PDA

View Full Version : Mark-8 boards on ebay



falter
September 27th, 2016, 04:18 PM
Wondering what you guys make of these and if they are legit:

http://m.ebay.ca/itm/302083673875?_mwBanner=1

The seller cancelled the auction. I did try to ask if they had some proveneance to back up whether they were original or not. I have seen some reproductions that got pretty close to original look. Wondered what you guys thought, and also what original bare boards would likely be worth? I'm prepared to go all the way if these are original and they ever come up again.

mwillegal
September 28th, 2016, 09:21 AM
They appear vintage to me - the only thing that gives me pause for concern, is the apparent shininess of the plating. The plating on most vintage PCBs, have acquired some amount of tarnish over the years. However the substrate appears to the be vintage, and that would be expensive to fake. I would want to get a view in person, before dropping a load of dough on something likely to be as expensive as those could become. Other things to check for, is unplated holes, as original Mark- 8 boards did not have plated holes.

My guess is that the auction was pulled because someone offerred a lot of money for the set.

regards,
Mike Willegal

dfnr2
September 28th, 2016, 09:32 AM
I was about to post about these boards, but since you already did, I'll just add a reply.

The seller contacted me about a month ago, after having bought several boards at an auction. These included some older and fairly rare unbuilt OSI boards, which is how he found me (through OSIweb.org). He put these up on Ebay, and asked me to let anyone who may be interested know about them, which is why I'm posting here. As far as I can tell, this is a complete, real, genuine set of bare Mark 8 boards.

I see he has re-listed the boards:

http://www.ebay.com/itm/MARK-8-circuit-boards-build-your-own-70s-vintage-computer-kit-/291890554372

Dave

falter
September 28th, 2016, 09:39 AM
Cool. Any thoughts on value? The only Mark-8 auction records I've found are 8-9 years ago and around $5200 for a complete system. I'd happily drop $5k on these (it'll kill my vintage budget for a year or two) if they're real.. not like I'm gonna get them anywhere else.

falter
September 28th, 2016, 09:42 AM
One other question.. if you wanted to 'fake' the substrate.. how would you do that? I was hoping to just build my own Mark-8 using the plans.. but I hate the look of modern PCBs. I was wondering if someone out there can reproduce something that looks closer to vintage than that.

falter
September 28th, 2016, 09:55 AM
And one final q.. would it be bad for value/crime against history to try and build these into a system? Given that they look to be unspoiled new?

Moonferret
September 28th, 2016, 10:07 AM
I'd been looking at Mark 8's during the past few days. These PCB's certainly look like the originals but I'd imagine they would go for big bucks. As for building them up into a complete machine, that's what they are for. Go for it! However, I guess I'm only saying that as I won't be able to afford them and if I could I'd probably chicken out on putting a soldering iron to them :)

I know there were some replica boards available a few years ago but does anybody know of anybody still selling board sets? I built a SCELBI 8B using a set of Mike's boards the other year and really enjoyed it. Would be nice to add a Mark 8 to my replica collection.

Cheers,
Dave

falter
September 28th, 2016, 11:04 AM
Yeah. I'm kinda torn.. if those are original they are pretty close to pristine.. how many pristine as-sold original pcbs are you likely to encounter? Feels like anything you did to them would take their value down, especially if you screwed anything up. And even if you built them into a machine it'd never be an 'original' Mark-8.. it'd be more like a new Delorean. Plus some original parts just aren't available as I've learned with my TVT project.

Hmmm. I had this debate when I saw an original set of TVT boards that I narrowly missed. $38 though. But even then, I was torn between 'oh it'll be an original TVT' and 'no.. it'll still be mostly a repro from 2016 and now no one will be able to see how the original boards came from factory.

Corey986
September 28th, 2016, 11:08 AM
One other question.. if you wanted to 'fake' the substrate.. how would you do that? I was hoping to just build my own Mark-8 using the plans.. but I hate the look of modern PCBs. I was wondering if someone out there can reproduce something that looks closer to vintage than that.

Faking substrate on boards without a solder mask is actually easy. It is not easy on boards that have a solder mask. Yellow modern boards can be dyed by boiling them in dye solution. They can even be tarnished using an oxidizer. I'm always very careful with valuable vintage boards that do not have a mask. You have to compare them to a known set to see if they have the same flexibility, colour tone and finish. This is very hard for Mark-8 systems since the boards were really made by many different companies with many different quality and manufacturing of boards.

Cheers,
Corey

falter
September 28th, 2016, 12:11 PM
Thanks Corey. I find with the new PCB stock I have they are a bit too brown.. but also the finish is very clean on one side. Just looks too modern. I'd like to learn more on how to get closer to a vintage look.. not to pull a scam but just so my replicas look better.

I'm assuming making those pcb manufacturer marks isn't hard by comparison to boiling dye into the fibreglass?

KC9UDX
September 28th, 2016, 02:21 PM
I wonder if you could sandblast one with a fine media before dying to make it look more like an older board. I think with low pressure the copper shouldn't blast off.

falter
September 28th, 2016, 02:24 PM
I'm really surprised there isn't like, 'vintage copper clad PCB stock' out there somewhere. There's virtually everything else but that. Ought to be a business there for someone enterprising enough.

Out of curiousity -- I've heard it said before that finding Mark-8 ICs and parts would be prohibitively difficult.. does anyone out there know which parts in particular are the toughest to get? I noticed there's not much in the way of 1101 RAMs, etc for example.

Moonferret
September 28th, 2016, 02:35 PM
Out of curiousity -- I've heard it said before that finding Mark-8 ICs and parts would be prohibitively difficult.. does anyone out there know which parts in particular are the toughest to get? I noticed there's not much in the way of 1101 RAMs, etc for example.

