IBM 1401 overview

Return to Main Page
Van Snyder's IBM 1401 Programming Hints

from Ed Thelen
Many of "us", the current crop of (post year 2000 ;-) 1401 programmers, regard Van Snyder as "Having forgotten more about the" 1401 than we will ever know.
OK, he had a head start, 1960s 1401 contact, and he wrote the 1401 Assembler running in ROPE ;-))
He has also driven from Los Angeles to check certain instructions on "our" 1401s - and donate 1401 artifacts.
Here is a collection Van sent Jan 27, 2015. (Over the years, he sent many separate items.)

IBM 1401 is a stored-program automatic computer oriented to small and medium businesses.
System overview

Memory
Peripherals
Registers
Instructions
Chaining
Programming tips for IBM 1401
The most useful IBM 1401 Reference manuals

1401 Principles of Operation - A24-1403-5, also known as "The Brownie Book" because of its brown cover.
Autocoder assembler. Reference number C24-3319.
Dataflow. Reference number G24-1377.
Many more at http://www.bitsavers.org/pdf/ibm/1401
Subjects

Program loading
Switched branches
Determine core size
Clear core
List of d-modifiers for BIN (5-character branch) instructions
Character Code Charts first in BCD order, then in collating sequence order.
B-field signs after arithmetic operations There's a shorter version of it on page 28 of the "Brownie Book," A24-1403. Dick Weaver gave me this version.
5 digit to 3 digit : 3 digit to 5 digit address conversions
Memory

Memory organization

The basic unit of information is a six-bit character, consisting of a four-bit digit part (8-4-2-1) and a two-bit zone part (B-A). The addressable unit of memory consists of a character, a word mark (W), and a parity ("check") bit (C). These are written in manuals, and appear on the computer console, as CBA8421W.
The memory cycle time is 11.5 microseconds.
Models were manufactured with 1400, 4000, 8000 and 16000 characters of memory.
Data representation
Fields of data are usually referenced by the character at the highest address. The character at the lowest address is indicated to be the end of the data field by having its word mark bit set.
Character fields are stored with the first character at the lowest address and the last character at the highest address of the field.
Numeric fields are stored with the high-order digit at the lowest address and the low-order digit at the highest address. The sign of the field is represented by zone bits in the low-order character: B-zone means negative, and any other combination, including no zone, is positive. Overflows are encoded by zone bits in the high-order character, in the order A, B, BA. During arithmetic operations the zone bits in other characters of the fields are ignored in operands, and cleared in results.
Addressing
Addresses are represented by three characters.
Addresses less than 1000 consist of three digits. Addresses less than 4000 consist of three digits, plus overflows in the high-order character. Addresses above 3999 have zone bits in the low-order character. Writing the zone bits in the low-order and then high-order character in the order BABA gives the thousands part of the address in binary.
Low zones High zones Thousands B A B A 0 0 0 0 0 0 0 0 1 1 0 0 1 0 2 0 0 1 1 3 0 1 0 0 4 0 1 0 1 5 0 1 1 0 6 0 1 1 1 7 1 0 0 0 8 1 0 0 1 9 1 0 1 0 10 1 0 1 1 11 1 1 0 0 12 1 1 0 1 13 1 1 1 0 14 1 1 1 1 15

Indexing
A special feature called "advanced progamming" provides three index registers. These are actually stored in memory, at locations 87-89, 92-94, and 97-99.
An address indicates indexing with zone bits in the middle character. If indexing is selected, the address used during execution of an instruction consists of the address specified in the instruction plus the contents of the specified index register. Indexing adds three cycles, and sometimes four, per address, to the instruction execution.
B A Index register 0 0 None 0 1 X1 1 0 X2 1 1 X3

Address arithmetic
Address arithmetic is modular, or "wraps around," at 16000. Therefore to subtract a number X from an address, add 16000-X to it.
Peripherals

IBM 1401 peripherals
Dozens of peripheral devices could be attached to an IBM 1401.
A basic system included a model 1402 card reader-punch and a 1403 printer. Bigger systems included model 729 or 7330 tape drives or model 1301, 1311 or 1405 disk drives. Peripherals for data communication, optical character recognition, check sorting, paper tape, ... were available.
1402 card reader-punch
The 1402 card reader-punch can read up to 800 cards per minute and punch up to 250 cards per minute.
Cards can have rectangular holes punched in 80 columns and twelve rows. The top two rows are called 12 and 11, or Y and X. The bottom ten rows are called 0-9. Rows 12 and 11 are called zone rows and rows 0-9 are called numeric rows.
Data are always read into positions 1-80 of memory, and always punched from positions 101-180 of memory.
Relationship between card punches and character representations

