Despite having bought it over a year ago, my Commodore 128D has not found its way onto my workbench as often as I’d intended. There are two specific reasons for this, the built in 1571 disk drive is temperamental to say the least and the fan that cools the machine’s power supply is not only noisy but cursed with a tedious whine.
Using the 128D therefore has been an exercise in irritation and frustration when it should be an enjoyable experience given the unit’s multiple computing personalities, a Commodore 128, 64 and CP/M capable machine all rolled into one.
I then became aware of the Commodore Flyer. This superb piece of kit acts as an internet modem and disk drive emulator, thus allowing me to forget about the flaky 1571 and transfer software into the 128D at my heart’s content.
I ordered a Flyer from Retroswitch and it arrived promptly from the States. It’s a really nicely engineered solution and I was quickly up and running, updating the firmware and pulling down disk images from the Commodore Online cloud service that I’d previously set up.
The guys over at Retroswitch have also written a Telnet client which works with the Flyer, although it will only run on a C64 or 128 in C64 mode. I loaded this up and pointed it at my BBS. It’s a little rough around the edges but it worked well enough for me to log in and check my mail, pretty cool with a C64!
As the Flyer is JiffyDOS compatible and the standard Commodore serial transfer rate is so woefully slow I decided to order a JiffyDOS upgrade kit for the C128D. On arrival this consisted of two chips, one to replace the kernel on the main board and one for the 1571 drive.
Once the kit had arrived I headed over to Jim Brain’s site and downloaded the installation instructions for the C128D. However it soon became apparent these instructions were meant for the C128DCR (cost reduced) model which differs considerably from the C128D internally.
I hunted around but couldn’t find installation instructions for my machine so I decided to trust my instincts and see if I could locate the relevant chips myself.
The drive chip was pretty obvious, being the only socketed chip of the correct size on the drive’s daughter card. Mine was labelled as a 310654-03. I gently prized it out with a flathead screwdriver and popped in the replacement.
I then discovered what I presumed was the 128′s Kernel chip, labelled as a 318020-03. Again I replaced the original chip with the new replacement, reassembled the case and powered on.
The first thing I noticed was that the 1571 was not springing in to life as it usually does on boot. However the 128 screen came up and there indeed JiffyDOS V6.01 (C) 1989 CMD was displayed. However there was no flashing cursor. I decided to completely disconnect the internal 1571 and try again. This time everything appeared to be working fine. A quick test using the Flyer to download a program from the cloud server saw an increase in speed from 45 to under 8 seconds, very nice.
I decided to replace the original drive chip and try again with the 1571 reconnected. Again, no cursor without a Runstop/Restore and even then no joy with either the Flyer or the 1571. Incidentally the C64 mode was working, but with no JiffyDOS support.
Concerned that in my haste I’d installed the chips in the wrong sockets I headed over to the excellent Lemon64 forum. There I learnt from the always helpful and knowledgeable members that the 128D requires the same set of chips as the plain C128. The chips I had were intended for the C128 DCR which is unfortunately also known as the 128D in America, despite being very different from the real European 128D. As with the C128 the 128D requires replacement chips for both the 128 & 64 Kernel, I believe these are combined on the 128DCR. Fortunately the original vendor has agreed to send me the correct chips at no extra charge.
Somewhat disheartened for now I removed the JiffyDOS chips and replaced the originals. As if that wasn’t irritating enough, when I rebooted the machine I was greeted with a black screen in 128 mode.
I consulted R Carlsen’s excellent document on common c128 faults and decided to swap the 6526′s in the U1 and U4 sockets. However on closer inspection it looks to me as though a previous poor repair to the U4 socket was a possible cause and it looked unlikely I’d be able to remove the chip without causing further damage. There’s melting to the actual socket, presumably from a soldering iron and some less than expert soldering to the board. I’m guessing that all that levering out and replacing of chips in the nearby U35 socket has damaged what was already a rather tenuous repair.
