THE CORE MEMORY FORUM

NCR Computers of the 20th Century => Mainframes => Topic started by: Somebody on May 15, 2006, 12:06:20 am



Title: NCR 315
Post by: Somebody on May 15, 2006, 12:06:20 am
Hello,

Has anyone worked with a 315? If so, I'd like to discuss various aspects of it. I never worked with a 315 myself, but as a teenager I worked at a place that had 2 of them. Seeing them in action is what made me decide to become a programmer. Unfortunately, I never got a chance to work with one, my experience is with IBM mainframes. I thought the 315 had the coolest looking console. The 315 has always been special to me because, had I never seen one, I may never have decided to enter the data processing field. Note I used the correct term for our profession, not the insipid "information technology" term  ...   ; ^ )         


Title: Re: NCR 315
Post by: Aleksandrs Guba on May 16, 2006, 09:21:04 pm
Hello, Somebody!

This forum just launched and therefore its popularity is far from the popularity of the site www.thecorememory.com, so please be patient...
You may enjoy 315 related photos present at the site. If you want, I may suggest where to find more.


Title: Re: NCR 315
Post by: Somebody on May 29, 2006, 08:07:17 pm
Hello, Somebody!

This forum just launched and therefore its popularity is far from the popularity of the site www.thecorememory.com, so please be patient...
You may enjoy 315 related photos present at the site. If you want, I may suggest where to find more.



Hi. Thanks for the reply.  I have enjoyed thecorememory's photos. It is a great site! Please feel free to suggest other sites where 315 photos can be found. I would really like to find detailed close-ups of the console lights and keyboard. Also, the 315 tape handlers I saw had reels in a side-by-side horizontal configuration, like CDC and IBM tape drives, yet all the 315 tape handler photos show the reels stacked vertically, which I assume were older units derived from the 304 series.


Title: Re: NCR 315
Post by: Aleksandrs Guba on May 31, 2006, 07:52:27 am
Hi. Please try the following link: http://www.bitsavers.org/pdf/ncr/ncr-315/ and let me know your opinion.


Title: Re: NCR 315
Post by: andyjl on September 08, 2006, 06:29:15 am
Hi Somebody:
I worked at NCR's central  London office (St. Alphage House) as a computer operator on an NCR 315. I also taught myself to program on the machine, both in NEAT (National's Electronic Autocoding Technique) and COBOL. Once I knew how to program I transferred to NCR's computer education group in Baker Street (close to the mythical 22A Baker Street where Sherlock Holmes "lived.") I then went to work for NCR's Programming Information Systems (an unfortunate acryonym, PIS) where I we wrote drivers for the NCR operating system for the NCR Elliott 4100 series.

Sorry it took so long to respond to your May but I only discovered the site while searching for something related to the NCR 315.

What specifically did you need to know? (And I'll see if I can remember it).

Regards
Andy Johnson-Laird
Forensic Software Analyst
Portland, Oregon, USA.


Title: Re: NCR 315
Post by: LeeLabs on September 09, 2006, 11:29:07 am
Hi Somebody,

I worked on the NCR 315 for a number of years, as an operator and programmer.  I still remember some details of the peripherals like NOF, CRAP (CRAM) etc.  If you need any information let me know.

Lee Labs


Title: Re: NCR 315
Post by: PGA on September 19, 2006, 04:59:32 pm
I was a customer engineer with NCR from 1963 through 1978. I was the "on site" CE at American Greetings in Cleveland, OH. They had (2) 315s with CRAM units. They were replaced with Century 200s in the late 60s which I also supported. ;D


Title: Re: NCR 315
Post by: Somebody on September 23, 2006, 04:51:49 am
Hi. Please try the following link: http://www.bitsavers.org/pdf/ncr/ncr-315/ and let me know your opinion.


Thanks for the link. Very useful information!


Title: Re: NCR 315
Post by: Somebody on September 23, 2006, 05:16:12 am
I was a customer engineer with NCR from 1963 through 1978. I was the "on site" CE at American Greetings in Cleveland, OH. They had (2) 315s with CRAM units. They were replaced with Century 200s in the late 60s which I also supported.   ;D

Wow, small world! I worked at the May Company in downtown Cleveland, and I frequently made trips to the 7th floor computer room to get reports. If memory serves, they had two 315s. I also remember they had an optical reader for cash register tapes, a paper tape reader, two printers, and I forget how many tape handlers. They had 2 different tape handler models, some with the reels stacked vertically and others with the reels configured horizontally. No CRAM. I thought those 315s looked really cool, especially the console. Did you ever do any CE duty at the May Co.? The 315s really got me interested in computers, and I still have a few 315 manuals that the EDP manager gave me.

How would you rate the reliability of the 315 and it's peripherals compared to other 60's mainframes?     

 

   


Title: Re: NCR 315
Post by: n8eyh with OCD-WM42 of Hilse on October 01, 2006, 03:03:22 pm
Hi, all of the core project member,
Good morning and good evening from Japan.
I am looking for the person who knows the architecture of NCR 315 CPU.

I was a software engineer of NCR 615 and 605 several decade ago.
Recently, I noticed differences of the computer architecture between NCR 615 and INTEL 80X86.
Especially, the index register of NCR 615 was located in the core memory, though there is all registers in SRAM based arrays of INTEL CPU. I heard that the architecture of the INTEL CPU was based on the IBM 360 machine.
I am wondering whether the architecture of NCR 615 is original or IBM 1401 compatibilty.
Because we, the NCR Century's software engineer, designed the software with data-move-reduced algorism, in order to
execute the command procedure rapidly with reducing the data fetch operation.
I would like to know the difference of architecture between 615 and 315 CPU in order to confirm the NCR traditional CPU architecture.
Because Mr. Donald E. Eckdahl, who had been one of fathers of NCR computer, has documented that his original  computer, Northrop Maddida, had been evaluated by Dr. John von Neumann in March, 1950.
The Maddida machine was the first product of CRC Corp., that was acquired as the electronics division  by NCR in 1952.