I think the tricky ones are the Signetics 8267's and 8263's

inotarobot
September 28th, 2016, 02:35 PM
I'd been looking at Mark 8's during the past few days. These PCB's certainly look like the originals but I'd imagine they would go for big bucks. As for building them up into a complete machine, that's what they are for. Go for it! However, I guess I'm only saying that as I won't be able to afford them and if I could I'd probably chicken out on putting a soldering iron to them :)

I know there were some replica boards available a few years ago but does anybody know of anybody still selling board sets? I built a SCELBI 8B using a set of Mike's boards the other year and really enjoyed it. Would be nice to add a Mark 8 to my replica collection.

Cheers,
Dave

Falter and Dave,

well if they are original, and they do look that way, spending up to $5k on the boards is one thing; but then you would need to find Guaranteed NEW OLD stock of the ics to use. That includes correct period date codes.

And certainly considering how RARE these blank boards are to come by, I would be checking, then doing a 'burn cycle time' run on, every IC on a quality IC tester before soldering them in place.

If the original uses IC sockets then you don't need to do the above step, but many early machines simply had the ICs directly fitted.

Then there is the issue of getting the period connectors and the matching interconnecting cable.

In my view, it matters NOT, if you built the machine in the 1970's or in 2016, IF and that's a big IF you use 100% original period parts that were all made in 1970's; then that machine is not a reproduction.

There are rare times via a Garage finds, you come across a sealed 'original kit' of some item, from the way back in the past.

So if you assemble and complete that 'kit' using only the tools and skills of the artisans from that past time, then I feel its no different than making it within a year or 2 of its manufacture.

Now one point I would say, is from when I built my first computer kit, a Motorola MEK6800 D2 in 1976, to now 2016, when I am about to build a brand newly designed surface mount component, Analog Computer, my electronic skills over that 40 years have vastly improved. ( well I still occasionally make the stupid mistakes).

So now if I was to get a brand new, un-assembled complete Motorola MEK6800 D2 kit, it would NOT end up being built exactly the same as my first one, 40 years back; as I cannot UNLEARN my computer and electronics skills.

So in this D2 kit example, if I could get a un-assembled D2 kit now and built it up, WOULD it be an original or a replica ???

Chuck(G)
September 28th, 2016, 03:39 PM
So in this D2 kit example, if I could get a un-assembled D2 kit now and built it up, WOULD it be an original or a replica ???

Well, suppose you rolled up the circuitry into a small FPGA?

If it works the same way, what's the difference?

Klyball
September 28th, 2016, 03:50 PM
I have some original unbuilt osi boards , and I made replica's so I could leave the original boards unbuilt and could play around without worry of wrecking them http://www.vcfed.org/forum/showthread.php?45450-SuperBoard-II-to-build-or-not-to-build/page2&highlight=superboard

I would do the same here (if I could afford to buy them)

falter
September 28th, 2016, 03:55 PM
Falter and Dave,

well if they are original, and they do look that way, spending up to $5k on the boards is one thing; but then you would need to find Guaranteed NEW OLD stock of the ics to use. That includes correct period date codes.

And certainly considering how RARE these blank boards are to come by, I would be checking, then doing a 'burn cycle time' run on, every IC on a quality IC tester before soldering them in place.

If the original uses IC sockets then you don't need to do the above step, but many early machines simply had the ICs directly fitted.

Then there is the issue of getting the period connectors and the matching interconnecting cable.

In my view, it matters NOT, if you built the machine in the 1970's or in 2016, IF and that's a big IF you use 100% original period parts that were all made in 1970's; then that machine is not a reproduction.

There are rare times via a Garage finds, you come across a sealed 'original kit' of some item, from the way back in the past.

So if you assemble and complete that 'kit' using only the tools and skills of the artisans from that past time, then I feel its no different than making it within a year or 2 of its manufacture.

Now one point I would say, is from when I built my first computer kit, a Motorola MEK6800 D2 in 1976, to now 2016, when I am about to build a brand newly designed surface mount component, Analog Computer, my electronic skills over that 40 years have vastly improved. ( well I still occasionally make the stupid mistakes).

So now if I was to get a brand new, un-assembled complete Motorola MEK6800 D2 kit, it would NOT end up being built exactly the same as my first one, 40 years back; as I cannot UNLEARN my computer and electronics skills.

So in this D2 kit example, if I could get a un-assembled D2 kit now and built it up, WOULD it be an original or a replica ???

I'm not sure one could ever get everything 100% vintage with a new Mark-8 build. Somewhere along the way you're going to get tripped up by modernity. I've had all kinds of problems with vintage caps, ICs with my TVT. It's taken 3 years just to get 80% of the way there. I even ran into an issue with the *wood* for the sides.. the wood they would have used for a prototype-similar case in the 70s was very different from wood harvested today. And on it goes. :)

falter
September 28th, 2016, 03:58 PM
I have some original unbuilt osi boards , and I made replica's so I could leave the original boards unbuilt and could play around without worry of wrecking them http://www.vcfed.org/forum/showthread.php?45450-SuperBoard-II-to-build-or-not-to-build/page2&highlight=superboard

I would do the same here (if I could afford to buy them)

Well, I think unbuilt OSI boards are probably less rare than unbuilt Mark-8 boards. I wouldn't have a problem building one if I needed to. But these.. I don't know.. I was all gung ho until I started thinking about it. What it would mean to alter their originality by putting an iron to them, or worse if I damaged them? The thing about an unbuilt kit is once built, it can't be unbuilt. There is value to having examples of things as they arrived to people back in the day. These probably belong in a museum. They're just too perfect to screw with, if they're real. Kind of hoping they aren't. :)

inotarobot
September 28th, 2016, 04:00 PM
Well, suppose you rolled up the circuitry into a small FPGA?

If it works the same way, what's the difference?

The difference is the LOOK, the FEEL, and even the smell of the old parts overheating, when you have accidentally install the wrong value component or installed one ic in back the front.

And finally there is the MOST important difference.

Most of today's suitably skilled electronics/computer younger foke, can build a system like a D2 in small FPGA, and get it running properly.

BUT how many of them have the skills and for that matter, the patience, to put together a back planed, 6 board system like say an Altar or S100 or EXORcisor Bus machine, and get all the board address and on board jumpers correct; then create or patch an O/S to suit. ?