While I had the case apart I took the opportunity to address the other issue with the 128D, the deeply irritating fan noise. I found a replacement fan that purported to be silent and quickly replaced the original, which is mounted on the underside of the power supply. On restarting the reassembled machine there was a considerable improvement. Silent no, considerable quieter and less whiny, definitely.
So for now I have what is essentially a rather bulky Commodore 64. I may attempt a repair, although I do have a flat C128, which rather ironically doesn’t work in C64 mode.
Still a Commodore 64 was exactly what I needed as I’d recently found a box of old Commodore 64 floppy disks in the loft…
The BBC Master that I recently snagged on Ebay has finally arrived, along with a Microvitec Cub monitor and an Opus dual disc drive unit. It arrived in a pretty grim state, the Master was fairly thick with dust and the monitor seriously grimy. I cleaned up the computer unit and will tackle the monitor tomorrow.
I knew the sensible thing to do would be to open up the computer and give it an internal clean whilst also checking for leaked batteries and unseated chips. However I instead plugged it all in and powered it up.
Fortunately everything worked, well almost. I was greeted on the screen by the text:-
This is not a language
I made the assumption that this was down to a failed CMOS battery and restarted the machine with the ‘R’ key pressed to reset the CMOS configuration. A quick search led me to the default configuration which I entered and following a CTRL/Break reset I was up and running. My guess is that I will be presented with the same issue when I next power up and a battery replacement will be in order.
A fair pile of old and pretty tatty disks also came with the computer so I attempted to read a few them with no luck. It seemed inevitable that some of them would have failed but I was surprised that I was unable to read any of them. Another quick search led me to realise I needed to change from the ADFS to DFS filing system to read them.
I successfully formatted some new disks in both 40 and 80 track mode but have as yet had no luck copying files backwards and forwards between the dual drives. I’m also having no luck using Omnidisk to transfer disk images onto floppy but I shall persevere.
Ideally I’d like to get the Yahtzee code transferred onto floppy so that I can continue development on the Master and relive the school computer studies class experience as closely as possible.
On the other hand I can imagine spending the last few days of the challenge cleaning up computer and monitor, replacing the battery pack and possibly installing some sort of solid state disk drive.
I have one definite project that I want to complete this year, building a multi-boot bridge machine. This is more of an enabler than a full on retro project and will likely prove far from taxing, hopefully taking only a few hours.
Once achieved however it will allow me to seriously de-clutter my workspace and this in turn will hopefully allow me to get on with some other retro goodness without the continual hindrance of being surrounded in junk.
Due to the ubiquitous nature of MS-DOS and the various subsequent flavours of Windows most of the retro computers in my collection rely on connecting to Wintel machines to facilitate file transfers. Some of the software provided is very specific about which version of DOS/Windows it will play ball with. I therefore have a collection of machines running early versions of DOS through to Windows XP. My aim is to narrow this down to one machine.
Helpfully Microsoft still provide minimum specifications for Windows For Workgroups and XP.
Windows For Workgroups:-
- 80286 microprocessor or better for version 3.1
- 80386sx microprocessor or better for version 3.11
- 2048 kilobytes (K) total memory for version 3.1
- 3 megabytes (MB) total memory for version 3.1 (2 MB with no network installed)
- 3 megabytes (MB) total memory for version 3.11 (4 MB is recommended)
- 6.2 MB of hard drive space (14.5 MB recommended)
- Pentium 233-megahertz (MHz) processor or faster (300 MHz is recommended)
- At least 64 megabytes (MB) of RAM (128 MB is recommended)
- At least 1.5 gigabytes (GB) of available space on the hard disk
- CD-ROM or DVD-ROM drive
- Keyboard and a Microsoft Mouse or some other compatible pointing device
- Video adapter and monitor with Super VGA (800 x 600)or higher resolution
I delved in to my collection of motherboards and cases and decided to pair an Abit BH6 motherboard with a 300 MHz Pentium 2 ripped from a dead Dell Dimension. The BH6 has a good combination of ISA and PCI slots, USB, standard serial and parallel ports. It’s had an interesting few years, some of them spent hanging on the wall as decoration (regularly vacuumed in a non sympathetic manner) and the last 2 or 3 in a pile of other MBs in a damp, non heated shed. I was therefore a little sceptical as to whether it would still work.