Who knows the architecture of NCR 315?
I am waiting for your advise.

Best regards,
Katsuhiko


Title: Re: NCR 315
Post by: NEXUS on October 01, 2006, 06:20:19 pm
Hi Katsuhiko,

I have not worked with the 315 Systems – my NCR years started with the Century Series – so unfortunately I cannot contemplate on your question; however, looking at the unique Century 350 pictures you posted to the site today, I take the chance to ask something different that possibly you may remember:

I think, but I am not sure at all,   that the 615-3xx Systems, the largest of the Century Series,  were sold only to Sumitomo Bank in Japan and that their operating software was built to custom in Japan by the Bank people possibly with the assistance of NCR Japan or vice versa.
Is that right? Was B1/B2/B3 Operating system still employed?
Any memories on that?

Best Regards.


Title: Re: NCR 315
Post by: n8eyh with OCD-WM42 of Hilse on October 02, 2006, 11:47:13 am
Hi, NEXUS and members,

I am appreciated of your reply and question!
I also started my career as a freshman of the software engineer at the software support division of NCR Japan in April, 1969. Nexus, we would be able to share our knowledge related NCR Century systems.
Because I studyed the programming technique of the NEAT/3 and Level 2 language initially, even though my just one year older freshman started to learn how to make a program from the educational version of NCR 315 NEAT language.

Then I would like to inspect the difference of CPU architecture between 315 and 615 asap, in order to confirm the correctness of generality of the data-move-reduced algorism on NCR 615.

Well, let me answer matters that you indicated in your reply, as follows:

1) The marketability of NCR 615-3XX in Japan and features of them;
    Finally, NCR Japan installed them to more than ten banking users and retailing users of Japan, I remember.
   They have been featured by the TOX operating real-time transaction processing kernel which we the young people had
   developed under Mr. Ikuo Akiyama who invented the TOX.

2) The designing and programming of NCR 615-3XX computer systems;
   The TOX kernel was a micro kernel structure supported by system tasks and provided the task management,
   the interrupt management, the memory management and the intertask communication management with handling the queue.
   The system tasks configured and provided the services of the external communication drives, the external storage drives,
   the mulfunction monitor drive and the error recovery drive.
   NCR young people developed the kernel and the system tasks.
   Almost users developed the application tasks and the batch programs.
   Earlier of 1980s, NCR would provide the software package of the transaction-center procedure engine for user application
   program modules, though I left NCR in 1983 at the end of NCR 5000 BAS system planning.

3) The smart-ability of the TOX kernel;
   As I mentioned before, Mr. Ikuo Akiyama who invented the arrayed queue theory to reduce the processing overhead of
   tasks and designed the protect mode B1/B2/B3 combinational operation with handling BAR/LAR registers of CPU.
   In order to eliminate the meaningless overhead of task management, Mr. Hiromichi Ishikawa invented and implemented
   the passive task theory for the TOX operating system in January, 1975.
   Recently, I noticed that the ACM Ada group started to argue the efficiency of the passive Ada task, initiated by
   A.N. Habermann's article of CMU-CS-80-103 in 1980.
   I think that Mr. Hiromichi Ishikawa advanced the theory almost five years earlier.
   By the way, as you know well, the object-oriented software design is now familier since 1980s. However, I can not confirm
   the OO operating system yet, even though the TOX kernel had supported the run-time object-orientation in 1975.
   Because the arrayed queue method can support the abstract data types, the passive task interface which is protected by
   five BAR/LAR register sets can support the inheritance, and the SEND-ACT system_calls which transfer the event signal
   can support the polymorphism, even though we did not advertise the usefulness of the information hiding.

That's the story of the TOX project for NCR 615-3XX through my memory related to the powerful formation of our youth
for the curiosity.

Now, I am trying to transfer these technologies and ideas of the software engineering to young people of the next generation,
until the final day of my life.

Who knows the architecture of NCR 315?
I am waiting for advise.

Sincerely yours,
Katsuhiko


Title: Re: NCR 315
Post by: NEXUS on October 02, 2006, 01:52:03 pm
Hello Katsuhiko,

Your detailed memoir of the NCR Century 3xx systems software development in Japan answers my question, and is a very valuable contribution to this forum.

I wish success in your research regarding the NCR 315 architecture. 

Best Regards.


Title: Re: NCR 315
Post by: Grondak on November 10, 2006, 07:33:58 pm
I worked for NCR from 1969 to 1979 and spent most of that time operating and programming 315s.  I started as an operator and taught myself NEAT programming, and even developed some improvements to the CRAMEX operating system.  I still have a credit-card-sized NEAT coding instructions list, and a couple of NCR programming templates.  For many years I had a 315 7-track tape drive in my garage until my wife asked me politely but firmly to get rid of it!  I also used to have a full set of 315 manuals, but I foolishly loaned them to somebody and never saw them again.  Oh well.

Yes, the cosole of the 315 was impressive -- it always reminded me of Star Trek.  I wrote several assemby programs punched onto single boot cards that would do tricks with the lights by loading certain values into the registers.  I could spell people's names; display symbols; make the lights chase in circles; etc.

I used to love making bogus CSUPRBOOT cards that unsuspecting junior operators would try to use to start the system.  Mine would display disaster messages like PM#35s or print silly stuff like, "Don't touch my keyboard -- your fingers are cold!"  A favorite trick was using printer interrupt as a way of branching to a hidden routine in memory.  So long as the printer was online, any attempt at rebooting would continually branch to that routine.  The only solution was to take the printer offline, then reboot.

Ah, the memories!





Title: Re: NCR 315
Post by: Grondak on November 10, 2006, 07:37:03 pm
Since we're talking about the 315, does anyone remember the ONEPAS COBOL compiler?  It was developed primarily by Jinx Higley (whatever became of her?) and superceded the two-pass COBOL complier that generated NEAT source code which then needed to be compiled into object code.  The problem was it was too little, too late.  By the time ONEPAS was really usable, most people had migrated to Century systems.