I doubt more than 0.5% of the current trained computer techs /engineers could. Or more correctly would want to. "My Grandpar did it so whay should I ?" they say.

falter
September 28th, 2016, 04:06 PM
That was one other thing I meant to ask. *Are* there Mark-8 examples out there known to have *not* been built from the kit boards, but rather via the plans? I know the artwork was available in the plans so it was theoretically possible to make your own boards.. wondering if anyone did.

mwillegal
September 28th, 2016, 05:00 PM
For something as scarce as a Mark 8, you are better off waiting for someone to make or making a reproduction, yourself, than mucking with original boards. Since the holes are not plated, and the artwork was published, you could etch your own PCBs at home, and it would be lot less expensive than buying a set of PCBs.

regards,
Mike Willegal


Yeah. I'm kinda torn.. if those are original they are pretty close to pristine.. how many pristine as-sold original pcbs are you likely to encounter? Feels like anything you did to them would take their value down, especially if you screwed anything up. And even if you built them into a machine it'd never be an 'original' Mark-8.. it'd be more like a new Delorean. Plus some original parts just aren't available as I've learned with my TVT project.

Hmmm. I had this debate when I saw an original set of TVT boards that I narrowly missed. $38 though. But even then, I was torn between 'oh it'll be an original TVT' and 'no.. it'll still be mostly a repro from 2016 and now no one will be able to see how the original boards came from factory.

Chuck(G)
September 28th, 2016, 05:22 PM
That was one other thing I meant to ask. *Are* there Mark-8 examples out there known to have *not* been built from the kit boards, but rather via the plans? I know the artwork was available in the plans so it was theoretically possible to make your own boards.. wondering if anyone did.

The Mark 8 isn't that dense, circuit board-wise. I can well imagine that there were those who simply used wire-wrap or even point-to-point wiring. PCBs were a little more involved back in those days. If you worked for a manufacturer, you were used to tape, india ink wite-out on mylar sheets, often photo-reduced before going to PCB. Back then, I had a Kepro PCB kit and made a few PCBs of my own, but between drilling and layout, it was easier to wire-wrap.

MarsMan2020
September 28th, 2016, 06:24 PM
There are high-quality scans of the original PCB artwork here - http://www.bytecollector.com/m8_docs.htm

Image to Gerber converter here - https://imagetogerber.wordpress.com/

I don't think these boards would be hard to etch at home, but having Gerber files would be nicer.

If I were doing it I'd use something like OpenCV to centroid a couple of the round pads and scale/rotate the scanned images to match the actual part pitches and be orthogonal and lined up through the 2 layers.

Drop a hole at each of the required locations in a Gerber editor and you're done.

falter
September 28th, 2016, 08:10 PM
I'm comfortable with using the original artwork.. provided I can get the correct dimensions. I had the thought to do something like what you suggested with my TVT project and then do soldermask and so on but I decided I wanted to stay true to the original design, even if it's not perfect. The ASCII encoder worked beautifully (thanks to Chuck and Dave et. al for their help!!). It just looks right.

My problem with doing a Mark-8, apart from getting scale right and the challenge of a dual sided board, is I really do not like the look of modern PCBs at all. For the TVT this isn't so critical since a wide array of PCB material was used by hobbyists.. but a Mark-8 has a certain look.

mwillegal
September 29th, 2016, 02:58 AM
Older PCB material was flourescent, which screws up modern PCB processes. Hence, it is not made anymore.
Finding enough old stock for a Mark 8 type project is likely to be challenging.

The artwork was published back in the day and is available on the web. Why not use what the original do it yourselfer had available. Dimensions are easily determined.

Regards,
Mike Willegal



I'm comfortable with using the original artwork.. provided I can get the correct dimensions. I had the thought to do something like what you suggested with my TVT project and then do soldermask and so on but I decided I wanted to stay true to the original design, even if it's not perfect. The ASCII encoder worked beautifully (thanks to Chuck and Dave et. al for their help!!). It just looks right.

My problem with doing a Mark-8, apart from getting scale right and the challenge of a dual sided board, is I really do not like the look of modern PCBs at all. For the TVT this isn't so critical since a wide array of PCB material was used by hobbyists.. but a Mark-8 has a certain look.

Moonferret
September 29th, 2016, 03:48 AM
For something as scarce as a Mark 8, you are better off waiting for someone to make or making a reproduction, yourself, than mucking with original boards.

Hmm, is this a hint as to what your next project will be? :)

Corey986
September 29th, 2016, 07:47 AM
33492
How old do you think this Scelbi board is?

Al Kossow
September 29th, 2016, 08:49 AM
33492
How old do you think this Scelbi board is?

not that old.

modern electroyltic and trim pots are too dark. old ones are lighter blue

extremely good registration of the holes.

You had to be working with a really good board house to get them to digitize your holes
with that accuracy. Back then, as Chuck said, you were using mylar and tape, and when you sent
your board out, they had to digitize the drill locations. Gerbers didn't exist in the hobby space.

at the price of gold back then, you'd gold plate even proto runs.


the other think looking at it is there is no water mark on the FR4

I'm assuming this is dyed modern stock?

hornbetw
September 29th, 2016, 09:16 AM
The consistent spacing/alignment of the tracks & components, plus the consistent "sharpness" of the 45 deg angles of the traces, show the board was laid out on a CAD system. It looks good, too good for a hand layout. Most of the old boards had a few "meandering" tracks on them.

Tom

Chuck(G)
September 29th, 2016, 09:26 AM
Agreed, though I do like the inclusion of the "top hat" diode. And any hobbyist of the time would have sprung for IC sockets.

glitch
September 29th, 2016, 10:43 AM
I'm really surprised there isn't like, 'vintage copper clad PCB stock' out there somewhere. There's virtually everything else but that. Ought to be a business there for someone enterprising enough.

