EEVblog #818 - Embedded Electronics - a conversation with Jack Ganssle (video)

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Winston

Lorenzo von Matterhorn
Joined
Jan 31, 2009
Messages
9,560
Reaction score
1,749
A really fascinating, spirited conversation with Jack Ganssle about embedded electronics, programming, and many other tech topics.

[video=youtube;1apCAzCTZdQ]https://www.youtube.com/watch?v=1apCAzCTZdQ[/video]

On his comment about an iPhone built with ENIAC technology, he estimated it would require 170 Vertical Assembly Buildings of volume to enclose and would cost 50 trillion dollars. That led me to find this where someone else asked the same question:

iPhone 6 implemented using ENIAC technology

https://www.reddit.com/r/estimation...ig_would_an_iphone_6_be_if_implemented_using/

A bit of Googling and some simple calculations reveal the following:

ENIAC

Clock Speed: 5000 Hz
FLOPs: ~455 (7 digit decimal)
Memory: ~83 bytes (200 decimal digits)
Power Consumption: ~150 000 W
Mass: ~27 000 kg
Area: 167 m2
Volume: ~64.8 m3

iPhone 6

Clock Speed: 2 800 000 000 Hz (x2 because of dual core)
FLOPs: ~76 800 000 000 (32 bit ~= 7 digit decimal)
Memory: 1 000 000 000 bytes
Power Consumption: 0.86 W (6.9 Wh battery being depleted in 8 hours)
Mass: 0.129 kg
Area: 0.0092 m3
Volume: 7.1 * 10-5 m3
In other words to match an iPhone 6's performance you would need roughly 560 000 ENIACs worth of clock speed, 17 million ENIACs worth of FLOPs and 12 million ENIACs worth of memory. Averaging those out we get around 10 million ENIACs.

10 000 000 ENIACs

Power Consumption: ~1 500 000 000 000 W (roughly 1/2 of the US' power consumption)
Mass: ~270 000 000 000 kg (roughly 60% of the mass of the entire human race)
Area: 1 670 000 000 m2 (5/4 the footprint of Los Angeles)
Volume: ~648 000 000 m3 (about twice the size of the great wall of china)
However as others have pointed out even with 10 000 000 ENIACs you would not be able to replicate the iPhone's performance because of the time it would take to transfer signals between ENIACs and because not all computing tasks can be parallelized.
 
Back
Top