Odkazy:
Počítač Whirlwind byl zkonstruován MIT jako hlavní počítač a součást projektu SAGE
Fakta:
Vzhledem k zadání, tedy že veškerá elektronika bude snadno dostupná byl počítač fyzycky rešen ve 2D prostoru. Z toho vzešly obrovské požadavky na zastavěnou plochu.
Počítač měl přes 5000 elektronek a 11 germaniových diod.
Jeden z neobvyklých vlastností bylo použití diodové matice k dekódování instrukcí. V té době se běžně používaly nestrukturované logocké obvody (Random Logic).
Obrázek 84.1. Instrukční slovo Whirlwind
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+ | operation | address | +---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+---+
Tabulka 84.1. Instrukce počítače Whirlwind
Oper. | Function | Binary Code | Octal | Dec | Time | Oper. |
---|---|---|---|---|---|---|
clc n | cycle left and clear (BR) | 11110-0 | 1.70 | 30 | 15+0.8n μs | clc |
clh n | cycle left and hold | 11110-1 | 1.70 | 30 | 15+0.8n μs | clh |
md x | multiply digits with no roundoff (p) | 11111 | 1.74 | 31 | 22μs | md |
si pqr | select in-out unit/stop | 00000 | 0.00 | 0 | 30μs | SI |
ill | illegal instruction | 00001 | 0.04 | 1 | n/a | ILL |
bi x | block transfer in | 00010 | 0.10 | 2 | max 16ms 1st; 16μs each word | BI |
rd | 00011 | 0.14 | 3 | 15μs | RD | |
bo x | block transfer out | 00100 | 0.20 | 4 | max 16ms 1st; 16μs each word | BO |
rc x | record | 00101 | 0.24 | 5 | 22μs | RC |
sd x | sum digits | 00110 | 0.30 | 6 | 22μs | SD |
cf pqr | change fields | 00111 | 0.34 | 7 | 15μs | CF |
ts x | transfer to storage | 01000 | 0.40 | 8 | 22μs | TS |
td x | transfer digits | 01001 | 0.44 | 9 | 29μs | TD |
ta x | transfer address | 01010 | 0.50 | 10 | 29μs | TA |
ck x | check | 01011 | 0.54 | 11 | 22μs | CK |
ab x | add B | 01100 | 0.60 | 12 | 29μs | AB |
ex x | exchange | 01101 | 0.64 | 13 | 29μs | EX |
cp x | conditional program | 01110 | 0.70 | 14 | 15μs | CP |
sp x | sub-program | 01111 | 0.74 | 15 | 15μs | SP |
ca x | clear and add | 10000 | 1.00 | 16 | 22μs | CA |
cs x | clear and subtract | 10001 | 1.04 | 17 | 22μs | CS |
ad | add | 10010 | 1.10 | 18 | 22μs | AD |
su | subtract | 10011 | 1.14 | 19 | 22μs | SU |
cm | clear and add magnitude | 10100 | 1.20 | 20 | 22μs | CM |
sa x | special add | 10101 | 1.24 | 21 | 26μs | SA |
ao x | add one | 10110 | 1.30 | 22 | 29μs | AO |
dm x | difference of magnitudes | 10111 | 1.34 | 23 | 22μs | DM |
mr x | multiply and roundoff | 11000 | 1.40 | 24 | 34-41μs | MR |
mh x | multiply and hold | 11001 | 1.44 | 25 | 34-41μs | MH |
dv x | divide | 11010 | 1.50 | 26 | 71μs | DV |
slr x | shift left and roundoff | 11011 0 | 1.54 | 27 | 15+0.8n μs | SLR |
slh x | shift left and hold | 11011 1 | 1.54 | 27 | 15+0.8n μs | SLH |
srr x | shift right and roundoff | 11100 0 | 1.60 | 28 | 15+0.8n μs | SRR |
srh x | shift right and hold | 11100 1 | 1.60 | 28 | 15+0.8n μs | SRH |
sf x | scale factor | 11101 | 1.64 | 29 | 30-78 μs | SF |