Binary Dinosaurs Computer Museum
button1Museum History button2Museum Updates button4Adverts&Reviews spaaaaace button5Moan, Bitch, Gripe scroll1
button6Inhabitants button7Reviews button8WOW! button9Contact button10Recursion 2017 scroll2
button11Links button17Floppy Recreation button13BDonFacebook button14CGE-UK 2004 button15WROCC 2006 scroll2
button16DECBOX button12Retro2017 button18Floppy Recreation spaaaaace spaaaaace scroll3
base blank_textbox

GBS Gerald McMullon
The following is the set of notes appended to an email from Gerald McMullon, ex of Grundy. He's given me permission to put it up verbatim since it's all previously published documentation. Thanks Gerald! Note that any phone numbers given below will have changed since there's been a couple of UK-wide phone number reorganisations since these notes were first published. I wouldn't be surprised if the addresses are no longer correct either :(
Nutshell history (from memory):-
The NewBrain project was one of many that Clive Sinclair started. Mike
Wakefield was the designer and Basil Smith the software engineer. When it
was shown that it could not be built and sold for the sub-£100 Clive
Sinclair wanted another design. The project got sold to Newbury Laboratories
who made printers and VDU terminals in a cream and brown colour. The
NewBrain was to be silver and orange. When the prototype case was run off it
was a few pounds to make brown and or cream (Newbury colours) but a
re-tooling to run with different ABS plastic colours. So the prototypes were
cream and brown. I think production went that way because no one told the
manufacturer to switch the colour.
Newbury Labs pushed the NewBrain for the BBC School Microcomputer project.
The BBC wanted a 'different' BASIC, colour and hardware ports to control
equipment. Acorn, Sinclair Electronics and others all claimed they could do
this but it was Acorn with the history of the Atom and the strong design of
the Proton that made it through. The Spectumn was later added in, under
strong pressure from Clive Sinclair himself. No other manufacturer had the
presents of character as those at Acorn and Sinclair himself.
Newbury re-organised, wanted to go back to their root of printers and drop
computers. Grundy, who already has a Superbrain clone running CP/M for some
reason took on the project and set up Grundy Business Systems.
For several years the NewBrain was always a promise and no delievery. This
was in keeping with all the other manufacturers who even took money, banked
it and did not deliver for months.
Christmas 82 saw no manufacturer meeting demand and UK facilities to
manufacturer too slow to react to demands.
GBS over 82/83 developed a range of plug on hardware for the NewBrain,
including floppy disc controller and CP/M 2.2 (CP/M 80), a paged memory
expansion that took the three 8K ROMS and 64Kb of RAM and added hardware for
parallel printers, hardware serial ports and 8-bit ports to control other
equipment. Nearing full manufacturing were 8 and 16 way concentrator boxes
that allowed 8 or 16 additional serial ports, an I/O module for parallel
printer port, hardware serial ports and 8-bit parallel ports and a ROM pack
for bespoke software application.
One project was the battery powered model that Vestric provided the software
for use in pharmacy stores. Later they used a standard NewBrain AD with with
an external battery module and ROM card.
A frequent criticism of the NewBrain was the calculator style keyboard. The
keys have centres placed identically to a modern PC keyboard and with full
mechanical bounce. The key shape was with vertical sides, so they looked
small. The space bar and return (enter) keys are small. The triple folded
back circuit board fitted in without much air space, making it one of the
most compact designs about that retained a standard spacing keys. The reason
given for the keyboard design was that it allowed for dust to be wiped away
without it getting in side the computer (and looking at the IBM keyboard I
am using I still consider that the NewBrain keyboard had a lot going for
it). I can wipe clean a NewBrain keyboard in a couple of minutes, about the
same amount of time it takes to get one row of keys off a 105 key keyboard!