I've got enough to just about tile a basketball court in the warehouse. We've used it for toner transfer boards, I used a fair bit of it in college for projects. It requires a cleaning to remove the copper oxide before you use it. If this is something people want, I can see about listing some of it. We've also got a 48" metal shear in the machine shop, so it can be sheared to desired size for e.g. OSI boards or Mark-8 boards. There's double-sided, single-sided, "clear" FR4 (natural resin color I guess, it's translucentish whiteish...you know, like old boards!), green FR4, and phenolic.

falter
September 29th, 2016, 02:19 PM
I would be all over that. Let me know if any of it becomew available for sale. I'm looking for single and double side with tht vintage green color like the original TVT and those Mark-8 boards.

snuci
September 29th, 2016, 02:27 PM
Count me in if someone makes a set of Mark 8 boards. I was considering those on eBay myself.

mwillegal
September 29th, 2016, 06:25 PM
Most original SCELBI PCBs did not have gold plated edge connectors and though the practice varied with builder, for the most part, only the SRAM boards were socketed.

A good image of an original SCELBI CPU board can be found on my SCELBI registry page.

I'm not so sure about doing a Mark-8 board set. I was wiring a SCELBI keyboard interface enclosure tonight and said to myself, "my next project is not going to require so much point to point wiring!". Between the two SCELBI chassis I have built and the peripherals and cabling, I must have soldered something like 500 point to point wires.



Regards,
Mike Willegal

Corey986
September 29th, 2016, 06:34 PM
not that old.

modern electroyltic and trim pots are too dark. old ones are lighter blue
I'm assuming this is dyed modern stock?

Yes dyed modern stock. BTW the Pots on two of my 1977 Persci drives are the same colour and I know they are original.

KC9UDX
September 29th, 2016, 07:31 PM
I have a machine from 1976 with the same colour ten turn pots.

falter
October 4th, 2016, 08:38 PM
I won!

I'm pretty shocked. I did a ton of research, talked to the seller about how these boards were found -- everything seems to point to them being legit. If they are, considering a Rev 0 Apple II board just sold recently for $1300 (and a later one with the dark green slots, no less), I gotta be pretty happy making off with these for $1500 and change! I thought I'd have to go to war against museums and so on. This leaves me with extra to pay for the ASR-33 I just bought. :)

I figure maybe it was a combination of things that held the price and competition down:

1) Hard to verify 100% they're real - maybe they aren't
2) They're bare boards
3) People probably weren't actively searching for them, given how rare they are, so they missed the auction?
4) People got psyched out by the anticipation of a bidding war
5) Maybe Mark-8s aren't as valuable as I assumed despite their rarity?

If they are legit, I'm very excited. This is the first truly ultra-rare thing I've gotten for my collection. I also have to hope the seller honors his end of it, I think he was expecting a much larger jackpot.

But assuming it goes through, and they are legit -- dare I build them? (once I get a h-ll of a lot more experience with other projects) I had a thought to just build a Mark-8 clone and display the real boards with them.

Ultimately these will end up in a museum, if I have my way. But not before I've enjoyed them a while first. :)

glitch
October 5th, 2016, 04:33 AM
Congrats! You've certainly picked up some really cool stuff this year :)

You might consider sending these off to be scanned into Gerber files. I've worked with Mile High Test (now a division of Gardien) out of Colorado for scanning before, they do an excellent job and the service is prompt. Grant Stockley recommended them -- that's who he used for his Altair clone boards. You can specify no clean-up and they'll leave crooked traces and off-center holes alone. See my OSI 495 prototype board as an example:

http://www.glitchwrks.com/2016/04/22/cloning-the-495

If you want help with the project, let me know -- I'd be glad to lend a hand! Personally I went ahead and added a reference to my website in the copper layer on the OSI 495 boards, just so no one ever tries to pass one off as an original old board (there are plenty of other differences, but I figure the URL makes it obvious even to eBay buyers).

Gerbers can then be used to manufacture boards at a commercial board house (I know of a few suppliers who will give you an OK price on a board that large), or you can print transparencies/toner transfer and do boards at home.

Moonferret
October 5th, 2016, 05:42 AM
Excellent!! Certainly went for less than I thought they would. Think you got a pretty good deal there.

I like Glitch's suggestion too. You could get a set of boards made up which you could use to build a machine and keep the originals in their untouched state. Of course, if you wanted to make several boards, I'd certainly buy a set :)

Cheers,
Dave

falter
October 5th, 2016, 06:35 AM
Thanks glitch! Yeah, it's been a pretty good year. I still want an Altair, but I might postpone that a while longer to absorb this (and preserve my marriage).

I'm curious though -- with the original artwork available (ie. via Bryan's site), why not just use that to reproduce boards, exactly as would have been in done in the day by those who declined to purchase them? I had some suggest to me doing something similar with my TVT boards, but I went with using the original artwork because to me that got closer to what the hobbyist would have done, and represented the actual work of the article's author. I'm a little nervous too about sending these away and risking them being lost. Doesn't happen very often but I tend to be lucky that way. :)

I've gotten a lot of conflicting advice re: build or not build. There is equal merit to both camps. The not build camp cites the rarity of the boards and the extreme rarity of finding unbuilt ones. They say worry about collector value. I kind of lean that way - thinking about how I cringed when that guy opened a brand new Atari 2600 box or that unopened Apple IIc. That's something you can't undo.

On the other hand, I recall an interview with a guy who owned a real Shelby Cobra and actually drove it. When asked why, he said 'what's the point of having it if you don't drive it?' Hmm. :)



Congrats! You've certainly picked up some really cool stuff this year :)

You might consider sending these off to be scanned into Gerber files. I've worked with Mile High Test (now a division of Gardien) out of Colorado for scanning before, they do an excellent job and the service is prompt. Grant Stockley recommended them -- that's who he used for his Altair clone boards. You can specify no clean-up and they'll leave crooked traces and off-center holes alone. See my OSI 495 prototype board as an example:

http://www.glitchwrks.com/2016/04/22/cloning-the-495

If you want help with the project, let me know -- I'd be glad to lend a hand! Personally I went ahead and added a reference to my website in the copper layer on the OSI 495 boards, just so no one ever tries to pass one off as an original old board (there are plenty of other differences, but I figure the URL makes it obvious even to eBay buyers).

