Way Back Computing IBM Automatic Sequence Controlled Calculator --ASCC-- in action (also known as the Harvard Mark 1)
This page is dedicated to the earlier days of computing. Click on the small version to see a larger version. Early computing devices were electro-mechanical, using relays, switches, gears, cams, and other such "hard wiring." The ASCC, or Harvard Mark 1, was such an electro-mechanical device. Started in 1939, the ASCC was completed in 1944. Below is a photo from, "A Manual of Operations for the Automatic Sequence Controlled Calculator," published in 1946 (perhaps the first "computer manual"). The relays and other devices are shown. Following is another picture of the ASCC being readied for its debut. Notice the drive shaft running underneath. Below is a photo of some relays from The Gallery's collection. These are identical to the relays shown in, "A Manual of Operations for the Automatic Sequence Controlled Calculator," Figure 16, p.55. Some of the devices are labelled IBM. All were probably spares for various equipment. Below is a photo of a panel of relays in The Gallery's collection. This panel looks very similar to the panels of relays shown in pictures of the ASCC/Mark 1 (above). All of the relays are stamped "IBM." Were such banks of relays used in other devices? Perhaps The Gallery has a piece of the ASCC? IBM wanted to build an all electronic computing device to outdo the electro-mechanical ASCC / Harvard Mark 1. The new device would use tubes instead of relays as its primary source of logic. Tubes had advantages and disadvantages (e.g., they were much faster than relays, but tended to burn out). The new, all electronic device was called the Selective Sequence Electronic Calculator (SSEC) and was completed in 1948. It was put on public display at 590 Madison Avenue, New York and functioned there for years. Below is a photo from an original 1948 SSEC marketing brochure in The Gallery's collection. Although the SSEC was "one of a kind," it was nonetheless marketed as a technical triumph and harbinger of what electronics (then synonymous with vacuum tubes) would bring in the future. An entire wall of vacuum tubes is seen on the right. Below, from the same brochure, is a floor plan description showing the parts of the SSEC. Below is a photo of the SSEC main console, as it sits today on private display. As tubes tended to burn out eventually, IBM began to assemble them into easily removeable modules, called pluggable units (PUs). If any one component in a PU failed, the entire PU was more easily identified and replaced. Below are some of the tubes in The Gallery's collection. One dual-tube pluggable unit is shown. Below is an advertisement from the tube era. It shows a pluggable unit and the first device to be constructed with them -- the IBM 604 Electronic Calculator, delivered in 1948. The 604 was an important milestone in electronic computing, as it pioneered standard circuits, mass production of electronic equipment (more than 5000 were built), and large scale deployment and servicing of computing devices. The 604 was "programmable," but did not store its program electronically, rather programs were "constructed" on wiring boards, a photo of such a hard-wire program board is shown at the end below. The 604 had a clock cycle of 50KHz (compare to modern PC giga-Hertz clocks). The 604 borrowed ideas from the SSEC and in turn ideas from the 604 influenced the IBM 701 (delivered in 1952 -- the first IBM mass produced stored-program computer). Below are some more advertisements from the tube era. Below is an example of a wiring plugboard, used to "program" calculators and other devices before electronically stored programs were common. Finally, below is IBM's first mass-produced stored program computer, the 701. It was first shipped in 1952 with an approximate monthly rental of $15,000, not including peripheral equipment ($US15,000 in 1952 translates to about $US107,000 in 2004). Instead of core memory (which IBM first used on a general purpose commercial computer on the 705), the 701 used CRT memory: a Type 706 Williams Tube Memory unit with 72 circular tubes can be seen at the right in the photo. Some references: "A Manual of Operation for the Automatic Sequence Controlled Calculator" (AKA Harvard Mark 1), Harvard University Press, 1946. "IBM Selective Sequence Electronic Calculator," 1948 marketing brochure. "IBM's Early Computers," Charles Bash, Lyle Johnson, John H. Palmer, Emerson Pugh, MIT Press, 1986. U.S. Dept. of Labor, Bureau of Labor and Statistics, inflation calculator: http://www.bls.gov/cpi/home.htm Return to The Gallery of Old Iron home page

Way Back Computing

IBM Automatic Sequence Controlled Calculator --ASCC-- in action
(also known as the Harvard Mark 1)

This page is dedicated to the earlier days of computing. Click on the small version to see a larger version.

