ENIAC vs. ENIAC-on-a-Chip IC implementation

[schwepes at moog.netaxs.com]

Why be surprised?  The second group had a far higher state of the art.
The first group, who actually built ENIAC, were probably actually shocked
that the damned thing actually worked.
The first Mercedes motorcar got ten miles to the gallon and had a top
speed of ten miles an hour. You can imagine all of the places in the
engine cycle from the carb to the transmission that could be cleaned up
from the original which was not much more than a modern golf cart.
bs


On Mon, 1 Sep 2008, William Blair wrote:

> >From the excellent book, "The First Computers - History and Architectures," an interesting comparison:
>
> "The ENIAC contained 17,468 vacuum tubes, 70,000 resistors, 10,000 capacitors, 7,200 crystal diodes, and 6,000 switches, it had a footprint of about 33 m x 1 m, occupied a room of 170 square meters, dissipated about 140-174 kW and weighed 30 tons. In contrast, he chip realization contains 174,569 transistors, measures 7.4 mm x 5.3 mm (the PGA package measures 3.6 cm by 3.6 cm), dissipates a few Watts (depending on how many units run in parallel and the clock speed), and weighs a few grams. Also, in terms of power requirement the comparison is striking. In addition to the AC power for the heaters of the tubes, the card reader and the card punch, the ENIAC required 78 different DC voltage levels to power 10 different types of vacuum tubes. The power equipment was housed in 7 panels which were separate from the ENIAC’s 40 panels. Special ventilating equipment consisted of an elaborate system of fans and blowers to keep the temperature inside the panels
>  below C. In contrast, the chip needs only one power supply of 5 V (or lower). The clock frequency used in the ENIAC was 100 kHz, while the one on the chip can easily run at 50 MHz or higher."
>
> http://www.ese.upenn.edu/~jan/eniacproj.html
>
>
>
>
>