Recent Posts

Hi Marv, you are right. I found an old picture of the inside of a 441 at ITEC Giessen.
I was probably thinking of a 395 or 400. I found it in this PDF.

http://www.thecorememory.com/Giessen_School_1965.pdf

At least the module 3 from above seems to be from a 441.

I remember things like 12AS7s…6SN7s…thyratrons…massive cooling fans…165-volt anode voltages...anodes, cathodes and grids. I don't remember any transistors in there. But I could be wrong.

Hi Marv,
I'm not the only one, who thinks the Computronic was transistorized.

http://www.ncr.org.uk/441%20computronic.html

Hi Wally,
I think you might have the Comp confused with some other system. Your timeframe is correct but the Comp was anything but leading-edge. It used old technology, nothing like the showcase EDP systems of the day.

The first NCR computer (315-100) I taught in the UK had a 10k memory (slabs) A slab was 12 bits which could consist of two alpha numeric characters or three numeric characters. I later taught the Century 100 which had 16k of memory (bytes). 
The laptop on which I am typing this message has 8Gb of RAM........how things have progressed!!!!!!

Hi Marv,
if I remember right, the Computronic was transistorized. But I'm not 100% sure.
It came on the market in 1959 and I think by that time transistors were used.
It is the same time, the NCR 304 was introduced and it was the industry's first
all-transistorized, or solid-state computer for general business in the market.
Regards Wally.

Erik,

I know this is an old topic but, if it still matters, that module you have is from an NCR class 441 Computronic. You can see the part number in one of the views. The Comp was an accounting machine based on a class 31, but with the added capability to multiply electronically. Very spiffy for its time.

The tubes, likely manufactured for NCR, all contained the company logo. Something about being "computer grade" if I recall.

Hi! RogerHallett,

Yes, exactly, the TMX is named by the BAS project team of Mr. David Nanda in E&M Dayton. Originally, the project team of Mr. Jerry Butler in E&M Columbia had modified the VERSADOS realtime OS, which is provided by Motorola,  with attaching several functions for the small controller based on MC68010 MPU which was different from the tower 1632 office systems. The BAS project manager tried to evaluate the availability of the Xenix that Mr. Bill Gates renamed the UNIX. However, E&M Dayton imported the OS architecture of VERSADOS from E&M Columbia into the BAS 5000 system, in order to install this modified VERSADOS through attaching several midleware functions for supporting the application friendliness to assist the transaction processing on the Branch Automation Systems. They called this application friendliness as the TMX (Transaction Management Executive). Originally the VERSADOS had been developed by the Motorola. When I attended the VERSADOS lecture room during a week by the Motorola instructor in the NCR Dayton Sugar camp in April, 1981. I was very surprised by the weight functionality of the VERSADOS comparing with the MTS 8 bits Intel MPU. I suspect that the Motorola strategy would be based on the simple procedure program like the 68xx MPU based application under the multi-tasking MC68010 MPU systems. However, on the other hand, the NCR system engineers might looked at this flexible OS like the versatile B4 executive on the powerful Century 300 processor, without detailed simulation of the performance of the TMX based MC68010 processing.
Although I am not sure the availability of LAN (might be OMNI-Net of NCR, 2Mbps transfer rate during the time), the BAS controller supported the IHDLC high speed communication and also the analog type communication for NCR C-270 and ATM etc.
I suspect that this IHDLC high speed lines would make high rate interruption for MC68010 MPU.
I experienced to discuss much time to make sure the performance of total volume processing for the banking transaction in the Japanese financial market. The product management had been very significant in the BAS development, I studied.
Best regards,
Katsuhiko

Dear RogerHallett,

  You are welcome. That's my presure to communicate one another.
I am remembering my old memory with NCR Dayton peoples to share the target throu our jobs.
Good memorial stories.
I am praying healthy lives of the BAS 5000 project all staffs.
I appreciate your notes, Dear Roger.
Best regards,
Katsuhiko

TMX (Transaction Management Executive) was used in Financial Banking Systems as the OS for the NCR BAS 5000 series online (branch automation) products. TMX allowed programs to be written in 'C' with calls to Fortran (and maybe other languages - not sure) if required, then all compiled into run-time. TMX was mutli-tasking, multi-thread if I recall correctly (some help please) and ran on the Motorolla platform in either the Tower box or a smaller custom built box. These controllers had a variety of propriety LAN comms protocols to connect the various banking peripherals (special keyboards, screens, swipe-card readers, passbook printers, back-office printers), and either IBM 3270 or SNA for Wide Area connection to the host. I am sure anyone from E&M Dayton would be able to add value to this post.

Prior to the BAS 5000, the MTS system had its own OS running on the Intel chip. Australia had some of the world's largest MTS and BAS networks.