Monday, September 15, 2014

Time to dust off that list....

This year has been a pretty good rescue year, just with the earlier PERQ rescue. Having just one like that in a year, would mean you could chalk it up as a good year.

So, having something else impressive happen would be epic... 

The birth of the next  epic rescue began when I was asked by a fellow Atlanta Historical Computing Society (AHCS) member, Kyle, if I would like to come help him with some repair work on a Sun 1/100.  He has been helping out a local collector who has, well a  massive collection of vintage equipment... at least that's what I'd say is the best way to describe it.   

To give you some perspective, to get to the area where the Sun 1 is (in a climate controlled building) you have to walk around two Cray 1's.    And within visual sight is an Apple 1...  so it's a fun place to visit. :-)  

So, since the Sun 1/100 is the #1 computer on my bucket list (to own and use), I jumped at the chance.   A picture of the actual unit is below:

The issue Kyle was working on was how to know if the unit was in good enough condition to get booting.  Since I'm the AHCS member with the oldest Sun knowledge, he thought of me.  So I brought over the oldest Sun Field Engineering Handbook references I had, and a collection of other old Sun docs that I have.  

(As an aside, I had kept a file of Sun 1 info, including some contact info for collectors that had Sun 1's at some time... I hadn't really chased much of the info down.  I was saving it for when I had an opportunity to either work on, or if I was very lucky, own one...and this opportunity got me started chasing the info down..)

Kyle had had time to examine the power supplies, and those all looked good. But there was an odd set of boards installed.  Two video cards was the major tip off that something was odd.   I had only seen one Sun 1 in person before, and that was at the Smithsonian.  The mouse matched what I had seen (and what I'd seen in other reference pictures).   Sun didn't actually initially provide a mouse (according to various references I have... it wasn't supported SW wise until Sun OS 1.0) 

FYI, the documentation on the mouse (and some old Sun docs in general) is a tad tricky to sort out.   This is because you have to be careful as to "which" Sun documents you're reading.  Is it related to 
the "Stanford University Network -- SUN"  university papers, or one of the two initial efforts to commercialize the "SUN" design, (The design was initially documented in the The SUN Workstation paper, and involved some of the same players that created the two companies). The two companies were VLSI Systems Inc., which focused on OEM relationships and Sun Microsystems... the two companies had amost the same founders... which also overlap the individuals from Stanford. It seems the Master's project initially spawned VLSI Systems Inc  which then somehow became Sun Microsystems.  
Stanford  SUN Workstation paper authors : 
Founders listed of VLSI Systems Inc :
Founders listed of Sun Microsystems :
It's not 100% clear to me at present what the exact legal relationships between Sun Microsystems and VLSI Systems Inc, or how they were legally related to Stanford ( and the SUN design from there).  One thing that does differentiate is that the Stanford and VLSI docs are usually written with "sun" in an all capital "SUN" format,  whereas Sun Microsystems used a mixed case, "Sun".   

So, anyway, in a doc labeled : "SUN-1 System Reference Manual" (from July 27, 1982, and listed as (c) Sun Microsystems Inc.) the following info is provided:

3.6.5. Mouse
       The back panel Mouse connector provides a hardware interface compatible with currently used mouse devices. Software support will not be provided for this interface until the release of the forthcoming Sun mouse product. If you need to use this interface before then, contact Sun for assistance.

This agrees with most of the additional supporting documentation I've been able to find.  However, Sun DID offer a mouse later (you can find information on it in the unpacking documentation -- 899-1109-01H System Managers Manual for the Sun Workstation Models 100U / 150U Rel 1.1 March 1984 )

4. Lift out the rubber and styrofoam packing material (the fit is fairly tight; try rocking it out). At this point you may find a mouse and mouse pad tucked along the side of the monitor; if you ordered an 84-MByte disk subsystem, these will be packed with it instead.

Additionally, anecdotal information shows that the Hawley Mark II Mouse was one available at the time and common at the time.   This mouse has been seen in pictures of Sun 1 units on display at museums, but to date, I've not found any definitive documentation from Sun indicating it was anyway officially connected to them.

Keyboard wise, the situation is even murkier. The Sun docs provide conflicting information on how many different types of keyboards were available for the Sun 1.   In the SUN-1 System Reference Manual" (from July 27, 1982, and listed as (c) Sun Microsystems Inc.) it lists two keyboard options : 

However, in the later editions there is no reference to a second type of keyboard.  There is reference to a "Klunker Keyboard" in some early Sun Field Engineering Handbooks.

So, at present, it's not clear whether or not there was one or two options here. However in my current research, I've yet to find any evidence of any but the keytronic models.   For the Micro Switch to have been an option, it would have to been a parallel model.  The earliest Micro Switch keyboards I've found for any Sun have been serial (as the one in the above picture...).  