Early computing devices were electro-mechanical, using relays, switches, gears, cams, and other such "hard wiring." The ASCC, or Harvard Mark 1, was such an electro-mechanical device. Started in 1939, the ASCC was completed in 1944. Below is a photo from, "A Manual of Operations for the Automatic Sequence Controlled Calculator," published in 1946 (perhaps the first "computer manual"). The relays and other devices are shown.

Following is another picture of the ASCC being readied for its debut. Notice the drive shaft running underneath.

Below is a photo of some relays from The Gallery's collection. These are identical to the relays shown in, "A Manual of Operations for the Automatic Sequence Controlled Calculator," Figure 16, p.55. Some of the devices are labelled IBM. All were probably spares for various equipment.

Below is a photo of a panel of relays in The Gallery's collection. This panel looks very similar to the panels of relays shown in pictures of the ASCC/Mark 1 (above). All of the relays are stamped "IBM." Were such banks of relays used in other devices? Perhaps The Gallery has a piece of the ASCC?

IBM wanted to build an all electronic computing device to outdo the electro-mechanical ASCC / Harvard Mark 1. The new device would use tubes instead of relays as its primary source of logic. Tubes had advantages and disadvantages (e.g., they were much faster than relays, but tended to burn out).

The new, all electronic device was called the Selective Sequence Electronic Calculator (SSEC) and was completed in 1948. It was put on public display at 590 Madison Avenue, New York and functioned there for years.

Below is a photo from an original 1948 SSEC marketing brochure in The Gallery's collection. Although the SSEC was "one of a kind," it was nonetheless marketed as a technical triumph and harbinger of what electronics (then synonymous with vacuum tubes) would bring in the future. An entire wall of vacuum tubes is seen on the right.

Below, from the same brochure, is a floor plan description showing the parts of the SSEC.

Below is a photo of the SSEC main console, as it sits today on private display.

As tubes tended to burn out eventually, IBM began to assemble them into easily removeable modules, called pluggable units (PUs). If any one component in a PU failed, the entire PU was more easily identified and replaced.

Below are some of the tubes in The Gallery's collection. One dual-tube pluggable unit is shown.

Below is an advertisement from the tube era. It shows a pluggable unit and the first device to be constructed with them -- the IBM 604 Electronic Calculator, delivered in 1948. The 604 was an important milestone in electronic computing, as it pioneered standard circuits, mass production of electronic equipment (more than 5000 were built), and large scale deployment and servicing of computing devices.

The 604 was "programmable," but did not store its program electronically, rather programs were "constructed" on wiring boards, a photo of such a hard-wire program board is shown at the end below. The 604 had a clock cycle of 50KHz (compare to modern PC giga-Hertz clocks). The 604 borrowed ideas from the SSEC and in turn ideas from the 604 influenced the IBM 701 (delivered in 1952 -- the first IBM mass produced stored-program computer).

Below are some more advertisements from the tube era.

Below is an example of a wiring plugboard, used to "program" calculators and other devices before electronically stored programs were common.

Finally, below is IBM's first mass-produced stored program computer, the 701. It was first shipped in 1952 with an approximate monthly rental of $15,000, not including peripheral equipment ($US15,000 in 1952 translates to about $US107,000 in 2004).

Instead of core memory (which IBM first used on a general purpose commercial computer on the 705), the 701 used CRT memory: a Type 706 Williams Tube Memory unit with 72 circular tubes can be seen at the right in the photo.

Some references:

"A Manual of Operation for the Automatic Sequence Controlled Calculator" (AKA Harvard Mark 1), Harvard University Press, 1946.
"IBM Selective Sequence Electronic Calculator," 1948 marketing brochure.
"IBM's Early Computers," Charles Bash, Lyle Johnson, John H. Palmer, Emerson Pugh, MIT Press, 1986.
U.S. Dept. of Labor, Bureau of Labor and Statistics, inflation calculator: http://www.bls.gov/cpi/home.htm

Return to The Gallery of Old Iron home page