[2012-01-04]
This embryonic site will eventually hold a great deal of hardware detail pertaining to the IBM S/360 Models 65 and 67. At the moment it is just a repository for the odd snippet where there is an immediate need to make information available for a particular purpose.

If you link to any particular part of this website be aware that the structure will change when I revamp it and "do it properly"! (The home URL will, of course, remain.) This is not likely to be for at least a year though. You are welcome to download anything on the site for reuse with suitable attribution - e.g. "Source: www.IBM360.info".

All the documents below are scans from my own hardcopies. To keep thing quick and tight I have used the lowest resolution that is easily readable. At some point I intend to high-res scan the main IBM documents (FETOMs, FEMMs, etc) for inclusion on the bitsavers site.

Whilst there will be much of general IBM S/360 interest my main focus at the moment is on the 2065 Model 65 and its enhanced sibling the FAA IBM 9020D 7201-02 Computing Element (CE). Whilst the FAA CE is generally understood to be based on the Model 65 I believe it may actually be closer to the Model 67 (it has a DAT frame). I have the FETOM and FEMM for the Model 65 but would welcome links to any other Model 65 or 67 documentation out there (there's nothing on bitsavers). I (Mark) can be contacted at:

    a   k     i   m   6   .   n   o
  m   r   @   b   3   0   i   f

When speaking (or thinking!) FETOM is pronounced as "fee-tom", FEMM as "fem", and FEMDM as "fem-dem".

[2012-01-04]
IBM S/360-65 FETOM section about Read-Only Storage (ROS): 2065 FETOM Chap 2 Sect 2

[2012-01-05]
IBM S/360-65 FETOM section CPU ROS description: 2065 FETOM Chap 1 Sect 4

[2012-01-05]
IBM 7201-02 FEMM section describing roller indications (lights): 7201 FEMM Table 2-2

[2012-01-05]
IBM S/360-65 FEMDM Diag 3-2 "Data Flows": 2065 FEMDM Diag 3-2
(This is a low-res digicam image. The diagram is ~A2 size and I will scan it properly when I get the opportunity to use a large format scanner.)

[2012-01-06]
IBM 7201-02 CROS Training Notes": 7201 CROS Training Notes
(Note: The 7201 has extra instructions and functions over and above the 2065. In particular it has an additional ROS "W field" coded in ROS bits 2-5 which are not used in the 2065. Some of the detail in the document is therefore not relevant to the 2065 but the principle of "how ROS works" is still entirely valid.)

[2012-01-07]
IBM S/360-50 FETOM Capacitor Read Only Storage (CROS): 2050 FETOM CROS
(Note 1: This is for the Model 50 not the Model 65 (it would be great to track down the equivalent FETOM for a Model 65). Other than the Model 50 words being 90 bits and the Model 65 being 100 bits there's not a lot of difference in the actual CROS itself. Again, the "how it works" principles are valid for the Model 65, but in this case the control fields are almost completly different as the Model 50 has a different architecture. For instance, it has a 32 bit serial adder whereas the Model 65 has a 64 bit parallel adder.)
(Note 2: There is an interesting mistake in Figure 9 on page 15. The Bit Plate (CROS Plane) illustrated is actually from a Model 65! It is correctly (for a 65) labelled as Bits 00-99 whereas Figure 7 (correctly for a 50) refers to Bits 00-89. You can see below an actual CROS plane from a Model 65 which confirms this.)

[2012-01-07]
Below are some annotated pictures of gates C and D of a S/360-65 illustrating various elements of the CROS.
(Click on the thumbnails to see the full size pictures, then click again to zoom in.)

[2012-01-07]
Below is an annotated picture of a Model 65 CROS plane.
(Click on the thumbnail to see the full size picture, then click again to zoom in.)

[2012-01-07]
Below is a composite picture of all the front panel rollers of an FAA 7201-02 Computing Element (CE)
(Click on the thumbnail to see the full size picture, then click again to zoom in.)