Gerbers can then be used to manufacture boards at a commercial board house (I know of a few suppliers who will give you an OK price on a board that large), or you can print transparencies/toner transfer and do boards at home.

falter
October 5th, 2016, 06:36 AM
Yeah I think we'll definitely be making some clone boards, one way or another. I actually had double sided copper clad on order and was going to attempt the challenge of making these boards straight off the plans. Probably will still do it for fun, but I will give serious thought to sending the originals for a scan to make it easier for others.


Excellent!! Certainly went for less than I thought they would. Think you got a pretty good deal there.

I like Glitch's suggestion too. You could get a set of boards made up which you could use to build a machine and keep the originals in their untouched state. Of course, if you wanted to make several boards, I'd certainly buy a set :)

Cheers,
Dave

dfnr2
October 5th, 2016, 09:15 AM
Congratulations. I honestly had not imagined them going above $1k. Shows how much I know.

Everyone's perspective is a bit different on the hobby. Personally, I appreciate the machines for the stories behind them, and the ingenuity and grit underlying their design. All that comes to life when they are built, maintained, repaired. If I keep a non-working machine, it's only because I entertain some hope of fixing it down the line. I wouldn't hesitate to build this up, but personally would have accurate Gerbers prepared from them first. For me, the dollar value of the boards is just the hurdle that must be overcome to obtain them. A set of unbuilt boards would be an itch to be scratched.


Thanks glitch! Yeah, it's been a pretty good year. I still want an Altair, but I might postpone that a while longer to absorb this (and preserve my marriage).

I'm curious though -- with the original artwork available (ie. via Bryan's site), why not just use that to reproduce boards, exactly as would have been in done in the day by those who declined to purchase them? I had some suggest to me doing something similar with my TVT boards, but I went with using the original artwork because to me that got closer to what the hobbyist would have done, and represented the actual work of the article's author. I'm a little nervous too about sending these away and risking them being lost. Doesn't happen very often but I tend to be lucky that way. :)


The benefit of using Mile High, or some other service, in additon to the actual scanning, is in the laborious cleanup of the data before and after converting to gerbers, drill hole placement, etc. Further, the art in the magazine has already been reproduced and may be subject to small distortions, and was then scanned to a TIFF, with further distortion by the scanning process. Mile high uses a special calibrated super high-res scanner that minimizes disortion, so their scan from the PCB will be truer to the original art, and so will the gerbers. Also, it's possible the production version may incorporate some fixes not present in the magazine artwork, with is likely the earliest Rev. Having true gerbers made from the PCB is a kind of archiving.

If you are at all willing to have Gerbers made from them, I would be willing to pitch in on having these boards scanned (which will not be cheap by any stretch). Perhaps we can get several members to sponsor the cost of scanning 1 board each, then free the gerbers for all. That would be sweet.

At any rate, I am looking forward to seeing you build up these boards (they deserve to be built), and sharing your experience with us here on the forum!

Dave

glitch
October 5th, 2016, 09:42 AM
Yeah, Gerber scanning was my way of getting over using a really rare OSI board. I could've laid out something more modern with probably equal effort but far less expense, but I like working with the old single-sided OSI 495 copies anyway.

W.R.T. Mile High Test's process, you actually get pretty clean Gerbers, not a bunch of basically Gerber "pixels". Even a huge board like the OSI 495 didn't crash KiCad or OrCAD on import. And as already mentioned, it's much higher resolution than consumer devices can produce -- IIRC 4000 DPI. I was going to photocopy-clone the OSI 495 until I figured out how much time the *drilling* would take, it was much easier to have the board fab'ed in China, and being as how I was able to bill the hours on day-job contract work rather than running a drill press, I probably saved money.

If you're thinking about having them scanned, just heavily insure the package going out and send it via a fast service. Carriers seem to take better care of items with high insured value/high priority status. I'd imagine you can cover the cost of scanning by selling a few sets of blank boards. That's what I ended up doing with the OSI 495 boards. I haven't released the Gerbers since I'm not sure of the legality behind it, OSI is defunct but it may be that someone still owns the IP and would think about a troll defense. Shouldn't be a problem with the Mark 8, since the layouts were released to the public years ago.

Chuck(G)
October 5th, 2016, 09:49 AM
Everyone's perspective is a bit different on the hobby. Personally, I appreciate the machines for the stories behind them, and the ingenuity and grit underlying their design.

Indeed--and that's why I value engineering documents, schematics and design reports above real hardware. Paper allows me to crawl into the designer's head to a certain extent and discern his intentions and thought process. A hunk of iron is, well, just a hunk of iron. Lights blink, stuff happens, but it tells me nothing about the human intellect behind the item. Sadly, it's the ephemera that tends to hit the dumpster when an organization is closing up shop.

Let's not forget that all this glorious hardware came out of the squishy wet stuff between someone's ears.

dfnr2
October 5th, 2016, 10:04 AM
Indeed--and that's why I value engineering documents, schematics and design reports above real hardware. Paper allows me to crawl into the designer's head to a certain extent and discern his intentions and thought process. A hunk of iron is, well, just a hunk of iron. Lights blink, stuff happens, but it tells me nothing about the human intellect behind the item. Sadly, it's the ephemera that tends to hit the dumpster when an organization is closing up shop.

Let's not forget that all this glorious hardware came out of the squishy wet stuff between someone's ears.

+1, Well said!

falter
October 5th, 2016, 04:40 PM
Really? :)

I mean, I'm an auction addict and I watch stuff on ebay all the time. I'm beginning to think ebay is just too disjointed and random to be a reliable indicator. I see run of the mill ELFs (Netronics, etc) go for $500-900 all day long. Even high serial number A2S1s pick up $800+ sometimes. I thought original, untouched boards from the 3rd (?) microprocessor based computer ever offered to the public, examples of which exist in fewer numbers than Apple Is, would almost certainly go above $2000. I mean, how many 'computers' put you in a club where there's only about a dozen or so other owners including museums? I didn't think bare boards would outprice an original completed system, but I would have thought original Mark-8 stuff would be way, way up there. Makes no sense to me. But whatever, I'll take it. The seller, as it turns out, was ecstatic. He's going to sell the OSI boards next. I am all over that backplane board, just FYI. :)

I'm kind of partial to the idea of building them.. seems kind of a sad fate to end up just sitting on display somewhere doing nothing. But once built.. that's it. Unless there are more out there somewhere waiting to be found by accident.

