[Coco] Bad memory reported by Disto Coco III RAM Test by Tony Distefano

[Michael R. Furman]

Thanks Kip, Richard, and Gene.

I figured out what the problem was.  

I was using a brand new CocoSDC and it has a a jumper to connect the CART line to Q so you can use it to autostart ROM images in it’s onboard flash.  Well when I opened up the CoCoSDC package the jumper fell out on the floor and I absently found a place to connect it to without thinking and it wound up in the AUTO position.  The CoCoSDC manual clearly says not to do this to use SDC DOS. Oops.



—

Michael R. Furman
Email: n6il at ocs.net
Phone: +1 (408) 480-5865


> On Nov 16, 2016, at 9:37 PM, Richard Goedeken <Richard at fascinationsoftware.com> wrote:
> 
> I would suggest running my RAM test program:
> 
> http://www.fascinationsoftware.com/media/Coco3_RAM_Test_v1.2.zip
> 
> It will run continuously and print multiple errors, and give you the values
> written and read back, so you may see a pattern.
> 
> Richard
> 
> 
> On 11/16/2016 09:25 AM, Kip Koon wrote:
>> Hi Michael,
>> Whenever a 512KB memory upgrade is installed in the Coco 3, the original 4 memory chips are removed and C65 & C66 are removed by cutting one leg and lifting up on the cap slightly.  That's the procedure I used when I added my 512KB memory board to my Coco 3. That upgraded Coco 3 is still running strong.  
>> As to the 8K address range that is bad, if an address line is truly bad, that 8K block would be bad everywhere, not just in one place I would think.  Maybe there is a logic problem somewhere maybe?
>> Can anyone shed more light on this situation?
>> 
>> Kip Koon
>> computerdoc at sc.rr.com
>> http://www.cocopedia.com/wiki/index.php/Kip_Koon
>> 
>> 
>>> -----Original Message-----
>>> From: Coco [mailto:coco-bounces at maltedmedia.com] On Behalf Of Michael R. Furman
>>> Sent: Wednesday, November 16, 2016 11:37 AM
>>> To: coco at maltedmedia.com
>>> Subject: [Coco] Bad memory reported by Disto Coco III RAM Test by Tony Distefano
>>> 
>>> Does anyone have a manual for the Disto Coco III RAM Test by Tony Distefano?
>>> 
>>> http://www.colorcomputerarchive.com/coco/Disks/Utilities/Disto%20Coco%203%20RAM%20Test%20%28Tony%20Distefano%29%2
>>> 0%28Coco%203%29.zip
>>> 
>>> I ran it on one of my Coco 3’s and it is returning an error, says that block 60 is bad, which I interpret to be 0x78000 to 0x79FFF
>>> 
>>> A second observation is that when loading certain programs in to BASIC from the CocoSDC those programs get corrupted at the same
>>> place every time.
>>> 
>>> I was looking at the schematics to try to figure this out and I am a bit confused.  What I see on the schematic is each of the 256k x 1
>>> chips on the 512 ram expansion board is attached to one data line and there are two sets of chips selected by WE0 or WE1.  The part I
>>> am confused about is what it means that a particular range of memory is bad.  If one data bit is bad across the entire 256k range I
>>> understand that one, that’s a bad chip. But it only seems to be bad in a particular 8k range of addresses, so maybe this is some issue
>>> with RAS or CAS or a flakey address line to only the WE1 set of chips?
>>> 
>>> What would happen if C65 or C66 still in place?  Why do they have those RC filters on RAS and CAS anyway?
>>>