A Tribute to the Dick Smith System 80
(aka Video Genie and PMC-80/81)

Micro 80 Review of the System 80

This review has been taken from the Australian-produced magazine Micro-80. Issue, 7 (JUNE 1980) pages 5 to 11. The pithy style is typical of that era in computing.

** SYSTEM 80 ** A review by Peter Hartley.

It is now twelve months since the Dick Smith organisation announced the System 80. Ever since, an amazing crop of tales, some tall, some true, have surrounded the new machine. We have been told of "World-wide shortages of memory chips", 'Dick Smith and Tandy Corporation have their lawyers at forty paces", "Factories and warehouses have suddenly burned to the Ground", and even "Microsoft have gone bankrupt...

The System 80 is NOW a fact - it IS real - we've touched one! like - some things we could quickly come to hate! And that is no more, nor any less than could be said for any other computer in the market.

Before getting into too many specifics, let's make our position quite clear. Micro-80 has purchased, over-the-counter, one Dick Smith System 80. Unless we state otherwise, the comments we make relate to this one unit, and may not he generally applicable to every System 80 sold, even though it is a mass-produced item.

The System 80 arrived in an adequate (just adequate enough to get it home, but not adequate enough to survive being sent to Sydney by carrier if it became unserviceable!) silver-grey box. Inside, the System 80 was snuggled up in polythene bag, well supported by two moulded expanded-polystyrene frames. A separate bag contained cassette connector leads (for the second cassette) and three instruction books were in a third bag.

The visual appearance of the System 80 is brilliant. Almost as good as the Sorcerer, and far more attractive than the TRS-80. The mains lead was of sufficient length for most applications, and we had the choice of using one of our Tandy monitors (the System 80 doesn't come with its own monitor) or using the inbuilt U.H.F. modulated output and a regular T.V. We took a punt and plugged in old faithful - wonder of wonders - the pin configuration is a Tandy standard! A couple of moments to reset the vertical hold and the screen said...

READY ?

typing in... produced..

Our first surprise - a blank screen, with another

READY ?

right at the bottom. Funny thing, that, we thought, and reached for the (CLEAR) key. Surprise number two. We couldn't find it. Above she key labelled (NEW LINE) (which is the same as the (ENTER) key of the TRS-80) was one labelled (bck/spc), but no right-pointing arrow (the TAB function) and no (CLEAR).

A further inspection revealed that the up and down arrow keys, which we have come to expect on the left hand side of the keyboard, were labelled (escp) and (cntrl). Pressing (cntrl) produced a line-feed, just as we thought it should do, but pressing (escp) produced a squared bracket symbol.

Immediately our thoughts were of our growing library of software - and especially of the games that used all four "arrow keys". Submarined'.

We then decided to inspect the character set, using the following routine (which would also test whether the video memory used the Tandy addresses)

l0 CLS

20 M=l5360

30 FORK=0TO255

40 POKEM,K

50 M=M+2

60 NEXT

70 PRINT@5l2,"";

80 INPUTD$

90 PRINTCHR$(23);

100 GOTOl00

The good news is that it worked, but there is one Hell of a lot of bad news. Some parts of the character set weren't TRS-80 (generally the useful things like arrows), and line 90 appeared not to work at all! However, when we pressed (BREAK) we were greeted with a double spaced (but otherwise normal-sized) READY. This caused a hurried grab for the manuals, and the subsequent discovery of a huge button on the rear of the System 80, labelled (most inappropriately we thought) VIDEO CUT. Pressing this causes the System 80 to switch from 64 character node to 32 character mode, but NOT IN A TANDY COMPATIBLE FORMAT.

When in 32 character mode, the TRS-80 displays ALTERNATE bytes of video memory. The System 80, however, displays either the FIRST 32 bytes of each line, or the LAST 32 bytes, according to the position of yet another switch - labelled (PAGE). Printing CHR$(23) on the TRS-80 trips a gate, to enable the 32 character format, but on the System 80, the same instruction merely causes the unit to insert a space between each character PRINTED AFTER THE CHR$(23)...

There are some other changes, too. There doesn't appear to be a L3 ERROR message. The Disk Basic Pointers, used in Level II, are addressed to SN ERROR instead. But, importantly, these pointers ARE there, so we are led to the conclusion that the machine will eventually (and we'll discuss "eventually" eventually!) be compatible with TRSDOS and other MICROSOFT DISK BASICs.

We discovered that the original READY message, at power-up, was in fact, the MEMORY SIZE? message heavily disguised. This was almost missed in the instruction manuals - about which we will also say more later. We also discovered that to execute a right TAB you must press, simultaneously, the (SHIFT), (CNTRL) AND (I) keys:

Because this unit is already on sale in other parts of the world as the "video genie" and probably under other names also, there are no on-screen copyright messages. These have been zeroed out in ROM. Nevertheless, it is the full Dartmouth/Microsoft Basic, and a small sticker beneath the unit proclaims that the BASIC interpreter is (C) l980 by MICROSOFT.

