H316 workspace now as GIT repository, website still alive!The website might seemed to be abandoned during the last years - but that is not true. I just did not do too much with my H316 and friends. It's all still there.
During the last days I have created a gitweb and git http access to my public repositories including all my H316 stuff which might be a good starting point for anyone wanting to compile and run software for the machine.
Here is the GIT repository!
516 running a bit, CCG's H316 now with 12K of coreI have tried to get the 516 working. I have tried only two of the four memory modules (8k words each) so far. There is at least one bad sense amplifier and one other ic. That's quite bad because I see no chance to get the right ICs to fix that. I'll put the missing ICs onto my "wanted" list.
The CCG's H316 has been fitted with 8k core which I kept from the H316R. I wired all together and powered up. Looks like the memories are working. I just don't have the time to get an ASR near the machine and boot CMT4 as there is no paper tape reader interface in the machine. But I'm quite confident that the computer will run fine with it's whopping 12k words of core memory with partial American military history.
Big changes, new computer!Something magic happened: I've got a DDP-516! I traded it for the H316R. So this machine is now gone.
I tried to get the DDP-516 into my office/workshop in Kiel. But it was impossible for two persons to get the empty rack without doors and side panels into the first floor. So the rack is waiting in the basement while the computer went upstairs in pieces.
Some new stuff and website reworkI've not been too busy with my H316 in the last time. That might explain the long time since my last entry on this page.
I bought a second CalComp 565 on eBay, works very nice. Has the higher resolution and is blue outside :-)
I also improved my plotting library. Nothing of it currently online here.
A new video has been put onto Youtube!
Attached Plotter and uploaded some filesI have recently attached a Calcomp 565 plotter to my H316. I made a custom interface and wrote some simple driver routines for it. It works fine. Some pictures and a movie can be found in the gallery.
Further I uploaded some more PDF files to my document section. Perhaps theres is something you are looking for...
H316R - ever heard about it?Hi folks,
here comes a new story of luck, organisational talents and happyness :-)
Perhaps some of you (the few readers of this page) have noticed the message concerning a "h316b" in Fort Worth, Texas in my guestbook. After having read that, I got in touch with the guy.
The guy is Andre D'Amours from Texas. I didn't know him before. Andre and some friends who visit the Nortex surplus store (must be a great place) on a regularly base have found the machine somewhere beneath a pile of other equipment. They pulled out the four memory modules for their personal decorative collection of old computer parts. The stunning price for all four core modules was $2.50 each. So they paid $10.00 for all memory modules. When I got in contact with Andre, I begged them to put the modules back in place and get the rest of the machine. So he and some friends went to Nortex and bought the rest of the computer. For another stunning $15.00. I recapitulate: Total hardware cost was $25.00 - bon marché en Etats Units...! I sent Andre the money and a bit more for his help. He also arranged professional packaging the machine for transport. I paid a huge amount of money for a 450% perfect package. And another huge sum for TNT to pick up the package and deliver it to my collection's address in Alfeld, Germany.
So, now the machine is here. It turned out that the machine is no H316B: It is a H316R, a ruggedized military version of the H316. I had never heard about that before!
The TNT service is not cheap but worth the money. They have great support and very friendly people working for them. I can clearly recommend them!
And, of course, I owe many many thanks to Andre, the nice man in Texas!!!
He had some work with the machine and did not put it on eBay even after I had explained him that the eBay value of the system would be much more than $50.00.
The crux of the matter is that I, as a student with very limited financial power, would not have been able to acquire the system for a usual eBay price.
Again: Andre, thank you very, very much!
And now the machine itself...
There are already some pics of it on a new gallery page.
The first thing I did was to pull out the memory modules and the CPU cards. I put the memory into my H316 and tried it out. There was a strange bootloader inside. I manually noted down the first 100(oct) locations of memory and keyed in a simple routine that tested the last 4K of memory if they are all zeroes. They were. So I entered my key-in loader, bootet my PAL-AP replacement and made a dump of all memory. Afterwards, I bootet CCT4, the CPU test. Immediate success. Fine.
