On Jul 30, 2007, at 9:17 PM, woodelf wrote: Sure they are: http://shorterlink.org/2848 They only get expensive when you're trying to match a particular type. If you're flexible about types, one can get them fairly cheaply. -Dave -- Dave McGuire Port Charlotte, FL Farewell Ophelia, 9/22/1991 - 7/25/2007

On Jul 30, 2007, at 10:59 PM, woodelf wrote: Ahh yes, that can be a rub. eBay is your friend, though...lots of overstock dealers dump huge volumes of new parts that way. That's true, but still...There are plenty of useful applications for such a machine, and better yet, someone who learns on such hardware will have a FAR superior understanding of computers than a graduate of one of today's "Javaschools". -Dave -- Dave McGuire Port Charlotte, FL Farewell Ophelia, 9/22/1991 - 7/25/2007

On 7/30/07, dwight elvey <dkelvey at hotmail.com> wrote: In the PDP world, there were diode bootstraps and rope-core bootstraps. Most DMA devices were smart enough on their own to easily load block zero into core at 0000 with little effort. I recall a PDP-11 bootstrap that depended on human reaction time being slower than the disk that could boot an RK06/7 in, essentially one word. You stuffed the right value into the CSR directly, then jumped to 0000000. Before you could type G 0000000 or whatever, the disk pulsed and the block loaded, leaving you a boot block to jump into. The full and proper auto-boot code wasn't even a dozen words anyway. OTOH, a TU58 boot on a PDP-11 or a TD8E boot on a PDP-8 is a much more involved thing to toggle in, but at least with a PDP-11 w/console ODT, one can have a smart host toggle that in for you (as if it were a human typing in the boot code one word at a time). -ethan

On 7/30/07, woodelf <bfranchuk at jetnet.ab.ca> wrote: Close... "The Art of Digital Design" by Prosser and Winkel (I'm home and staring at the book). Well... owning an -8/i, my only complaint of it is how difficult/expensive it is to expand the memory. Peripherals aren't bad, but it takes 8K internally (I think I have all the parts I need to take mine from 4K to 8K), but the external 24K is rare and expensive. I own one or more of nearly every style of -8 (lacking -8/f, -8/m, WT78), but even so, a TTL and/or an AMD29xx-based -8 is still fascinating to me. I already happen to have a small stack of '181 ALUs, and need to locate a few '83s to keep my pile of TTL -8s working. The F20 uses the '183 adder, so hopefully I can find some at the same place that has '83s. modern "by 8" SRAMs instead of the now-ancient 6147 4Kx1 SRAM, I think I already have 95% of the needed parts in my parts bins. FWIW, I don't see as much value in building a TTL -8 design that doesn't have a memory extension register, since I could just build an IM6100 board if I wanted a 4K TTY-only PDP-8. I'd want to expand the design to have at least 8K, so why not go on to 32K. At least with 8K, OS/8 is possible, but why implement one memory extension bit when 3 bits just isn't that much harder. I disagree. I have tubs of C&K switches with various sizes and shapes of paddles. What isn't cheap is *new* switches with *particular* toggles/handles. Those are easily $5 each. Surplus "match only themselves" can be had for $0.50 at Hamfests or places like BG Micro. I've built Elf toggle panels for $5. A PDP-8 needn't cost more than $15 unless you want very specific switches to match an existing design. -ethan

On 7/30/07, woodelf <bfranchuk at jetnet.ab.ca> wrote: I certainly didn't mean to "put you in your place" - just an honest disagreement about switch costs. The most I ever paid for switches is for my FP6120 - I needed to match specific paddles and a specific soldering hole footprint and paid about $4/switch. OTOH, I've bought fistfuls of similar (but not compatible, unfortunately) C&K switches for $0.25 at Dayton, and for $0.50 at Mansfield. The $0.50 switches are going into a PDP-11/70 front panel - the bodies are compatible, but I still need a double-handful of "real" PDP-11/70 rockers to go in them. My recollection of programming boot PROMs for our 1982-era MC68000 board was yes... we used to need 6309s (like on an Apple II disk card), but routinely used whatever we could get with the same pinout and geometry (256 x 8). I have an old Data I/O, but it's packed away and I haven't powered it on in years. Mostly, I've been programming with a UP600A that might or might not burn PROMs. I hadn't gotten as far as programming PROMs in my F20 flights of fancy. -ethan

On Jul 31, 2007, at 12:41 AM, woodelf wrote: I can program pretty much everything except 1702s here. -Dave -- Dave McGuire Port Charlotte, FL Farewell Ophelia, 9/22/1991 - 7/25/2007

On 30 Jul 2007 at 19:53, woodelf wrote: I wonder how much of an 8/i would fit in a current FPGA? Cheers, Chuck

On 31 Jul 2007 at 1:11, woodelf wrote: That was my thought. Why fool with IM6120s when you fit the durned thing in a teacup? How about a handheld 8/i? Maybe a few MB of CF for storage and an LCD display for a terminal... Might make an interesting coversation piece. Cheers, Chuck

On Jul 31, 2007, at 2:54 AM, Chuck Guzis wrote: How much of an 8/i? I think you mean "how many 8/is"! -Dave -- Dave McGuire Port Charlotte, FL Farewell Ophelia, 9/22/1991 - 7/25/2007

On Jul 31, 2007, at 11:03 AM, Ethan Dicks wrote: The hardware is a no-brainer (I'm visualizing something like the DA11 interprocessor buffer but with appropriate architectural adjustments for the PDP-8) but...the software...thinking of that on an -8 just makes my head hurt. -Dave -- Dave McGuire Port Charlotte, FL Farewell Ophelia, 9/22/1991 - 7/25/2007

On 31 Jul 2007 at 11:19, Dave McGuire wrote: I understand that the PDP-11, while a good candidate for FPGA residency, suffers from a lack of freely available software. Is this true? Cheers, Chuck