80 Dahlgren 



computers into ballistic work, and we developed a contract with MIT to have 

 them operate for us the Bush Differential Analyzer. They ran a great many of 

 our basic trajectories for the firing tables and bombing tables, and they sent us 

 the output. We polished it up and developed the rest of the subsidiary data that 

 was essential to the firing tables — all the so-called differential effects. Then we 

 saw the need for having a large-scale calculator of our own. 



I had some consultations with Bell Labs, with Harvard Computation 

 Laboratory, MIT, and Princeton. As a result, through the Bureau of Ordnance, 

 a contract was given to the Harvard Computation Laboratory for the develop- 

 ment of the MARK II Calculator. That was the first one [large-scale computer]. 



Incidendy, about the same time or maybe prior to the MARK II, I had 

 consultation with Dr. Eckert at the Nautical Almanac Office, who later went 

 with IBM. IBM developed a couple of small relay computers. They were about 

 4 or 5 feet long, 4 feet high, and a couple of feet wide. They built three of them, 

 and they placed two of them at Aberdeen and one with us. They called it the 

 799 Relay Computer, and they had us experiment with it to see what we could 

 actually do with it in the way of practical computation. We used that to do a lot 

 of our computation work with IBM cards before we got the MARK II. 



Then Aiken developed the MARK II for us and shipped it down. What a time 

 we had getting the bugs worked out of that! The programming was rather 

 difficult. There were lots of failures, inaccuracies, and a great deal of difficulty 

 in troubleshooting, but we got a lot out of it, nevertheless. 



That program was followed by a research program through the Harvard 

 Computation Laboratory to investigate the feasibility of a fully electronic cal- 

 culator. Aberdeen, at the same time, had developed an electronic computer 

 called the Eniac. Of course, I was also in consultation with them and made 

 frequent trips to Harvard and kept in touch with Aiken. 



As a result of the research contract with the Harvard Computation Labora- 

 tory, we were able to have the MARK III developed. The MARK I was a relay 

 calculator. It was completely based on relay and tape — punched tape. The 

 MARK III was electronic with a magnetic drum memory and magnetic tape. 



The next development was the NORC [Naval Ordnance Research Cal- 

 culator]. Rather let's say the next development at Dahlgren was the fact that the 

 Bureau decided to locate the NORC at Dahlgren. That really put the Computa- 

 tion Laboratory in "big-time" calculations because the NORC was so far beyond 

 anything else at that time. 



About that time I was approached by Norden Laboratories. The decision [to 

 get the NORC] was pending when Norden approached me to come with them, 

 and I talked with Admiral Parsons and Admiral Schoeffel about it, and they 

 proposed that I stay at Dahlgren until the NORC situation was fully firmed up. 

 The NORC was delivered very shortly after I left in 1954. That was really, I 

 think, one of the turning points in the history of the Laboratory. 