Heat was expelled by the use of a well designed heat sink that also provided
anchor points for the three circuit boards and backplane. The NewBrain runs
The directors and their teams were working on several new models in spring
of 83. The arrival the Apricot systems with magic numbers of 4(6?)MHz 16-bit
8088 processors and 256Kb of RAM with full screen colour and faster
performance in some tasks gave all 8-bit home micro manufacturers pause for
thought. A color NewBrain processor unit was under design and a 16-bit 8086
co-processor model as well as changing the VF display to LCD, like the
There was also considerations for a CMOS low powered battery operated
version. The heavy battery module would run a NewBrain for about 30 minutes,
under CMOS this would have been hours (like a Husky). Some home-builders had
replaced all components possible with CMOS alternatives to achive a cool
running low power consumption machine.
In truth GBS had a hard time getting the design to production qualities of
the disc controller and expansion modules and the metal box enclosure and
system power supply.
A range of CP/M software now included NewBrain installation (for screen
mapping) and was available on NewBrain format floppy diskettes. Preachtree
was in the process of badging up their database and account suite for the
Until a faster processor and faster screen mapping hardware with colour was
made there was little interest in the games market at GBS. To make a good
game required assemblier skills and an understanding of NewBRain I/O streams
and after that effort it was still black and white.
Third party manufacturers started to use the NewBrain and there was even a
SCSI disc drive with a star network to support NewBrains to run office
accounts, database and word processing through the serial port to load
software and to save data.
Support for third party development was not high. Often it wasn't needed,
but there was no backing and it was the general attitude that developers
should do this because the basic system was good. However, lack of software
meant lack of sales and lack of sales meant lack of new software for the
During the summer of 83 the manufacturers of the basic system modules where
kitted out with components, ready to solider onto the circuit boards.
Purchasing the kit made a large hole into the cash flow. The summer was a
lean time and many consumers gave up on the wait for the promised modules.
The advanced order book and projected sales for Christmas looked very good.
Even with this in mind Grundy, Thorn EMI (the main manufacturer of the
circuit board assembly) and Barclays (the bankers - I believe) decided to
pull the plug. Tangerine, Oric, Lynx, Acorn and Sinclair all got further
funding until Christmas (and later) and although three of these Cambridge
companies quickly followed Grundy there designs where less expanable and
flexible and fell for much the same reasons of a contracted home market, the
successs of the BBC Model B, the cheapness of the Spectrum and the business
world moving to IBM clones.
Eventually Tradecom purchased GBS because they had a contract to delivery
microcomputers to schools and training centres in Holland. They created a
switching system for loading software and to run CP/M on RAM upgraded
machines (more memory, but no extra hardware). Further development was not
part of Tradecom's brief, but they did turn the handle on the current (1983)
component stock and make available the disc controller and expansion module
in 83 and 84.
I and a former GBS contractor, John Kelly, supported some further
developments in hardware expansions, software and a NewBrain user group
(NBUG) with over 1200 members at it's peak. There were strong Dutch and
Danish user groups and Open #Stream in the UK. Material was still sent out
to a few members as late as 1987.
Unique points of the NewBrain design:-
2 cassette ports, 2 serial ports, TV and composite video output
Modular system that was added to externally, no opening up cases
Expandablity was not limited to 5 or 6 slots inside a box or a couple of
Compact design, even with the expansion modules it remained smaller than the
Stream I/O that allowed software to write out to monitor, built in vaccuum
florescent (sp?) display (VF), cassette, disc, serial, parallel ports

Footnote (off topic - more general):-
When the UK lost all of it's manufacturing of home and business computers,
leaving the Acorn RISC systems as the last 'desktop' and Psion, who moved to
useable PDA designs, in the handheld market place we now have no UK
owned/based industry left. Maybe you have an insight into why they all
failed following the boon years of 81/82.
Conversely why has the PalmPilot appeared and continues to win adding to the
downfall of the alternative Psion design but killing off CE and the
Franklin/Xircom Rex? The new Microsoft software and hardware design may make
a come back to the original Pocket PC and Intel's prototype PDA and those of
Psion may be manufactured, or something no so far disclosed appears and
wipes the market.
