NMOS-CMOSdif/ .

Am I the only one who's thinking - Hey- Cash Register... 6502... win

-Gordon

A cash register is a fairly obvious application for a low-cost MPU like the 6502.

In the old days, simple mechanical adders had to suffice, and those had an awful lot of moving parts to lubricate and avoid bending. They also had to be replaced (in the UK) for the conversion from £sd to decimal currency - initially with slightly newer mechanical registers, hastily adapted (eg. by putting £.p stickers over moulded $.¢ markings) from designs for other decimal currencies. I think some shops simply rounded their prices to the nearest 2.5p, which corresponded to 6d in old money, and continued to use their old registers (as forty sixpences made £1, it was easy to convert on the fly, and the accounting would be correct in the end).

The 6502 came along too late for use with £sd, but it could have been programmed almost equally easily for both the old and new currencies. Where it came into its own would have been with barcode readers (initially the province of supermarkets) allowing very rapid and accurate accounting at the checkout. Not only the price but the name of the item could be listed on the receipt, and a summary of the exact items sold could be aggregated to guide restocking efforts, reducing the need to take daily and even hourly inventories of the shelves for that purpose. Weight scales could be tied into the system to automatically calculate the net price given the price per pound. Eventually cheque printers were added to the ensemble, meaning that the customer only had to check the printed amount, sign, and present the bank guarantee card for impressing, instead of having to laboriously write the correct trading name and the amount in words and figures...

Today's POS systems do basically the same job with a million times more processing power, fancy GUIs and probably a hundred times more system failures per month. Only now they interface with debit card readers instead of cheque printers.

Always fun to walk in to a store front with a modern computer, running a modern OS, on a flat screen display, with a 3270 emulator connected to the back office POS system.

I went into a local toy store at lunch and bought a vintage Star Wars action figure. The cash register was pretty old. Only had the 8 or so LED's to show the running total along with a paper receipt (the paper advanced every time he added another item).

After reading this thread, I wondered if there was a 6502 inside that machine. Next time I go in I will see if I can notice the make/model.

(call me weird...but I love old 70/80's cash registers)

Mind you, I've also seen shops being run without a register per se, just a cash-box and a solar-powered calculator. In the more deprived districts of Liverpool, you used what you could get, not what was ideal for the job.

The cash register is mostly designed to protect the business as much as it is designed to handle totaling receipts.

All of those "Free cookie if you don't get a receipt" signs you see are technique to deal with employee shrinkage and such, not so much to get you a free cookie.

I thought that was for avoiding taxes.. if they don't produce a receipt then it is (or was, see below) easier to keep the sale under the radar. In Italy it's actually illegal to not get a receipt, if the Guardia di Finanza catches you coming out of a shop (and you've walked less than 50 meters) without a receipt, then there's a big fine for you. And a slightly smaller fine for the shop. Not making this up, it's happened to several people I know. Including a friend's father who didn't have the receipt for an espresso (~ 30 US cents) - he was fined approximately US $60. The shopkeeper half of that. That was some years ago, I bet the fine is higher (the price of coffee too..)

In Norway the authorities now forces shops to install highly sophisticated cash registers in order to make sure that no taxes (or VAT) are avoided - not that it really was a problem in the first place - but the net result is that, due to the very high cost of these registers, small shops are going out of business simply because they don't have enough money to buy them.

My understanding is that cash registers were invented to reduce fraud by the staff. It was a huge and successful business - see NCR, where also Thomas J Watson got his big idea.

Of course, it might be that these days there are different motivations. But in the US, yes, I think it might well still be to ensure sales are rung up accurately.