Because of certain hardware changes there are some other small change in the ROM. These relate to the line-printer operation (on the System 80, the line printer is port addressed only) and to the operation of port 255 (used on the TRS-80 for the more sensible 32/64 graphics as well as for cassette output on both units). There is also a change to the READ logic. We have not yet been
able to determine the long-term consequences of this alteration, but it is almost certainly of little consequence.

** ** However, the important news is that all the TRS-80 routines are in the System 80, and ARE located at the same addresses.

** ** The BASIC in the System 80 IS TANDY Level II Basic.

** ** There is, therefore, no reason why any TRS-80 software, either in BASIC or machine/language should not work PROVIDED that is does not require arrow symbols in its video display and also provided that the user will not be required to input either a (CLEAR) or a (TAB) from the keyboard, and provided that the use of large graphics is not absolutely essential.

In some circumstances the (ShifT) (CILRL) , (I) method of generating a TAB will not work. Although the function creates an identical result when executing a string input, it does not generate the same responses in the keyboard memory, and some of our authors use direct peeking into this memory for speed.

The System 80 boasts an inbuilt cassette recorder. We used to think this was a good idea. Now that we own a System 80, we KNOW that it's a BLOODY SILLY IDEA - unless it works perfectly - and the one on our on unit (and according to reports from readers and others all System 80's are the same in this regard) is totally R.S.

Because there is no loudspeaker or other audio-outlet (not even a socket for an earpiece) it gets amazingly frustrating when the System-80 refuses to load a perfectly good tape, since you cannot find out where you are without getting another tape recorder anyway, which seems to us rather self-defeating!!!!

Because loading the System 80 is even more fussy and inconsistent than the unmodified Tandy TRS-80 (yes, even such a terrible thing is possible - and Dick Smith has done it!) - thank you Fort Worth for the XRX III! - you are forced to use the second cassette port. Fortunately the plug is again compatible with the TRS-80. Unfortunately, SYSTEM will not work on #-2, so if you are trying to load a machine language tape you will resort to ripping the System 80 apart - to see what gives - and you'll find the most horrible tape recorder East of Aden. In fact, we would say that if this unit were manufactured in Japan, (where they give export a capital E) they probably wouldn't allow the unit to leave the country!

Both the mechanism and the electronics of the cassette are woeful. The major reason for our unit failing to load is a grotty little switch, vaguely connected to the record button. This is supposed to switch between play and record. It might succeed if the connection didn't flap about like a barn door in a gale - 'and since the connection is made of spring steel, it's kind of hard to bend it without breaking it.

The erase/bias oscillator is a radio-technician's idea of purgatory, visibly degrading the otherwise excellent (superior to Tandy) video output whenever you go into record mode. (Transmits well on F.M. too.') If you use the UHF modulated output, you might as well turn the T.V. off while you record, because you'll only get upset, otherwise. (Has the F.C.C. seen this yet?)

There appears to be some sort of constant-volume device, but it certainly doesn't make for easy loading.

The rest of the construction seems to be a strange mixture of good ideas and conservative design, kind of screwed up with odd bits of penny-pinching.

For example, the Keyboard assembly uses a high-quality keyswitch, of the type used in office adding-machines, and is held down by no fewer than EIGHT screws. Yet, the associated metal-work has provision for nine extra keys (and they left two out?) and the pcb. would easily accommodate these. Then there just isn't any physical support for the centre of the keyboard assembly, which will be very necessary when the cassette deck almost begs you to smash your fist through it.

The interconnection between the boards (keyboard and two main p.c.b's) does not use flexible printed-circuit, but steel (!!!) wires in a very rigid clear vinyl former - soldered at one end and plugged in to what looks like a pcb. edge-connector at the other. Strong it is. Electronically, we suspect that it might create some high resistances. Physically - well, it won't break easily, but be prepared to spend two hours re-connecting the socket end if it comes unplugged.

The power-supply is in its own self-contained module. This may be a convenience of manufacture since the System 80 is destined for world-wide distribution. The module has a mesh grid in just the right place to stop little fingers from poking where they might get hurt. At first we thought that this may have been screening to prevent some of the spurious R.F. interference from getting into the mains, but since the screen isn't earthed, and isn't continuous, it seems more likely that it was an after-thought to stop people like us from looking it the power-supply board - it is not the most beautiful piece of wiring that we've ever seen - but it does work.

One nit-picking comment here. The screws that hold the power-supply cover grid down look exactly like those that hold the keyboard down. Unfortunately, if you put a keyboard screw into a power-supply screw-hole, you wont notice, until you find that one of the keyboard screws won't do up tightly/ By then it is too late, because the power-supply screw-hole is now too big for the right screw. Why not standardise?

