NetNews Usenet Archive 1992 #19/NN_1992_19.iso/spool/comp/sys/amiga/hardware/14663

Internet Message Format | 1992-08-29 | 3.6 KB

Path: sparky!uunet!sun-barr!cs.utexas.edu!ut-emx!ibmchs!auschs!awdprime.austin.ibm.com!calais.austin.ibm.com!jpietro
From: [e-mail] (Jeffrey A. Pietro)
Newsgroups: comp.sys.amiga.hardware
Subject: Re: VXL-30: df1 trouble
Message-ID: <[e-mail]>
Date: 28 Aug 92 18:06:57 GMT
References: <[e-mail]>
Sender: [e-mail] (USENET News)
Reply-To: [e-mail]
Organization: IBM Austin, TX
Lines: 72

In article <[e-mail]>, [e-mail]
(sanford tong) writes:
> Subject: VXL-30: df1 trouble
> Date: Thu, 27 Aug 1992 22:54:49 GMT
>
>     Hi, folks!
>
>         I've finally installed a VXL-30 into my A500. It seems to run
> without any problems, except when i choose the FASTROM option.
> (ie: Kickstart -> 32-RAM). When i use this, the df1: light turns on
(like when
> you put a disk in), then comes up with df1:bad on WB. This happens
even when the
> drive is EMPTY. I haven't tried it with a disk inside, but i shudder
to think.
>
>     Has anyone else had this problem, or know what the story is (might be)?
>
>     (All other options (cache, burst, etc) don't have this problem)
>


I've got a similar problem as well.  I've got a B2000 with MicroBotics'
HardFrame/2000 (with Quantam 105 Meg HD) and the VXL-030 w/RAM*32 board.

When I use the FastROM option (either with SetVXL or SetCPU), my system
runs a little flaky:  system hangs, keys will repeat, or total loss of input
from the keyboard.

This doesn't happen all the time, but one or more of them will occur
given enough
time.  I have not been able to recreate this without using the FastROM option.

One side note:  when using SetCPU to enable the FastROM option, the memory test
program (MBRTest2) lists the RAM*32 board as 16 bit wide memory.  Disabling the
FastROM changes it back (at least to MBRTest2) to 32 bit wide memory.
There seems
to be little difference in performance (as reported by AIBB4.5):  some a little
better, some a little worse.  SetCPU, however, appears to use less FAST
RAM than
SetVXL does (about 270K compared to 512K)--anybody else notice this???
[By the way,
I do have a 68030 w/MMU, although it is an XC68030 which has performed
flawlessly
compared to the RC68030 (what fails with one fails with the other)].

Another side note:  when I first received my RAM*32 board, Burst mode would not
work when Caches were enabled.  I could have one, but not the other enabled.  I
called MicroBotics and they had me check Chip U3 (???) on my VXL-030
accellerator--it's
found directly above the "L" in the "VXL-030" logo.  If this chip is
ANYTHING OTHER
than a 74AS86 made by National Semiconductor, it is too noisy for the board and
should be replaced.  I sent my board back to MicroBotics for repair and
not only did
they replace the chip, but the soldered a jumper on the back of the
board so that
the black jumper wire to the RAM*32 board is no longer needed (very
NICE!!!).  My board
was a Version 2 board before I got the RAM*32 with upgrade.

Now, my only problem appears to be the flakiness related to using the
FastROM option.
Anybody else have this problem (or better yet--get it corrected)?

+-------------------------+------------------------------------------+
| Jeffrey A. Pietro       | EMail: [e-mail] |
| Bull HN, Inc.           +-----------------------+------------------+
| c/o IBM (int zip: 9582) | VOICE: [phone] |    Listen to     |
| 11400 Burnet Road       | FAX:   [phone] |  Rush Limbaugh!  |
| Austin, TX  78758       |                       |     "Ditto!"     |
+-------------------------+-----------------------+------------------+