I would be more than happy to cover the cost of scanning the boards -- I realize this is a one-off opportunity. What do they typically charge? And I would not expect anything in return for sharing them. My only hesitation is that I have lost stuff with couriers on occasion. I don't care about the money -- but losing the boards themselves.. ugh that would suck!! But you know, I'm almost tempted to have the seller ship them to Mile High right off, and then have Mile High boot them up here. They're gonna have to be in transit anyway.

My other thought was to just do what I did with the TVT plans and try with the published ones. See how close I get and compare to original. I was told the original produced boards have errors on them that were uncorrected from the plans. I could ask Bryan to ask Jon about the differences. Definitely there would be some distortions -- I ran smack into that with the TVT and really had to play with the scale to get it to come out right. And it's not perfect at all.


Congratulations. I honestly had not imagined them going above $1k. Shows how much I know.

Everyone's perspective is a bit different on the hobby. Personally, I appreciate the machines for the stories behind them, and the ingenuity and grit underlying their design. All that comes to life when they are built, maintained, repaired. If I keep a non-working machine, it's only because I entertain some hope of fixing it down the line. I wouldn't hesitate to build this up, but personally would have accurate Gerbers prepared from them first. For me, the dollar value of the boards is just the hurdle that must be overcome to obtain them. A set of unbuilt boards would be an itch to be scratched.



The benefit of using Mile High, or some other service, in additon to the actual scanning, is in the laborious cleanup of the data before and after converting to gerbers, drill hole placement, etc. Further, the art in the magazine has already been reproduced and may be subject to small distortions, and was then scanned to a TIFF, with further distortion by the scanning process. Mile high uses a special calibrated super high-res scanner that minimizes disortion, so their scan from the PCB will be truer to the original art, and so will the gerbers. Also, it's possible the production version may incorporate some fixes not present in the magazine artwork, with is likely the earliest Rev. Having true gerbers made from the PCB is a kind of archiving.

If you are at all willing to have Gerbers made from them, I would be willing to pitch in on having these boards scanned (which will not be cheap by any stretch). Perhaps we can get several members to sponsor the cost of scanning 1 board each, then free the gerbers for all. That would be sweet.

At any rate, I am looking forward to seeing you build up these boards (they deserve to be built), and sharing your experience with us here on the forum!

Dave

mwillegal
October 6th, 2016, 07:35 AM
This is a super rare item that is relatively easily reproduced. Over time, I changed my thinking for these sort of super rare items, which used to I used to think should be made to run. For something so rare, I would NOT build them, but use them to make clone PCBs. Building a clone will give you nearly the same experience without risking damaging or reducing the value of the original boards.

If you decide to build it up, no matter how you approach it, it will never be quite the same as a unit built in the mid-70's.

regards,
Mike Willegal

Corey986
October 6th, 2016, 08:09 AM
This is a super rare item that is relatively easily reproduced. Over time, I changed my thinking for these sort of super rare items, which used to I used to think should be made to run. For something so rare, I would NOT build them, but use them to make clone PCBs. Building a clone will give you nearly the same experience without risking damaging or reducing the value of the original boards.

If you decide to build it up, no matter how you approach it, it will never be quite the same as a unit built in the mid-70's.

regards,
Mike Willegal

100% agree with Mike, I would also be willing to go in on a set of scanned boards from these. Don't bother with trying to reproduce them from scans of the original articles when you have access to the real boards to send out and be converted to gerbers.

Chuck(G)
October 6th, 2016, 08:09 AM
If you decide to build it up, no matter how you approach it, it will never be quite the same as a unit built in the mid-70's.

How so, Mike? One can still find vintage TTL and other components if one looks hard enough. 60-40 solder is still the same stuff it was in 1950. PCBs can still be fabbed the hard way. So what's different? Heck, people still build relay computers.

One thing is that is different is the experience of the observer. Something on the order of "You can never hear Bach's music the way Bach did--there's been too much water under your bridge--you've heard music that Bach never did and it has affected your perception".

I don't think that you can fix that.

Corey986
October 6th, 2016, 08:16 AM
Well the difference is in the bring up process.

In the 70's when building a machine, the odds on having a chip tester and logic analyzer for the home builder was not really an option. This makes it much easier to bring up the machine. When I built my Scelbi replica, I used a roll of solder from the 1970's, chips from 1974 or earlier and even wired the chassis with NOS wire from 1973. However when troubleshooting a problem with the 8008 CPU and the clock, I was able to put a 16 channel logic analyzer on the CPU and record the entire process of a program running to figure out my issue. This would have been something a hobbyist in the 70's would have needed to use a scope on and a lot cigarettes/beer.

Cheers,
Corey

falter
October 6th, 2016, 08:28 AM
Good points all around. Rest assured nothing will happen to these boards for the forseeable future.

It's such a tough call. Building a Mark-8 from original boards is something I doubt anyone else will even have the opportunity to do in the forseeable future. But building them means they are never original again. I went through this whole debate with my MDS keyboard for my TVT project - I was hesitant about painting it to match the prototype because it couldn't be undone. But in that case I already knew another had surfaced months before and I figured probably one would surface again. Further, an MDS keyboard without the machine is useless anyway. Mark-8 boards though.. I don't know how likely we're going to come across unbuilt ones like these again. Probably not any time soon. But then I've said that and been proven wrong repeatedly about other rare items.

I'm actually finding replication to be preferable to acquiring originals. I have a few very vintage, original systems in my collection, but I feel like I have no rapport with them because I never had the experience of building them. Plus, I get cagey about firing them up because sooner or later something's going to blow.

There are four RAM boards among the Mark-8 boards I've bought.. maybe I build a replica and build just one of those to give it a little cheekiness.

