Talk:Research Machines 380Z

Can 380Z/480Z software be emulated on the PC? Drutt 14:09, 1 March 2007 (UTC)

Name should be changed to RML 380Z. 86.151.21.56 18:35, 15 March 2007 (UTC)
 * Was that it's official name? The box itself just says Research Machines 380Z on the front. Drutt 20:18, 17 March 2007 (UTC)
 * I have moved this article to Research Machines 380Z, being the official name of the machine. RML 380Z and RM 380Z redirect here. Sidefall 15:19, 25 August 2007 (UTC)

Was colour video output ever an option? 18:35, 15 March 2007 (UTC) Yes RGB output — Preceding unsigned comment added by 94.67.100.205 (talk) 16:51, 9 January 2012 (UTC)

Should the RM 280Z and Link 480Z also be covered by this article? Drutt 20:18, 17 March 2007 (UTC)

System hardware designed by Mike Fischer, software by Dr David Small. 280Z was a hobby system, just the CPU and VDU boards, No case or PSU. 380Z was the initially supplied in a grey/blue box (in pretty small numbers. The more robust black box was supplied by Vero Cases. The sales manager from Vero was John Bowden, who became the sales manager at RML. Hardware wise the system initially loaded software via cassette - the firmaware was called COS (cassette operating system). Early systems had a 40x23 VDU line display. This was later updated to 80 characters wide. A separate hi-res graphics card was developed (sorry don't know the screen resoultions). These would drive TTL RGB monitors (mostly Microvitec), but an optional card was available to drive analogue RGB monitor (mostly converted ferguson TV's). Disks were introduced in the form of single and double sided 5.25" and 8" floppies. 5.25" single / twin drives inside the 380Z case. 8" floppies in a rather large heavy extenal case. Initial drives were BASF. Also produced were a PIO board with a wire-wrap prtotyping area; An IEEE-488 interface board and finally an A/D convertor board (which has a very expensive Ferranti A/D chip on it). A network interface was developed - based on the Zilog SIO chip - called Z-Net. This was a poor man's Ethernet system, 1Mbs CSMA/CD running down RG58 or URM43 co-axial cable. The Network as called the Chain 64 and used 380Z's as servers (initially using 5Mb 8" winchester disks). Servers ran CP/NET which was a DR Networking product based on MP/M. Stations were 480Z's which were basically designed as workstations, but were developed later as stand-alone systems with floppy disk drives. The 480Z booted CP/M across the network using a priorietry remote boot protocol and were primarily sold as diskless workspations. Interestingly, the first prototype Chain-64 network used the IEEE488 intrumentation bus. —Preceding unsigned comment added by 217.155.14.86 (talk) 10:31, 31 August 2007 (UTC)

Anyone else remember CTRL-F, J103 ? 71.77.10.170 (talk) 22:09, 18 August 2011 (UTC)

Bus
This was a long time ago... memory might not be 100% correct. I understood the bus was S100, not a ribbon cable. The machine powered up to a screen full of numbers (probably in hex). I was told at the time this was some sort of memory dump. J100 and J103 were typed to get the thing into Basic - one never knew which one would work. I was always told J stood for 'jump to'. I always remember the dual 8" that we had having 400K capacity each, but could be wrong. That much space seemed enormous at the time! The discs were turned by a mains synchronous motor via belt. That always struck me as not exactly crystal controlled. IIRC most of the machines we used had 1k 'graphics', one had a lot more, perhaps 16k. 'Graphics' for the 1k machines really meant what is now known as arial-alt, ie a font where each character consists of 6 blocks in many combinations. Although this enabled graphics of a sort using text, there were 2 issues with that. 1st not all combinations were present in the font! 2nd I remember it being unworkable to calculate each block & translating that into the font - but my programming skills were hardly good at the time. Lastly I remember getting it to run at 4x the speed by peeking & poking basic to change the word commands to 1 or 2 letter words. I made 2 versions of this, one for max speed & the other for compatibility with existing programs - but it was still necessary to go through the basic program looking for variable names that would now be misinterpreted as commands by the machine. The BBC B used this same approach to shrink & speed up basic programs. 82.30.111.65 (talk) 00:51, 30 December 2019 (UTC)