exxos's Atari, Amiga & retro forum

@alexh @Maximilian

Thanks for the info both. Actually you're right it should work with KS1.3/WB1.3, I just thought I'd give something higher a try, BUT I'll see what I can do first with the basics :)

Part of my reasoning was that the document that explains how to partition shows it being done under WB3.1 though... Will do some more digging.

Interestingly the replacement Fat Agnus arrived today and it's no change (Remember the Agnus socket is also new), the trapdoor RAM is still invisible, though the clock works fine.

I will upgrade to 1MB on board this evening and report back.

Since I have changed or swapped out what I think is everything relating to the trapdoor RAM, there is either a hidden issue in the RAM board or the connectors, I just can't say.

I did identify a possible issue with a RAM IC on the board using my thermal cam and I may quickly swap it out just to see if it helps, but I don't think it'll change anything.

I see what you mean about the Gary signal too now.

If you have a Kickstart 3.1 (or later) ROM for 16-bit Amiga's (e.g. v40.63+) then I would recommend you used that and it's corresponding WB (AmigaOS).

Lots of things need OS2+ and life under KS1.3/WB1.3 is only fun for about 5 seconds

alexh wrote: 13 Feb 2025 16:02 If you have a Kickstart 3.1 (or later) ROM for 16-bit Amiga's (e.g. v40.63+) then I would recommend you used that and it's corresponding WB (AmigaOS).

Lots of things need OS2+ and life under KS1.3/WB1.3 is only fun for about 5 seconds

I have a couple of BooBip 27C400 emulators that I can put 4 ROM images on, but I can't find any that work for 3.1, which is quite annoying. If I can sort that out I'll be good.

Incidentally the 41256 that I thought was bad actually is, it fails instantly in my RAM tester, so I'll replace it and see if that helps.

OK, so it's fixed...

Excuse the upsidedowness of the image, but IC3 was the culprit:


This was the IC I picked out with my thermal camera. It was subtly different to the others (Ignore IC6 that was one that was previously replaced, just not by me), just a bit cooler and very hard to spot.

This was the test result:


So I put a socket and known good 41256 IC in and boom:

All memory is present and correct in ADK and DiagROM.

I just need to figure out a working KS 3.1 ROM to put in one of my 27C400 emulators and we're off to the races.

But there is a lesson to be learned. If somebody says it works, it may not.

Glad you sorted it in the end @rubber_jonnie (and very relieved it wasn't a fault with the Amiga from me) :D

:bravo:

Yay! It's always the memory. ;-)

JezC wrote: 13 Feb 2025 17:57 Glad you sorted it in the end @rubber_jonnie (and very relieved it wasn't a fault with the Amiga from me) :D

:bravo:

I feel like I put you through the mill unnecessarily thinking the Miggy was bad so sorry about that.

At least it's sorted in the end.

stephen_usher wrote: 13 Feb 2025 19:58 Yay! It's always the memory. ;-)

Yeah, bloody annoying as it took a really close look at the RAM with my thermal camera to suss it out and I'm more kind of surprised that one chip out of 16 prevented all the RAM from being seen on the card.

Well, I guess that the ROM memory test routines just write a value to certain boundary addresses and then read it back and only mark the memory as there is the read value matched the written one.

After some fun and games I finally got the BooBips working with KS 2.04 and can boot to WB2.

Not sure why but whatever ROM I programmed in it wouldn't work, even byte swapped, but now it's OK. Need to try a higher version next.