I also found an Nvidia TNT 2 graphics card, two 128MB sticks of PC100 RAM, a 3.5″ FDD and a CD-ROM drive. I assembled all of the components into a suitable case and was somewhat surprised when greeted by a successful POST.
I decided in advance that the easiest way to boot into various OS’s would be to use Compact Flash cards. I have a number of them lying around in various sizes from 32MB up to 16GB and have had great success with them in the past.
I had therefore ordered an IDE to CF adapter with a back plate fitting so that the card could be swapped out easily. In retrospect a 5.25 Bay model would have been even more convenient so I may yet get one of those.
Once fitted and hooked up with a suitable cable I set about installing Windows for Workgroups, Windows 95 and 98 onto separate cards. The installations went pretty smoothly, I put DOS 6.22 and WFW on a 32 MB card, Windows 95 on a 1 GB card and 98SE on a 4 GB card. One other useful aspect of this method is that the cards can be easily removed and inserted into a card reader on another machine allowing easy transfer of files.
Having discovered some old floppy disks whilst rooting around for components for this machine it wasn’t long before it was called in to action. I installed a 5.25″ FDD into a spare bay and inserted the Windows 95 CF card. Once booted into Windows 95 I set about attempting to recover some old files as detailed here.
The build has therefore been successful, I’ll also do an XP installation and then I’ll be able to simply swap cards and boot into whichever OS I need. I shall also add an internal CF adapter with a 4 GB CF card onto which I will install all of the retro software that I need so that it’s always available to each OS.
The machine is reasonably quiet as the Pentium is passively cooled although the graphics card has a small fan, I may therefore root around for a fan-less model.
A sound card would be nice but not essential and I may see if I can hook up the two Voodoo 2 3D cards that I used to run in SLI mode.
An illness in the family has rather curtailed my retro activities this year. Any grand projects are unlikely to reach fruition, however I’m still tinkering with odds and sods.
I finally got around to Telnet enabling my Nostromo BBS, which was previously dial-up only. Trouble is I may now therefore actually get some visitors and that in turns means I ought to add some content.
I hadn’t realised, but my Broadband supplier at work, Demon, actually furnishes me with a static IP address so the process was very simple. I had previously assumed it was a dynamically allocated address. I also still have my original ‘tenner a month’ dial-up account with Demon, purely to retain the associated e-mail address. Next year will mark twenty years since I first signed up for this account and it occurred to me it would appropriate to create a suitably retro-style website on the included web space.
So you can now visit the hand coded Nostromo BBS Website replete with snazzy background, marquee, page counter and other early 90′s Internet goodies. Twenty years, I can’t believe it!
Since contracting this retro computing bug, after prolonged exposure to Urbancamo, I’ve had a yearning to try to recreate one of my earliest and fondest computing memories. That memory, when recalled, still has the ability to send a little shiver of excitement down my spine. Picture the scene if you will, a musty old office in the maths block of my local upper school. Two spotty teenage boys, myself and the aforementioned Urbancamo. In one dingy corner of the room an ASR 33 Teletype, on the window sill an acoustic coupler and telephone and on a piece of paper, a phone number.
Thinking we were Matthew Broderick from WarGames we eagerly dialled the number inserted the handset into the coupler and waited excitedly for the Teletype to burst into life. Sure enough the ASR 33 started clanking away and we were in! Not NASA or anything like that you understand, but a nearby college upon which we were able to play a game called Shark Attack! A momentous moment in my computing history and the first and last time that I ever used an acoustic coupler.
I have for some time therefore been on the lookout for a working acoustic coupler that I could use to try to recreate some of that experience. After failing to bring back to life a very old Anderson Jacobson ADC 212 model, I was recently lucky enough to sport an Epson CX-21 on Ebay.
