On Mon, Jun 20, 2016 at 10:07 PM, ben <bfranchuk at jetnet.ab.ca> wrote:
> Do you use Static or Dynamic ram with the FPGA's?
I've done both.
You indicated that you wanted 5V I/O. AFAIK, there haven't been any
new FPGAs made in many years that have even 5V-tolerant I/O, let alone
actual 5V I/O. Some really old FPGAs may still be in production, but
are not very cost-effective. The latest midrange to high-end FPGAs
aren't even 3.3V-tolerant. However, the economy FPGAs such as Spartan
6 and Artix 7 still support 3.3V I/O, and are quite inexpensive for
the amount of resources provided.
For 5V-tolerance, it is usually adequate to use 3.3V I/O with series
resistors to limit the current. Xilinx specifies a maximum rated
current for the clamp diodes. This works fine when interfacing actual
TTL (or TTL-compatible MOS) parts. It is NOT adequate for driving 5V
CMOS, such as CD4000 series, because the FPGA won't drive above 3.3V,
and the 5V CMOS inputs typically are specified for Vih min of 90% of
Vdd, which is 4.5V.
The series resistor does slow down the signal, which usually isn't a
problem with TTL since TTL is quite slow by FPGA standards. Where it
is a problem, an nFET voltage clamp can be used instead.
Due to the news about the MacOS name change, it's becoming quite hard to
Google for older MacOS stuff. Was it ever possible to netboot MacOS 8.1 or
earlier? I have A/UX 3 running nicely on a Quadra 700, now, but now I want
to dual boot it with MacOS, but I don't have a CDROM. Taking out the drive
and putting it on my other 68k Mac (a Centris 660AV) and installing MacOS
still gives me some weird issues that I suspect are related to having
installed it on different hardware.
If I can't do a network based install, I'll probably just steal a longer
SCSI cable, use a molex power splitter to add another 5V power cable, and
then install from CDROM while the system is half-open. Then I'll just
button it up afterwards. The factor SCSI cable in my Quadra 700 has only
one connector for a drive.
-Swift
Changing thread title and invoking filter.
Thanks,
- Ian
On Mon, Jun 20, 2016 at 10:50 AM, Swift Griggs <swiftgriggs at gmail.com>
wrote:
> > And yet, now that significant chunks of the Linux underpinnings are
> > being combined into one purpose-written close-knit chunk, designed by a
> > single team, the same sort of people that praise *BSD for its conceptual
> > unity are harshly damning the thing bringing comparable unity to Linux.
> > Odd, that.
>
> It's not the same thing, IMO. People aren't slamming Linux+systemd for
> unifying their team (I've not even seen the harshest systemd critics
> mention this even in passing). They aren't upset because of the greater
> "conceptual unity", either. They are upset because it's breaking faith
> with "the unix way" (creating a giant all-consuming mega-daemon with
> equally heinous binary opaque supporting-crap ala journald) and going
> their own way (a hard right toward Bloatville with a couple of stops near
> Lake Clueless if you ask me), and they are, in the opinion of many, being
> jerks with the implementation of their planned schism.
>
> -Swift
>
>
--
Ian Finder
(206) 395-MIPS
ian.finder at gmail.com
> From: Paul Birkel
>> I will upload the content to the CHW (and add the DB9 pinouts, too).
> Yes, please.
OK, done; see:
http://gunkies.org/wiki/DEC_asynchronous_serial_line_pinout
and it references the new:
http://gunkies.org/wiki/EIA_RS-232_serial_line_interface
I'd be grateful for any feedback on how I can improve the page(s) - more data
to add, thinks to explain better, etc.
Note: I added the DB9 pinouts, _but_ I have never made an actual DEC->DB9
cable, and am unlikely to (see below), so these have not been experimentally
tested. I'm pretty sure they're right (I checked them against some online
tables), but 'the difference between theory and practise', etc, etc. So if
someone does make any DB9 cables from this page, and can confirm that they
works, I'd be very grateful! :-)
I personally don't recommend making DEC->DB9 cables. DB25P<->DB9S adaptors are
cheap and easy to find on eBay, and if you make everything DB25, all you'll
need are a few DB25P<->DB9S adaptors to connect to PCs, and all your other
cabling activity (e.g. connecting one PDP-11 to another) will be simple, since
everything will be standardized on DB25s; no having to keep two kinds of every
cable.
Noel
Is there an electronic copy of this floating around? My (ex-library) copy
is missing all of chapter 11, "What is there to calculate?. (And the last
page of the previous chapter). The pages weren't ripped out, they were
missing when it was bound. Very annoying, I enjoyed the book right up
until it crashed, so to speak.
Two, also ex-library, copies are listed on Amazon, and I hesitate to get
another copy with the same problem. There are others, of course, at
outrageous prices. Or maybe I don't realize the significance of '1st
edition, not ex-library'.
