That is indeed an interesting article; thank you for bringing it to my attention.
However, I do have a few issues with some of the comments made in that article.
Re: para. 2: "While it is easy to buy the parts of a modern PC - motherboard, graphics card and processor - the sheer complexity would defeat any attempt to build one starting with electronic components such as resistors, capacitors and chips".
Now although I would not disagree with that comment, simply obtaining the parts for a modern computer is the relatively simple part of such an exercise. It is having the technical knowledge of which particular parts to obtain and how to get them all efficiently to work together that is the main challenge. This has been over-looked, despite the fact that it was just as relevant in years gone by as it is now.
Re: para. 4: "Is it still possible to construct an 8-bit machine from a pile of parts?"
Probably: the SN74181 (or a variant) is still available, as are 1k x 1-bit DRAMs and the SN7489 64-bit static memory chip. These are the key components of a discrete processing unit; the other parts can be basic SN74xx combinational and sequential logic elements.
Quote: "The difference between a pile of parts and a working computer is solder - the metal alloy used to connect components to circuit boards."
Although not untrue, it misses the essential difference: a computer must have a program to make it 'do something'. A "pile of (electronic) parts" can be put together in such a manner that all the hardware is present that comprises a computer, but without a program to run - and run without crashing - it remains nothing more than
a "pile of parts".
The policy of fitting in small parts first, then the bigger parts around them seems to be contradictory to anyone who has assembled individual electronic parts onto a form of assembly board. Advance planning is needed and it is nearly always the case that the larger parts will require location first, simply because it is much easier to fit small parts around large ones than vice-versa.
And finally, the comment from Mr Honniball: "Half of the stuff here is to frighten the circuit boards into working: they know the treatment they it will get if they do not work properly!" induces me to comment that if he has staff who report to him, I can only hope that he restricts that attitude to the hardware he comes into contact with!
In conclusion, an interesting article and it did indeed bring back many memories of working with early computer systems that used punched paper tape, several rows of 'bit-switches', custom-designed keyboards and monochrome (green screen) monitors.
Al.
However, I do have a few issues with some of the comments made in that article.
Re: para. 2: "While it is easy to buy the parts of a modern PC - motherboard, graphics card and processor - the sheer complexity would defeat any attempt to build one starting with electronic components such as resistors, capacitors and chips".
Now although I would not disagree with that comment, simply obtaining the parts for a modern computer is the relatively simple part of such an exercise. It is having the technical knowledge of which particular parts to obtain and how to get them all efficiently to work together that is the main challenge. This has been over-looked, despite the fact that it was just as relevant in years gone by as it is now.
Re: para. 4: "Is it still possible to construct an 8-bit machine from a pile of parts?"
Probably: the SN74181 (or a variant) is still available, as are 1k x 1-bit DRAMs and the SN7489 64-bit static memory chip. These are the key components of a discrete processing unit; the other parts can be basic SN74xx combinational and sequential logic elements.
Quote: "The difference between a pile of parts and a working computer is solder - the metal alloy used to connect components to circuit boards."
Although not untrue, it misses the essential difference: a computer must have a program to make it 'do something'. A "pile of (electronic) parts" can be put together in such a manner that all the hardware is present that comprises a computer, but without a program to run - and run without crashing - it remains nothing more than
a "pile of parts".
The policy of fitting in small parts first, then the bigger parts around them seems to be contradictory to anyone who has assembled individual electronic parts onto a form of assembly board. Advance planning is needed and it is nearly always the case that the larger parts will require location first, simply because it is much easier to fit small parts around large ones than vice-versa.
And finally, the comment from Mr Honniball: "Half of the stuff here is to frighten the circuit boards into working: they know the treatment they it will get if they do not work properly!" induces me to comment that if he has staff who report to him, I can only hope that he restricts that attitude to the hardware he comes into contact with!
In conclusion, an interesting article and it did indeed bring back many memories of working with early computer systems that used punched paper tape, several rows of 'bit-switches', custom-designed keyboards and monochrome (green screen) monitors.
Al.
