course	mod	topic
cit-115	wk. 3	Is newer better?
module learning objectives
CCAC.115.LT.1: Categories of computers (personal, mobile, servers, mainframes, supercomputers, embedded) and examples of computer usage (SOHO, mobile, power user, enterprise) CCAC.115.LT.4: Input devices and characteristics (keyboard, pointing devices, voice input, digital cameras, scanners, game controllers) CCAC.115.LT.5: Output devices and characteristics (displays, printers, audio) CCAC.115.LT.6: Storage devices and characteristics (magnetic disks, optical disks, static-state, removable)

By Casey Fleser - Flickr: Evolution (34 / 365), CC BY 2.0, Link

computer fundamentals > change and remain project

Is newer better?

The "Change and Remain" project

Original telephones manufactured by AT&T were designed to last for decades. Modern electronics is far more multi-purposed, and often designated as at end-of-life in 5 years or less.

The newer-is-better notion is very lucrative, indeed, and has driven the development of consumer electronic capabilities many predicted were physically impossible to produce at the scale of global commofification.


core question	In what circumstances does gearing up with the state-of-the-art technology add productive value to a computing endeavor/business/project?
supporting questions	What do we mean by "new" tech? Current year? half-decade? decade? Since systems have bottlenecks, not all components need to be upgraded to increase computational throughput. To ably design a computer system, what data can we generate to help us decide which components of our particular computer systems are critical enough to upgrade from base configuration? [explore a possible spreadsheet tool]
key points	The way we experience computers in our day-to-day lives has, indeed, changed dramatically over the past several decades. But not all components and their underlying technologies change at the same rate, if at all. In fact, computer scientists and engineers have designed chips whose speed has roughly doubled every two years. Gordon Moore modeled this trend mathematically and wrote a now-famous paper about it, leading us to describe this growth pattern in transistor density as Moore's Law. Since component evolutions have developed at different paces, so-called bottlenecks or choke points along the complex processing pathways embedded inside modern computers. A practical implication of this phenomenon is that if one wants improved system performance, only a subset of components need to be upgraded. Alternatively, one might, in fact, upgrade a component needlessly which was not "holding up the show" resulting in no actual overall system performance. By analogy: Given a desire to increase throughput of cars from downtown Pittsburgh to the airport, adding another lane of traffic before and after the Fort Pitt tunnel would likely not increase overall traffic flow significantly. Since traffic systems are complex, such a change could have some unexpected impacts on component functioning.

core question

In what circumstances does gearing up with the state-of-the-art technology add productive value to a computing endeavor/business/project?

supporting questions

What do we mean by "new" tech? Current year? half-decade? decade?

Since systems have bottlenecks, not all components need to be upgraded to increase computational throughput. To ably design a computer system, what data can we generate to help us decide which components of our particular computer systems are critical enough to upgrade from base configuration? [explore a possible spreadsheet tool]

key points

The way we experience computers in our day-to-day lives has, indeed, changed dramatically over the past several decades. But not all components and their underlying technologies change at the same rate, if at all.

In fact, computer scientists and engineers have designed chips whose speed has roughly doubled every two years. Gordon Moore modeled this trend mathematically and wrote a now-famous paper about it, leading us to describe this growth pattern in transistor density as Moore's Law.

Since component evolutions have developed at different paces, so-called bottlenecks or choke points along the complex processing pathways embedded inside modern computers. A practical implication of this phenomenon is that if one wants improved system performance, only a subset of components need to be upgraded. Alternatively, one might, in fact, upgrade a component needlessly which was not "holding up the show" resulting in no actual overall system performance.

By analogy: Given a desire to increase throughput of cars from downtown Pittsburgh to the airport, adding another lane of traffic before and after the Fort Pitt tunnel would likely not increase overall traffic flow significantly. Since traffic systems are complex, such a change could have some unexpected impacts on component functioning.

War Games (1983) and Jurassic Park (1993) Tech Comparison

After watching the two film clips twice, populate the following table with facts about the capabilities and characteristics of the computers shown in the clips

War Games (1983)	Jurassic Park (1993)
Describe: Computer display, user interface, computational power, communication systems, "speed"	Describe: Computer display, user interface, computational power, communication systems, "speed"

Generate component evolution data

Create a spreadsheet that allows you to support a legitimate answer to the core question in this module, namely: When is buying newer computer components advantageous to improve overall performance?

Your spreadsheet should probably have at least this set of columns plus additional ones to support your core calculations (such as adding a column for storing units of some measurement separate from its value: store "100" in a column separate from "mhz").

computer component changes through time spreadsheet template

You should try creating this spreadsheet from scratch. Use an internet search engine to help you format cells in libre office sheets or excel and insert basic formulas to calculate cost per cycle, etc.


resources	Timeline of computer history Moore's Law in The Wikipedia Gordon Moore's original 1965 paper in Electronics Magazine and the related Intel corporation newsroom feature article The Law that has driven digital life by Jo Twist via the BBC Moore's law lecture notes by Prof Bajramovic at UMSL

Heart of the matter: What should we upgrade?

Choose a computing environment and make the case to the system design committee about which components should be optimized for your general use case and which can remain static as they were in year ***INSERT BENCHMARK YEAR***.

Use quantitative figures (numbers) you generated and specific device models to make your case. Feel free to make a table or chart or fancy list.

Submitting

Use a computer or paper/pencil. Please include a paper copy of your work in your folder.

Choose your context (or better yet, compare two!)

A pure science microbiology research laboratory
An e-commerce business like TruTech tools which sells quality HVAC diagnostic gear
A university like CCAC buying boxes for a general computer lab
An approved field of your interest

Extension ideas

Create a graph in your spreadsheet of choice depicting the change in price per computing cycle of your chosen models
Turn your graph into a graphic and use it in an infographic creator program online
Compute the size of room needed if one were filling it completely with IMSAI 8080 boxes to roughly equal the compute power of a modern hand-held computer of your choice. Document your calculations and visualize your result creatively so we can share it!