- vcf-midatlantic - lists.vcfed.org

2:00am BASIC epiphany
by Evan Koblentz 31 Aug '18

31 Aug '18

I had a spontaneous idea tonight for removing two lines of my Lego robot code. As usual, this epiphany on my part is probably TOTALLY FRIGGIN' OBVIOUS to most of you, but I'm proud that I figured it out on my own. :) Before, on the forklift robot, this was my code for checking the upper/lower touch sensors (limit switches) and moving the forklift motor appropriately. I had it at 9 lines (not counting REMarks) which was greatly reduced from Dan R.'s original 16 lines. 3010 IF PEEK (49249) > 127 AND PEEK(L) = 64 THEN RETURN 3020 IF PEEK (49249) > 127 THEN POKE L,16 3030 IF PEEK(L) = 80 THEN RETURN 3040 IF PEEK(49249) > 127 THEN GOTO 3020 3050 IF PEEK(49250) > 127 AND PEEK(L) = 128 THEN RETURN 3060 IF PEEK(49250) > 127 THEN POKE L,32 3070 IF PEEK(L) = 160 THEN RETURN 3080 IF PEEK(49250) > 127 THEN GOTO 3060 3090 RETURN Background: In that code, PEEK(49249) in Applesoft checks joystick button 0 and PEEK(49250) checks button 1. If their value is greater than 127 then you know they're currently being pressed. PEEK(L) checks the value at the address of the interface card. 64/128 are the sensors attached to ports (bits) 6 and 7, respectively. POKE L,16/32 turns on the motor (in opposite directions) that's attached via a connector spanned to ports 4/5. Lines 3030 and 3070 stop the motors if the sensors become active ** while ** the motor is moving, because then the total value being read from the interface box is sensor + the motor. The idea that popped into my head tonight: you can combine lines 3010/3020 into one, and also lines 3050/3060 into one, thereby reducing the routine by two lines and having the same effect. 3010 IF PEEK (49249) > 127 AND PEEK(L) <> 64 THEN POKE L,16 3020 IF PEEK(L) = 80 THEN RETURN 3030 IF PEEK(49249) > 127 THEN GOTO 3010 3040 IF PEEK(49250) > 127 AND PEEK(L) <> 128 POKE L,32 3050 IF PEEK(L) = 160 THEN RETURN 3060 IF PEEK(49250) > 127 THEN GOTO 3040 3070 RETURN So instead of saying, "Exit the routine if the lower/upper touch sensors are active, which we know is true if the value at address L is 64 or 128," instead I'm saying, "Turn on the motor ** ONLY IF ** the lower/upper touch sensors aren't active." I can't test it until this weekend at the soonest, but I am ** pretty sure ** it'll work -- or at least I don't see any obvious reason why it would ** not ** work. After all, the forklift motor can't turn on when either sensor is active, because I'm not programmatically allowing it -- right? If it works, then I'll convert it to IBM BASIC too. The button PEEKs become STRIG (stick trigger) values, and the POKEs become OUT. Saving two lines of interpreted / non-compiled BASIC isn't going to have any significant impact on program execution speed or disk space, but it makes the code that much easier for non-programmers to understand, and it gives me that much more confidence of about being than just another English major. :)

2 3

Working Tandy 1000 or 2000
by Jonathan Gevaryahu 31 Aug '18

31 Aug '18

Does VCF have a working Tandy 1000 or 2000 on display or in the warehouse? I need someone with access to a working Tandy 1000/2000 to do some poking from basic of the sound chip, since there's an issue I'm running into with a possible incompatibility of the Tandy 1000/2000 NCR8496 PSG (sound chip) and the more common SN76489/SN76496 PSG which the colecovision etc used. -- Jonathan Gevaryahu jgevaryahu(a)gmail.com jgevaryahu(a)hotmail.com