A blank column corresponds to no bits set.
A punch in the 12 row corresponds to both B and A zone bits set.
A punch in the 11 row corresponds to only the B zone bit set.
A punch in the zero row, accompanied by punches in no additional rows other than 12 or 11, corresponds to the 8 and 2 bits set.
A punch in the zero row accompanied by punches in rows 1-9 corresponds to the A zone bit set.
A punch in rows 1-9, accompanied by punches in no additional rows other than 12 or 11,corresponds to the digit part of the character having the same binary value, 0001 - 1001.
Two punches in numeric rows other than zero, provided one of them is in row 8 and other is in one of rows 3-7, corresponds to bit 8 set, along with the binary representation of the other row.
Punches in rows 0, 2 and 8 correspond to a character having A82 representation.
The only character that cannot be represented as a card punch combination consists of only the A zone bit; a no-cost special feature allowed this to be represented by a 2-8 punch.
1403 printer
The 1403 printer consists of a printing mechanism and a paper-advance mechanism.
The printing mechanism consists of a bank of 132 hammers, a ribbon, and a chain of characters. Horizontal spacing of characters is ten per inch. The chain is contained within a removable cartridge. The chain is spun in front of the paper at high speed. When a character on the chain is at a print position at which that character is to be printed, the hammer strikes the paper from the back. The ribbon is positioned between the paper and the chain, resulting in an ink impression of the character when the hammer strikes.
Three standard chains were available. The A chain is for business applications, and the H chain is for scientific applications. These chains had six copies of a 48-character set. With these chains, the printer could print up to 600 lines per minute.
A numeric chain, having eighteen copies of only the digits plus six special characters allowed printing at up to 1200 lines per minute.
Characters having 5-8, 6-8 or 7-8 representations (with or without zones) cannot be printed without special chains. The representations of the A and H chains, in Paul Pierce's ASCII encoding of BCD tapes, are
8421 in hex, A chain 8421 in hex, H chain Zone BA | 0 1 2 3 4 5 6 7 8 9 A B C D E F | 0 1 2 3 4 5 6 7 8 9 A B C D E F none 00 | 1 2 3 4 5 6 7 8 9 0 # @ : > { | 1 2 3 4 5 6 7 8 9 0 = ' : > { zero 01 | ^ / S T U V W X Y Z | , % ~ \ " | ^ / S T U V W X Y Z | , ( ~ \ " 11 10 | - J K L M N O P Q R ! $ * ] ; _ | - J K L M N O P Q R ! $ * ] ; _ 12 11 | & A B C D E F G H I ? . ) [ < } | + A B C D E F G H I ? . ) [ < }
The only differences here are B-A (12) [&,+], 3-8 [#,=], 4-8 [@,'] and A-4-8 (0-4-8) [%,(]. On a real A chain, B-A-4-8 (12-4-8) printed as a character named "lozenge," a little box with pointed corners.
Data are always printed from memory positions 201-332.
The paper advance mechanism can be set to advance paper at six or eight lines per inch. It can also skip to a position indicated by a punch on a paper control tape at 75 inches per second.
Other peripherals
None of the other peripherals transfer data to or from dedicated positions in memory.
Registers

IBM 1401 registers
IBM 1401 registers

Address registers
The IBM 1401 processor has three address registers that the program can affect, and one optional address register the program cannot affect or examine. All address registers can be examined from the console.
Index registers are actually memory fields, not proper registers.
I-address register
The I-address register contains the instruction address. It is incremented by the normal sequence of instruction execution, or changed by a branching instruction.
A-address register
The A-address register contains the address of the first operand of an operation. It is initially set from the A field of an instruction, and may be changed during execution of the instruction.
B-address register
The B-address register contains the address of the second operand and/or destination result of an operation. It is initially set from the B field of an instruction, and may be changed during execution of the instruction. If the advanced programming feature is installed, after an instruction that branches is executed, it contains the address following the branch instruction. This is useful for subroutine linkage.
O-address register
If the process overlap special feature is installed, the 1401 processor includes the O-address register, which is used during overlapped input/output operations. It is not affected by and cannot be examined by the program.
Data registers
The 1401 processor includes three one-character data registers. They are used during instruction execution but cannot be otherwise affected or examined by the program.
B register
The B register is the register through which all data pass between the processor and memory.
A register
The A register contains a character from the first operand of an operation. During instruction fetching the B register is copied to the A register.
OP register
The OP register contains the instruction OP code.
Instruction length register
The instruction length register contains the instruction length.
Instructions

IBM 1401 instructions

Instruction representation
Instructions are represented by consecutive characters in memory.
The first character has a word mark. With three exceptions, the representation continues through characters at increasing addresses, up to but not including the next character that has a word mark. Instructions can have lengths of one, two, four, five, seven, or more than seven characters. Instruction lengths of three or six result in address decoding errors, which cause the processor to stop with the "check stop" lamp illuminated.
The three instructions that do not need to be followed by a character with a word mark are five-character branch with blank D modifier, seven-character set word mark, and seven-character clear storage and branch. Instructions and their formats are summarized below.
Instruction fields
Instructions have from one to four fields.
OP code field
The first character of every instruction is the operation code, or the OP code. It specifies the operation to be performed.
A address field
If the instruction length is four or more, the second, third and fourth characters of the instruction are the A address field. If the first character is % or @ this is a device address. Otherwise, this is the memory address of a branch target in a branching instruction or an operand of an operation.
B address field
If the instruction length is seven or more, the fifth, sixth and seventh characters are the B address field. This is the address of an operand of an instruction.
D modifier field
If the instruction length is two, five, or more than seven, the final character is the D modifier field. It further specifies or modifies the operation. The D modifier of an instruction is taken from the A register. Therefore, if an instruction expects to use a D modifier character, but it doesn't have enough characters, the last one fetched is used as the D modifier. Instructions with operation codes of B or V use a D modifier if the instruction is one character or more than six characters. Instructions with operation codes of L or M use a D modifier if the A address field of the instruction begins with % or @ and the instruction is longer than six characters.
Instruction summary
The Dataflow manual provides detailed information about the instruction fetch phase, and about the method of operation of most instructions. It was not updated when the overall logic scheme was changed at serial number 25,000, and it does not include input/output.
The Principles of Operation manual provides detailed information about each instruction.