There may be something wrong still with the business and marketing in the
There might be a general consumer purchasing altitude issue, as VHS
continues and V2000 and Beta and even 8mm (and Hi8) and DVC all became
redundant. [Sony's new micro-DVC cassette is 80% of the size of DVC and uses
MPEG2 compression, so DVC that killed Hi8 is now killed in a few years by
yet another format]. Why APS failed to beat 35mm off the shelves (although
most of the advances in the APS design have been put back into 35mm designs
and also the film type). etc.
I wonder if the consumer will hit back and not get the latest MHz, Mb, Gb
system even if it is less than a quarter the cost of what they have and more
than double the capacity and performance. I have seen no killer app for many
years and only want what I have with the bugs removed and the interface to
be more consistant.
12 years old but has the checked facts (# for £ in places)

Other possible outlets interested people / groups for the NewBrain:-
(now at 2 Butcher close Milton Cambridge CB4 6ED)
The NewBrain Users Group, 190 members, on every continent.
Newsletters (back issues available), PD software, commercial
programs at cost, NewBrain hardware, other hardware at cost.
Started in November 1983. Software and hardware design and
developement. Subscription #9 for current newsletters from
number 19. Number 24 out.
Geoff Riding
20 Westwood Road
Heald Green
Cheshire SK8 3JW tel 061 437 3696 (?)
NewBrain user group started by Philip Crooks about 4 years
ago, now about 80 members.
NBUG and OPEN #STREAM will tie up this year into one user group: NewBrain
Owners Group 'NOG' or 'NBOG' for short!
Tradecom International B.V.
Postbus 60
2396 ZH Koudekerk A/D Rijn
Netherlands tel: 01714 4300 Jack van der Schrier
Current holders of NewBrains, last manufacturer, selling off
John Kelly
1 Bridge Street
Whaddon, Nr Royston Herts SG5 8SG tel 0223 207237 (?)
Design engineer, NewBrain repairs, printer switch boxes,
to Centronics box, EPROM blowers.
AJ Electronics
Sproughton House
Suffolk IP8 3AW tel: (0473) 461600 (?)
NewBrain repairs. Has large stock of PCBs, circuits and
NewBrain equipment in different stages of repair.
Tyepro Limited
30 Campkin Way, Cambridge CB4 2NG tel 0223 322394 (?)
Design engineer. NewBrain graphics dumps, printer switch
serial to Centronics boxes.
Angela Enterprises
4 Ninnings Lane
Rabley Heath
Welwyn Herts AL6 9TD tel (0438) 812439 (?)
NewBrain software, books, some hardware. Retailer.
493 London Road
Camberley Surrey tel 0276 684774
8K Battery backed RAM card with software for the NewBrain.
CData Computes Ltd
9 Sea Road
East Sussex TN40 1EE tel 0424 212591 (?)
Repairs of NewBrains. Agents for Tradecom. 128K upgrade to
standard Model A/AD.
What can a NewBrain be used for?
Most NewBrain owners have shelved their machines. The s/h price is about
#35-40. For the AD model (with 16 character display) you get a very accurate
programmable calculator that can be dedicated to do other tasks (number
cruching, control, printer buffer, etc) with programs loaded from tape
or in EPROM.
With a disk controller and expansion interface module you have the use of
of RAM, 256x640 graphics, up to four drives of 800K each, three comms RS232
ports, an RS232 printer port, Centroincs printer port (yes up to 5 printers
be connected at once), two cassette ports, monitor and UHF (so two screens)
outputs, and built in 16 character VF display. The NewBrain is faster than
Amstrad PCW, although lacks the GSX graphics interface to CP/M. The
program area under CP/M is 58.5Kb (PCW is 61K) but you can have a printer
buffer of 256 bytes to 16Kb at no extra cost. A s/h disk controller (with
PSU) is about #80 (or you can build one for about #65, pcbs and cases from
NBUG). A s/h expansion interface module (with PSU) is also about #80. New
drives are only #130, so s/h 5.25" 800K and 200K drives should be alot
Monitors new at #90, s/h from #30. I know that an Atari ST520 with drive is
only #243, but the NewBrain is portable, can have a battery module fitted
is a good Z80 development system.