It’s in lovely condition, and the original NiCad battery even appears to hold a reasonable charge. It came with an Epson HX-20 in a custom-made case which no doubt accounts for its excellent condition. There was a little blooming on the rubber cups but I cleaned this off with a mild detergent solution and the whole unit looks almost new.
I will of course also need an old telephone with suitable handset and fortunately we still have a couple of 1970′s models at work that thankfully were never thrown out. These old BT models were built to last and the one I chose cleaned up really nicely, in fact it looks so cool I may use it in place of our current home set.
The CX-21 only has two options to select, half or full-duplex and answer or originate mode. It has a standard DB25 port and I have the appropriate cable to hook it up to my PX-8, an Epson 724 cable. This is a modem cable with DB25 at one end and an 8 pin Mini DIN connector at the other. I have two options for terminal software on the PX-8, both previously downloaded, Kermit and Mex. The PX-8 does have a rudimentary terminal application on the built-in ROM but I’ve never had much success with it.
After charging the CX-21 and PX-8 I hooked up the TF-20 FDD and copied over all the software I’d need to the PX-8′s RAM disk so I had the minimal amount of stuff to take home from where I would be calling back to the box running my BBS.
I then spent rather too long trying to get my head around whether I was originating or answering the call, clearly I was originating it and the BBS box was answering but for some unknown reason I’d convinced myself that the coupler would have to ‘answer’ the handshaking attempts of the remote modem.
Nevertheless once sanity had been restored I set the coupler to ‘originate’ mode at full-duplex and dialled the BBS with the telephone. After several attempts the ready light finally illuminated on the coupler and it seemed a connection had been established. However once I’d connected via the terminal software all I received was a screen full of garbage.
I checked all my settings, made sure background noise was minimal and continued to make many more attempts to create a good connection. Eventually I discovered that contrary to my instinct to push the handset more deeply and securely into the coupler, lifting the mouthpiece end slightly out of the cup resulted in a much better connection! Whether the proximity of the transducers was causing some sort of distortion I don’t know but whatever the reason I now had a working connection and was reliably able to re-create it. There are still a few erroneous characters coming through but I can live with that.
I shot a quick video showing the process below, all I need now are a couple of ASR 33′s.
I had for some time been eagerly awaiting Apple’s update to its MacBook line of computers having managed to resist the allure of a new machine for quite a few years. My 2.4 Ghz Core 2 Duo MacBook is still working well but it’s a looking a little sad what with the broken plastic at the edge of the right hand wrist rest (apparently quite a common failure), the self inflicted white spots on the LCD and the general misalignment of the lid following a couple of drops.
If these issues weren’t reason enough my wife has been using the MacBook for a while (invariably forgetting to charge it) and for her requirements, mainly web browsing and mail it’s still perfectly adequate. Quite frankly that’s the only green light I needed, so having attempted to justify my extravagance let’s talk about the Retina MacBook Pro.
Not cheap is it, especially here in the UK where the base model comes in at £1799 or roughly $2795 at today’s exchange rate. Still I suppose that’s what you get when you are, as Stephen Fry put it, ‘an early adopting sillyhead‘. I ordered my 2.3 GHz model with 8 GBs of RAM as soon as the UK Apple Store came back online following the WWDC Keynote. Months of pent-up desire for a new machine meant there was no hovering over the confirmation button as with previous purchases, my blood was up, resistance would have been futile.
My shipping notification duly arrived on the 13th of June and I watched with some bemusement over the following days as my MacBook winged its way to old Blighty via China, Korea, Kazakhstan, Poland, Germany and… Castle Donnington. The packaged arrived, earlier than estimated, on the 18th June. Having seen the subsequent shortage in supply it seems I was very lucky to receive it so quickly with some who ordered at similar times to myself still waiting.
Coming as I was from the aforementioned MacBook the first thing that struck me was just how thin and relatively light this new model is and yet it manages to avoid feeling flimsy or insubstantial, quite a trick to pull off. Of course the thing that demands your attention next is the screen, impossibly thin and extraordinarily clear, bright and sharp. Gorgeous blacks on pristine whites and vivid colours all rendered with superb accuracy.