OP, D Autocoder Lengths Description Branch Special

1 R 1 4 Read a card If A No

1, C R 2 5 Read column binary If A Yes

2 W 1 4 Print If A No

2, S W 2 5 Print, suppress spacing If A No

2, ) W 2 5 Print word marks If A No

3 WR 1 4 Print and read a card If A No

4 P 1 4 Punch a card If A No

4, C P 2 5 Punch column binary If A Yes

5 RP 1 4 Read and punch If A No

5, C RP 2 5 Read and punch, column binary If A Yes

6 WP 1 4 Print and punch If A No

7 WRP 1 4 Read, print and punch If A No

8 SRF 1 4 Start read feed If A No

9 SPF 1 4 Start punch feed If A No

# MA 1 4 7 Modify address No If > 4k

@ M 1 4 7 Multiply No Yes

/ CS 1 4 7 Clear storage To A if B No

S S 1 4 7 Subtract No No

T T 1 4 7 Translate No 1410, 1460

U, A CU 5 Diagnostic read tape No No

U, B BSP 5 Backspace tape No No

U, E SKP 5 Skip and erase tape No No

U, M WTM 5 Write tape mark No No

U, R RWD 5 Rewind tape No No

U, U RWU 5 Rewind tape and unload No No

V, 1 BW 1 >6 Branch if word mark Yes No

V, 2 BWZ 1 >6 Branch if no zone Yes No

V, B BWZ 1 >6 Branch if AB zone Yes No

V, K BM 1 >6 Branch if B zone Yes No

V, S BWZ 1 >6 Branch if A zone Yes No

W, d BBE 1 >6 Branch if bit equal Yes Yes

Y MZ 1 4 7 Move zone No No

Z MCS 1 4 7 Move and suppress zero No No

, SW 1 4 7 Set word mark No No

% D 1 4 7 Divide No Yes

K, 1248 SS 2 5 Select 1402 stacker If A No

K, L SS 2 5 1009 load mode If A 1009

K, $ SS 2 5 Overlap on If A Yes

K, A SS 2 5 1009 suppress alarm If A 1009

K, C SS 2 5 DDC read request If A DDC

K, D SS 2 5 1009 set receive If A 1009

K, D SS 2 5 DDC write request If A DDC

K, E SS 2 5 1009 set transmit If A 1009

K, E SS 2 5 DDC reset If A DDC

K, F SS 2 5 1009 send EOF If A 1009

K, . SS 2 5 1402 Overlap off If A Yes

K, ) SS 2 5 1402 Reset overlap If A Yes

L LCA 1 4 7 Load to A-field word mark No No

L, R RTW >6 Read tape with word marks No Tape

L, W RTW >6 Write tape with word marks No Tape

M MCW 1 4 6 Move to A- or B-field word mark No No

M, R RT >6 Read tape No Tape

M, W WT >6 Write tape No Tape

N NOP any No operation No No

P MCM 1 4 7 Move record No Yes

Q SAR 1 4 Store A-address register No Yes

! ZS 1 4 7 Zero and subtract No No

A A 1 4 7 Add No No

B B 4 Branch, 4-characters Yes No

B, A BLC 5 Branch if last card Yes No

B, B-G BSS 5 Branch if sense switch Yes Yes

B, other BIN 5 Branch on indicator, see next table Yes Some

B, other BCE 1 >6 Branch if character equal Yes No

C C 1 4 7 Compare No No

D MN 1 4 7 Move numeric No No

E MCE 1 4 7 Move and edit No Some

F, d CC 2 5 Carriage control No No

H SBR 1 4 7 Store B-address register No Yes

? ZA 1 4 7 Zero and add No No

) CW 1 4 7 Clear word mark No No

D-modifier indicators for five-character branch

D Char Autocoder Unit Meaning blank BIN 1401 Unconditional -- no word mark needed in next char 1 BIN 1009 in run condition A24-3071-2 I-66 1011 punch in backspace operation 1011 reader parity indicator on 1231 auto select 1285 error 1412 control-check indicator on 1418 late read 1419 document to be read 1428 late read or late reading mode change 1445 printer error 7340 hypertape unusual end 7740 transmission complete with abnormal status DDC transmission error A24-3071-2 I-14 2 BIN 1009 buffer available A24-3071-2 I-66 1011 punch ready 1011 reader ready 1231 full buffer 1285 end of line 1412 reader-not-ready signal on 1418 ready to engage 1419 document under read head (PDS 4) 1428 ready to engage 7340 hypertape normal end 7740 transmission complete successfully DDC transmission ended by GMWM A24-3071-2 I-14 3 BIN 1009 good transmission occurred A24-3071-2 I-66 1011 punch not ready to receive data 1231 ready to read 1285 reader transporting 1412 read-check indicator on 1418 document under selected read station 1419 valid amount field 1428 document under selected read station 7340 hypertape control unit 7641 busy 7740 receive request DDC read request A24-3071-2 I-14 4 BIN 1009 reply-bad indicator on A24-3071-2 I-66 1011 punch not ready to read 1231 empty hopper 1285 marked line 1412 amount-field indicator on 1419 valid process-control field 7340 hypertape attention 7740 7740 attention DDC write request A24-3071-2 I-14 5 BIN 1009 error reply acknowledgement A24-3071-2 I-66 1011 punch overextended 1231 read error or overrun 1285 header information 1412 process-control indicator on 1418 document end 1419 valid account-number field 1428 document end 1445 printer busy 6 BIN 1009 program attention required A24-3071-2 I-66 1011 punch supply reel low or chad box full 1231 timing mark check 1285 ready to read a line 1412 account-number indicator on 1418 character on line 1419 valid transit-number field 1428 character on line 1445 carriage busy DDC write in progress A24-3071-2 I-14 7 BIN 1009 end of message A24-3071-2 I-66 1285 reader ready 1412 transit-number indicator on 1418 empty hopper and transport (end of file) 1419 valid serial-number field 1428 empty hopper and transport (end of file) 1445 carriage channel 9 DDC read in progress A24-3071-2 I-14 8 BIN 1009 end of file A24-3071-2 I-66 1285 end of file 1412 document-spacing-check indicator on 1418 ready to read 1419 auto-select 1428 ready to read 1448 carriage channel 12 DDC system A stopped A24-3071-2 I-14 9 BIN 1403 carriage tape channel 9 0 BIN 1404 validity error #/= @/' BCV 1403 carriage tape channel 12 : > BIN 1448 end of block { ^ / BU 1401 unequal compare S BE 1401 equal compare T BL 1401 low compare U BH 1401 high compare V BIN 1301 disk error 1311 disk error 1405 read or write parity check or read back check error W BIN 1301 wrong-length record 1311 wrong-length record 1405 wrong-length record X BIN 1301 uniqual-address compare 1311 uniqual-address compare 1405 unequal-address compare Y BIN 1301 any disk-unit error condition 1311 any disk-unit error condition 1405 any disk-unit error condition Z BAV 1401 arithmetic overflow | BIN 1403 printer error (record mark) , %/( BIN 1401 process check with process check switch off (%) 1440 I/O check stop switch off (maybe 1440 only) ~ \ BIN 1301 access busy 1311 access busy " - J BIN 1419 I/O channel-busy indicator TAU tape busy SIO serial input-output busy K BEF TAU end of reel (writing) or tape mark (reading) L BER TAU tape error M N BIN 1301 access inoperable 1311 access inoperable 1405 access inoperable O P BIN 1403 printer busy Q BIN 1407 inquiry request R BIN 1403 carriage busy ! 1402 punch error $ * BIN 1407 inquiry clear ] ; _ &/+ BIN 1442 last card (Reader unit 2) A BLC 1402 last card if SS A is on B BSS 1401 SS B BIN 7340 hypertape attention response C BSS 1401 SS C D BSS 1401 SS D E BSS 1401 SS E BIN 7340 hypertape end response F BSS 1401 SS F G BSS 1401 SS G H BIN 1402 reader busy I BIN 1402 punch busy ? BIN 1402 reader error . ) [ < BIN 1448 early warning }
Chaining

IBM 1401 address chaining
An important feature of the IBM 1401 that is useful for reducing the size of programs is address chaining.
As an insruction of length greater than one is fetched, the second character is put into the first character of the A-address register. If there are third and fourth characters, they are put into the second and third characters of the A-address register.
If the instruction length is one, the A-address register is not changed.
If the instruction OP code is not L, M, Q or H, the second through fourth characters of the instruction are also put into the B-address register as they are fetched. Otherwise, the B-address register is not changed at this point.
As an instruction of length greater than four is fetched, the fifth character is put into the first character of the B-address register. If there are sixth and seventh characters, they are put into the second and third characters of the B-address register.
The Dataflow manual provides detailed information about the instruction fetch phase.
Because of this method of changing the contents of address registers, if the contents of an address register after executing the previous instruction are the contents desired at the beginning of the next instruction, the address field need not be present. It is not possible to provide the B-address field and omit the A-address field.
Chaining is frequently used to move or add adjacent fields, but it has many other uses. One that is not obvious is to chain BCE or BWZ instructions.
The Principles of Operation manual provides detailed information about the contents of address registers after each instruction is executed.
Program loading

Loading programs in IBM 1401
Programs can be loaded into IBM 1401 from cards, tape or the serial I/O adapter.
Loading programs from cards
To load a program from cards, put the cards in the card reader. Press check reset and start reset on the 1401 console, then press check reset and load on the 1402.
The effect of pressing the load key on the 1402 is to set a word mark in location 1, clear all word marks in locations 2-80, read a card, and branch to location 1.
Since every instruction has to begin with a word mark, the first card loaded has to start setting word marks for the rest of the program on that card. Fortunately, a seven-character set word mark instruction is one of three instructions that do not need to be terminated by a word mark in the next character.
Very simple loader
A very simple loader, used for the diagnostic decks, loads one instruction or data field from each card. Bootstrapping the loader requires a very simple program on the first card:
001 sw 8,12 008 r 1
Which on the first card reads
,0080121001
Each of the next cards loads one instruction or data field:
001 lca abc,xyz 008 r 1 012 * first character to load ... abc * last character to load at xyz
which on the card reads
LABCXYZ1001...
The last card clears the read area and branches to the program using a seven-character clear storage and branch instruction:
001 cs xyz,80
which on the card reads
/xyz080