Title: Re: NCR 315
Post by: zster on December 06, 2006, 08:33:52 pm
I worked as an operator ar a bank in Michigan in the mid 70's. I started out on a 315 then a  Century 201, then a Criterion 8500.

The computer room was over a parking garage and next to some very active railroad tracks. Every time a train would go by, the whole building would shake. Eventually one of the of the boards would jiggle loose causeing a "PE".  SOP was to go around with a rubber mallet and gently tap all of the boards to reseat them. It usually worked.

Durable heat generating hardware.

I don't get all nostalgic about crams though. They were evil and caused me much grief.

New Start 39234


-Mike


Title: Re: NCR RMC 315
Post by: RMC-LADPC on January 04, 2007, 09:54:29 am
I went to Dayton on the 315-3 with Martin Osborn and John Roland in the mid sixties.  We three returned and supported the 315-3 system at Pt. Mugu with 7-334s, 2-340s, 4-353-1,1-1442 IBM reader/punch, etc.  A year later, Martin relocated to Sparks, Nevada and John Roland was canned.
For the next three years I supported the site alone later replacing the old 315-3 system with a faster RMC-315-502 system with 8-333 (CDC) tape drives, 2-340s, 40K long rod memory, 4-353-1 crams, 1-1442 IBM reader/punch, etc.  The RMC system ran very well as did the 353-1 CRAMs once they were tuned up.  Cracked memory stacks, worming memory rods by off center solenoids was my main problem. When the site was replaced by an IBM 360-65, I transferred to the LA office where I worked at the NCR corporate Data Center with many great guys for several years.
At one time, we had four complete RMC systems, two with several 353-3 CRAMs, dozens of CDC 333 tape drives.  Later we added two 615-300 systems with the newerr 200 IPS CDC 635 tape drives and several rows of CDC 658 disk drives, CDC 647 printers, 315 emulators, etc.  A real fun place to work as lots of hardware to play with.  We even had a few Quantor microfiche units to blow our minds on.
We had several RMC 315s in the Los Angeles region all of which ran very hard for most of their productive lives. We also had our share of 615-100, 200s and 300s but I had the most fun during the 315 era.  The 615-300s were great machines but I hated the smaller systems as the NCR
dual disk drives were an absolute nightmare even making the 353-1 CRAMs look like heaven.
Head crashes galore and bad head wiring harnesses bring back late night nightmares.  The later
ceramic heads were not near as bad as the originals and actually ran well if cleaned on a regular basis.
The CDC 657 dual drawer disk drives were not too bad as long as one did not mind getting oil all over ones hands and clothes.  Ripped rubber diaphragms allowing oil to blow all over the blinking sensor lights was common place and a bitch to clean up.  Once the oil ran down into drive belt area,
life really started to suck again making a 353-1 CRAM look good to me.

Great times never to return.  I always considered myself fortunate to have experienced those days.

The Criter class was OK but not as fun to work on as the earlier CPUs as machine language writing was not to be as one was too distant to be effective.  (-01999) still comes to mind.

RVDaba


Title: Re: NCR 315
Post by: jswallow on January 10, 2007, 09:50:31 pm
somebody,
the thing great abpout the console was that it had an "Address Register", a group of lights that displayed the current address. this was good for program debbuging in a step by step mode. if I wrote a program for a customer, I would include a compare that met all possibility followed by GOTO 9669. IN the documentation i would say if a large "X" appeared on the console, don't call me, call hardware repair.

Jack Swallow 


Title: Re: NCR 315
Post by: paddyt on January 11, 2007, 04:15:51 am
Hello,

Has anyone worked with a 315? If so, I'd like to discuss various aspects of it. I never worked with a 315 myself, but as a teenager I worked at a place that had 2 of them. Seeing them in action is what made me decide to become a programmer. Unfortunately, I never got a chance to work with one, my experience is with IBM mainframes. I thought the 315 had the coolest looking console. The 315 has always been special to me because, had I never seen one, I may never have decided to enter the data processing field. Note I used the correct term for our profession, not the insipid "information technology" term  ...   ; ^ )         
Hi,

My first full time job, as an 18 year old in 1971, was doing rotating shifts as a 315 operator, and later in Data Centre operations management.
I spent a lot of hours fighting with tape drives, drum printers, card & paper tape readers etc.
Happy to answer any questions I can remember the answers to :-)


Title: Re: NCR 315
Post by: jswallow on January 14, 2007, 12:01:51 am
I installed a 315 on the 3rd floor of a building using a crane to put it through a large window. As the mainframe was comming off the truck it fell off the tailgate and landed flat on it's back. We tipped it up an lifted it to the 3rd floor. After all the pieces were hooked together the hardware guys turned it on and it ran just fine. No debug time was logged. We jokingly called Dayton and hinted they make that the first step in unpacking. That drop had reseated any boards that lossened during transit.
Jack Swallow


Title: Re: NCR 315
Post by: Jan Heek on March 14, 2007, 07:39:54 pm
Just discovered this site and felt, as a former NCR315-expert, I should join this community. I'm from The Netherlands and started as a programmer with NCR's Data Processing Center at The Hague. This was a high-tech building on the famous Laan van Meerdervoort, at the corner of the Zoutmanstraat, right across the now well known Peace Palace, the international Court of Law.
Started as a NEAT-assembler and BEST-programmer in 1969 and - after several NCR in house courses - became analist/programmer in 1971. In 1972 I changed my position to a company at Amsterdam, called Infonet Automation Services (not the US-based Infonet). They had two NCR315's for database publishing purposes (state-of-the-art encyclopedea production at that time). Because of my assembler skills I was one of the first programmers of the PDP-11 series (24,34,45,70) that came to The Netherlands. By 1978 the NCR315's were dismantled but I still possess the drum of one of the CRAM units. As for the early days....
In 1969 I found it a challange to have the NCR315 run at its highest possible speed by programming the read-commands for the CRAM units in such a way that they didn't release the CRAM-card unless I said so. For a major file-merge operation this resulted in all 4 CRAM-units having their vacume-pumps running at maximum power, nobody could stay in the computer-room because of the noise! Another trick I found out was the use of only 4 assembler instructions to read the papertape resulting in maximum speed of the reading process (in case you didn't read within a certain time-limit the tapebreak would be activated resulting in a slow reading). Those were the days .... your program on punched cards or papertape, scheduling a testrun at least 8 hours in advance and if you discovered a bug in your source, you had to wait another 8 hours for the next test run. Most of the testing was done in the evening or during the night for that reason. I nearly lived at that computer room!

