I just purchased two 3TB USB 3 external drives at a great price. I am going to extract the drives for NAS use.
In doing this, I became curious as to the capacity evolution and how the ability to store so much data can be done and still stay in a standard 3.5″ form factor? Back in the day, a massive 40GB drive was the same shape and size (although the connections differ, IDE vs SATA).
I understand that the head manipulates a small magnetic field on the platter and the basics of how HDDs work. I do not know what has changed to allow the same principal to apply and have greater gains in capacity.
So how is it that I can now store so much more data on platters and everything still fits in a 3.5″ form factor (or large capacity in a 2.5″ for that matter)?
To record more data in the same size package requires chiefly, higher bit density. More bits per track, and more tracks per inch. Perpendicular recording allows the bits to take up much less space (probably 10% of the space required for previous methods). But advances in micro-mechanics, microelectronics, magnetics, and signal processing are chiefly responsible for the current day bit and track densities.
The first hard drive I owned (circa late 1970’s) had a capacity of a whopping 5MB (yes Megabytes). I believe the specs were: 2 platters – 2 sides (4 surfaces), 300 tracks per surface, 17 sectors per track, 256 bytes per sector (4 * 300 * 17 * 256 = 5,222,400 Bytes).
That’s about 120 tracks per inch. Compare that to present day 236,000 tracks per inch and you can see where the increased capacity comes from.
It was a full size 5.25 inch form factor “bare” (internal) drive … The same width and height as a stack of 2 typical present day internal DVD drives, with a length (depth) of about 30% longer (slightly smaller than 6 x 3.5 x 10 inches). It weighed about what a full size HD-Video laptop weighs today.
At the time, it was pre-production, for developmental use. Definitely state-of-the-art at the time. It was manufactured by Shugart Associates, and was packaged inside a full-size Samsonite briefcase. Complete with cables, documentation, and SASI host adapter circuit board (SASI was predecessor to SCSI). The case had a “foam rubber” insert with 2 “cutouts”, 1 for the drive and 1 for everything else (the SASI host adapter circuit board was about 8 x 10 inches).
The whole kit was marketed as “Shugart’s Case for Success”. I don’t remember how much it cost, but I’m sure it was probably at least $1000.00 or perhaps $1000.00’s. I don’t remember how much it cost because it didn’t cost me anything… I won it as a “door prize” at a seminar that was hosted by a Shugart distributor. It is the only thing of any significance I have ever won. Pretty sad (5MB hard drive) by today’s standards.
There’s no one technology that has made this possible. There’s a lot of different technologies that have. Its pretty hard to pin down one but a few examples of technologies that made a huge difference in data density would be perpendicular recording and significantly better head tracking made possible by more sensitive micro electronic devices through the use of giant magnetotesistance. They’re basically making individual bits smaller and detecting them better.
While we’re pretty close to the current theoratical limits on data density, new technology like HAMR will probably keep improving storage density.
In addition to the technologies mentioned above, I would like to add “Analog signal processing” and “A/D converters” that has been tremendously enhanced over the past 2 decades.
If you think about it, HDD still rotates with the same speed and for comparable tracks density (that becomes much harder to increase), the only solution for you becomes to record more data in the same track.
This means that you should be able to induce and detect much faster magnetic field variations. This is only possible by faster analog electronics and faster A/D converters.
To quote from tom’sHARDWARE a partial answer may be:
“After one drive however, Seagate developed a 2 GB drive with dual actuators. The drive had an actuator mechanism at opposing corners which theoretically provided double the throughput and operated independently of each other. The drive had an RPM rate of 7200 RPM and had a track density of 3000 tracks per inch.
For comparison sake, the new Cheetah 15K.7 has a track density of 165,000 tracks per inch and Seagate’s latest 2 TB drive has 236,000 tracks per inch.”