Autocoder loader format
The Autocoder assembler produces cards in which data to be loaded are in columns 1-39, and instructions to load those data, and set word marks, are in 40-71. Usually, the instruction in 40-46 is a load instruction. Usually, the instructions in 47-67 are set word mark instructions, although sometimes the first one is a clear word mark instruction. Usually the instruction in 68-71 is read a card and branch to 40.
To bootstrap this loader, Autocoder produces the following card:
....5...10...15...20...25...30...35...40...45...50...55...60...65...70. ,008015,022029,036040,047054,061068,072/061039 ,0010011040 1 1 1 1 1 1 1 1 1 1 1 1
A simpler one is
....5...10...15...20...25...30...35...40...45...50...55...60...65...70. ,008047/047046 ,054061,068072,0010401040 1 1 1 1 1 1 1
If a program has overlays, it's useful not to clear the word marks in the read area, so as to be able to continue loading. To enter a portion of a program without clearing the read area, columns 68-71 contain a branch instruction.
The last card has a clear storage and branch instruction in columns 40-47.
This is described in detail on page 51 in the Autocoder assembler reference manual.
Loading programs from tape
To load a program from tape, mount the tape on unit 1 and make it ready, press check reset and start reset on the 1401 console, and then press tape load on the 1401 console.
When the tape load key is pressed, the 1401 reads one record from tape unit 1, with word marks, into location 1, and branches to location 1. It's not necessary to bootstrap this process in the same way as for loading from cards because records can be read from tape with word marks. It may be necessary, however, to provide that one record of the program reads the next one.
Autocoder provides a loader that runs entirely in locations 1-80. The first record clears core:
....5...10...15...20...25...30...35...40...45...50...55...60...65...70. ,200/000H008V0052001L051115L/100099L%U1001R/007199 CLEAR STORAGE 1 1 1 1 1 11 1 1

The second record contains the loader for the remaining records:
....5...10...15...20...25...30...35...40...45...50...55...60...65...70. U%U1B.L%U1020RB001L.020 1 11 1 1 1

which is executed beginning in location 7. Subsequent records load data into core, and then load the next record by branching to location 7.
This is described in detail in on page 51 the Autocoder assembler reference manual.
Other formats for loading from cards
The number of ways to load programs from cards is limited only by your imagination. The 1401 Fortran IV compiler uses a format in which word marks are represented on cards by the same character as is used on tape. There is no "read a card with word marks" instruction, so a loader program has to be provided. Here's one:
job Alternative boot loader -- uses SBR ctl 6611 * * Boot loader. On each program card * CC 1-3 give first address to load. * CC 4-5 give number of non-word-separator characters to process. * CC 7-70 are characters to be loaded, or word separator * characters 0-5-8 preceding characters to get word marks. * Load CS xxx,080 atop read - read&6 to enter program, thus * 09907~/xxx080 * or change the A-address at done&1 - done&3 to enter the program * without clearing the read area, thus * 13103xxx * * This runs in 96-199. The start-up code from 200-218 is cleared. * sfx 1 org 96 k001 dc 001 read r sbr move&3,007 Starting from address & 1 loop mcw 3,cwsw&3 Move ,,to,, address to cw or sw mcw cw Move CW op code to cworsw mcw 3 Move ,,to,, address to move s read,5 Decr count of chars to process ma read-1,003 Bump ,,to,, address done bm read,5 Read another card if all moved move mcw 0-0,save Pick up two characters mcw save-1,0-0 Store character, B gets cc 1-3 cwsw cw 0-0 Either SW or CW, A gets cc 1-3 addrs ma k001,move&3 Bump from address bce *&5,save-1,~ word separator, 0-5-8 b loop Process next character s save-1 Clobber word separator bce addrs,save,~ word separator, 0-5-8 mcw sw,cwsw Prepare to set WM b move cw cw sw sw save equ 80 Use cc 79-80 for work first sw 79 cs sw 1,4 cs read,* xfr first
Switched branches

Switched Branches
There are several ways for programs to change their paths of execution, what would now be done by setting and testing boolean or logical variables.
Word marks as flags
Word marks can be set or cleared with four-character instructions, and tested with an eight-character instruction.
sw flag turn on the branch cw flag turn off the branch bw xyz,flag branch to xyz if flag has a word mark flag dcw #1

Characters as flags
Characters can be used as flags, but this is somewhat less efficient than using word marks, since seven-character instructions are needed to change them, and a value to put into them has to be stored somewhere. An advantage is that they can have more than two values.
mcw @x@,flag turn on the branch mcw @y@,flag turn off the branch, maybe turn a different one on mcw @z@,flag turn off the branch, maybe turn a different one on bce xyz,flag,x branch to xyz if flag is x flag dcw #1

Modify the OP code
The op code can be changed between branch and NOP
mcw branch,switch turn on the branch mcw nop,switch turn off the branch switch b xyz branch if the OP code is not NOP
The values to set into the OP code can be ordinary literal values, or OP codes of instructions that don't change. A program always has branches for other purposes. NOP is less commonly needed.
Switch branch with word marks
This is the most efficient method. Branches can be switched by setting and clearing word marks, either at the OP code of the D-modifier.
Initially on branch
An initially on branch is implemented by a NOP followed by a branch. The branch is switched by setting or clearing the word mark under its OP code.
nop switch b xyz branch if switch has a word mark, else part of NOP cw switch turn off the branch sw switch turn it back on

