This article may be of interest to some of you out there:

https://www.bbc.com/news/science-environment-44554891

How times have changed!

-------

"Dr Sumner says the developments since then have been "astonishing". According to his calculations, the prototype machine that first ran in Manchester on 21 June 1948 could store 1024 bits of information, each "bit" being a zero or one.

In contrast, a smartphone today might have 64 gigabytes of storage, which is 500 million times as much.

The 1948 machine was fast for its time, working at a rate of roughly 1,000 instructions per second.

Yet a modern processor for a laptop or tablet might reach 20 or 30 million times this speed, says Dr Sumner, and the instructions involved would be much more complicated."

Here's why:

https://en.wikipedia.org/wiki/Moore%27s_law

In 1980 whilst I was with Telecom I was involved in the early ARE 11 telephone exchange switching conversions where a processor took ALL the dialled digits from subscribers & controlled the switching of the crossbar switching plant.

It had 2 data stores with one for customer data (SCS) and the other for switching/routing data (TRS).

Each store was duplicated and each had a "massive" memory capacity of 128kB all done via ferrite core logic on a rack that was 10' 6" high and about 36" wide. So these 2 racks held ALL the data required to switch a 20,000 line exchange.

Now 38 years later my Mac Book Air has 256Gb solid state hard drive is about 15mm thick x 320mm x 230mm & weighs nothing.

Computing has seen massive changes & there is no doubt it will continue to evolve even faster.

Lindsay

The Baby:

https://www.youtube.com/watch?v=cozcXiSSkwE

The development of the first hard disk.

https://www.youtube.com/watch?v=tFRBD2uN9PY

This was a huge breakthrough. Computing changed from batch mode to random access. Interesting to see the light-bulb moment - it really was one of those ideas where it's very hard to believe no-one had thought of it before.

Look at the SIZE of that thing! And consider that your smallest USB thumbdrive outperforms and outstores it a hundred times over!

Yes impressive for its day. Of course it was based on the valve computers that had been developed during WWII to determine the wheel settings on the Lorenz and Siemens encryption units for the German teleprinter networks. Colossus as it was known had several versions but they were still secret after 1945 so certain members of the development team could not say that they knew that a valve computer would work because they had done it before.

Perhaps it is not widely known that there was a lot of similar computer development being done by what is now the Australian Dept of Defence in Adelaide at the same time. In the mid 1980s when I was working for DSTO at Woomera, I had to set up a computerised file registery system which required that I go through a massive card index. I was surprised to find entries for several series of files on computer development and also radar. What was even more surprising was that these files were still sitting in the Registry but had not been accessed for more than 30 years. I then spent many hours reading through them and identifying those that I thought should be retained for historical interest. There was a cull of the archive after the registry system went into production but as far as I know some of these old files were retained. Most of the files were marked unclassified or just confidential so there was no security issues with the information.

Before electronic/valve computers there had been relay based logic calculators developed in the 1920s and '30s. Some of these followed sequential instructions stored in a 'register' so they could be 'programmed'. A development of this were gunnery control systems in WWII that took data from optical and later radar range finders, adjusted for encoded weapon and ammunition parameters, factored in weather conditions and then displayed range and bearing settings on dials on each gun.

Going further back to 1918, some automatic telephone exchanges in big cities such as London had a 'Director' unit that stored the number dialled, counted the first three digits dialled, determined which exchange the called number was connected to and then selected the appropriate trunk line to the required exchange. The Director used what we would now call a 'look-up table' in 'firmware' that was programmed to match a trunk link to the pattern of the three digit code. It was all done with relays and uniselectors in a box about the size of a three drawer filing cabinet. The later Crossbar switches previously mentioned by Kakadumh were just an improved version of the Director.

Andrew

Not surprised that the British did it; they must be the smartest country on earth: think of their intellectual greats, like Alan Turing. Wouldn't be surprised if England win the soccer game today because soccer is, in the end, a game of brains (even though the Europeans don't like us using the word "soccer", we'll do it anyway.)

The gunnery control devices mentioned by Relayautomatic were analogue computers full of operational amplifiers performing addition, multiplication etc. Their setting up included having to adjust power supplies and amplifiers zero balance with a series of colour coded knobs and screwdriver slot potentiometers, the sequence remembered by:

Red and white, then the blues, wait a minute and do the screws.

Harold

Ian Robertson. That video is interesting but not the first hard drive. CSIRAC had one in 1949! https://en.wikipedia.org/wiki/CSIRAC

Yes but the CSIRAC device was not random access. Random access was the breakthrough.

Having looked at that entry on CSIRAC, I wouldn't mind the task of getting it running again!