Just to make any discussion a bit more interesting, what would you suggest
along similar lines? The two giant books on IBM (detailing "pre-360", and
"360") were quite fun too.
bri
> I'm puzzled as to how one could drive both interlaced and
> non-interlaced monitors off the same video signal - wouldn't the
> interlaced one need a video signal which has 'odd lines, then a
> vertical retrace, then even lines, then a vertical retrace'?
So to sort of answer my own question, interlaced and non-interlaced video
signals are indeed different.
It turn out that 1024x768 was defined by IBM as XGA, and it was originally an
interlaced format - although a non- interlaced version was done later. So my
laptop quite possibly really is producing interlaced video...
Although how a monitor is supposed to tell whether a signal is interlaced, or
non-interlaced, is not clear - there's certainly no pin on the VGA connector
which says so! :-)
> Anyway, so which one is the one which is the number to look at when
> considering if the refresh rate is so high it might be dangerous to an
> old CRT monitor?
> E.g. my HP M50 manual says "Setting the screen resolution/refresh rate
> combination higher than 1024x768 at 60 Hz can damage the display."
Since the monitor I'm using is called an "Ultra VGA 1024", I'm going to assume
it can handle 1042x768, and just stop worrying about it... If it melts down
the monitor, it melts down the monitor! :-)
> From: Jochen Kunz
> Sounds like an interlaced video mode. No surprise that the LCD can't do
> this.
Yes, as soon as I realized it probably really was interlaced video, it became
obvious why none of my LCD monitors would display it.
Noel
I have a friend looking for 2 DEC VRC21-KA monitors in Europe.
I have one here, but will pack it only as a last resort.
If you have one of two you want to sell or trade, please let me know off
list. Also where you are located.
Thanks, Paul
So I'm using an old (very old! :-) laptop for a console for some of my
computers. The built-in screen on it is this miserable 800x600 thing (I said
it was _old_ :-), so I'm trying to hook up an old 15" display (of which I have
tons) and run it at 1024x768 at least (the highest resolution the old display
hardware in the laptop will support). However...
The first couple of LCD's I plugged in, they worked fine, but when I went to
change from 800x600 to 1024x768, I got error messages about 'can't handle
that display format' (or words to that effect). Nothing about what parameter
was out of bounds, alas. Finally I found one that _did_ give me the numbers,
and it said the retrace was 35.6 Khz / 44 Hz.
OK, looking at some specs, it makes sense those LCD monitors didn't work - 44
Hz vertical retrace (which I gather is the same as the refresh rate, in things
like 1024x768/60Hz - sorry about the newbiesh questions, but I want to make
sure I'm not making a bad assumption) is _below_ the input specs on them.
(Although why the refresh rate on an _LCD_ would have a lower bound that high
doesn't make sense to me - so the screen updates less often, what's the
problem? It's not like a CRT, where it could cause flicker.)
So I finally found an old CRT monitor that worked - but now I have another
problem! It reports the video as being 35.6 Khz / 87 Hz! Which makes me
worry, because I'm not sure that particular monitor can handle an 87 Hz
refresh.
(Parenthetically, what exactly is the mechanism that causes damage if you run
an old CRT monitor at too high a refresh rate? I assume the excessive speed
generates too much heat somewhere, and causes transistors to fail, or
something like that?)
Anyway, back to the monitor - I'm wondering if that monitor is reporting a bad
number on the vertical retrace, and in fact it actually is 44 Hz? Because I
found this equation to calulate the vertical retrace frequency from i) the
horizontal retrace frequency, and ii) the number of lines. So I plug 35.6 Khz
(which agrees with the other monitor, note) and 768 into that formula, and
get... 44 Hz!
So is there some bizarre interlace mode, or something, that could
legitimately cause confusion over the vertical retrace? Or is the second
monitor just confused?
I tried to look up the video controller chip (a Cirrus CL-GD 7543 'Viking'),
to see if it could even _do_ 88 Hz, but I was unable to find any
documentation about it online.
Also, if it really is 44 Hz, can anyone point me at an old LCD display that
will handle that? (The CRT I'm currently using - another one that supposedly
handles up to 100 Hz vertical retrace - takes up too much room! And there's
no point getting something new - and larger - since the machine won't do
anything past 1024x768 anyway.)
Noel
Hi
The great TK revival continues apace.
There's a TK50 in my VAX 4000 running really well. Purrs like a cat
I have TK70 in the RT-11 (11/83 QED) Machine that also runs.
TK70 should read the TK50 tapes. It tries but complains about cant read
the directory.
So whats a common format I can write on a TK50 tape in a TK50 Drive (VMS)
then take it over to the RT-11 box and have it read it.?
Where is all this going? Well I need to end up with RSTS on an 11/83.
So a bootable TK50 with RSTS install files on it is the goal.
Meanwhile transfer by tape is useful.
Rod