Initially off branch
An initially off branch is implemented as a branch-on-indicator using an indicator that is never set by the processor, such as a group mark. The branch is switched by setting or clearing the word mark under the D-modifier.
switch bin xyz,} branch if D-modifier has a word mark cw switch&4 turn off the branch sw switch&4 turn on the branch
Determine core size

Determining actual IBM 1401 core size
When a one- or four-character clear storage instruction (/) is executed, if the value of the B-address register before execution is greater than 99 its value after execution is x99, where x is one less than the hundreds-thousands part of the value of the B-address register when the instruction began execution. For example, after CS 345 the B-address register contains 299. After CS 345 followed by CS, the B-address register contains 199.
If the value of the B-address register when execution of the instruction begins is less than 100, its value after execution is the address of the last installed character. This value will be 1399, 3999, 7999, 11999 or 15999, depending on the amount of core installed.
If the indexing and store address feature is installed, the B-address register can be stored after a clear storage instruction. If it is stored after one clears starting from any address in the range 0-99, it contains the address of the highest-addressed character.
Clear core

Clearing core in IBM 1401

Clearing fewer than 101 characters, down to x00
To clear fewer than 101 characters, down to x00, use CS xtu, where tu are the tens and units digits of the top address to clear.
Clearing a few hundreds of characters, down to x00
To clear a few hundred characters, down to x00, use CS ztu, where z is greater than x but not much greater, followed by z-x+1 one-address clear storage instructions. These are chained onto the first one. This is frequently used to clear the print area:
cs 332 cs

Clearing many, or an indeterminate number, of hundreds of characters, down to x00
There are three ways to clear core down to x00 using loops.
Using compare
Assume an address constant contains w99, where w is x-1.
loop cs top sbr loop&3 c loop&3,adcon bu loop ... adcon dsa w99

This requires 23 characters. If the code needs to be reused, put the top address to be cleared into loop&3 using
mcw &top,loop&3
or
sbr loop&3,top
This increases the requirement to 33 or 30 characters, respectively.
Using BCE
If the prior contents of a character in the bottom century to clear are known, and known not to be blank, it can be tested to determine whether that century has been cleared:
mcw *-6,xyz loop cs top sbr top&3 bce loop,xyz,M
This requires 23 characters.
If the value of xyz is known, that value can be tested at the bottom of the loop, and the MCW is not needed. This brings the requirement down to 16 characters.
Using BW
If it is known that a character in the bottom century to be cleared contains a word mark, it can be tested to determine whether that century has been cleared:
sw xyz loop cs top sbr top&3 bw loop,xyz
This requires 20 characters.
If it is known that xyz has a word mark, the first instruction is not needed. This brings the requirement down to 16 characters.
Using an index register
So as not to require to change the A-address field at loop&3, one can use an index register, say X3:
loop cs 0&x3 sbr x3 ...
This requires putting the top address to clear into X3 instead of loop&3 before starting the loop.
Clearing down to xyz
To clear down to xyz, where xyz is not x00, the easiest way is to use an index register. Assume clearing is to start at xtu, where tu is greater than yz.
sbr x3,xtu sw xyz loop mcw @ @,0&x3 sbr x3 cw 1&x3 bw loop,xyz
If X3 already contains an appropriate address, perhaps as a result of using X3 to clear down to y00 where y equals x+1, the first instruction can be deleted.
List of d-modifiers for BIN (5-character branch) instructions