3 14

Re: [vcf-midatlantic] Schematic work needed
by Evan Koblentz 30 Aug '18

30 Aug '18

> There's one for $40/free shipping right now: > > https://www.ebay.com/itm/232755988436 Hmm. That's weird, I knew it as part #9750, so I searched for that term. Did not know to search for #70455, which apparently is the same exact thing: https://www.bricklink.com/v2/catalog/catalogitem.page?P=70455#T=C&C=11. > The receptacles that the various Lego things plug into, in the control > box? Those are a little formed contact with the housing of the control > box making up the connector shell, from what I remember. You could > probably cob something together that'd work, but you can't just go buy > that connector, as far as I know. Oh, didn't know that. I do know that 3mm mini-banana plugs fit perfectly into those holes. So the matching 3mm receptacles would do fine rather than using the Lego part. Speaking of which: I take back what I said earlier about "100% perfect". I meant 100% functional, not 100% in appearance. It could be powered by a Raspberry Pi for all I care. I just want to give people a way to run the vintage software (Lego Logo; BASIC; etc.) on their Apple II or IBM PC and connect it to the period 1980s Lego motors/sensors. Is there a easier/modern way to create the middleware hardware?

2 2

Re: [vcf-midatlantic] Schematic work needed
by Evan Koblentz 30 Aug '18

30 Aug '18

>> this is for a presentation slide, not to build the thing No. This is for anyone to view so they can make the card for their own use. On Wed, Aug 29, 2018, 2:19 PM RETRO Innovations via vcf-midatlantic < vcf-midatlantic(a)lists.vcfed.org> wrote: > On 8/29/2018 10:07 AM, Bill Degnan via vcf-midatlantic wrote: > > this is for a presentation slide, not to build the thing > > And, if that's the goal, then I don't see why 100% perfection is needed, > and thus I agree the effort is much more manageable. However, I will > again note, that's not what the original request stated. > > In fact, I would agree with you that simply a block diagram should > suffice for such a slide. > > Jim > > >

3 4

Configuring the Processor Tech 3P+S as an IMSAI SIO-2
by systems_glitch 30 Aug '18

30 Aug '18

link: http://www.glitchwrks.com/2018/08/30/3ps-imsai-sio I recently repaired a Processor Tech 3P+S I/O board for the VCFed Museum. This 3P+S is going in an IMSAI, so it was to be configured as compatible with the first channel of an IMSAI SIO-2. The 3P+S is a super versatile I/O board, much of the configuration is left to the user to jumper in as they want, but making it look like an IMSAI SIO-2 did require a cut-and-jump. But, after fixing the control/data register ordering and taking the bitrate base clock from the 2 MHz pin (pin 49 on S-100) instead of the system clock, it works! I was able to pull my IMSAI SIO-2, drop in this modified 3P+S, and run the software I'd been running on the SIO-2. The 3P+S is on my short list for "S-100 boards to reproduce." It's a very handy serial interface, being as how it can be made to look like most other manufacturers' boards. Thanks, Jonathan

1 0

Re: [vcf-midatlantic] the first pDa
by William Dudley 30 Aug '18

30 Aug '18

My understanding is that Wikipedia "management/rules" will accept information if backed by citations. This has resulted in the following stupidity: An expert in a field makes an edit to a Wikipedia article. It is rejected for lack of an accompanying citation. The same expert cites an article published in a magazine, that HE WROTE, about the topic. So -- he's citing *himself*. Expert's Wikipedia edit is now accepted because it has an accompanying citation. All of which suggests that if Evan makes an edit, and cites HIS OWN BOOK, the edit will be accepted. Bill Dudley This email is free of malware because I run Linux. On Wed, Aug 29, 2018 at 11:20 AM, Neil Cherry via vcf-midatlantic < vcf-midatlantic(a)lists.vcfed.org> wrote: > > > ---------- Forwarded message ---------- > From: Neil Cherry <ncherry(a)linuxha.com> > To: vcf-midatlantic <vcf-midatlantic(a)lists.vcfed.org> > Cc: > Bcc: > Date: Wed, 29 Aug 2018 11:20:33 -0400 > Subject: Re: [vcf-midatlantic] the first pDa > On 08/29/2018 12:50 AM, Evan Koblentz via vcf-midatlantic wrote: > >> >> https://en.wikipedia.org/wiki/Personal_digital_assistant >>> >>> who is going to correct this? >>> >> >> Obviously I'm the guy, but I am very un-motivated to argue with Wikipedia >> editors. >> > > Amen to that! > > -- > Linux Home Automation Neil Cherry ncherry(a)linuxha.com > http://www.linuxha.com/ Main site > http://linuxha.blogspot.com/ My HA Blog > Author of: Linux Smart Homes For Dummies > >