Greetings from Jan Heek


Title: Re: NCR 315
Post by: RonS on June 21, 2007, 10:05:41 pm
What an interesting site – and such memories.
My wife and I met in 1961 when we were recruited by Elliot Automation in England to work on the NCR 315. Elliot had just secured the contract to build 315’s for the European market. We both knew nothing about computers and we cut our teeth on the 315 processor, line printer, tapes, CRAM, core memory etc.
In 1965 we immigrated to the USA and worked for NCR at Hawthorne, CA and then Rancho Bernardo until we retired- my wife in 1988 and me in 1989.
Just a few incidents I remember that you may (or may not) find interesting:

While we were working at Elliot’s the CRAM’s were generating a lot of interest and we received a visit from the Duke of Edinborough (the Queens husband) specifically to look at the CRAMs. He stayed a very short time because, we found out later, the noise of about 20 CRAM’s and about 20 line printers all running at the same time, was just too much for him.

In the USA a group of visitors came to view a mock up of the 315. Each of the cabinets was opened up for their inspection – and in the third one was a young engineer, sitting on a three legged stool- holding an abacus. The young engineer eventually became General Manager of the Rancho Bernardo plant.

We built a number of 315’s for the Sumitomo bank in Japan. I believe they were our biggest customer. At that time the Japanese engineers had a reputation for being very exacting and precise, especially in following instructions. We had to send them two large wooden crates of cables, and just for the hell of it, we had someone who spoke Japanese stencil “Important, Open other box first”, on each crate. Then we waited to see what confusion we had caused.  Instead, about a month later we received a polite, but formal request from Japan asking us to build a computer system that would be “more suitable for their environment”. With the request was a very detailed set of specifications and drawings for a computer system – made entirely of bamboo.

Ron





Title: Re: NCR 315
Post by: n8eyh with OCD-WM42 of Hilse on June 22, 2007, 05:28:53 am
Dear Ron,

  Good evening and good morning from Yokohama, Japan.
When I noticed the original sentences of yuor message in this morning, what does it mean the '#8211' keyword of your last sentence? Even though you have corrected last sentence now.

I could not find any such numerical string through google search.
Now, I am going to transfer your good memorial story to the FED member for 315RMCs in Sumitomo Bank since mid 1960s.
They will read this page of the web site.
If some one contact to you, will you please share each other the long story between southern CA and Osaka Japan since delivery?

Best regards,
Katsuhiko
a former software engineer


Title: Re: NCR 315
Post by: Aleksandrs Guba on July 01, 2007, 12:10:48 pm
Hi All,

Today an ‘Archive/Hardware Manuals’ page of the site have been considerably uploaded with NCR 315 manuals/instructions.

Enjoy!

Aleksandrs.


Title: Re: NCR 315
Post by: davew on July 11, 2007, 10:37:08 pm
Hi - Interesting to find this site and this discussion.

My name is Dave Williams and I was a service engineer on 315's between 1970 and 1975 in the UK. I started my working life as an apprentice electrican at Markham Colliery, Chesterfield. By the time I had completed by apprenticeship I had acquired an ONC in Mining and Electrical Engineering (and my working mate at the pit had just been killed there). I decided thereupon to get out of mining, and there were adverts in Wireless World offering CE jobs to anyone who had an ONC in anything. I got the job and started in May 1970 in the vaults of the Clydesdale Bank in Glasgow. As the trainee I seem to remember that my main job was oiling and adjusting a row of cheque sorters. After 3 months I was sent on the 315 training course to the Brent training centre on the North Circular Road (NCR) in London. This was where the Brent Shopping Centre is now. I spent 9 months in digs in Hendon and explored the delights of London nightlife! On completeing the course I was sent to work at the Coalite Chemical plant at Bolsover (just a mile from the pit where I started). The main thing I remember from there was the number of random system crashes. We eventually traced them to the fact that the fumes from the Coke and Chemical plant was rotting the interconnection pins and the rotary switches on the CRAMS and Tape Units. The fumes also ate our oscilliscope! Eventually I was sent to be the on-site engineer at Lockheed Brakes in Leamington Spa. There were two of us covering our 315 - one would do days from 0730 to 1530 (including maintenance from 0730 to 0900) and the other would be on-call from 1530 through to 0730. In an adjoining room was an ICL 190something with another two engineers. They had a byline in refurbishing antique motorbikes in the computer room on nights! With this posting (because of the on-call) I got given a company car - a Hillman Imp - which I used to get in trouble about because I bent it several times. Eventually I lost my digs and so decided to quit NCR, and got a job with Ferranti.

It's all a long time ago.


Title: Re: NCR 315
Post by: Roger on August 01, 2007, 09:39:13 pm
What a fascinating site, IU have just found it today.
I started programming on a 315 back in 1966, with H.H. Robertson in England. I well remember gonig to St Alphage House in London and to the North Circular office for testing, and to training in Greenford.
I also remember going to Chesterfield to Coalite & Chemical a couple of times for collaboration on 315 development.
Small world!