 The keyboard types after the Type 1 are better documented.  
The current list I have is:
  • Sun Type 2. Key Tronic capacitive switches.
  • Sun Type 3. OAK linear switches.
  • Sun Type 4. Key Tronic capacitive switches.
  • Sun Type 5. Standard Fujitsu rubber dome internals.
  • Sun Type 5c 
  • Sun Type 6. Standard Fujitsu rubber dome design in Sun's layout.
  • Sun Type 7. Standard Fujitsu rubber dome design in Sun's layout.

Lastly, as I mentioned before, when we opened up the machine, we found that it had two video cards. And that didn't make much sense.   Later model Sun's did support multiple video cards, but since the Sun 1/100 was an all-in-one, it seems that this wouldn't be a standard configuration.

So, after our first couple of days working on the box, we had realized a few things:  1) That it's not the correct keyboard, and 2) that the previous owner had put in an odd set of boards. 

At this point, I decided it was time to start researching in earnest, including reaching out to the contacts that I had stashed away... it was time to see what information I could pull together.

To be continued....

Friday, September 5, 2014

A PERQ in time... saves... 1

The first rescue for 2014 was a pretty spectacular one.   

During the prep for VCF SE 2.0 (which is local for me), we got a contact from Andrew who was local and had a PERQ computer that he was considering displaying and potentially offering for sale.   I posted back and encouraged him to display, since I was interested in seeing one first hand. 

See, I am familiar with the name, and to some degree the machine... I didn't earn the nickname "cesspool of useless information" for nothing :-) .  However, most of my information wasn't first hand but rather came from some passionate postings (perhaps you could even call them "musings") discussing them.   

These postings I had read on one of the computer rescue lists that exist. (it's a great one and discusses rescuing just about everything. It started with a focus on Sun Microsystems Computers : )  It's one of many communities that exist where you can find that you're not alone in rescuing computers.   Like any other community, it can be a surprisingly encouraging place to many folks who feel disconnected and alone.   I believe people are built to look for places to experience life together with others, and this is definitely one where that happens.

The  VCF SE show planning continued and ultimately the PERQ didn't get displayed, but the owner did join the Atlanta Historical Computing Society (AHCS) mailing list. 

A month or so after the show, he posted saying that he was moving.  His wife had given him a year to do "something" with the PERQ, and since it had sat largely idle, it was time to try to find it a good home.  He asked that folks contact him after 10 am the next day.    So at 10:00:01, I gave him a call.  He had received an email expressing interest, but told me that the suggested time windows for pick up wouldn't work for that fellow.  He needed someone to come over in the middle of the week.  I told him that was something I could do, as long as it didn't need to be "today" and that we could work out a day.  

As I was making arrangements to pick up the machine, I knew that ultimately I wouldn't be keeping this machine, even though it is a very rare one.  I simply didn't have the passion that I knew others did for this machine.   So I posted to the rescue mailing list where I had read about them, and asked to be reminded who was (or who were) the appropriate people.   

Later that night, I had my reply, and almost caused Skeezics Boondogle (aka Chris) to have a heart attack by telling him that it was his if he wanted it. (The details of his overall reaction will be the topic of a future post's a story in and of itself.)  His initial response started with :
Okay... hyperventilating... take a breath... 

And my reply included this :

So, here's the deal, if you want it, it's yours... yes, you read that correctly.  

Resulted in  :

It's a good thing I'm on beta blockers.  I think my heart might have exploded otherwise.  Joy!  Oh frabjous day!
Holy cow... 

So a few days later, I convinced a co-worker (Jeff) to take a "road trip" with me to rescue the machine.  It was a first for him.  To him the machine is "antique", since it was manufactured before he was born.

We headed off with general direction, address and GPS to direct us in. (the general location was Dallas GA, which "looked" closer than it ultimately was)

As we started out, we grabbed some lunch and saw this license plate, which we took as a good omen.  Almost PERQ.... and notice the Dallas :-)

So as we proceeded we called to try to make sure they'd be there when we arrived.  We learned that we'd have to wait a short time (though, we did get to learn first hand that one of their neighbors had a talented child, and looked like he was turning into a good 3 point shooter :-) )  

When they arrived home in a matching pair of Smart Cars, they graciously invited us in and we found the PERQ up a short flight of stairs, ready for us to cart off.   (Though Jeff and I did have the same thought running thru our heads, their front yard was close to a 40% grade.... and we had NO idea how heavy a PERQ actually was... we were thinking... 1980's.... all Steel... going to be a beast...but in actuality it's got a lot of plastic and isn't nearly as heavy as it looks... honestly...)  And this is when we realized that it was at least a PERQ 1 chassis (at least from the pictures we had seen on-line..)