In terms of performance, I’ve had experience of SSDs having upgraded my Mac Pro but there they are restricted by the ageing SATA II interface so it was pleasing to see them operating nearer their full potential in the MacBook. Blackmagic Disk Speed Test showed write speeds averaging around 390 MB/s with Read speeds at 435 MB/s. Both results based on 1 GB file sizes.
Geekbench returned a score of 11065, pretty impressive for such a light and compact machine. I decided to try a game which I had previously downloaded for the Mac Pro via the App Store, namely Dirt 2. It’s not the most up to date game but I have no desire to install Windows on the MacBook as I will not be doing a great deal of gaming on it. The game ran at 2880 x 1800, the native Retina resolution, although it was stuttering slightly. The fans did get noticeably louder as the unit heated up and the battery charge took a severe hit. All I suppose to be expected, dropping the resolution to 1440 x 900 resulted in silky smooth gameplay, video here.
With two Thunderbolt ports and an HDMI port it occurred to me I could try to hook up three external displays. I plugged my two 24″ Apple Cinema Displays into the Thunderbolt ports and a Samsung TV into the HDMI and everything worked without a hitch. Initially I only mirrored the displays although I will also create a single desktop across them and test performance, it would be interested to know what ATI’s Eyefinity under Windows would make of such a set up.
Returning to the Retina display, I have mostly run it at its native resolution where it performs best, although non Retina enabled apps do look quite blurry, particularly when seen alongside those which are Retina compatible. Clearly Apple is hoping to force the issue and rely on 3rd party developers to release compatible software ASAP, although to be fair even some Apple software is I believe still not Retina optimised.
I’ve also used the ’1680 x 1050 mode’ which I quite like. Retina enabled apps still look pin sharp plus you get a little more real estate which can be useful. In the course of creating this blog entry I’ve used Grab on the MacBook to take some screenshots and when you open these captured images on a non Retina machine it really brings home to you the resolution of simple dialogue boxes. Click on the image of the Display Properties Pane shown opposite to view it at full resolution.
Choosing non standard resolutions does come with a performance hit as the screens are actually rendered at much higher resolutions and then scaled to fit. A couple of other negative points to consider, firstly battery life. Apple claims typical life of around 7 hours with light use. In my experience my machine is not achieving this, I would suggest it’s somewhere between 5 – 6 hours, although to be fair that is only an estimate. Fortunately the battery seems to charge quite quickly.
The other major issue for me is the sharp edges to the palm rests which results in quite painful indentations in the sides of your hands after only 20 or so minutes use. I really wish Apple would address this issue, yes the clean lines look lovely but surely it’s not beyond the wit of the designers to soften the edge and maintain the clean aesthetic.
So overall I’m very pleased with the new Retina MacBook Pro, it’s light, thin and gorgeous to look at with plenty of power and just a few minor niggles. I’m fully expecting this thing to truly come alive with the release of Mountain Lion which I’m eagerly anticipating.
This year’s Retrochallenge will soon be upon us, running for the entire month of July there’s still time to sign up and participate. Head over to the Retrochallenge Website and register your interest.
This year I’m hoping to hook up my recently acquired Epson acoustic coupler to my PX-8 and with the aid of an old telephone dial in to one of the few remaining dial up services out there, and no doubt my own BBS.
In my previous blog entry I posted some scans of some promotional literature for the PX-8 and it occurred to me that I’m now in possession of the all the hardware shown in the image opposite. I thought therefore it might be interesting to try to recreate the shot.
Other options are to create another instalment for my Silent Running text adventure created with Inform.
Following the previous upgrades to my 2006 Mac Pro I finally decided to take the plunge and swap out the original dual core Xeon CPU’s for some quad core ones. A matched pair of Xeon 5355′s came up on Ebay at a sensible price and from a reputable seller so I snapped them up. The seller actually had four pairs for sale and they were sold pretty quickly so there is clearly still some demand for 5355′s.
