04 October 2019, 11:12 | #1 |
Registered User
Join Date: Jan 2019
Location: Sundsvall
Posts: 37
|
Amiga 4000 - strange chipram behaviour
Hi,
I have this A4000 with an odd problem.. A known working (Actually two different) chipmem module, standard jumper settings. No chipmem in diagrom. Sometimes 1 MB is found .. If J214 is changed to pin 1-2 (VBB/MA10). 1 MB of chipram is found every time. If J214 is changed to pin 1-2 (VBB/MA10). And a 4 MB module is used, 2 MB chipram is found. What can be wrong...? :/ Also noticed this morning that fastram has siezed to exist ... CAS/RAS/CRAS/CCAS/BWsignals are checked ok for continuety... |
04 October 2019, 11:20 | #2 |
ex. demoscener "Bigmama"
Join Date: Jun 2012
Location: Fyn / Denmark
Posts: 1,635
|
Your jumper setting is the same in both scenarios, is that correct?
Do you have corrosion damage in that area from leaking battery/capacitors? |
04 October 2019, 12:31 | #3 |
PSPUAE DEV
|
That sounds normal to me, A4000 is picky on modules used for Chip RAM.
Switching jump always forces it to 1MB. What happens if you remove all RAM, you should get Green screen. No green scren means board damage. For Fast RAM, hooverphonique is correct. Leakage over RAM latch chip with cause that. |
04 October 2019, 12:45 | #4 | |
Registered User
Join Date: Jan 2019
Location: Sundsvall
Posts: 37
|
Quote:
Well.. yes and no.. If I use standard, factory settings. Sometimes when i boot the machine (with diagrom) it gives me flashing red screen and no chipram. Next day... it MIGHT give the charateristic green flashing and 1 MB found chipram. IF i change the J214 (VBB/MA10) jumper to none standard setting (jumper between 1-2 instead of VBB to Alice 2-3) I get 1 MB chipram every time. IF i change the J214 (VBB/MA10) jumper to none standard setting (jumper between 1-2 instead of VBB to Alice 2-3) AND change the SIMM to a 4 MB single sided SIMM i get 2 MB och chipram every time. So in both the later scenarios the jumper settings is the same. Yes. |
|
04 October 2019, 12:48 | #5 | |
Registered User
Join Date: Jan 2019
Location: Sundsvall
Posts: 37
|
Quote:
I'm mainly using DiagROM when fault tracing... If I remove the chip ram SIMM i do get the same result as I most of the times get when I start it with standard jumper settings. The SIMMs are known working in my other A4000. All traces to both chip and fastram sockets are tested ok. .. Board is cleaned 2, 3 and 4 times... :P |
|
04 October 2019, 15:32 | #6 |
Registered User
Join Date: Jan 2019
Location: Sundsvall
Posts: 37
|
Actually, also fastmem is behaving odd...
Memory in first slot, 4 MB Memory in first and second slot, sometimes 0 MB, sometimes 8 MB. Memory in first, second and third slot, 4 MB. Memory in all slots, 0 MB. Memory in first and third or forth, sometimes 4 MB sometimes 8 MB........... Haha.. wtf.. |
04 October 2019, 16:30 | #7 |
ex. demoscener "Bigmama"
Join Date: Jun 2012
Location: Fyn / Denmark
Posts: 1,635
|
Ok.. first, don't mess with J214, it switches between a 2 and an 8 MB Alice, and the latter never existed.
Also, for 2mb chipram you need a double-rank simm, i.e. 2, 8, 32, etc, MB. Use a 2 MB. If you use a single-rank (1MB, 4MB, 16MB, etc), it'll only detect 1 MB of chipram. Regarding fastram, what size module(s) are you swapping around? If double-rank, only every other slot may be populated. Last edited by hooverphonique; 04 October 2019 at 16:37. Reason: add comment about single-rank |
04 October 2019, 17:33 | #8 | |
Registered User
Join Date: Jan 2019
Location: Sundsvall
Posts: 37
|
Quote:
My other two 2 MB simms are double and works/worked as a charm. Tested in my other A4000. The fastram aswell is known working, four single sided 4 mb modules. Everything was working A-OK before my 3660 accelerator went bust..... |
|
04 October 2019, 17:34 | #9 |
Registered User
Join Date: Jan 2019
Location: Sundsvall
Posts: 37
|
So there IS something wrong with some component on the motherboard or maybe a trace.