[2012-01-07]
Below is an illustration of the differences in ROS field (micro-order) structure between the standard and FAA 65s. It can be seen that what where 'spare' bits in the standard 65 have been allocated new functions in the FAA 65. It is significant to note that nothing has been taken away from the standard machine's ROS.
(Click on the thumbnail to see the full size picture, then click again to zoom in.)

[2012-01-08]
Below is the ROS Data Flow diagram from the FAA CE FETOM
(Click on the thumbnail to see the full size picture, then click again to zoom in.)

[2012-01-08]
IBM FAA 9020 7201-02 FETOM CE ROS: 7201 FETOM CE ROS
(This FAA 7201 text should be compared with the similar 2065 text posted above. I have highlighted the substantive differences using "double dots" and it can be seen that 99% of the text is identical. The differences are: a) the FAA machine is called a CE (Computing Element) instead of a CPU; b) ROSPAR is called PROSAR (more accurate?); c) ROS spare bits are used for a new W field (increasing the number of fields from 21 to 22) and to extend the E field by one bit.)

[2012-01-09]
IBM 9020 Design Data Appendix C provides an example of microprogram control: 9020 Design Data Appendix C

[2012-01-10]
Below is the page from the 7201-02 Parts Catalog covering the front panel rollers. There is a total of 72 separate printed roller legends ("decals") with part numbers in two numerical groups: 573XXXX and 578XXXX. I am guessing that '573' decals relate to the standard 2065 and haven't changed for the 7201, whereas the '578' decals are new for the 7201. No great revelations here but the information may assist in defining the differences between the 2065 and the 7201.
(Click on the thumbnail to see the full size picture, then click again to zoom in.)

[2012-01-10]
IBM 9020 System Principle of Operations ("POO") Appendix G: 9020 POO Appendix G
(The IBM 7201 Computing Element [CE] is an enhanced version of the IBM S/360-65 CPU. It is the key component of the FAA 9020 system. This extract from the POO describes the differences between "9020-mode" and "360-mode". It can be seen that when in 360-mode the 7201 functions in exactly the same way as a 2065.)

[2012-01-12]
IBM 9020 7201-02 Control Field Listing: 7201 Control Field Listings
(This is a definitive listing of all the ROS micro-orders and their function. This document along with the CAS micro-program listings [CAS pages] fully defines the operation of the CPU. I have added some explanatory notes on page i.)

[2012-01-13]
CROS bit listings for the 7201-02. The third column headed "CLD" gives the number of the CAS page the block appears on (e.g. QY051) and the 'grid reference' of the block on the page (e.g. N8). The main part of the page shows the actual bits (0 or 1) broken down into fields and parity bits. Note the only unused bit in each ROS word is the second bit (bit 01) which is set to 0 in every ROS word.

Bit listing for CROS plane 01 (0): QZ001
Bit listing for CROS plane 02 (1): QZ011
Bit listing for CROS plane 03 (2): QZ021
Bit listing for CROS plane 04 (3): QZ031
Bit listing for CROS plane 05 (4): QZ041
Bit listing for CROS plane 06 (5): QZ051
Bit listing for CROS plane 07 (6): QZ061
Bit listing for CROS plane 08 (7): QZ071
Bit listing for CROS plane 09 (8): QZ081
Bit listing for CROS plane 10 (9): QZ091
Bit listing for CROS plane 11 (A): QZ101
Bit listing for CROS plane 12 (B): QZ111
Bit listing for CROS plane 13 (C): QZ121
Bit listing for CROS plane 14 (D): QZ131
Bit listing for CROS plane 15 (E): QZ141
Bit listing for CROS plane 16 (F): QZ151

[2012-01-17]
CAS microprogram listings for the 7201-02. See the "IBM 9020 7201-02 Control Field Listing" for details of how to interpret these CAS blocks.

