Hi folks,
I have or don't have a problem with one of the OmniUSB board. I just don't know.
So is there anybody who has tried to use a board in a pdp8/a? Or is there
someone with a board and a pdp8/a at hands to give it a try?
This is the story: I distributed most of the OmniUSB boards last year. And did
not get very much feedback from the people who bought them. I know only of one
party having successfully used the board. All boards have been checked with a
special CPLD firmware and a checkout program on the pdp8 which together provide
complete test coverage for all signal ways going to the pdp8 and back. Even the
DIP switches were tested individually. So I strongly assume that all boards I
shipped are working fine. But.... A few weeks ago I got informed of a board just
not doing anything in a pdp8/a. I gave hints how to use it etc. Then I asked to
send the board back to me to check what is wrong with it. The board arrived and
I put it into my development 8/e - works perfectly. I also did my checkout
routines - everything seems to be perfectly fine. Now I have no idea what could
be wrong. Something with the 8/a? Or a general 8/a issue? The 8/a has the same
CPU as the 8/e afaik. Could there be some subtle differences between 8/a and /e?
Following some detailed info about how OmniUSB talks to the Omnibus:
These are the signals from and to the Omnibus:
> -- Omnibus connections
> io_pause : in std_logic;
> ts3 : in std_logic;
> tp3_n : in std_logic;
> sw_n : in std_logic;
> initialize_n : in std_logic;
> md : in std_logic_vector (3 to 11);
> DATA_R : in std_logic_vector(4 to 11);
> DATA_D : out std_logic_vector(4 to 11);
> internal_io_d : out std_logic;
> c0_d : out std_logic;
> c1_d : out std_logic;
> skip_d : out std_logic;
> int_rqst_d : out std_logic;
The most operations are enabled by a signal named io_active which is:
> io_active <= io_pause and ts3;
e.g.
> if (match_rx = '1' or match_tx = '1') and io_active = '1' then
> internal_io_d <= '1';
> end if;
for driving the internal_io Omnibus line
Might this be a problem? Is there anything different with 8/a bus timing? Is the
TS3 signal unavailable? Is the io_pause signal unavailable?
I appreciate any hint.
Kind regards
Philipp
> Message: 9
> Date: Fri, 21 Mar 2014 10:06:59 -0600
> From: Roe Peterson <roe at liveblockauctions.com>
> To: General Discussion: On-Topic Posts Only <cctech at classiccmp.org>;
> Subject: Re: PDP-8a Restoration/RL01 issues
> Message-ID: <532C63A3.9040302 at liveblockauctions.com>
> Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
>
> On 03/21/2014 09:55 AM, Ethan Dicks wrote:
>> On Thu, Mar 20, 2014 at 4:47 PM, Anders Sandahl <anders at abc80.net> wrote:
>>> Ok, I got the parts and putted the things together.
>
> It might be a good idea to cover the window on the eprom in your third
> photo...
>
I'll fix that.
/Anders
We have 18 paid units for the group buy, which is enough for the buy to
proceed. There are still several folks who expressed interest, but have
not paid.
If you want one of these, now is the time! Here are the details again, in
case you missed them:
The 32K RAM kit is for an Omnibus card that should work in the PDP-8/e, 8/f,
8/m, and 8/a:
- Complete kit, including battery, PCB and parts. Soldering required.
Includes IC sockets for the RAM kit, but not for the prototype area.
- The amount is $95, including shipping. If you are outside the USA,
please add $20 for shipping.
- Please make Paypal payment to this email address. It will show my
wife's name, Hildebrand. (My initials are vrs, and that email is just
my initials at msn.com.)
- Ordering for this project will close March 31, 2014.
- The kits are expected to ship by May 8th. (Long PCB turn time is used to
keep the price down.)
- Please allow up to 10-days for delivery.
Some features of the kit:
- Battery backup with long life.
- Inhibit on power fail to prevent memory corruption.
