Colossus

Introduction

The building of Colossus (Mark II) was one of the most remarkable engineering events in history. It was a race against time to put together various pieces of a new and daunting puzzle. What England did during the early 1940's, required measures that have never been taken before at that point. The first attempts at building a computing machine was the Bombe, which consisted of electromagnetic relays for increased switching. With painstaking work, the Bombe ended up being successful against the Enigma machine, however the Allies had to deal with a more intense form of enemy encryption, code named "Fish." Fish was derived from 'Sagefisch' or SawFish, which was what the Germans called their Luftwaffe's teleprinter cipher device but most importantly it was a general term for Germany's most secretive high-command wireless communication. Fish was a non-Morse 5-digit International Teleprinter Code. Bletchley Park focused on the Lorenz SZ40/42, given the code name "Tunny," which was one of the main targets of Fish and was in full use by the Germans by 1941.

Another Blunder

The Germans made quite a few mistakes while trying to keep their communication secret. In a pivotal instance, a German encipher clerk encoded the same message twice by using the same key. A young mathematician at BP, William Tutte, had luckily guessed that this was the case. Colonel Tiltman of BP was then able to obtain the plaintext message along with its key. Tutte was further able to know the structure of the Tunny, which added to the pot of miracles that was needed to break the German code. In 1942, Major Tester set up another section at BP exclusively dedicated to this endeavor of breaking the Tunny machine. Even though this new group did have some manual successes and was able to read some messages in a reasonable time, it still wasn't enough. The new section at BP was Hut F and headed by Max Newman, another mathematician. Machinery, essentially another improved Bombe, was built for this area to help with the computational process for cracking enemy encoded messages. Even with extra machinery, however, it was quickly realized that wasn't enough as well.

Stepping Into the Unknown

As early as 1920, the electronic valve had been shown to perform switching in a millionth of a second. This notion provided the springboard that was needed. Surely, it was time to embark on a new project and so it was called the 'Heath-Robinson' project. This project was unique and cutting edge at the time. It would yield a one of a kind machine using a tape reader to send pulses of light to a collection of photoelectric cells. Of course, in doing anything new there were inevitable problems. The electronic valves were not reliable and gave off a lot of heat thus experimental machines would have a tendency to smoke. Another problem was that because of the speed required, the tape used would tear on occasion.

Improvements were made to the design. One of the most notable was from Tommy Flowers, an engineer, who suggested to have internal data storage. This however, required over 1,500 electronic valves - which is how the name Colossus came about. As skeptical as this operation was, the engineers at Dollis Hill, a manufacturing plant, went ahead and built Colossus I. Even though multiple engineers contributed to this development only a few esoteric individuals knew about all of its parts and its exact purpose. Alan Turing was actually one of the chief consultants for this project. After working intensely in Hut 8 against the German U-boat fleet and after a visit to the U.S., Turing aided Max Newman greatly from a scientific sense but did not directly work on Colossus. Nevertheless, beginning in February 1943, the Colossus was finally built and installed at BP by December of the same year. The reliability factor increased when the machine was left on continuously. Thus, Fish encrypted messages were being broken at an average rate of 300 per month.

World's First Programmable and Electronic Computer

Fish traffic and strength increased from the end of 1943 well into 1944. A realization was made in March 1944, that Colossus I needed to be improved. So, another project was set up for the development of a successor called Colossus Mark II. There were many reasons for the expediency of this project. The Allies had invasion plans into Europe and knowledge of Enigma sources was helpful into determining how much the enemy knew about this. Furthermore, as the tide was just beginning to turn on the Germans, they began to grow in fear of an invasion and thus sought to increase security as much as possible. The Germans also were paranoid of an increase in disloyalty and espionage. The goal for this project was to deliver Mark II by June 1, 1944. This was a lucky coincidence as being only 5 days before D-Day. Mark II was about 5 times as fast as Colossus I, allowed programmatic flexibility and would contain 2,500 electronic valves. Thus the world's first programmable and electronic computer was born.

Colossus Mark II [4d]

Conclusion

Multiple German high-command communication links began to be broken as far back as May 1943. A total of 10 Colossus Mark II computers were constructed in conjunction with Tunny machines for deciphering, during the war. Unfortunately, these computing machines were destroyed out of secrecy right after the war. One of the unique characteristics of Colossus was that it was too complex for individual success from say someone like an Alan Turing alone, but was a success from sheer teamwork. This was a brilliant engineering feat and an 'aristocracy of the talents.'