6 7

the first pDa
by Bill Degnan 30 Aug '18

30 Aug '18

https://en.wikipedia.org/wiki/Personal_digital_assistant who is going to correct this?

4 4

Re: [vcf-midatlantic] Schematic work needed
by systems_glitch 29 Aug '18

29 Aug '18

I'd had the ISA card in the shop for a while and didn't get around to mapping it out. I've explained to Evan that it's not a trivial process. I offered to depopulate the board and send it off to Mile High Test for a scan + conversion to Gerbers, which are guaranteed to be 100% accurate, but that costs around $200-250. It's the same process I've used to reproduce various Ohio Scientific and S-100 boards. Thanks, Jonathan On Wed, Aug 29, 2018 at 2:58 AM RETRO Innovations via vcf-midatlantic < vcf-midatlantic(a)lists.vcfed.org> wrote: > On 8/29/2018 1:15 AM, Evan Koblentz via vcf-midatlantic wrote: > > I need someone to make two (relatively simple) schematics for me. One > > is the IBM card for the Lego kit, and one is the Lego interface box. > > > > Here is one side of the IBM card: > > http://snarc.net/lego_ibm_1.jpg > > > > Here is the other side: > > http://snarc.net/lego_ibm_2.jpg > > > > Here's a picture of the top of the interface board: > > > http://4.bp.blogspot.com/-DGOlTkGXj_Q/U8-T2uBG5iI/AAAAAAAAATY/7ra6gVkSKg8/s… > > > > > > Here's the bottom of the interface board: > > > http://lukazi.blogspot.com/2014/07/lego-legos-first-programmable-product.ht… > > > > > > Here's a document Dan R. made of the board a year or two ago: > > http://snarc.net/9750.png > > > > Now here is the catch..... I absolutely MUST have a complete, perfect, > > finished schematic of the IBM card and the interface board by a week > > before Maker Faire -- so let's say the deadline is Friday, Sept. 14. > > Do you have a backup plan? I say that because I professionally feel > that creating a working and tested schematic within your timeline is > going to be nearly impossible. > > * The first board is two sided, and those typically need to be > unsoldered to ensure all traces under the ICs have been mapped. One > can try to use a multi-meter, but there are pitfalls > * Even if the schematic is doable, there's no way to truly ensure it's > perfect unless one make a spin of the PCB, populates it, and tests > it. unless one wants to spend $250.00, it takes about 2 weeks to > spin a PCB. > * Though the other board looks to be single sided, all of the items on > the board must be described and validated. The toroids in the > center and right seem particularly problematic, as one needs to > determine the uH, turns, and such of the items in order to be > perfect. As well, those TO-220 transistors (or maybe they are FETs) > sandwiched in between the connectors need to be desoldered or > somehow moved so the markings can be read. The LEDs need to be > scoped to determine their mA rating, etc. > > I'm not trying to rain on the parade, but I do want to inject some > realism into the request timeline. Doing such a reverse engineer will > probably take a month or so, assuming someone squeezes it into their > normal workflow and they do this regularly. > > Someone already made a schematic of the Apple II card: > > > http://lukazi.blogspot.com/2014/07/lego-legos-first-programmable-product.ht… > > > > Regrettably, it looks like the Apple uses a different schematic > completely (It uses a VIA instead of the IBM PC discrete TTL) and so > this schematic will not help. > > Jim > > > -- > RETRO Innovations, Contemporary Gear for Classic Systems > www.go4retro.com > store.go4retro.com > >