NBUG, The NewBrain Users Group has over 105 volumes of Public Domain
Each volume is up to 188K of software (the capacity of a 200K formatted
diskette). All the source code for the NewBrain software is available. Many
programs can be supplied on cassette for users without disks. Some are
Public Domain programs tried, tested and modified for the NewBrain's CP/M

A first draft (2160 words) by Gerald McMullon editor of NBUG, The NewBrain
User's Group. 36 Armitage Way Cambridge CB4 2UE
A Short History of the NewBrain
The NewBrain started its evolution with Sinclair Radionics; R&D has always
been in Cambridge. Sir Clive was aiming at a sub-#100 machine where-as the
NewBrain was designed for sub-#200. With the collapse of Sinclair Radionics
the project (and designers) were taken over by Newbury Laboratories. Here
keyboard layout and casing were completed along with the basic hardware and
software philosophy.
Many of the decisions wre taken by default. For example, the colour of the
NewBrain. The brown and cream is a good choice, and matched the Newbury
colours. This was choosen however, because the manufacturers of the casing
cream and brown plastic available when the prototypes were made. Other
(the intended red and grey) would have cost more money. When the time came
order the production of the cases no one changed the colour.
The NewBrain was announced in 1980, although only a few prototypes had been
built. Besides a mains only machine a battery operating version with 16
character VF display was built with internal RAM of 18K, externally
up to 4Mb.
About this time the BBC was looking for a computer for a new television
The NewBrain was a strong contender, but the Managing Director pulled out
leaving Acorn to get the contract.
Grundy Takes on the NewBrain
When the MD left to join The Grundy Group Newbury's new director wanted to
streamline and eventually the NewBrain followed to Grundy who set up Grundy
Bussiness Systems Ltd.
Marketing was always considered to be a weak point of the NewBrain.
complained the R&D didn't produce what the market wanted. The sales
and later with the expansion of the company, the Technical support staff had
training about the NewBrain. No reserch was conducted as to where the
could best fit into the consumer market. When it had to be explained to the
sales people that graphics on a VBrain is a joke (the VBrain has no VDU /
screen hardware), that Centronics printer port on the Expansion Interface
module does not require user definable baud rate, that it was not possible
produce a proportional spacing device driver for all printers, etc, you can
begin to understand why lack of training was a real problem.
GBS was formed in November 1981 and marketing announced various NewBrain
based upon an altered specification. The screen editor, maths pack and
were outlined and remains unaltered, except for the removal of bugs.
Many machines were 'loaned' out to the press and to software authors.
Unfortunately many of these machines were pre-graphics and pre-80 column
(the latter was developed at the last minute in 1982). It was difficult to
persuade professional software houses to write for the un-expanded NewBrain.
They wanted a complete disk based system in their hands first, and even then
they had their doubts.
From the summer of 1982 marketing announced a January delivery date for the
CP/M system. When told that the best possible date was April they didn't
to know. Infact the disk controller was ready before April, but could not be
released until it was working with the Expansion Interface module. In the
only minor changes were required because of this. By the shut down of GBS
expansion interface module was still not finished although about 300 had
sold. Certainly the fault of R&D.
Christmas 1982 saw a rush of NewBrains being ordered. Demand could not be
Distributors ordered 10 then asked for 500 per month. So after Christmas the
decision to increase production to 10000 per month was forced through. Some
correctly forecasted a dead period after Christmas but were over-ruled.
