Posted By
TLC
on 2009-06-19
11:43:01
|  Re: faulty plus4
Okay.
When you turn the computer on for the first time, you'd normally see nothing (ie. black screen, or no valid videosignal at all) if there wasn't the CPU, the Kernal ROM and some ram (this latter one being important because of the organization of the Kernal; it uses subroutine calls that would never return without some stack ie. ram). You can perfectly simulate that situation by turning the computer on (I'm talking about turning it on after a long period of power-off) whilst holding reset. Upon power-up, the TED starts up from some default state -- PAL mode, blank screen, border is black -- but the rest of the init is done by the reset routine (which is in the Kernal ROM) and finished by the init of Basic 3.5 (which is in the Basic ROM).
Consequently, if (upon power-up) you can see a non-blank, non-black screen, that is perfect proof for the CPU, the Kernal ROM, and some of the RAM being at least partially correct. Moreover... if you can get some prompt by pressing run/stop + reset, that means all of the above do work correctly "enough" (TEDMON needs no Basic ROM but only the Kernal ROM + some RAM to be present... the computer is supposed to boot up with no Basic ROM at all, although in TEDMON only, which is still a nice feature). ...As we could see, this was not the case here.
The screenshot shows that wherever the TED fetches the screen data from, that area of the RAM has not been initialized by the Kernal. The screen code of @ is 0, whilst the screen code of the other typical character (code #191 here: http://www.df.lth.se/~triad/krad/recode/pet3.jpg) is $ff. The color code of black is 0, the color code of flashing bright green is $ff. If you'd check how the memory content of a Plus/4 looks like after power-up, you'd see similar patterns (mostly 00s and FFs in a row after each other + some other typical values inbetween). That may mean two things: either the Kernal ROM couldn't clear the screen (bad Kernal ROM, or bad RAM, or something around them that prevent them to work correctly), or it could indeed do that, but the TED fetches data from the wrong place (= bad Kernal ROM (wrong default values written upon TED initialization) or bad TED).
From that point and on, I would do two things: first, try booting up a ROM cartridge (game cartridge or such). If it boots, albeit with strange symptoms, that suggests a slightly faulty TED, not the Kernal. From the other hand, if it works correctly, or if it doesn't boot up at all, that might suggest a broken Kernal.
...Ultimately, I'd start testing the chips (the Kernal, the TED, and maybe the FPLA, in that order) by trying if they work in a known-to-work Plus/4... I know that's bad news... At least, you seem to have a working CPU (which is important). For the test, both a C-16, C-116 or a Plus/4 are/would be suitable.
|