Same behaviour with 3 different working cpu cards. |
04 October 2019, 23:14 | #10 | |
Registered User
Join Date: Dec 2015
Location: USA
Posts: 2,946
|
Quote:
What he ^ says. Don't screw with it. With a working 8MB SIMM you don't have to do anything but plug it in. Ignore that post you're talking about, where the guy goes, "I'm out of 2MB SIMMs, I'll just put an 8MB one in and set this jumper", and yes you can set the jumpers incorrectly and it still *might* work, but just set them as the manual states. |
|
05 October 2019, 07:02 | #11 | |
Registered User
Join Date: Jan 2019
Location: Sundsvall
Posts: 37
|
Quote:
But as it is now, it doesnt work. Thats why I would like some help of you experienced guys. And to get a little guide in WHY Alice finds 2 MB of ram if I flip the jumper and just 1 MB in original state... The jumper just connects Alice DRA10, with U213 I9 and U212 I_CK. I know that the jumper setting is wrong, no need to point that out. I wouldnt just change it for fun. But Im very interested in why it makes it work 'a little better'.... Maybe I'll have to be more clear about that it isnt working, and that im trying to get it to work. Its not a weekend joy to drink beer and change jumper settings. None the less, other hints are welcome. Last edited by Cav; 05 October 2019 at 07:08. |
|
05 October 2019, 09:23 | #12 | |
Registered User
Join Date: Sep 2008
Location: Somewhere
Posts: 223
|
Quote:
|
|
05 October 2019, 10:35 | #13 |
Registered User
Join Date: Jan 2019
Location: Sundsvall
Posts: 37
|
|
05 October 2019, 19:47 | #14 |
Registered User
Join Date: Dec 2015
Location: USA
Posts: 2,946
|
|
05 October 2019, 20:04 | #15 |
Registered User
Join Date: Jan 2019
Location: Sundsvall
Posts: 37
|
|
06 October 2019, 07:31 | #16 |
Registered User
Join Date: Jan 2019
Location: Sundsvall
Posts: 37
|
From Wikipedia..
†/RAS1 and /RAS3 are only used on two-rank SIMMS: 2, 8, 32, and 128 MB. https://en.wikipedia.org/wiki/SIMM Does it makes sense that I have an issue with RAS1 & 3 as it works ok a larger single sided simm? |
13 October 2019, 08:55 | #17 |
Registered User
Join Date: Jan 2019
Location: Sundsvall
Posts: 37
|
First issue solved.
Change of U891 and reflow of U212. 2 MB chipram and 16 MB fastram consequent. Now its only the graphic issue left.... :/ |
14 October 2019, 03:46 | #18 | |
Registered User
Join Date: Jun 2019
Location: BNE AU
Posts: 216
|
Quote:
Sounds like some of the connections in the SIMM slot isn't making proper contact. IIRC the 4000 needs to see the previous slot before looking at the next one. So if it doesn't see memory in the 2nd slot, it won't look at the 3rd one. I had a broken part in my 2nd slot which was preventing the memory from being seated properly, so I had to put some non conductive inserts between the memory to get it to recognise properly. I've got to replace the slot itself one day! |
|
Currently Active Users Viewing This Thread: 1 (0 members and 1 guests) | |
Thread Tools | |
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
diagROM expected behaviour with bad chipram | project23 | support.Hardware | 2 | 20 May 2019 19:06 |
Strange Prefs (non) behaviour. | khph_re | support.WinUAE | 4 | 02 October 2010 14:34 |
Strange guru behaviour! | deejaya | support.Other | 11 | 31 January 2009 21:01 |
Strange behaviour in p. illusions | _ThEcRoW | support.Games | 4 | 18 June 2007 18:17 |
Strange A1200 behaviour | manicx | support.Hardware | 39 | 09 November 2005 08:32 |
|
|