Software Technical Manuals were produced and 20000 Beginner's Guides. The
Guides were to be given away with the machine, but marketing said the could
sell them, but very few were sold. R&D wanted the Technical manual to be
for under #20, marketing put a #50 tag on it.
Marketing were keen to sell a machine like the Grundy 8200 (a Superbrain
clone), somewhat like an Amstrad PCW. A 64K twin 800K drive CP/M system with
choice of printer ports for sub-#1000. Marketing considered that the modular
design and flexiblity of the NewBrain was too difficult to explain to the
dealers let alone the general public. They also considered the NewBrain too
expensive, yet put the price up in February 1983. Dealers were not informed
the available software and only a few visits were made by sales staff. Great
efforts were made to produce good technical notes, which were free issued to
owners. The dealers were not informed of their availablity.
Peripherals could have been ready. The original NewBrain had problems with
loading from some cassette recorders. A modification was made, althought not
put into production (marketing didn't supply details of the complaints to
Production to force this modification onto new machines). Quality tape
recorders were on 26 weeks lead time from Japan with a demand of 120 per
Orders were not placed. Monitors were ordered in small batches, costing up
#40 per unit more than if a quantity order were placed with so many per
taken. Some complaints about the NewBrain's screen relate to poor monitors,
the use of colour televisions.
A range of printers were considered, but decisions were not taken. GBS could
have had the small Tandy plotter before Oric and Micro-Perpherials got it. A
computer controlled cassette recorder was under development, as well as a
Software or Soft-where?
Quality programs were simply not available. The first programs were
in BASIC. They showed poor programming practises and even worse NewBrain
implementation (would not run on all NewBrains). The first good program was
Chris Whittington's Chess, followed by Draughts and Backgammon. Grundy also
commissioned an extensive assembler, a marco-driven word processor and Comal
(as the next language).
With CP/M discussions were under way to get the NewBrain as an installation
option with many well known programs (dBaseII, Spellbinder, Plannarcalc,
Masterplanner, Superfile, TCL Pascal, Pegasus software, Peachtree products
others). Peachtree software was very important as these ran from CP/M Net
MP/M and the NewBrain has the software to be then centre of a star network
32 other NewBrains.
Other Modules
The Network module was completed, and ready for production after the
Interface and Disk controller modules. MP/M was not finished before the
collapse of GBS in August 1983.
A cut down version of the Network module was the Perpherial interface
This provids two hardware serial ports, Centronics and parallel user port.
Intended for use as a terminal and for laboratory control applications.
The Memory Expansion Module worked in the first design form. Each module
two boards. These could be RAM or ROM (up to 128K each), real-time clock,
more perpherial interfaces, battey backed CMOS RAM etc. Up to 4Mb of extra
RAM/ROM could be fitted in this way to all NewBrains that have
Expansion Interface modules.
The modular nature of the NewBrain means lots of boxes, each with their own
power input. A case was being designed to house four modules, giving support
the NewBrain (keyboard module) and a monitor. The power supply being a
mode with noise filtering. This would allow users to select whether to have
more memory cards or networking capablities.
Colour circuits and better analogue to digital convertors were being
The Model C (for colour) would allow a choice of 4096 colours on a 250x640
screen and have the Expansion interface module built into the current
size case. The next model would be a Processor module with external keyboard
and better built-in screen intended for use with the desk-top system.
An 8088 module was under consideration to provide MSDOS.
The fall of GBS
The collapse began in November 1982 when customers were promised expansions
that R&D could not supply. Three of the directors were pulling against each
other and this lead to over production in the spring of 1983. Staff were cut
back, the MD left, then the R&D directors were sacked and the Chief
failed to obtain backing from the banks. The Grundy Group refused to support
GBS, even though the order book exceeded both stock value and debts. Had the
company been allowed to trade until January 1984 then I am sure that the
NewBrain would have had the Amstrad PCW success.
The collapse was followed by the sale of many incomplete and prototype
NewBrains and modules.