CAS microprogram QA pages: QA001
CAS microprogram QB pages: QB001
CAS microprogram QC pages: QC001
CAS microprogram QE pages: QE001
CAS microprogram QF pages: QF001
CAS microprogram QG pages: QG001
CAS microprogram QJ pages: QJ001
CAS microprogram QK pages: QK001
CAS microprogram QP pages: QP001
CAS microprogram QQ pages: QQ001
CAS microprogram QQ pages: QQ501 [missing page from previous file]
CAS microprogram QS pages: QS001
CAS microprogram QT pages: QT001
CAS microprogram QU pages: QU001
CAS microprogram QY pages: QY001

[2012-01-17]
Below is the CPU data flow diagram from the 7201-02 FEMDM
(Click on the thumbnail to see the full size picture, then click again to zoom in.)

[2012-01-17]
IBM 9020 7201-02 FEMDM SS (Storage to Storage) instruction flow charts: 7201 FEMDM two CAS Flows
(This is Diagram 5-305 [SS instruction Decimal Multiply] and Diagram 5-306 [SS instruction Decimal Divide] from the 7201-02 FEMDM (Field Engineering Maintenance Diagram Manual). They show the CAS microprogram data flows at a somewhat higher level than the CAS microprogram pages themselves. These are just examples of reasonably complex instructions - there are similar diagrams for all the instructions (op codes). These instructions are NOT 'special 9020' instructions and I would guess that they are the same on the 2065 CPU and the 7201 CE.)

[2012-01-20]
One of the key things I want to do with this website is to document the differences between the 'standard' IBM S/360-65 and the FAA IBM S/360-65. The former is known as the 2065 CPU and the latter as the 7201-02 CE. Here are the details of the front panels of the two machines. Very similar but a few differences. I will document the differences in detail later.
(Click on the thumbnail to see the full size picture, then click again to zoom in.)

[2012-01-27]

The '9020' was a large computer complex based on IBM S/360 technology. The largest 9020 configurations consisted of a 9020D CCC (Central Computer Complex) and 9020E DCC (Display Channel Complex). A maximum configuration CCC/DCC complex contained no less than 12 IBM S/360 mainframes! Not all FAA ARTCCs (Air Route Traffic Control Centers), of which there were 20 (plus one in the UK), had the maximum configuration. The UK system had 'only' 8 mainframes - 4 x S/360-50 and 4 x S/360-65. This schematic shows the maximum configuration, with the mainframe boxes highlighted in blue.
(Click on the thumbnail to see the full size picture, then click again to zoom in.)

[2012-01-29]
IBM 9020 7201-02 FEMDM sections 0 to 4: 7201 FEMDM Sections 0 to 4
IBM 9020 7201-02 FEMDM sections 6 to X: 7201 FEMDM Sections 6 to X
(This is the complete 7201-02 FEMDM (Field Engineering Maintenance Diagram Manual) except for Section 5, which has the flow diagrams for individual op code instructions. Full details of what is in the 'missing' section can be found in the index in Section 0. Section 5 is by far the largest section and the only reason I have not included it is that I haven't yet had chance to scan it all in. It will be posted soon.)

[2012-02-05]
IBM 9020 7201-02 FETOM.pdf: 7201 FETOM.pdf
(This is the complete 7201-02 FETOM (Field Engineering Theory of Operation Manual). It is a large file [~50MB] consisting of 556 pages.)

[2012-02-09]
IBM S360/65 2065 FETOM.pdf: 2065 FETOM.pdf
(This is the complete 2065 FETOM (Field Engineering Theory of Operation Manual). It is a large file [~50MB] consisting of 500 pages.)

[2012-02-09]
Below are some pictures of the final load of scrap S/360 frames leaving the site of the UK 9020D system for the melting pot.
(Click on the thumbnails to see the full size pictures, then click again to zoom in.)

[2012-02-18]
IBM 9020 7201-02 FEMDM chapter 5: 7201 FEMDM Section 5
(This is the 7201-02 FEMDM (Field Engineering Maintenance Diagram Manual) Section 5.)

[2012-02-24]
I am in the process of transferring some old IBM 800bpi NRZI program and data tapes to the PC. I need the tapes to end up on the PC in "AWS" format, which is described here www.cbttape.org/awstape .

I have transferred the first program tape (which is physically labelled "NON LIBRARY LOADER CD017 REVISION 28") to the PC, where it exists as a single 5,903,231 byte EBCDIC binary file (tape01.bin).