- Easy to build, with all through-hole parts.
- Provides full 32K memory in a single card slot.
- Cooperation with any mix of core memory or ROM boards.
- Compatibility with the Omnibus.
- Most of the full Omnibus board area is laid out for prototyping.
- High quality board with gold fingers, plated holes, silk screen and
solder-mask.
You can read Steve's article about the prototype here:
http://www.tronola.com/html/ram_for_pdp-8e.html
Thanks,
Vince
Apparently there's a project to revive it.
http://www.geekdot.com/helios-ng
Source on OSnews:
http://www.osnews.com/comments/27631
The first comment is a nice little paean to the Atari Abaq which
people might enjoy - with a link to a Youtube video of boot-up.
--
Liam Proven * Profile: http://lproven.livejournal.com/profile
Email: lproven at cix.co.uk * GMail/G+/Twitter/Flickr/Facebook: lproven
MSN: lproven at hotmail.com * Skype/AIM/Yahoo/LinkedIn: liamproven
Tel: +44 20-8685-0498 * Cell: +44 7939-087884
I did a comparison of the bus input receiver on the 26S10 vs 8641 a while back.
Here is the plot: http://ak6dn.dyndns.org/stuff/26S10vs8641.jpg
As you can see, there is about a 0.5V difference in the receiver threshold.
May make a difference in some configurations, probably not in most.
Also the 26S10 is a faster device on the bus transmitter output.
This may cause undesired reflections/ringing under some bus loading conditions.
Ideally the transition time should be longer than the electrical length of the bus.
With a 26S10 the maximum unibus length would be much shorter than with a legacy 8641.
Again, may make a difference in some configurations, probably not in most.
Don
-----Original Message-----
>From: Vincent Slyngstad <vrs at msn.com>
>Sent: Mar 19, 2014 12:07 PM
>To: "General Discussion: On-Topic and Off-Topic Posts" <cctalk at classiccmp.org>
>Subject: Re: DEC Unibus Interface ICs
>
>From: Paul Birkel: Monday, March 17, 2014 12:13 PM
>> Unibus Handbook Figure 1-30 does specify the RC-equivalent input for this
>> IC but I have not as yet tried to cross-walk it against well-known
>> SN74xx-series chips. I imagine that I'm the umpteenth person to encounter
>> this problem, so before I try to rediscover fire I'm hoping that someone
>> else could share their torch of knowledge :->?
>
>I remain a fan of the AM26S10 for this, though like everything else, it fails
>to meet the detailed specifications.
>
> Vince
Hi,
The QBus field guide is pretty specific when it comes to the M7609, but rather terse when it comes to the M7621 description. Reading through the KA630 and 650 CPU module user's guides, it seems both CPU's communicate with RAM via CD and the 50 pin ribbon cable.
Is the main difference between the uVax memory boards just access speed, or are they matched by design to the various CPU's?
ie. can an 8Mb board from a KA650 CPU be used with a KA630 CPU?
thanks
--
alex
Hey,
A friend of mine has had an unassembled Heathkit H89 in original packaging.
He wants to know if it is worth any more or less because it was never
assembled. If anyone could help, we'd appreciate it!
Thanks
Joe.
All,
Many of you know I took in Pat's E10K. Sadly, I just don't have the time or
physical power requirements to get it going. I need to find it a good home.
Here is what I know...
64 400Mhz CPU x 64G of RAM E10K. Comes with an assortment of PCI-based and
sbus-based IO boards. Has S1 boot drives. It has 2x U5 SSPs that have the
SSP software and run and boot. It comes with several spares including power
supplies, system control boards and system boards. It's crammed with
various IO boards of the FCAL, ethernet and SCSI varieties.
It's located in Indianapolis, IN 46219 in my garage. You'll need a truck
with a lift to pick it up.
--
-Jon
Jonathan Katz, Indianapolis, IN.