The NewBrain moves to Holland
Tradecom, after a long pause obtained rights to the NewBrain stock. Tradecom
had a contract with the Dutch Government to supply computers, monitors,
printers, software to several centres and schools. The NewBrain was the
part of this project.
Tradecom never developed the NewBrain. Several projects that they looked at
were produced elsewhere. The gas-plasma display was interfaced to the
by a Norwegian company, the mechanical network switch box of Tradecom was
replaced by an electronic switch box developed in Denmark. The internal RAM
card was upgraded to 128K, which Tradecom bought the rights to from the
This provided Tradecom with a mobile classroom. A central, standard,
with disk controller twin drives connected to 16 128K NewBrains downloaded
CP/M to the satilites. So each user ran a 64K CP/M system via a fast
controlled serial port as back-up (ie in place of disk drives).
Later each NewBrain was fitted with the Velotype keyboard (as seen on
Tomorrow's World) and used to train young people word processing skills.
Tradecom did make a box for the disk controller and twin drives at
double the UK price. If the expansion interface was fitted it had to be
There were plans for production in India, but it seems that in the end that
the BBC Model B was choosen.
Getting information out of Tradecom since 1985 has been fruitless, and I
understand that they are selling off remaining stocks (not all working).
Other Developments
Kuma continued supporting the NewBrain until about 1985 then stopped
even their own products for it.
Some of the ex-GBS staff continued with the development of software and
hardware for the NewBrain.
A side ways paged ROM/RAM card called Micropage, was developed, giving up to
64K ROM or 48K ROM and 8K RAM with new software fitted. This allows the
NewBrain to be set up for dedicated tasks. Programs such as HiSoft Pascal,
HiSoft Devpac Z80, Pro-Ed word processing, Chess and Typing Tutor have been
popular choices for this card, as well as owners' own programs.
MPE-FORTH in EPROM was the first proper alternative language to BASIC.
This has been followed by a Public Domain FORTH that is the first 96K
for the NewBrain (loaded from disk). Although interfaces to several
are now available (Pro Pascal & Pro Fortran, Microsoft Fortran, HiSoft and
Turbo Pascal FTL-Modula-2 and BCPL) for the 96K disk system.
A joystick to fit into the COMMS port of the standard NewBrain was the next
development, with controlling software and some programs.
A Z80 PIO board giving BBC type user port, parallel port and Centronics
port gave users a cheap option for parallel printers, a faster cheaper
and connection to BBC type perpherials.
Because of the shortage of Expansion modules a Perpherial box (PI BOX) was
developed. This contains hardware RS232 comms port (for a modem), RS232
port, Centronics printer, 8K battery backed CMOS RAM, BBC type user port, an
extra parallel port (as the PIO Board) and software in EPROM. In true
fashion this can be connected to a standard NewBrain or a NewBrain with disc
controller. This box is particularly useful to turn the NewBrain into an
terminal with NewBrain extensions. You can run a NewBrain program and switch
between it and terminal mode, or transmit the output from a program to the
host. The PI BOX is #89.
An EPROM blower for the BBC type user port provided a cheap developement
for NewBrain Micropage and other ROM cards.
An NBUG member developed a sound board, that connects to the parallel port
the PI BOX or PIO board.
Remaining stocks are available from NBUG, whose members continue to develop
software and sometimes hardware.
Infotek Design Ltd have just completed a Eurocard (165x100mm) single board
computer with an HD64180 6MHz CPU, 384K RAM, 128K EPROM, controls up to four
800K drives and runs the CP/M super-set Z-system software. They use a
with PI BOX as well as an Amstrad 6128, PCW8256 and an IBM PC as terminals.
NBUG is developing software for the PI BOX that allows this board to be the
fast intelligent back-up for the NewBrain. Starting prices from #199+VAT.

All images and text © Adrian Graham 1999-2024 unless otherwise noted using words. Also on