This is not much use on its own as it consists of 1,597 contiguous variable length records but there is no way of knowing where one record ends and the next begins. On the original NRZI tapes the records are separated by 0.6" inter-record gaps, but the program I used to transfer the data does not insert anything in the PC file to represent these gaps. However, the tape analysis program that I used did output information that enabled me to create a block information file block01.csf

This file lists, for each block: 1) The block number; 2) The block size; 3) The block start offset into the file; and 4) The block end offset into the file. Some of this information is redundant, but it enabled me to verify the accuracy of the block information by cross checking. There is enough information in these two files to enable an AWS format file to be generated.

[2012-02-24]
The file referred to above contains 1,597 blocks so it is impractical to manually edit the .BIN file to create a .AWS file. However, I also transferred a smaller file that consisted of only four blocks. Having such a small number of blocks it was possible to manually edit the file to create an .AWS file. Here it is: FLUT01.aws . I used Hex Edit to edit the .BIN file. The .AWS file can be viewed using AWSBrowse which can be found halfway down this page.

"FLUT" is the Fault Location test Utilty Tape. It is stand-alone (i.e. doesn't require an OS) and can be directly IPL'd into a S/360 system.

[2012-02-26]
Modified the server web.config file to permit the FLUT01.aws file (see previous entry) to be downloaded instead of just getting an IIS error message!

I have two program manuals for FLUT, one from 1967 and one from 1974. Hopefully they will provide the information needed to successfully run FLUT (a standalone program) under the Hercules emulator. I am keen to do this as one of FLUT's functions is to print detailed information from FLT tapes. I want to be able to do this with the FLT tapes I have as there is a lot of bit-level information about the S/360-65 ROS that I'm interested in.

This is the 1967 document: FLUT 1967
This is the 1974 document: FLUT 1974

Note that the terms "CE", "IOCE", and "SE" are synonymous with "CPU", "Channel", and "Main Storage" respectively.

[2012-02-28]
Thanks to software wizard Camiel Vanderhoeven (Operation Blinkenlights) I now have Tape01 in .AWS format: Tape01.aws .

Viewing this tape image in AWSBrowse reveals that it is essentially a 9020 system maintenance tape. The middle word of each of the 12-byte blocks is the 'identity' of an individual maintenance program, of which there are just over 500 on the tape. The blocks following these 12-byte headers begin with the same program identity.

The structure of the tape is essentially as pictured here: 9020 System Maintenance Tape

The main part of this tape (which is stand-alone IPL'able on a 9020 system) consists of a large number of hardware diagnostic routines (or "sections" as they are called), most of which check out the operation of individual CPU instruction. I have the full user manual for this tape plus the source code of all the sections - to be uploaded in due course.

[2012-03-16]
IBM 9020 7201-02 PC (Parts Catalog): 7201 PC.pdf
(This is the complete Parts Catalog for the 9020D Computing Element. To my mind the Parts Catalog is almost a work of art in itself, as well as being a model of clarity and utility.)

[2012-03-17]
Here is a page from the CE Parts Catalog where I have highlighted the Dynamic Address Translation frame that is not found as part of the standard Model 65.
(Click on the thumbnail to see the full size picture)

[2012-03-22]
FAA Engineering Requirement document for what was to become the IBM 9020 System: FAA-ER-606-063.pdf
(This FAA Requirements Spec from 1963 was downloaded from the Kent State University website at this link. I don't know the provenance of the annotations.)

[2012-03-22]
IBM 9020 Design Data: 9020 Design Data.pdf
(This document was submitted to the Federal Aviation Agency by IBM in response to Section 3.11.5 of the above Requirements Specification and contains the Design Data for the IBM 9020D and 9020E Data Processing Systems.)

[2012-06-26]
IBM 2065 FEMM: IBM 2065 FEMM.pdf
(This is the complete Field Engineering Maintenance Manual for the IBM 2065 - the S/360 Model 65 CPU.)


This is an IBM 9020D Compute Element, model 7201-02, with the covers open. It is a complete S/360-65 with a few added bells and whistles!

IBM System/360 Model 50