There are a multitude of videos available on Youtube detailing the relatively simple procedure of removing the old Xeons and replacing them. All you need are the new processors, some good thermal paste, a couple of Phillips screwdrivers and a 3mm Allen/Hex key which needs to ideally be at least 6″ long. You’ll also need something to clean the old thermal paste off of the heat sinks, isopropyl alcohol is ideal, anything else risks leaving residue which will negatively impact on the heat transfer between CPU and heat sink. Optional extras are a can of compressed air for cleaning out any dust build up and an anti-static wrist strap.
I won’t go into the full procedure here as others have already documented it well. A couple of points I would make are firstly many who have performed the upgrade refer to issues removing the memory cage screws, I had no such problems but did use a small jeweller’s screwdriver which allowed vertical orientation and full purchase with the screw head.
Secondly, I was surprised by how thick the factory application of thermal compound was when I removed the heat sinks, I had expected a thin veneer but it was definitely thicker than that. This rather threw me as to how much compound I should apply on re-installation.
In the end I decided to apply a thin line horizontally across the CPU’s with an additional smaller vertical line making a cross. I didn’t spread out the compound but instead allowed the pressure of re-attaching the heat sink to flatten it out.
Reassembly was straightforward, I even managed to complete the whole operation without dropping one screw. The machine promptly booted up and I ran Geekbench to assess what performance improvement the new Xeons were giving.
With the original Xeon 5150′s Geekbench would typical return a score in the low 5000′s. With the 5355′s the score has leapt to 9456, a very useful improvement in performance bringing my 2006 Mac Pro within touching distance of a 2012 iMac Core i7.
The only thing left to do now is monitor the CPU temperatures for a while to ensure the thermal compound is working well. There are a number of different apps out there that claim to give accurate read outs of the Pro’s thermal sensors, however in my experience they all seem to give different results. I’ve therefore decided to trust the software that I’m familiar with, namely Marcel Bresink’s aptly named ‘Temperature Monitor”.
This is showing the cores running at an average of about 40°C under light load* which is around 30% higher than the previous processors. Considering there are twice as many cores and that the thermal compound has not broken in yet this seems pretty reasonable to me. Also a considerable upswing in the ambient temperature that has coincided with the upgrade must be taken into account.
I decided to download SMC Fan Control and crank the systems fans up by a couple of hundred RPM just to be on the safe side, they are still whisper quiet. There is one last issue that needed resolving, when you perform this upgrade and click on About This Mac you’re likely to get a processor unknown reading although System Report will show the correct number of cores. There are a couple of solutions to this, you can try upgrading the 1,1 firmware to 2,1 using this utility. Or you can do what I did which was download the CPU injector Kext from here and install it with a kext utility, I used Kext Helper b7 here.
*After three months of running smoothly the CPU temperatures under light load are now averaging around 32°C so clearly the thermal compound has bedded in well and I’ve been able to leave the fans running at factory settings. I recently installed Windows 7 and have been enjoying a number of games, the machine remains stable and reliable.
I’m still enjoying playing with my Psion MC 400. I wanted to find a reliable way of getting data into and out of the machine. I believe there was a portable FDD available which I would absolutely love to get hold of but I think the chances of finding one of those are very slim. As far as I’m aware there is no way of reading and writing to the SSDs in anything other than a Psion unit.
My attention therefore turned to the RS232 interface. I was somewhat concerned when I first noticed that the Psion employed a 9 Pin mini DIN connection for the Rs232 port, this connector in the same configuration is found on the Epson PX16 and I’ve never managed to build a working cable for it.
However I discovered on this site that the serial cable that Psion sold for the Series 3/3A is compatible. The cable comes in two parts, a pod with hardwired cable with a special connector for the Series 3 and a second cable with a DB9 serial connector at one end and the required mini DIN at the other end which plugs into the pod. This second part of the cable is exactly what you need for serial transfers to and from the MC 400.
Psion link software is built into the MC 400 but I needed to download PsiWin for the PC end. I found various versions here and downloaded both version 1.1 and 2.3.3. I installed the former on my Windows 98 Wyse terminal and the latter on Windows XP running under Parallels on my Mac Pro.