Title: Re: NCR 315
Post by: devriesbwmi40 on September 14, 2007, 06:17:27 am
I was trained as a CE on the 315 starting in 1962  in Dayton. After some OJT in the Dayton Data center I returned to Grand Rapids MI to work at Old kent Bank.  It was a system with 10 K of memory, 2 CRAM, 1 332 tape drive, a punch card reader, 340 line printer and 2 402 check sorters, We also had 12 450 proof machines with 40 pockets.

I recall sitting on a stool with cold air from the raised floor blowing up my back as I would troulbe shoot the printed circuit cards in the 3 processor bays.  We had a few problems with the PCS banjos on the top of the processor cabinet. those copper wires would break and have to be resoldered.  Memory planes could be repaired by running new copper wires through the cores if they broke.  That was more common  during shipment.  We had a full set of spare PIB's in our on site shop.  We had 3 CE's and we were there 24 hours a day. Mon - Fri   We seemed to have quite a few power supply failures. 

The Card reader needed a lot of adjustment and of course the CRAMS were a high maintenance item. bearings went bad, blowers came apart. the heads were very expensive, Listed at $25,000.00

We serviced 7 systems from the Grand Rapids office. At the end of thier life, one of our customers gave me the system and I took it home and set it up in my garage. I guess I did it just to prove I could, but it took up so much room, I scrapped it out after a few months.  If you have a specific question, I will try to answer it.

I often think about the fact that the 316 Memory of 10 K was the size of a Double wide Refriidgerator and now I have a 2 GIG  USB the size of my finger.


Title: Re: NCR 315
Post by: JimT on September 18, 2007, 08:12:55 pm
Another newbie here.  RonS told me about this site last week.  Ron - Nice to see you at the picnic.  I started with NCR at Hawthorne in 1964 in the hardware development area.  My first project was the 315-2 CRAM and then I moved over to the 615-100.  From there I worked on a lot of processors and later the memory systems.  We transferred down to Rancho Bernardo in 1970.  In 1986 I transferred to Manufacturing and had to build and deliver what Engineering had released.  In 1998 NCR sold their Manufacturing to Solectron, who I have worked for ever since.  I still work with the same people in the same location.  It has been quite a ride.

Jim Taylor


Title: Re: NCR 315
Post by: JimT on September 18, 2007, 08:34:29 pm
Some clarification on the 615-3XX – First let me say that I know nothing about software, I was always in hardware.  The 615-350 system was unique to NCR-Japan.  The 615-300 was a general release.  The –350 used the TOX operating system (Transaction Oriented eXecutive) and it tied three –300’s together through the memory system.  The memories had up to 9 input ports and up to 8 memory modules each (618-920, 618-930, 618-935 and 618-940).  Three –300’s could be tied into each memory sub-system.  Each –300 had three inputs to the memory.  As I recall they were called the E, I and I/O ports.  The E was the execution connection and the I was the instruction setup unit.  In a 615-350 system there was a high level System Memory.  The nine ports of this memory were tied to the Execution portion of each of the nine 300’s.  Over the years I had a chance to be at both the Tokyo and Osaka installations of Sumitomo bank.  The Osaka installation was very impressive!

Jim


Title: Re: NCR 315
Post by: n8eyh with OCD-WM42 of Hilse on September 29, 2007, 06:14:03 am
Dear Jim T,
 I am very happy to share our memory through working for the 615-3XX contributed on the Banking Systems Automation.
I am sure that we have met together at the computer center of Sumitomo Bank in Osaka around 1974, because you remembered correctly the design concept of the 615-3XX.
Now we, RSX OB member, the team of TOX development and the staff member of the Sumitomo Bank's OB, have been together at Tokyo in every Autum season since 2004 in order to talk with Mr. Akiyama and confirm our good health one another.
I rememer that the RSX-3XX executive, that was a kernel of the TOX, was initiated through the B1 Resident executive and was managing all of TOX task. Then our OB club name is called the RSX, because it's familier for us rather than the TOX.
All the staff member of software team are still now fine on health, and I heard that Mr. Ishii who was a manager of the field engineering staff for the 615-350 in Osaka site is still in good health. I can communicate all of them through their mailing net list.

Well, I remember the advantage of the E/I/IO ports of the 615-3XX memory bay as the multi-processing feature and the performance of the MCU(Memory Correction Unit ?) for the System Memory.
Finally, in order to get the higher stability of 615-350, 615-300 unit's portions of 615-350 had been connected fuctionally together through NOLA(500Kbps serial transmittion lines) adapters, instead of MCUs after inspecting their performance deeply.

I am appreciated of your posting because I enjoyed to remember our NCR international collaboration to provide the economical innovation.

Have a nice day,
Katsuhiko


Title: Re: NCR 315
Post by: JimT on October 01, 2007, 08:08:21 pm
The Memory Correction Unit to which you refer was actually called the Error Correction Unit, or ECU.  It was a 618-940.  My team designed and implemented that unit, but it was not a good match for core memories.  Its design capability was to error correct any single bit error, but that is not the failure tendency of a core memory.  It cost about 180-200 nanoseconds on each memory cycle, which turned out to be a bad trade off since it did not do as good a job correcting core memories as it did IC memories.  As I recall all units were eventually disconnected from the systems.  You may particularly remember one of my visits there.  Sumitomo Bank (I believe the Osaka installation) had a 618-940 that had been damaged at the receiving dock in Japan.  I was checking the error correcting displays on the unit when it brought the on-line system down (NOT GOOD!).  It turned out the sheet metal had been bent and when I rotated the rotary switched it shorted to part of the frame, therefore shorting out the logic.  It shouldn’t have happened, but it did.  I never touched another on-line system after that.  I also traveled to Japan for 8500 memory problems and later I went as a Mfg rep for 9800-2 problems.  My last trip there was somewhere around 1991 with Bob Crawford.  I believe that each time I was there I met and coordinated with Mr. Ito and Mr. Akiyama.  I worked closely with Terry Terasawa and Terry Tanaka.