The owner shared the story of how he had purchased this at a garage sale in an area that was very close to where we had started from.  He had actually fit it into one of the Smart Cars, it just barely fit.  He told us the owner had used them in his business, which he believed was an architectural firm.  He also told us that as he understood it, the previous owner either "had" or "did" own a bunch more of them.    That got us even more excited, and we wondered if Skeezics would survive learning that even more of the rare machines might be available.   

So, Jeff and I got started and carried the PERQ down the stairs, down the front steps, and finally huffed and puffed it up the front yard / hill and set it down on the tailgate of my truck.    During the trip, at least one piece of a step broke off underneath me... not due to weight, but due to the repairs was a rental house. A few minutes to wrap it in quilts/blankets, put a tarp over top, bungee it in and we were off.    After retracting our trek (and trying to avoid Atlanta traffic) and we safely made it back home.   There we could appreciate the rescue in much more detail.

With some better lighting, you can see that the unit is excellent cosmetic condition.  One of the pieces that rarely survives unscathed is the front bottom grill, since the front panel is a single very large piece of plastic.  After removing it, you see that the machine is simply contains 3 boards (in the previous picture).  

There are a number of fairly unique aspects of this machine. First, it's CPU is actually not implemented by a single microprocessor, but a "board" of TTL logic.   The three boards in the system are color coded so you don't place them in the wrong slot (a great idea that I've not seen on other bus based systems) White:CPU, Green:Memory, Red:IOB , Yellow:OIO.

Since I knew I was going to take the machine apart for shipping, I called Skeezics and asked him about details of taking it apart.  After simply taking a couple of screws to remove the front panel the boards have simple latches that help disconnect them. 

If you look the picture with the cover off above, on the left side is a 14" Shugart hard disk (that's belt driven). The boards are on the right.  Pictures of the individual boards are below.

CPU Board 

Memory board (taken in two pictures to try to provide some more details)

IOB (taken in two pictures to try to provide some more details)

Skeezics was also glad to hear that the rescue also included some software, all on 8" floppies (It's fairly "sizable" amount, at least for a PERQ... the SW is as rare or rarer than the HW)

Of course the unit also included the keyboard, and digitizer tablet (no mouse came with the PERQ, it was tablet and puck).   But the rescue didn't include a puck.  The PERQ keyboard has my favorite ACTUAL key, the OOPS key. 

And with all of the PERQs, you got a monitor that matched the video card (it's all memory mapped, and portrait and landscape versions required different boards)   This one had someone at least take it apart, and they didn't quite put it back together correctly.  It shouldn't have a gap big enough for your finger... (all the bezels are plastic... basically if it's dark brown, it's plastic on the PERQ). 

On the side of the monitor, you can see the original PERQ logo, and the original three rivers logo.  For those of you who don't know it, Three Rivers is a reference to the Pittsburgh area (where they were located/founded initially) Behind the monitor in this picture you can see part of the tablet (the light tan square corner).

As a footnote, after I got the machine safely home, I was able to get some further information from Andrew about the person he acquired the machine from, who's name is Vic.   Andrew had the company name as well... ASSI (it's not quite where we live, but further north in Calhoun.)  I did some web searching, found out the company still existed, and indeed Vic was the owner and still around.  I gave him a call and explained my situation, and inquired as to whether or not he had any other PERQ HW or SW.  After his expression of disbelief that I had found him and was asking about PERQs (which he did recall fondly),  he told me sadly that the unit I rescued was the last one that they had.  After chatting with him, I looked around the company web site and found this on the history page. This may actually be a picture of the rescued unit from 1985.

Thursday, September 4, 2014

Welcome to the Vintage Computer Expo...

When I mention to folks that I "rescue" vintage computers it always seems to result in a confused look (esp. from non-computer folks ). The conversation  then normally staggers along as I try to explain in detail what that mean,  (because engineers do that :-) ) and that frequently generates an even more glazed look in response.   So, I fall back on my quick explanation that the experience is something akin to restoring vintage automobiles. Socially that allows the conversation to continue, and If I start with this, I've learned that I can even prevent the glazed look, awkward response and uncomfortable pause. However, in both cases, while providing "a common reference" this explanation has never really satisfied doesn't capture the experience very well and has never been something that I've ever been happy with.

So, in an effort to explain / educate / rationalize the craziness that I get involved in, I've begun this blog.   Here you will see stories of computers I've "rescued", stories of others who have a similar passion...what's involved in their restoration, some of the amazing engineering re-discovered, and the simple joy of bringing them back to life.  

You'll also see that just because a computer is old doesn't mean it's useless, just like people.