Scor­pio News


July–September 1987, Volume 1, Issue 3

Page 42 of 67

Upgrading aGM813to 256K by W.H. Turner Introduction

This article describes how to add more memory to « Gemini GHB13 CPU board and what to do with it.

There was not a single good Gemini Galaxy 2 computert” possibility of using it for Eache (as in CP/M Plus), and the pos processor), that made me do it.

Whstever the season, the additional nenocy boarde looked vary expengive end used valuable bus slots which made the whole exercise 4 lot Lese attracts golucton was to seplace the existing 4k bit MAM with the pis competibie 236k Bie’RaMe which quadeupled the aemory for an outlay of only £35.

ries of com} up to 512k bytes of menory via 19 address Lines but of fo address the Suk bytes of onboard memory. Access to the additi Ifnes ie controlled by two 7413189 TIL RAMs used as sixteen 8 bit mapping registers between the top four Z80 address Lines and the top seven bus addres Bach mapping register controls a 4k byte block off the 64k byte Z80 space. Each address generated by the £80 selects a 4k byte block of Remory via one of these registers and the 6-bit value written inte the mapping Fegiscer determines which 4k block out of the Sizk bytes is actually accessed. ult dg that any 4k byte block of memory can be mapped at any 4k byce “yin the ouk byte ‘space of the 2a whould mention jefined and

eed in extending my memory when everyor was But the cost of buying an additional memory board Ieecnatively, che complenity of producing my vn Dot’ it that things very far) Te was only after an engine friend suggeated replacing the existing 6ek bit RAMs on the processor with 256K Dit RAMS that T began to look at the problem more seriously.

For a long time I proceastinated, believing that if T attempted even minor surgery on the processor board I could end up with a very dead computer. ually 1 took « firm grip on myself and removed the GHE13 board from the computer to look at the possibilities, besides at that point T had already Bought the 256k bit chips! Looking at ‘the board and the circuit diagram \tfully supplied by Gemini T worked outs plan of action.

Not eing very dmaginative when tt cones to bardvare I felt that I shuld ae far Able, duplicate the circuite iy in use on the processor boar: idefonal by mounting the additional chips

‘could be easily connected to the

fenory. Furthe

which the board would be ei{d back into ite slot in the card cage, the cables Feconnected and the syste booted back up to check that Lt still worked, If Je Suddenly stopped working then I would only have to check the changes performed in the Last stage, looking for pine not inverted in sockets, dry jointe, solder bridges etc. Further T would not cut any tracks ao everything could hopefully be returned to its original state if necessary.

This is an OCR’d version of the scanned page and likely contains recognition errors.

Page 42 of 67