Title: Re: NCR 315
Post by: n8eyh with OCD-WM42 of Hilse on October 05, 2007, 12:43:26 pm
Dear Jim Taylor,
  I am appreciated of your correct memory of our NCR Large Sytem Design Era around 1980s.
Yes, we might share the ECU problem of C-350 and the bus fault by memory error of 8570MP or 8550MP(?)
in Osaka of Sumitomo Bank.
I believe that the TOX operating system had been enhanced sophisticatedly by those problems since NCR-315 days.
I remember, for example, when the fault bell was ringing by the rod memory error or the CRAM memory film stick of NCR 315 RMC during online process computing, every field engneers used to run and gather to the console of the failed computer, in order to confirm the status and the cause. Of course, the bank staffs had already switched the service to the altered computer immediately. Then, there was no problem for notices by end-users at ATMs.
On the other hand, news paper used to introduce that the Giants computer of NCR competitor stopped their services around a hour in order to dicide the restart and recovery point for the altered computer.
I believe that such smart designs came from the 'All-for-One' good tackled partnership with Rancho Bernardo staffs, Mr. M. Ito and Mr. I. Akiyama who was an original designer and coached us, young trainees, to grow to developers of the Century TOX soon.

I rememer that I was encouraged by the presentation of Mr. Tom Tang for the bank staffs with us, related to the NCR 9800 concept, when he visited to the Osaka computer center of Sumitomo Bank. Because he introduced new idea related to computing virtually on the Disk-less large scaled flat memory for the Data-Base handling of the post-Criterion computer.

Anyway, next month will cause us to change to young, because we, the TOX delopment staff, are going to gather around Mr. I. Akiyama in Tokyo.

Have a nice weekend,
Katsuhiko


Title: Re: NCR 315
Post by: JimT on October 06, 2007, 12:11:46 am
Dear Katsuhiko-san  - From your description I am sure we have crossed paths.  My first visit to Japan was for over two weeks on the 618-940, including teaching logic classes for it.  Larry Allman accompanied me at the time.  By the second visit I was working in the Memory group.  There were two Criterion memory problems, but fortunately they were understood and under control shortly before my visit.  Both times Tom Tang sent me.  I will not be able to pass any stories to him as he died of a heart attack several years ago at his cabin in Lake Tahoe.  I worked in his organization for a long time, and he was responsible for a lot of our development – especially with NCR-Japan.  I don’t know that Akiyama-san will remember me, but please pass on my greetings to him.  I met with him and Ito-san several times in Japan.  The TOX systems were responsible for a lot of NCR’s revenue and system development.

I have a picture taken in 1984 with Tom Tang, Chuck Exley, Rocky Young and myself.  I will see if I can find it and see if it can be posted on this board.


Title: Re: NCR 315
Post by: n8eyh with OCD-WM42 of Hilse on October 07, 2007, 04:49:26 am
Dear Jim,
  I am appreciated of your posting of your memory.
I am very sorry that Mr. Tom Tang passed away. Because I got much good encouragement from his technical concept.
I am praying his good watching from Heaven to our junior activities.

Well, I informed your message to Mr. Akiyama and a group of former field engineers for Sumitomo Bank systems.
I am hoping a new message for you soon directly or indirectly.
From my friend of the TOX team, I heard that Mr. Akiyama is enjoying to play golf for his health often in these years.

It's a beautiful sky on Sunday, today.
Have a good day,
Katsuhiko


Title: Re: NCR 315
Post by: JimT on October 07, 2007, 06:54:36 am
I hope this attachment works.  If so you will see a picture taken in June of 1984.  Left to right is Chuck Exley, myself, Rocky Young and Tom Tang.


Title: Re: NCR 315
Post by: n8eyh with OCD-WM42 of Hilse on October 08, 2007, 08:24:21 am
Dear Jim Taylor,
  Good evening and good afternoon from Yokohama.
With successful openning, I am appreciated of your memorial award photo
with engineers contributed for the NCR computer development.

And I am appreciated of Mr. Aleksandrs Guba who is contributing for us
to share our brilliant young powered days through this web site.
Because my brain is now flashing back to creative days around 30 years old,
in order to transfer our technical experience and theory to next generation
people even though my body is just in 62 years old situation.

I am very happy that I have worked and been educated in NCR.

Best regards,
Katsuhiko Hirai


Title: Re: NCR 315
Post by: n8eyh with OCD-WM42 of Hilse on October 12, 2007, 07:39:01 am
Dear Jim,
  Good evening and good afternoon from Yokohama, Japan.
Today, I received a message, from Mr. Akiyama, that he is watching
these pages with his curiosity deeply.
He is waiting for our updates of discussing project through this NCR core memory.

By the way, now I am planning to attach the MMU fuctional circuit to my 16 bits MPU
with getting help of my friend who is a hardware engineer.
I am hoping to rebuild the feature of TOX-II on the micro-micro processor, in order to
confirm the performance of task-switch-less kernel architecture
under the register stack machine, even though it would have been already inspected
by the NCR post-8600 computer after I left NCR.

To trell the truth, I would like to know the experience of Mr. Akiyama, who tried to
analyze the performance of BAL/LAR handling for the multi-programming
on NCR 315-RMC of the Sumitomo Bank in 1960s.

I heard his trial story from my instructor lady through discussing after completion of our freshmen trainning.
She said to me that this result of inspection had caused him to imagine the capability of TOX-OS concept.
I am wishing to feel his passion when he found the pure light of NCR realtime processing archtecture.
Because, in the mcro-process based OS, I could not find a technology similar to TOX operating kernel.
I suppose that the TOX concept would provide more economical processing and contribute to
the reduction of CO-2 problem on this globe by the reduced energy on the computer systems.

Best regards,
Katsuhiko


Title: Re: NCR 315
Post by: Somebody on October 14, 2007, 04:17:42 am
Hello!