BIN D modifiers D cha r Punch Bits Unit Meaning blank none C------ 1401 Unconditional -- no word mark needed in the next char 1 1 ------1 1009 in run condition A24-3071-2 I-66 1011 punch in backspace operation 1011 reader parity indicator on 1231 auto select 1285 error 1412 control-check indicator on 1418 late read 1419 document to be read 1428 late read or late reading mode change 1445 printer error 7340 hypertape unusual end 7740 transmission complete with abnormal status DDC transmission error A24-3071-2 I-14 2 2 -----2- 1009 buffer available A24-3071-2 I-66 1011 punch ready 1011 reader ready 1231 full buffer 1285 end of line 1412 reader-not-ready signal on 1418 ready to engage 1419 document under read head (PDS 4) 1428 ready to engage 7340 hypertape normal end 7740 transmission complete successfully DDC transmission ended by GMWM A24-3071-2 I-14 3 3 C----21 1009 good transmission occurred A24-3071-2 I-66 1011 punch not ready to receive data 1231 ready to read 1285 reader transporting 1412 read-check indicator on 1418 document under selected read station 1419 valid amount field 1428 document under selected read station 7340 hypertape control unit 7641 busy 7740 receive request DDC read request A24-3071-2 I-14 4 4 ----4-- 1009 reply-bad indicator on A24-3071-2 I-66 1011 punch not ready to read 1231 empty hopper 1285 marked line 1412 amount-field indicator on 1419 valid process-control field 7340 hypertape attention 7740 7740 attention DDC write request A24-3071-2 I-14 5 5 C---4-1 1009 error reply acknowledgement A24-3071-2 I-66 1011 punch overextended 1231 read error or overrun 1285 header information 1412 process-control indicator on 1418 document end 1419 valid account-number field 1428 document end 1445 printer busy 6 6 C---42- 1009 program attention required A24-3071-2 I-66 1011 punch supply reel low or chad box full 1231 timing mark check 1285 ready to read a line 1412 account-number indicator on 1418 character on line 1419 valid transit-number field 1428 character on line 1445 carriage busy DDC write in progress A24-3071-2 I-14 7 7 ----421 1009 end of message A24-3071-2 I-66 1285 reader ready 1412 transit-number indicator on 1418 empty hopper and transport (end of file) 1419 valid serial-number field 1428 empty hopper and transport (end of file) 1445 carriage channel 9 DDC read in progress A24-3071-2 I-14 8 8 ---8--- 1009 end of file A24-3071-2 I-66 1285 end of file 1412 document-spacing-check indicator on 1418 ready to read 1419 auto-select 1428 ready to read 1448 carriage channel 12 DDC system A stopped A24-3071-2 I-14 9 BC9 9 C--8--1 1403 carriage tape channel 9 0 0 C--8-2- 1404 validity error = 3-8 ---8-21 ' BCV 4-8 C--84-- 1403 carriage tape channel 12 : 5-8 ---84-1 > 6-8 ---842- 1448 end of block { 7-8 C--8421 ^ none --A---- / BU 0-1 C-A---1 1401 unequal compare S BE 0-2 C-A--2- 1401 equal compare T BL 0-3 --A--21 1401 low compare U BH 0-4 C-A-4-- 1401 high compare V 0-5 --A-4-1 1301 disk error 1311 disk error 1405 read or write parity check or read back check error W 0-6 --A-42- 1301 wrong-length record 1311 wrong-length record 1405 wrong-length record X 0-7 C-A-421 1301 uniqual-address compare 1311 uniqual-address compare 1405 unequal-address compare Y 0-8 C-A8--- 1301 any disk-unit error condition 1311 any disk-unit error condition 1405 any disk-unit error condition Z BAV 0-9 --A8--1 1401 arithmetic overflow | 0-2-8 --A8-2- 1403 printer error (record mark) , 0-3-8 C-A8-21 ( 0-4-8 --A84-- 1401 process check with process check switch off (%) 1440 I/O check stop switch off (maybe 1440 only) ~ 0-5-8 C-A84-1 \ 0-6-8 C-A842- 1301 access busy 1311 access busy " 0-7-8 --A8421 - 11 -B----- J 11-1 CB----1 1419 I/O channel-busy indicator TAU tape busy SIO serial input-output busy K BEF 11-2 CB---2- TAU end of reel (writing) or tape mark (reading) L BER 11-3 -B---21 TAU tape error M 11-4 CB--4-- N 11-5 -B--4-1 1301 access inoperable 1311 access inoperable 1405 access inoperable O 11-6 -B--42- P 11-7 CB--421 1403 printer busy Q 11-8 CB-8--- 1407 inquiry request R BPCB 11-9 -B-8--1 1403 carriage busy ! 11-0 -B-8-2- 1402 punch error $ 11-3-8 CB-8-21 * 11-4-8 -B-84-- 1407 inquiry clear ] 11-5-8 CB-84-1 ; 11-6-8 CB-842- _ 11-7-8 -B-8421 + 12 CBA---- 1442 last card (Reader unit 2) A BLC 12-1 -BA---1 1402 last card if SS A is on B BSS 12-2 -BA---2 1401 SS B 7340 hypertape attention response C BSS 12-3 CBA--21 1401 SS C D BSS 12-4 -BA-4-- 1401 SS D E BSS 12-5 CBA-4-1 1401 SS E 7340 hypertape end response F BSS 12-6 CBA-42- 1401 SS F G BSS 12-7 -BA-421 1401 SS G H 12-8 -BA8--- 1402 reader busy I 12-9 CBA8--1 1402 punch busy ? 12-0 CBA8-2- 1402 reader error . 12-3-8 -BA8-21 ) 12-4-8 CBA84-- [ 12-5-8 -BA84-1 < 12-6-8 -BA842- 1448 early warning } 12-7-8 CBA8421 # Pierce's alternate BCD encoding: # # Numeric b 1 2 3 4 5 6 7 8 9 0 3 4 5 6 7 # Rows b 8 8 8 8 8 # =========================================== # No zone 1 2 3 4 5 6 7 8 9 0 = ' : > { 000-017 = 00-0F # 0 zone A ^ / S T U V W X Y Z | , ( ~ \ " 020-037 = 10-1F # 11 zone B - J K L M N O P Q R ! $ * ] ; _ 040-057 = 20-2F # 12 zone BA + A B C D E F G H I ? . ) [ < } 060-077 = 30-3F # # | is 0-2-8, and ^ has no input punch (it may punch as zero # on output). On other systems it may be 2-8.
Character Code Charts first in BCD order, then in collating sequence order.