The next thing was to try the CPU cards. After putting them in, the ASR output did not work. I looked at the ASR card and saw the broken IC. Replaced the ASR card with my old one for now and tried out CCT4. Worked flawlessly. The CPU works! So most of the most important pieces of the machine are in working condition.
In this moment, while I write this test, my H316 chassis with the new CPU and memory in it perform CMT5, the core memory test without errors. So I seem to have got a system in great condition :-)
Today I plan to power up the new chassis, with all endangered devices disconnected. Perhaps the switching (!) PSU works....
There are some strange peripherals in the chassis, I don't know what they could have been for. Have to further investigate that.
That's it for now, more to come soon.
H316 on the roadI have had a course in processor design. Everybody had to introduce one technology or processor design. So my choice was not too difficult :-)
H316 is perfect for someone who missed the test like me:
No cache, no stack, no superscalar architecture, no branch prediction and so on. All ugly and hard to explain things are missing in this great computer.
My presentation was held on January 16th in Bremen. Only half of the audience was asleep, so it must have been one of the more interesting presentations in that series :-)
Some images to proof that I really moved my H316 over 180km can be found in a small gallery.
I delayed this news entry because I did not want to write about that before the machine is reinstalled in its original place and up and running again.... It sits in its rack again. No damages, no additional scratches, not much rain on it. Everything went smoothly...
Printing 2006I have got a new printer!
Mr. Albert Richheimer offered me his working 1979 Data Products 2230 drum printer. Wow!
I fetched it from Switzerland in the last week before Christmas where I spent a very nice evening with Albert and his wife Irene. They served "confit de canard" (duck legs in tin cans!), very nice french food.
The next morning we disassembled the printer and put it into my Mother's car. It's a bit heavy, but with two persons and a wooden brush shaft, we managed to load the car.
The customs at the border were very friendly (!). They just asked me "What's that?" and I answered "An old Printer!", no further questions. I even did not have to show my ID.
At home I had to ask my friend Jörg to help me with the machine. Finally I have reassembled it in the location where it stands now. And it still works!
But there were some interface issues....
The printer itself has a so-called Data Products Interface, a simple parallel latched (not clocked) data interface with ready and acknowledge signals etc. Albert had a small box constructed which does some decent magic to convert from a standard parallel Centronics port to the DPI.
But it did not on my H316 and PC: The interface lost every second character. After thinking a bit, I came to the conclusion that the DPI's busy signal comes too late so that the host has already transmitted a second character.
The box is made with WireWrap and still has a lot of free space in it. So I took my wrapping tool and added another 47221 monoflop, triggered by the strobe signal and ORed the output with the busy signal from the old interface to form a new busy signal. Now the busy signal on the Centronics interface is forced high immediately after strobe for about 50usec disregarding the printer's state.
My modification worked perfectly on the first try. Now I can print. Via PC oder H316. I just plug in the printer and send data. Fine!
Off course, I know that it's not too complicated to interface the DPI to the H316 IO bus. But as you might know, I already have made a Centronics interface for the H316, so I can use that and it's drivers.
I have just taken a few pictures of the machine, which can be found in a new gallery!
And a small (not so the file...!) video sequence, showing the machine working. This can be found here.
Many thanks again to Albert Richheimer for the printer, the nice evening and the cookies! I enjoyed my stay and will visit you again next time I pass Switzerland!
And what about my guestbook?!?Nobody writes into my newly installed guestbook :-(
Perhaps there are no visitors? Or all of them dislike the site? I don't know.
Just waiting with my empty guest book.....
Some PSU repairToday I have spent the night to repair one of my H316 power supply units. It seems to be in working condition and the machine connected to it behaves well.
I have made a handfull pictures. You can find them in the gallery.
A trip to UKLast weekend I spent two days with Adrian Wise (http://www.series16.adrianwise.co.uk) who has two Honeywell machines, a H316 and a DDP-516.
I brought him one of the famous GNT 4604 paper tape reader and punch units so he is now able to read and punch paper tape in a convenient way. We also read in some of his paper tapes. Now there is a lunar lander game written in FORTRAN IV around. The best find for me was the magtape I/O library. I am still missing the magtape utilities but if I need to, I will go without them or write some own routines.