As noted before, I never actually worked on a 315 but my late 1960's job at May Company Department Stores in Cleveland sometimes required me to visit the computer room which had a compliment of two 315s (both 10K if memory serves), two printers, an OCR reader for cash register rolls (a 420?), paper tape reader/punch, and several tape handlers. Every document I've ever seen regarding 315-era NCR tape handlers show them with tape reels stacked vertically, one above the other. However, the tape handlers I saw had horizontal (side by side) tape reels with "IBM-type" vertical vacuum columns. I don't remember if they had an NCR logo on them. Did NCR supply such tape handlers for their 315 systems? If not, were 315-compatible tape handlers made by other manufacturers? The tape handlers in question looked suspiciously similar to same-era Control Data (CDC) units.

Does anyone know where to find good, sharp, close-up pictures of a 315 and 315-RMC console? All the pics I've seen so far are too small to show much detail.

Best regards to all!


Title: Re: NCR 315
Post by: Aleksandrs Guba on October 14, 2007, 01:33:49 pm
Hi Somebody,

I believe I have some pictures of NCR 315 console, which may satisfy you. By the way, today I've posted 22 photos of NCR 315 electronic boards to "The Gallery/Electronic Boards" page of the site.

Enjoy.


Title: Re: NCR 315
Post by: Somebody on October 27, 2007, 04:57:25 am
Hi Somebody,

I believe I have some pictures of NCR 315 console, which may satisfy you. By the way, today I've posted 22 photos of NCR 315 electronic boards to "The Gallery/Electronic Boards" page of the site.

Enjoy.


Hi Aleksandrs,

Can you please post the 315 console pics to your website?

Your website brings back great memories of a marvelous time when computers and programmers were considered to be special, not a dime a dozen. Thank you for a great website!


Title: Re: NCR 315
Post by: Aleksandrs Guba on October 27, 2007, 03:14:45 pm
Hi Somebody,

Please find attached NCR 315 Console image. Hope it will suite your needs. I will put some bigger images to the site within few days.

Enjoy!


Title: Re: NCR 315
Post by: Aleksandrs Guba on October 28, 2007, 06:42:54 pm
Hi Somebody,

You may find bigger images of NCR 315 Console and NCR 315 Console Display here http://www.thecorememory.com/html/computers.html as two last items under NCR 315 second row. Unfortunately photo of the NCR 315 Console Display is too big for pop-up image (you may see just part of it), so if that is what you need, I will send it to you as a separate file. Just let me know.

Aleksandrs


Title: Re: NCR 315
Post by: Somebody on October 29, 2007, 03:03:33 am
Hi Somebody,

You may find bigger images of NCR 315 Console and NCR 315 Console Display here http://www.thecorememory.com/html/computers.html as two last items under NCR 315 second row. Unfortunately photo of the NCR 315 Console Display is too big for pop-up image (you may see just part of it), so if that is what you need, I will send it to you as a separate file. Just let me know.

Aleksandrs

Hi Aleksandrs,

Thank you for those great pictures. Sure, I'd be glad to receive a complete picture of the console lights (you're right, pop-up is truncated). If you wish, you can e-mail it to me at:  somebody@fastmail.fm

Thank you so much! Have a great day.
 


Title: Re: NCR 315
Post by: RetrospectUK on June 15, 2008, 10:09:38 pm
I have put a film of the 315 and CRAM onto the NCR History website at www.ncr.org.uk . The film is about 15 minutes long.

I hope that you enjoy it.

Ian


Title: Re: NCR 315
Post by: sameasitev on January 16, 2010, 08:23:54 pm
I started my career in computers on an NCR 315 back around 1965,, i have a photo of me at the keyboard of the 315 at the Moscow narodny bank in London.  It was taken for a Computer Weekly article on the bank and it's new computer.  Sadly the photo is a bit the worse for wear.  But maybe some of you may be interested.  I'm still working in computers, programming still, 45 years later.
http://www.flickr.com/photos/sameasitev/268019800/sizes/l/

 


Title: Re: NCR 315
Post by: NEXUS on January 16, 2010, 11:00:22 pm
The picture is "history",  and in good condition.
I see the CRAM unit, and I can still sense the field engineer's stress of those days!
Congratulations for being active in programming after all these years.
Regards.





Title: Re: NCR 315
Post by: nlcatter on February 09, 2010, 06:30:52 am
I still  had some CRAM cards until about 7 years ago.

wrote diagnostics on 315 emulator



Title: Re: NCR 315
Post by: terrymoz on February 16, 2010, 02:29:43 pm
My name is Terry Mozley, I was a 315 Instructor at the NCR Training School located at
Borehamwood England. I went to Dayton in 1964 for an upgrade course on the 315-100 and
visited the manufacturing plant on El Segundo Boulavard Hawthorne Ca, On my return to the UK
to teach the 315-100 the school had moved to the North circular Road at Brent. I continued
teaching the 315-100 and in 1968 went to Hawthorne Ca to learn the Century 100 from the
design engineers, I self taught myself the short rod memory for the century system (615) and
taught the other international engineers who were also in Hawthorne at that time. I returned to
England after nine months in Ca to teach the Century 100 system at the training school at
Brent. In 1971 it was decided to relocate the training school from Brent to Dundee in Scotland
and I became Manager of the Dundee school. Unfortunately the move did not work out too well
for me on a personal basis and I reluctantly left NCR in 1972. My career then took me to Control Data
where I managed the London Computer Technology dept of Control Data Institute in London for four years
and then into the IBM plug compatible market with Memorex. I finished my career with Memorex
in 1992 after spending the last five years in Milpitas Ca working as a financial analyst.
I am now fully retired and enjoying life living in Berkhamsted England but still have fond memories
of my time with NCR







     


Title: Re: NCR 315
Post by: JimT on February 20, 2010, 01:03:28 am
Hi Terry.  I worked at El Segundo from 1964 through 1970, when I transferred with the company to Rancho Bernardo.  I started on the 353-2 CRAM and then went to the 615-100 development in 1965.  I remember the classes you spoke of, and I taught at least part of the 615-100 class.  I was the UM on that product until I moved to the 615-101/151 in about 1970 or 1971.