The System 80 purchased by Micro-80 was a 16K unit. Only one I.C. was in a socket, and that was the Z80 itself. We didn't like this, because it makes servicing harder than it need be. We sincerely hope that the 4K versions DO use sockets. Dick Smith and other retailers sell memory upgrade kits at very sensible prices, but the ability of your typical 4116 to withstand the average computer hobbiest's soldering is extremely limited and single prime 4ll6's currently cost around the $14 mark.

The rest of the internal construction was excellent. Pressure of time has precluded our attempting to analyse any of the circuitry, but the tracks on the two main pcb's were well etched and appeared solid, as did all of the joints. Interconnection between the inbuilt cassette and the main unit was through a small but solid plug, as were the connections from the power-supply module. There was no evidence of any dry joints or solder-balls, and with the exceptions noted, the System 80 appeared to be a solidly constructed piece of equipment.

Now to the instruction manuals. We thought it surprising that something so very long in the production pipeline as the System 80 would have any need for addendum sheets in Its manuals, but addendums there are - and some of them are very basic ones at that - like how to protect your high memory... how to access machine-language subroutines, etc.

On the whole the manuals are adequate, but we think that most owners quickly be heading off to their local Tandy Store for copies of the TRS-80 literature , to obtain adequate expansions on many of the commands. There is evidence of an attempt on the part of the manual's authors to emulate David Lien's witty and pithy style (a La Level I handbook), but it doesn't come off. At least one reader has been trying to get technical literature and mud-maps from Dick Smith Electronics - so far, we understand, without any great success, although there is some hint of "eventually..."

Which brings us to that very question: how long is eventually? Garry Cratt, Service Manager with Dick Smith Electronics, was in Hong Kong during the first week of June, talking to the production engineers at the plant of origin. We understand that he had a whole bag of TRS-80 software that just wouldn't load/run on the System 80, and was trying to (a) solve the cassette problems, and (b) get some of the "improvements" removed (like getting those two missing keys put back). The end results are not known to us at Micro-80, but this does show that Dick Smith Electronics have their hearts in the right place.

The System 80 was announced as having Sl00 compatibility. It doesn't have any compatibility at present. The one outlet port is System 80 50 pin bus, presently unique.

Until we heard of Garry Cratt's visit to U.K. we were very doubtful of D.S.E.'s long-term plans for the System 80. In fact , we had gained the impression that they had lost all interest in the project. We knew of some 500-700 units coming into Australia. We knew of a sign in some Dick Smith Electronics stores implying that the wait for any upward expansion of the System 80 would be indefinite. As late as ten days prior to the Sydney Home Computing Show we were advised by a spokesman for Dick Smith Electronics that they would not be exhibiting the System 80, although, in the event they did. (Perhaps because their stand was next to TANDY?) We heard of an internal Dick Smith Electronics memo asking their Store Managers whether they would be able to sell an additional 500 units Australia-wide if an order were placed. (500 units! Surely if this computer were even 75% of what it was announced to be, they would be talking of 5 or 6 times that figure?) Just mention "upward expansion" to a Dick Smith Electronics store assistant, and you'll be very suddenly told that the "real home computer, of course, is the Exidy Sorcerer

However, one reader has supplied us with a stack of photocopies - purporting to be circuits of the Sl00 expansion for the System 80. In fact, the accompanying notes even gave the unit a code number, but the circuits appeared to be of individual S100 boards from within. Nevertheless, one Dick Smith Electronics source has suggested that the wait may only be a couple of months. Certainly, if Dick Smith Electronics don't come out with an expansion, it's London to a Brick that someone else will.

Remember, this computer is coming from Hong Kong (allegedly from a Toy Factory!) and is clearly designed for massive quantity production. Within two years, we predict that the System 80 (or Video Genie, or Alpha One, or what-have-you) will be all over the world, under a dozen brand-names, and you will probably be seeing the C.B. boom (and crash?) all over again with a new product. There will be a dozen competing expansions available - EVENTUALLY.

We have been advised of an impending price rise on the System 80 which will bring the 16K version up from $699 (Aust.) to $750 on August 1st.

Now let us do some sums...

Now, even that sum puts the Dick Smith System 80 in a very favourable light. Even if you buy a decent tape deck fron your local big W and put it in the System 80, you end up at least $150.00 ahead - and that pays for you next 16k of RAM chips.

To sum up then, the System 80 is not without it's faults. Dick Smith Electronics seems to be trying to fix these problems...the machine has an identical computing capacity to the TRS-80, for a lot less dollars. For any hobbyist who is not looking for upward expansion in the immediate future, or for someone who wants to make thier fortune writing software for the System 80/Alpha one/what-have-you. a bargain! If you want immediate reliability and expandability, we can only say that you were told.