I was reading on Bryan's site that an estimated 7500 plans were sold and that Jon Titus estimated that 1000-2000 machines were built. But he also said only 400 board sets were sold. So presumably.. somebody out there built these without original boards.. wondering if any examples of totally homebrew Mark-8s have surfaced anywhere? And would they be worth what one with a commercial board set would be?

mwillegal
October 6th, 2016, 08:45 AM
Even if you manage to find all circa 1974 components, including hardware (screws, nuts, wire, connectors and whatever), if you went to sell it, it would be a Mark 8, built with vintage components in 2016. It will never be the same as a Mark 8 built in 1974, because it doesn't have the history of a unit built in 1974. Just the way I would look at it.

regards,
Mike Willegal


How so, Mike? One can still find vintage TTL and other components if one looks hard enough. 60-40 solder is still the same stuff it was in 1950. PCBs can still be fabbed the hard way. So what's different? Heck, people still build relay computers.

One thing is that is different is the experience of the observer. Something on the order of "You can never hear Bach's music the way Bach did--there's been too much water under your bridge--you've heard music that Bach never did and it has affected your perception".

I don't think that you can fix that.

Chuck(G)
October 6th, 2016, 09:06 AM
In a blind A-B test, a board assembled from vintage components yesterday would be very difficult to tell from a board built 40 years ago. So no objective difference.

But then, I'd consider that a skillful clone of the Mona Lisa would be just the same as the original, objectively.

I'm not a big believer in ghosts. I've seen too much fakery with people claiming history that didn't happen--especially with "this widget belonged to someone long dead" claims.

falter
October 6th, 2016, 12:21 PM
True enough. Plus.. we've established the street value of original Mark 8 boards is $1500. So I'm not sure I'm all that worried about collector value, unlike with an Apple I. Frankly I've never cared about provenance.. if I can get essentially the same machine and it wasn't owned by X so it's worth less.. bonus!

Chuck(G)
October 6th, 2016, 01:38 PM
Now, an original MCM/70 might be worth having...

snuci
October 6th, 2016, 02:02 PM
Now, an original MCM/70 might be worth having...

I've been looking for one of those since I started collecting mainly because of it's Canadian lineage.

dfnr2
October 6th, 2016, 02:21 PM
I would be more than happy to cover the cost of scanning the boards -- I realize this is a one-off opportunity. What do they typically charge? And I would not expect anything in return for sharing them. My only hesitation is that I have lost stuff with couriers on occasion. I don't care about the money -- but losing the boards themselves.. ugh that would suck!! But you know, I'm almost tempted to have the seller ship them to Mile High right off, and then have Mile High boot them up here. They're gonna have to be in transit anyway.


I would call the Mile High Test guys and talk to them. I got a quote for about $250 for an OSI 505 board, which I thought was a bargain. The OSI board is about double the size of the Mark 8 boards, so perhaps we're looking at $150 per board. For sure, with a near-perfect reproduction, you will be a lot less tempted to build the original set, philosophy aside.

Dave

Chuck(G)
October 6th, 2016, 02:21 PM
Anent the MCM/70...

I find the thing fascinating, not only because of its early introduction (with the 8008 CPU), but that it implemented a dialect of APL. Pretty impressive. Might also qualify as the first portable computer, since it did have an option for battery power.

falter
October 6th, 2016, 03:48 PM
I would call the Mile High Test guys and talk to them. I got a quote for about $250 for an OSI 505 board, which I thought was a bargain. The OSI board is about double the size of the Mark 8 boards, so perhaps we're looking at $150 per board. For sure, with a near-perfect reproduction, you will be a lot less tempted to build the original set, philosophy aside.

Dave

Oh. I didn't realize it was quite that much.. :) Yeah I'd be happy to do that but it'll be a while.. other priorities. :)

falter
October 6th, 2016, 04:07 PM
I'm thinking more about this debate regarding collector value, and just playing Devil's Advocate here.

We've postulated that the boards should not be built into a system because it will harm value. But will it?

It's not like the boards are worth $10000. At $1500, I have a hard time believing anything short of destroying them would reduce their value. If they're festooned with vintage parts and assembled into a working computer, I gotta think the end value is more than the boards alone.

Further, from a collecting standpoint, I've been on both ends of the transaction. There's only a handful of Mark-8s known to exist. They aren't something you can reliably wait on to appear on ebay. The last one to sell was almost a decade ago. If you're a collector living in the now and a period correct 2016 built Mark-8 comes up with original boards -- maybe you value it less than a 1974 one, but who's to say there isn't a bidding war regardless? Buyers have to weigh originality vs. the likelihood of seeing another, better one come up.

And then there's the experience of building a Mark-8 with original boards. What's that worth? Should I care more about what some stranger thinks about value or should I place the emphasis on my own enjoyment?

Not saying I'm leaning this way.. just talking it through. :)

inotarobot
October 6th, 2016, 04:13 PM
+1, Well said!

I'll second that !!!

mwillegal
October 7th, 2016, 06:56 AM
Just one final comment from me.

How we value something today, may not be the same as how something is valued tomorrow. Apple 1's are a classic example of that, going from 20K to 500,000K in a couple of years. Once you build them or alter them, you can't entirely undo it.

In the end, I respect differing opinions and decisions on how to care for these rare items.


regards,
Mike Willegal

qubitz
October 7th, 2016, 07:08 AM
Congrats on the win! I've chased many vintage computing items and unbuilt Mark-8 boards was near the top of my list for a long time. I was tempted to bid until the last minute (checking the countdown while at a concert) and I actually had a snipe set at a much higher number. I ended up deciding not to bid, but already I was going through the same emotions as you: would I build it, or keep them as is?

In the end I agree with Mike 100%. How we value these things today will certainly be different from how we value them in the future. I have no question that we will value unbuilt Mark-8 boards much greater in the future than we do today.