Title: Re: NCR 315
Post by: terrymoz on February 20, 2010, 11:33:12 am
Hi Jim  Good to hear from you, we must have met during that very first training session on
the 615-100 in 1968. As I said in my previous post I in fact taught the 16K memory module
on that course after self learning it from documentation and talking to the engineers.
I very much enjoyed my 9 month  stay in the factory at El Segundo, the enthusism
that prevailed amongst the people working on the 615 was infectious and they were good
times, Other names I remember among the international staff at the time who were also in
El segundo are Gunter Seitz, Hubert Wudke and Tom Stafford, all of which I have sadly lost
touch with. Are you still working Jim, or are you retired like me and reminiscing about the good
times in the past with NCR?


Title: Re: NCR 315
Post by: terrymoz on February 20, 2010, 01:29:23 pm
Hi Jim Reminiscing about the first 615 course at El Segundo has set me thinking.
I cannot recall why I taught the Core memory module on that course. Was I asked
or did I volunteer and why did not the engineers want to teach it. Perhaps as UM
you know more about that time than I. I cant remember the names of the engineers
to whom I spoke at that time but your name Jim Taylor is very familiar. If my memory
serves me correctly I believe the other parts of the course were taught by engineers
who were part of your team?


Title: Re: NCR 315
Post by: JimT on February 20, 2010, 07:55:26 pm
I remember Hubert Wudke and Tom Stafford, but I haven't heard about them in a long time.  I worked with Hubert for awhile.  I retired from NCR in 1998 when they sold their manufacturing operations to Solectron.  However, I stayed at the RB plant and worked with the same people until January 2009 when I retired for real.  You probably remember K.R. (Bob) Crawford.  He managed the field support organization at Hawthorne and RB.  I still see him periodically.  As to the class - there were several different Engineering organizations.  For processors there was the project group and the product group.  I was in the product group, but my management came from the project group.  I reported to Norm Gerstner, who technically reported to Gil Robert.  Then I transferred to  George Foster, who reported to Tom Tang, and Tom reported to Bill Campillo.  One of the reasons that I taught the processor class is that I changed the detailed implementation of the design done by the project group.  I did not change the design itself, but I went through a laborious process to minimize the logic to get it to fit into the backpanel space that we had.  When I got done the project engineers didn't recognize it anymore.  I'm guessing that Doug Starich did the I/O portion.  As for the rod memory, I think there were two issues.  I don't think they had a working level person that would have been suited to teaching the class.  Herb Lee could have done it, but I also think they were still working on improving the design.  We didn't get the rod memory working well until the night before the press release, so there continued to be a lot of design improvement activity in the group for several years.

I can't remember for sure when they officially made me the UM of the 615-100, but I believe it was after the world wide announcement on March 5, 1968.  You probably recognize my name as I signed off on every drawing and DR for the 615-100 processor.  Being the UM did not include managing all of the development, but it did include all of the ongoing engineering and support activities.  Over the years I went back and forth between Continuation Engineering (product support) and Development Engineering.


Title: Re: NCR 315
Post by: terrymoz on February 21, 2010, 04:08:46 am
I do remember Bob Crawford and I also have not seen or heard from Hubert Wudke or Tom Stafford
for some time. I also remember when the first customer 615s were shipped to the UK, at that time
due to my involvement with the short rod memory module in El Segundo I was often called upon
to support the systems and recall spending many a long hour putting current probes on the horizontal
drive wires and changing the transistors if the wave was not square, Quite often the memory ran OK on memory test but dropped a bit when running the  compiler. If my memory is correct the memory was a two wire system the horizontal wire was pulsed and the vertical wire was pulsed in the plateau of the
horizontal pulse, If the rod flipped (stored a 1) a pulse was induced on the plateau of the horizontal
pulse. Without doubt  leading edge technology at that time. Later in my career I became involved
with IBM computer hardware technologies of the 370 series and 303X series all with solid state memories, more reliable perhaps but not as challenging as the 615 rod memory.     


Title: Re: NCR 315
Post by: sameasitev on February 21, 2010, 11:33:38 pm
The first computer I worked on was an NCR 315 back in the mid sixties.  It was at the Moscow Narodny Bank in London.  Being the type of guy who hangs on to odd relics I thought you may get a kick out of my CRAM card.
This was used to help when cutting notches for replacement cards after a head crash.  This is the front (business end) and back (instructions).


Title: Re: NCR 315
Post by: donowens on November 26, 2015, 05:35:09 pm
I started working on the 315 as an apprentice in Indianapolis. In January, I was sent to service the 315 at the Indiana National Bank, reporting to Frank Murphy. He had me working on the CRAM units from day one, but I also assisted with service on the rest of the system before long. In addition to the Indiana National, we also had Merchants National Bank. Between the two site, there were seven Cram units, all my responsibility to service.

In September, I was sent to Dayton for training on the 315, finishing in May 1964. During this time, I received a transfer offer as a Technical Writer for the 353 family and was sent to Hawthorne. It was there I met Jim Taylor, better known to this forum as JimT. I wrote the service and parts manuals for the 353-2 unit. I also wrote manuals for the 420-2.

A side note about the 315 photos in the Gallery - the fourth picture from the end is not related to the 315, it is a 308 (if my memory is right), which NCR oem'ed from either 3M or CDC. It was an octal base machine rather than hexadecimal, and only had paper tape, at least the one I worked on. For some reason, I remember I received a routine to be used for debug and had to key the whole thing in as octal characters - big pain.


Title: Re: NCR 315
Post by: donowens on November 26, 2015, 08:08:40 pm
My memory indeed failed me again! The 308 I mentioned before is actually a 310!