Here is a first draft. Anything wrong/missing architecturally?
Ethernet Bus Interface Board
2013-12-21 aek
+--------------+
| 4Mx16 BBSRAM |
+------+ | AS6C1616 |
| | +------+-------+
LEDs - + ARM | |
SD -- + uCtl | JTAG +------+-------+
USB -- + +-----------+ | -> Simulated Peripheral Registers
ENET - + +-----------+ FPGA | Bus Monitor / Adr Mapper
SER?-- | | ARM intf | | Bus Master Controler
+------+ +------+-------+
|
+------+-------+
| |
| 5v<->3.3v |
| Bus Buffers |
+--------------+
The entire address space of the host is maintained in the SRAM
ARM performs DMA by placing the data to be transfered where it
will appear on the host. If the local memory is the system memory,
we're done, just clean up the DMA registers. If not the FPGA does DMA
to synchronize local memory with core. 3 cycle data break side effects
will have to be investigated. Does any driver watch the value of ADR,WC
in memory while a DMA is occurring?
Bus transactions are snooped to maintain consistency on the PDP-8
where some core memory is on the Omnibus. This may work OK on small
11's, too but won't be the right thing to do on an 11/70 where
memory transactions occur that are invisible to the Unibus.
It should be possible to simulate all the front panel functions of
an Omnibus machine and to capture all of the state in real time for
all of the major registers from watching Omnibus transactions.
It is also possible to adapt the board to be a dedicated bus analyzer or
soft front panel or peripheral/ memory exerciser for any supported bus
architecture.
All registers and supported peripherals are obviously soft. The configuration
is stored on the SD, along with the FPGA bitstream. Everything should be
reconfigurable through transactions across the Ethernet by updating files on
the SD card.
Battery-backed SRAM was chosen over FRAM because of cost.
I'm not sure if a local serial interface would be that useful other than
for debugging messages.
There are a small number of LEDs hung off of a STP08DP05
Small local SPI NOVRAM for configuration/bitstream storage? SD hot-swap?
Larger local storage? Is performance an issue on Omnibus/Qbus/Unibus CPUs?
They are slow enough busses that they'll saturate in the couple MB/sec range.
Write-through track caching in SRAM?
Block mode on QBus
.. and on and on
Ethernet Peripheral Controller
This is the mirror of the Bus Interface Board. It provides block-level data
that the Bus Interface sends into the computer. In its simplest form, there
is no hardware at all, it is just a simulation running on a computer on the
network. The second simplest is hardware that understands the Ethernet
peripheral command protocol but has a simple enough interface to the physical
device that no FPGA is required. It may also not need any local storage. The
most complicated could look like the Bus Interface Board, with local SRAM and
FPGA. A Diskferret-like forensic disk interface would be an example of this.
I'm looking for a BC80J-20 cable... that's the one that goes from the
RL controller card to the first drive, with a 40 pin header on one end
and that big quick-release connector on the other end.
(I am tired of failures and damage to the rigged ribbon cable/IDC
header I've been using, and now that I'm chasing a fault in my two-
RL02 system, I can't swap drives without taking the cover off to get
to the header connectors inside).
thanks
Charles
Ask and you shall receive..
4 BNC to PC VGA, new, qty 4, approx 4 feet long, $15 each
3 BNC to PC VGA, used, qty 1, approx 3 feet long, $10
5 BNC to PC VGA, used, qty 1, approx 5 feet long, $15
4 BNC to 15-pin Rasterops PN 9000-0059, new but dusty, $15 (will not work on
PC, wrong end)
4 BNC to 15-pin Rasterops PN 9000-0001, new, $15 (will not work on PC, wrong
end)
Shipping $6 for the 1st cable, and $3 each additional cable ground in US.
Cindy Croxton
Electronics Plus
1613 Water Street
Kerrville, TX 78028
(830)792-3400 phone (830)792-3404 fax
AOL IM elcpls
_____
No virus found in this message.
Checked by AVG - www.avg.com
Version: 2014.0.4259 / Virus Database: 3658/6978 - Release Date: 01/05/14
As I don't have a logic analyser to help me with my analysis of the MFM
interface, can anyone tell me how the MicroVAX 2000 determines disk size? I
know it is related to the number of heads, but what does the drive do when
the host asks for a head that does not exist? How does it report this back?
The only thing I can think of is the Ready signal, could that be it?
Regards
Rob
From: Liam Proven <lproven at gmail.com>
> So, an ad-hominem attack is your idea of a balanced reply, is it?
Can't attack the message, attack the messenger. Good to see my assessment
was spot on.
KJ
> I'm just wondering if anyone knows what a typical value for a single unit
> with nothing fancy is? I note a recent auction had two of them together
for
> $1400 - Am I correct to imply from that that the value of an individual
unit
> is probably less than $700 (that set was unique due to the systems being
> linked and that control box they had, I think).
A working, complete, 1975-6 SWTPc with top and bezel, original backplane,
original CPU and RAM with an MP-S card is worth $1000+. That would be the
bare minimum system. $700 is too low for a working system. The newer 6809
SWTPc's would be worth less, maybe you can get one of these for $700.
Bill
I've seen a couple of vague references to 3rd party memory cards for the
VAX 11/730, equivalent to the M8750, that used denser DRAM chips in order
to use less power. Unfortunately, my google-fu has turned up nothing.
Does anyone know who made these cards and what part number they would be
(assuming they exist at all)?
KJ