After ensuring the Com port settings on the Win 98 box were correct and connecting up the cable I launched the Link application on the Psion and then PsiWin on the PC. The link was immediately established and I was soon able to drag and drop files between the machines. I was very impressed with the PsiWin software.
Then using a USB to Serial converter cable I hooked the Psion up to my Mac Pro and launched PsiWin 2.3.3 under XP. Much to my surprise this also worked well. The only issue I have is that the PsiWin software is designed to convert the Psion format files to Windows friendly ones during the transfer process, however neither version of PsiWin seems to understand the files produced by the MC 400 and conversion therefore fails.
One other thing I wanted to try was to connect to my BBS using the built in Terminal application. Using a null modem adapter I was able to connect the Psion serial cable to my US Robotics modem. Modem options including Baud rate, frequency, pulse/tone dialling etc can be set within the terminal application. After a few aborted attempts I found the correct settings and successfully connected to my BBS. I made a short video of the process and another quick video of general MC 400 operations:-
Probably in my top five of most wanted retro machines I’ve finally acquired a Psion MC 400 mobile computer. Originally released back in 1989 this ill-fated machine, described by Time Magazine in their All Time 100 Gadgets as one of five gadgets ahead of their time, was initially priced in the UK at £845.00 and according to some sources sold less than a thousand units.
It is therefore unsurprisingly quite rare. Way ahead of its time it featured the first touchpad, removable storage in the form of sold state disks (SSDs) and a claimed battery life from 8 AA alkaline batteries of 60 hours.
Lack of compatibility, proprietory interfaces and a high price no doubt all contributed to its lack of success but you have to admire the manner in which its designers and engineers pushed the boundaries to create a gorgeous product that is reminiscent of what Apple do so well.
The Psion is powered by an Intel 80C86 processor and has 256 Kb of memory. For external connectivity there’s a standard parallel port, a 9 pin mini DIN RS232 port and Psion’s fast serial link, a precursor if you like to USB. The parallel and RS232 ports are contained within a removable module that slides into one of the two identical sockets found on the rear of the machine. Other modules including a modem were available, my MC 400 has a dummy module in the second socket.
There are four sockets for SSDs, two on either side of the machine and these are compatible with the SSDs used in the Series 3. The touchpad mimics the screen, so the position of your finger on the pad will correspond with the cursor position on the screen and the whole pad is clickable. In practise it doesn’t work particularly well, needing a lot more pressure to move the cursor than we’ve grown accustomed to with modern machines and accuracy is poor. Fortunately, navigating the GUI with the keyboard is pretty efficient once you’ve learnt the required key combinations.
The screen is a non-backlit retardation film LCD with a 640 x 400 resolution. It’s crisp but needs careful positioning of lighting for optimal viewing. The contrast can be adjusted by way of a dedicated function key found to the right side of the touchpad.
The MC400 runs the Psion developed EPOC graphical operating system, a preemptive multitasking OS which would in later years evolve into the Symbian OS adopted by Nokia for its early smartphones.
A suite of software applications are supplied on SSD, a text editor, diary, personal database, calculator, file manager, terminal emulator and word processor. Not a great deal of other software was available as far as I can tell although my MC 400 did come with an additional spreadsheet on SSD.
There were two NiCad battery packs available or you can use 8 standard AA type alkaline batteries which reportedly give an extraordinary 60 hours plus of usage. I’m currently using rechargeable Duracell AA’s and will be interested to see how long they last.
The included Link software can be used in conjunction with Psion’s PsiWin software (available here) and a suitable cable to transfer data backwards and forwards between the Psion and a PC via the RS232 interface. I was concerned it would be difficult to locate or build a suitable cable for the non-standard 9 pin mini DIN but discovered here that you can use a Series S3/S3A’s lead.
I plan therefore to acquire one of these cables and see if I can get PsiWin working and also if I can hook up a modem. I’m also going to look at doing some programming with OPL, Psion’s structured programming language.
Some useful resources:-