7 8

Re: [vcf-midatlantic] Schematic work needed
by Bill Degnan 29 Aug '18

29 Aug '18

this is for a presentation slide, not to build the thing On Wed, Aug 29, 2018 at 11:05 AM RETRO Innovations via vcf-midatlantic < vcf-midatlantic(a)lists.vcfed.org> wrote: > On 8/29/2018 9:49 AM, Kyle Owen via vcf-midatlantic wrote: > > Why would ohming out this board not yield a 100% correct schematic? I see > > no tricky components that would confuse an ohmmeter except the address > > switch, which could be unsoldered. > > Many of the signals go under the ICs. Sometimes, you can hold to a > light and see through the PCB and validate where they go, but other > times, I have missed wires connecting to more than 1 pin of the IC under > the IC. Ohming, if applied verbosely for each signal around the IC > compared to each IC pin, should find all of them, but I don't think I > would promise 100% perfection without spinning a PCB and testing the > completed unit. > > Maybe everyone else is better at ohming out boards and the hidden > signals under ICs never bother anyone else, but I've gotten hit twice by > them on two different efforts. > > Jim >

2 1

Re: [vcf-midatlantic] Schematic work needed
by alan＠alanlee.org 29 Aug '18

29 Aug '18

Sure. If Evan can wait till next week and Eagle or Altium is fine. Sent from my Verizon LG Smartphone ------ Original message------From: systems_glitch via vcf-midatlanticDate: Wed, Aug 29, 2018 11:48 AMTo: vcf-midatlantic;Cc: systems_glitch;Subject:Re: [vcf-midatlantic] Schematic work needed Sounds like we've got plenty of volunteers, then! :P Thanks, Jonathan On Wed, Aug 29, 2018 at 11:45 AM alan--- via vcf-midatlantic < vcf-midatlantic(a)lists.vcfed.org> wrote: > It's two 8 bit opposing latches and some address decoding glue. If one > knew the port address, you could create the schematic blind and be 98% > correct. > Sent from my Verizon LG Smartphone > ------ Original message------From: Kyle Owen via vcf-midatlanticDate: Wed, > Aug 29, 2018 11:17 AMTo: vcf-midatlantic;Cc: Kyle Owen;Subject:Re: > [vcf-midatlantic] Schematic work needed > On Wed, Aug 29, 2018 at 10:05 AM RETRO Innovations via vcf-midatlantic < > vcf-midatlantic(a)lists.vcfed.org> wrote: > > > > > Many of the signals go under the ICs. Sometimes, you can hold to a > > light and see through the PCB and validate where they go, but other > > times, I have missed wires connecting to more than 1 pin of the IC under > > the IC. Ohming, if applied verbosely for each signal around the IC > > compared to each IC pin, should find all of them, but I don't think I > > would promise 100% perfection without spinning a PCB and testing the > > completed unit. > > > > Maybe everyone else is better at ohming out boards and the hidden > > signals under ICs never bother anyone else, but I've gotten hit twice by > > them on two different efforts. > > > > Yes. If you do not completely ohm out each pin/component, I could very much > imagine generating an incomplete schematic. The netlist generated by ohming > out the board should have every pin on every component by the time you're > done. If not, you've missed something. With some conditional formatting, > it's possible to highlight possible errors like duplicate entries, too many > pins per component, etc. > > Hidden traces are no problem *if* you are not relying on visuals, but > rather the ohmmeter. > > And if you're using a Fluke, don't forget to hold down the yellow button > when turning it on, lest you have it automatically power off while you > attempt to ohm out the board. Ask me how I know... :) > > Kyle >

1 0