UK Vintage Radio Repair and Restoration Discussion Forum - View Single Post

TonyDuell · 2nd Feb 2019, 7:23 pm

I seem to remember that the TI99/4A had very little RAM on the processor bus (256 bytes or something). There was 16K of RAM on the video chip (a TMS9918 in the States, possibly a TMS9927 over here) and BASIC programs were stored in that and then interpretted by the processor, being read a byte at a time via the video chip

This mean no user machine code on the standard machine (there was nowhere to store it, you couldn't run machine code programs via the video chip) and it was rather slow.

I think one of the expansion cartridges added conventional RAM which made the machine a lot more useable.

Incidentally the same video chip was used in the Tatung Einstein computer but that machine had 64K RAM on the processor bus (as well as the 16K on the video side) so it was a lot more useable. The video RAM on that machine also stored the definitions of user-defined keys I think.

2nd Feb 2019, 7:23 pm	#5
TonyDuell Dekatron Join Date: Jun 2015 Location: Biggin Hill, London, UK. Posts: 5,208	Re: Texas Ti99/4a out of hibernation. I seem to remember that the TI99/4A had very little RAM on the processor bus (256 bytes or something). There was 16K of RAM on the video chip (a TMS9918 in the States, possibly a TMS9927 over here) and BASIC programs were stored in that and then interpretted by the processor, being read a byte at a time via the video chip This mean no user machine code on the standard machine (there was nowhere to store it, you couldn't run machine code programs via the video chip) and it was rather slow. I think one of the expansion cartridges added conventional RAM which made the machine a lot more useable. Incidentally the same video chip was used in the Tatung Einstein computer but that machine had 64K RAM on the processor bus (as well as the 16K on the video side) so it was a lot more useable. The video RAM on that machine also stored the definitions of user-defined keys I think.