Cameron

falter
October 7th, 2016, 07:49 AM
I had no plan at all to buy Mark-8 stuff. I assumed it would be completely beyond my reach, and that was if it ever appeared. I set my bid at $6500 but I was sure it would go crazy. I think a lot of people probably were thinking like we did -- ie. 'what do we do with these?'. It really provokes deep philosophical questions. I almost was kind of hoping I lost so I'd be off the hook. :)




Congrats on the win! I've chased many vintage computing items and unbuilt Mark-8 boards was near the top of my list for a long time. I was tempted to bid until the last minute (checking the countdown while at a concert) and I actually had a snipe set at a much higher number. I ended up deciding not to bid, but already I was going through the same emotions as you: would I build it, or keep them as is?

In the end I agree with Mike 100%. How we value these things today will certainly be different from how we value them in the future. I have no question that we will value unbuilt Mark-8 boards much greater in the future than we do today.

Cameron

8008guy
October 12th, 2016, 09:16 PM
Personally I'd build the machine. Unlike the apple 1 the mark 8 is a hobbiest built machine. The parts came from where ever you got them. So there is no right or wrong... Mine is built on later production boards. I also used a molex back plane. Wiring the boards together is a big mistake. It makes trouble shooting a big pain. The boards are laid out to accommodate the connectors.

I also designed a video card that plugs into mine, 16 by 32 characters. It was based on some early articles in byte.

Btw Brian was in town this summer and we got together with Jon for lunch. We had a nice trip down memory lane.

If you have questions I can probably get Jon on the forum to answer them directly.

Len

falter
October 13th, 2016, 09:56 AM
Thanks!

If I did it, and I'm not sure I have the wherewithal to, I'd want the date codes to be right at least. I've yet to see a 1975 or earlier 8263 show up anywhere. I've also been looking around for my clone project and cannot find correct vintage 1101 RAMs anywhere. So those are some hurdles. Plus, I just got photos from the forwarding service I use (this was a US only auction) and the boards have their original Techniques bags! Hard to think of messing with that. But we'll see.

falter
November 9th, 2016, 05:19 PM
Boards are here!

There was a bit of a delay as I use a freight forwarder from the US and wanted to bundle it with other stuff I bought to save on shipping. Pretty awesome having these pieces of history in my hands. Definitely the crown jewels of my collection. Quite a bit smaller than I visualized. No doubt at all that they are legitimate.

I *think* they are untouched.. but they do have some scratches and every now and again it kind of looks like solder remnants are there. Maybe plating? One board has clearly been repaired.. not sure if that was factory to correct mistakes.

Anyway I've posted my Drive link so you get the high detail. I'll work on getting them scanned asap.

https://drive.google.com/folderview?id=0B4pq0-BHd2x6U1RKREJPTDFZQ00

Roland Huisman
November 13th, 2016, 01:36 AM
I always wondered if my Mark 8 boards were original or were home made back then.
The machine is 100% sure built in the early seventies.
http://technischmuseum.nl/devices/Mark%208/Mark%208.html

But if I compare them to the ebay pictures to my boards they
might be original. Never really cared because it's old anyway.
But my PCB's also show the NVF logo as well...

Regards, Roland

Corey986
November 13th, 2016, 03:28 AM
I always wondered if my Mark 8 boards were original or were home made back then.
The machine is 100% sure built in the early seventies.
http://technischmuseum.nl/devices/Mark%208/Mark%208.html


I would say original, even if it was a home etched set, the Mark-8 isn't like a Scelbi or Altair. They gave out plans and wanted people to make their own boards. Ordering a set of complete boards was just a convenience. Boards were also manufactured by sources other than those associated with the magazine article. Think of the Mark-8 as the first "Open-source" computer.

Cheers,
Corey

falter
November 13th, 2016, 09:04 AM
Cool! It's interesting, seeing the ICs someone in Europe used.

I'm curious about all the extra 7401s on the multiplexer board and what they do.

Chuck(G)
November 13th, 2016, 09:21 AM
It's curious that while there's interest in early 8008 efforts, there's none that I can detect for Datapoint 2200 systems.

Want to know where the ISA and "little endian" conventions for the 8008 came from? look no further (http://bitsavers.informatik.uni-stuttgart.de/pdf/datapoint/2200/2200_Reference_Manual.pdf). What's remarkable is that many of the opcodes (binary encoding) are the same as those of the 8008.

Roland Huisman
November 13th, 2016, 11:05 AM
Cool! It's interesting, seeing the ICs someone in Europe used.
Hmm I've never thought of that difference between US and Europe...


I'm curious about all the extra 7401s on the multiplexer board and what they do.
Well I have the documentation of the machine in some of the move boxes. So I can tell you later on.
The machine was built by two brothers (Jaap and Henk Berkhoudt). The machine was highly expanded by them.
It has a HUGE amount of 14KB memory (the blue boards). They hated the toggle switches and made the push button panel.

34302

It has also an EPROM in it with a tape loader, a casette interface, a 'graphics' card to connect to a monitor etc...

The machine doesn't look like the 'standard' mark 8... But I really love it like it is.

Regards, Roland

KC9UDX
November 13th, 2016, 12:35 PM
It's curious that while there's interest in early 8008 efforts, there's none that I can detect for Datapoint 2200 systems.

Want to know where the ISA and "little endian" conventions for the 8008 came from? look no further (http://bitsavers.informatik.uni-stuttgart.de/pdf/datapoint/2200/2200_Reference_Manual.pdf). What's remarkable is that many of the opcodes (binary encoding) are the same as those of the 8008.

I think we're supposed to believe that the Datapoint 2200 is totally irrelevant, so that the 4004 can be considered the ultimate progenitor of the processors used in Windows X machines.

Chuck(G)
November 13th, 2016, 02:12 PM
...but the 4004 was developed in parallel with the 8008; it wasn't a progenitor of anything, save other 4 bit MCUs. At least that's the way I think of it.

KC9UDX
November 13th, 2016, 02:35 PM
That's the way I think of it too. I don't even think it influenced any other 4 bit uPs.

Yet others insist that the 4004 is hugely significant as we wouldn't have the latest "x86”, and hence, any useful computers, without it.