Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!sdd.hp.com!spool.mu.edu!uwm.edu!psuvax1!ukma!seismo!dimacs.rutgers.edu!rutgers!cbmvax!daveh
From: daveh@cbmvax.commodore.com (Dave Haynie)
Newsgroups: comp.sys.amiga.hardware
Subject: Re: Possible A3000 Scram Problem
Keywords: scram 3000 problem
Message-ID: <19993@cbmvax.commodore.com>
Date: 20 Mar 91 14:32:51 GMT
Article-I.D.: cbmvax.19993
References: <1991Mar19.000852.1808@medsys.uucp>
Reply-To: daveh@cbmvax.commodore.com (Dave Haynie)
Organization: Commodore, West Chester, PA
Lines: 57

In article <1991Mar19.000852.1808@medsys.uucp> wendell@medsys.uucp (Wendell Dingus) writes:
>Is this a ram problem, or what? I recently upgraded my A3000/16 to 6 megs,
>but 6 of the 256x4 Scrams were page mode, and I returned them to be replaced
>with scram type. I went ahead and installed the other 3 meg while waiting
>on the return, and things worked fine. 

So, far, so good.

>Now I install the final 1 meg, and the program REBOOT gives a yellow screen, 

I don't know anything about any REBOOT program.  It may be perfectly OK, but
you shouldn't base the performance of your system on any single program.  By
the way, a yellow screen generally indicates an exception taken before you
get to the point in the OS where alerts work.  That could, of course, go along
with what you're seeing down below.

>Also, I've been having some sporadic gurus that just didn't happen before.. 
>Sometimes bquite bad, 3 or 4 crashes in a couple hours! 

The most likely thing is that one of the last RAM chips you installed is
either defective, or not quite installed correctly.

>Dave, I understand that SetRamsey is totally wrong in how it names the 
>registers, but nonetheless, I can turn on everything it supports, and the 
>machine still functions properly.

Well, you have just pointed out that it doesn't function properly.  But I
think what you're claiming is that you can turn on page-detect mode (what
SetRamsey calls "static column" mode), and the system works OK.  You're wrong.
It may appear to work OK, but it will be using incorrect RAM timing every so
often with hard disk use.  On some systems, this will cause immediate failures.
In other systems, you'll get occasional failures, which show up as an increased
rate of memory failures (which usually generate level 3 and 4 exceptions).  You
may only get failures when the machine gets hot.  Unless you have something 
like 60ns DRAM in there, you will be overdriving your memory.  

Technically speaking, what happens is that RAS precharge to the DRAM, on 
occasions during disk DMA, gets only 40ns, rather than the 80ns it's supposed
to get.  

>Anyway, is one of the new chips bad? 

Failures without page-detect mode set most likely indicates a bad memory chip
or an incorrectly-inserted memory chip.  Failures with page-detect mode set
would be considered expected behavior.

>Are there any A3000 ram test programs?

Unfortunately, none that I know of, yet.

>Wendell Dingus                            UseNet: ...uunet!medsys!wendell


-- 
Dave Haynie Commodore-Amiga (Amiga 3000) "The Crew That Never Rests"
   {uunet|pyramid|rutgers}!cbmvax!daveh      PLINK: hazy     BIX: hazy
	"What works for me might work for you"	-Jimmy Buffett