B-field signs after arithmetic operations
5 digit to 3 digit : 3 digit to 5 digit address conversions

On Tue, 2015-01-27 at 11:05 -0800, Ken Shirriff wrote: > What's the easiest way to convert a 5-digit number to a 3-character > address in a program? > > I understand how the zone bits encode the upper digits of the address. > E.g. if I have the number 12345, the corresponding address is 34E (or > something like that). But how can I do this in code? I can think of a > couple ugly ways to do it: > > > 1. Do various comparisons and set the appropriate zone bits. E.g. test > if address >= 12000, 8000, 4000 and set zone bits. Then test if the > remainder is >= 3000, 2000, 1000. > 2. Use the MA (modify address) instruction in a loop, subtracting 1000 > from the number and adding 1000 to the base address each time. > > > But is there an easy way to do this? > > > Thanks, > Ken Van Snyder responded with the following The attachments have my collection of routines to convert 3-to-5 and 5-to-3. -------------------------------------------------------------- job Convert five-digit number to an address ctl 6611 x1 equ 089 five equ 205 three equ 210 org 336 zones dcw @2skb@ adr dcw #5 number dcw #5 start sw three-2 lca &15999,number outer mcw number,adr Number to convert b conv Convert it to an address mcw adr,three mcs number,five w s &1000,number Subtract 1000 bm two,number q. done b outer no two lca &15999,number outer2 mcw number,adr Number to convert b conv2 Convert it to an address mcw adr,three mcs number,five w s &1000,number Subtract 1000 bm fast,number q. done b outer2 no fast lca &15999,number outer3 mcw number,adr Number to convert b conv3 Convert it to an address mcw adr,three mcs number,five w s &1000,number Subtract 1000 bm done,number q. done b outer3 no done h done ltorg* * * Smallest routine to convert the five-digit number in adr to * a three character address, in place. Needs the digit part * of the last character of the address of ZONES to be nine. * conv sbr convx&3 bav *&1 turn off overflow loop a @96@,adr-3 bav loop mz adr-4,adr mn adr-3,*&4 mz zones-0,adr-2 convx b 0-0 ltorg* * * One that works similarly, but uses X1 and doesn't need ZONES * to be at a particular location. Saving and restoring X1 * would make it bigger and slower. * conv2 sbr conv2x&3 sbr x1,0 bav *&1 turn off overflow loop2 a @96@,adr-3 bav loop2 mz adr-4,adr mn adr-3,x1 mz zones-9&x1,adr-2 conv2x b 0-0 ltorg* * * Fastest routine to convert the five-digit number in adr to * a three character address, in place. Uses X1. Would be * slower and 14 characters longer if it saved and restored x1. * conv3 sbr conv3x&3 mcw adr-3,x1 mcw @0@ mz hzone-15&x1,adr-2 mz lzone-15&x1,adr conv3x b 0-0 hzone dcw @9zri9zri9zri9zri@ lzone dcw @9999zzzzrrrriiii@ end start ----------------------------------------------------- and ------------------------------------------------------ job Convert three-character address to five digits ctl 6611 from equ 205 start lca @999@,from First address to convert outer mcw from,to b c3to5 Convert it mcw to,215 mcw from,to b c3to5s Convert it mcw to,225 mcw from,to b c3to5f Convert it mcw to,235 mcw from,to b c3to5x Convert it mcw to,245 w ma a1k,from Bump it by 1k c from,@999@ Q. Wrapped around bu outer No done h done a1k dsa 1000 ltorg* * * Fast routine to convert the three-character address at * ,,to,, to a five-digit address, in place. * c3to5 sbr exit&3 s to-3 Clear thousands digits mz to,to-4 Low zone to accumulate loop1 bwz loop1x,to-4,2 Q. low zone remaining a @?4@,to-3 Yes, add 4k per low zone b loop1 loop1x mz to-2,to-4 High zone to accumulate loop2 bwz loop2x,to-4,2 Q. high zone remaining a @?1@,to-3 Yes, add 1k per high zone b loop2 loop2x za to&1 Clear zones exit b 0-0 to dcw #5 dc #1 ltorg* * * Smallest routine to convert the three-character address at * ,,to,, to a five-digit address, in place. * c3to5s sbr exits&3 mcw k00,to-3 Zero high digits, leave no zone loop1s za to&1,test&1 Copy digits only c to,test Q. Address all digits exits be 0-0 Yes, zones all gone ma am1000,to MA -1000 from address a k1,to-3 Bump thousands b loop1s test dcw #3 dc #1 k00 dcw 00 am1000 dsa 16000-1000 -1000 k1 dcw 1 * * Faster routine to covert the three-character address at * ,,to,, to a five-digit address, in place. * c3to5f sbr exitf&3 mcw k00,to-3 bwz lowf,to-2,2 bwz high2,to-2,k bwz high1,to-2,s a *-6,to-3 high2 a *-6,to-3 high1 a *-6,to-3 lowf bwz donef,to,2 bwz low2,to,k bwz low1,to,s a &4,to-3 low2 a &4,to-3 low1 a &4,to-3 donef za to&1 Clear zones exitf b 0-0 ltorg* * * Fastest routine to covert the three-character address at ,,to,, * to a five-digit address, in place, but it uses all three index * registers. * c3to5x sbr exitx&3 lca @0020000400006@,99 mz to-2,*&3 Set index register for thousands mcw thou,to-3 mz to,*&3 Index register for 4-thousands a thou4,to-3 za to&1 Clear zones exitx b 0-0 thou dcw 00 dcw 01 dcw 02 dcw 03 thou4 dcw 00 dcw 04 dcw 08 dcw 12 end start