Adrian's machines are only partially working at the moment:
The newly-acquired DDP-516 has a working teletype interface and nearly stable working 8K core memory (needs some fine tuning because of a replaced sense amplifier). The CPU has some faults but these don't frighten Adrian. The 1000 cps electrostatic paper tape reader isn't corrently working as well. I think it needs some service after twenty years of service.
Adrian's H316 also has a slight problem with the paper tape reader. The modified ASR interface (9600 baud, full duplex) works but seems to run a bit too slow so it doesn't accept data on full speed i.e. when the next byte arrives immediately after the stop bits. A little gap between the sent out bytes has been our workaround for this weekend. The core memory (8k) works well and passes the tests after slightly correcting the core voltage to its original value (stack at 14.75 volts). Adrian has a very cool selfmade 8k semiconductor memory for this machine which didn't work properly this weekend. But the CPU itself passed all tests.
All in all he has two very nice machines which will be perfectly working again.
Despite the fact that there was a bit trouble with the hardware we still had some fun playing around and discovering old software.
I spent two very nice days with him and his family who all are very likeable. I'm pretty sure that this will not be our last meeting in real life.
Thanks again for everything!
New stuff around!First I'd like to send you to the page where I collect the material I've received from Leonard Kleinrock from UCLA.
Further, Mr. Kirchhoff has sent me first version of the movie he shot at my site.
I have put it up for download! See in the download section (Hachti and H316 in action).
And the first set of Mr. Kirchhoff's photographs is also online. You can find it here.
Enjoy the stuff. Hope you like it :-)
PhotoshhotingOn monday Prof. Dr. Jens Kirchhoff from the Computerkabinett Göttingen visited my collection to take a lot of photos and fixate some H316 action on video. We had a nice time together, about 4+ hours in a small room with 1.5kW of illumination and the Honeywell :-) On a sunny and very warm spring day.
Mr. Kirchhoff is going to cut the video material to a nice presentation of what's going on my site. The video won't contain too many technical details, it rather tries to catch the atmosphere of the heavy metal. When the video has been finished I will put a download onto this website.
In the meantime you can visit the latest gallery where you find some randomly taken images and some impressions from the shooting on monday evening.
Repair beginsToday I have started to investigate my vacuum switch problems. See the appropriate gallery for details.
Good news is that I have a chance to get the missing electronics.
A Magtape unit!Yesterday I have picked up the hardest and heaviest piece of "computer hardware" I've ever owned:
An end-1960's Honeywell "Very Low Cost (VLC)" 1/2" magnetic tape drive.
Its restoration will be a bigger task so the drive has its own page.
Some photos of the haul and my first steps with the machine can be found in the gallery.
A second machine!I did not believe ever to see another Honeywell H316. A few day ago I found a hint on the Website of the Computerkabinett Göttingen an interesting collection of good old stuff.
When I went there I found the H316 CPU. The power supply is missing. No backplane wirelists at all. But the machine has a tape controller for the Honeywell 7-track tape drive I will get from Göttingen, too. The tape drive and controller will be mine while the CPU and documentation are on loan for the next two years. But I am the only one around who knows to operate that machine. And the only one with a spare PSU in stock.
Perhaps I will still find more interesting stuff belonging to that machine. Will keep you up to date. Take a look!
Happy new year!All my Honeywell stuff moved house. The H316 is now located in Alfeld (near Hanover). Now I have some more room to build up the system. Everything still works perfectly, nothing has broken during the transport (puh...).
Still have to clean out a lot of unused/not any more needed stuff, then everything will be fine.
A first few pics from the new location are already online. You can see them here as part of my gallery.
What to do with my Honeywell now?I have reached a point of "don't know what to do with the equipment". My Honeywell runs flawlessly and does everything I want. Only issue still to fix is the power supply. The voltage sequencing and the overvoltage protection don't work correctly.
At the moment I can:
Who has any ideas? I want your proposals!
It's printing!I have built a centronics printer interface for my Honeywell. It works on address '0 and takes 8 bit parallel data. It also can initially send cr/lf and the addressing is completely reprogrammable. I used the eproms because I had enough of them...
I have made two pictures: front back
Happy new year!Thanks to Adrian Wise I have my paper tape punch up and running. My Honeywell ist getting more complete...
The new website is online. Hope you like it.