Slashdot Mirror
Seagate Ships Billionth Hard Drive
Lucas123 writes "Seagate's first drive, shipped in 1979 was the ST506, which had a capacity of 5MB and cost a cool $1,500 — or $300 per megabyte. Today, a typical Seagate holds 1TB and cost just 1/5000th of a cent ($0.0002) per megabyte. Seagate, which claims to be the first company to ship a billion drives, says all those drives amounted to 79 million terabytes of capacity, enough for 158 billion hours of digital video or 1.2 trillion hours of MP3 songs." Update: 04/23 14:56 GMT by CT : The quoted fraction is wrong. Someone complain to ComputerWorld. Update: 04/23 15:13 GMT by CT : TY. The site is corrected to say "just 1/50th of a cent ($0.0002) per megabyte." The universal equation is once again balanced.
It's funny how it always seems as if the next drive we purchase offers virtually limitless and impossible to use storage space but is never really enough.
While todays hard drives may be much larger, its not going to be long before we move on. I remember when I got my first 100mb HD and thinking "wow this is it ill never need any more storage than this". But now we know that as HD capacity increases so will the features and size software and media. Think of how big the first windows distro was and how big Vista is. Soon we'll all have HD DVD rips and real life quality music filling our new 100TB HDs
;)
In short, we as consumers don't need to worry about how to use this multitude of ever expanding space; software and media companies will do it for us.
You're right. The ST506 was full-height, (remember the squeaky monkey-like noise it made?) the ST225 was half-height - somewhere in my basement, I still have an ST225 I paid $250 for.
No folly is more costly than the folly of intolerant idealism. - Winston Churchill
158 billion hours is a shade over 18 million years. If you had a camera fixed to record for the past 18 million years you'd only have started in the Miocene era ... it'd all look really quite modern. It'd have been a bit more grassy, but there'd be recognisable mammals like deer and wolves, birds like ducks and grouse.
It sounds a like long time, but it really isn't.
http://twitter.com/onion2k
Everybody I know has some vendor they swear by, and some vendor they think is just terrible. I know people who think Western Digital is the best, and that Maxtor is crap. I know people who say the exact opposite. None of these people buy enough hard drives to have any real say in which one is better than the other. Google probably buys enough drives, but they don't buy the consumer level desktop drives either, so I don't know if I'd trust their opinion much either.
Anthropic principle: We see the universe the way it is because if it were different we would not be here to see it.
Yeah, but nobody expected to be ripping entire DVD collections to HD for use in a media server back then. Now it's economical to do it, especially if you've already got a PC doing DVR work. At less than 20c/GB, it's less than a dollar to rip a typical movie (without the extras and ads) to a server. That compares favorably to putting discs into a jukebox, and has the advantage of speed and playing multiple streams at once.
Now that HD content is out, we need the capacities to go up another order of magnitude so that storing HD is as easy*/cheap as SD.
*I buy discs, but download the rips. My setup is only 720p, so it's easier to get someone else's recode at 720p than do it myself, and it takes less space on my server. With 2TB in DVDs and recorded content off TiVo/OTA, I'm always worried about bumping into the limit on my unRaid box and having to buy more drives.
Is it just my observation, or are there way too many stupid people in the world?
Just thinking about how much of that storage is filled by redundant data blows my mind. It seems like such an inefficient structure. Imagine how much could be saved if there was only one copy of each song (lossless, why not), each movie, etc, and instead the trillions of dollars spent on storage, we spent slightly less trillions to build up massive networking infrastructure and a few server farms that make it all accessible on the fly. Obviously unrealistic, but a fascinating idea. I have approaching 2.5TB of media at home, but the vast majority of it just sits there essentially never used. I only need it locally because my home network has the bandwidth to access it whenever I want. But even so, I only use a small part of that bandwidth an hour or two a day at most. Getting rid of redundant storage could realistically reduce storage needs 99% (ever see a torrent with 100 seeds? All the time), and bandwidth consumption wouldn't be too many times greater (by some measures) than YouTube uses, because by far most of the time we aren't consuming highly dense media. You'd need a world with completely free culture, though. Just a thought.
Their lawyers must work out the royalties, but consumers get a very nice copyright exemption. Dunno about P2P, but it might also be covered.
It's also important to remember inflation. $1,500 in 1979 is over $4700 today, so the cost was more like $1K per megabyte.
Karma: -2147483648 (Mostly affected by integer overflow)
... we called a 5.25 hard drive a "mini-winnie" since the established 8 inch hard drive at that time was called a Winchester .
Back then the two CP/M Z-80 "micro computers" at university lab where I did my class work used 8 inch floppies. Real floppy disk Users dismissed mini floppies not only because of it's paltry storage capacity but because some pinhead decided to reduce the disk rotation speed of the mini floppy by one half thus reducing its data transmission rate. At least that's how I remember it.
Some other graybeard is gonna have to take over for me now cuz I gotta go chase some kids off my lawn...
"Where's my other sock?" - A. Einstein
And it had the passive backplane (which meant that the processor was on an ISA card and plugged into one slot, and the RAM was on another ISA card and plugged into another slot)!
And the full-length HDD/FDD/serial port card (WTF?) had not just one but *two* monster ribbon cables connecting to the hard drives in order to achieve the staggering data throughput of, nearly, a megabyte a second! Beat that, SATA!
Mine ended up getting skipped. I wish I'd known how much in demand they are now, I'd have kept it...
Man, I remember in the early 90's being given the manual for the ST506 drive controller so I could write the "bare metal" interface to actually write the drivers for an OS my prof was writing for his research work.
:-P
Pretty cool shit, push bytes into a couple of registers to make the damned thing seek to a given track. Service the interrupt. Push in a couple of other bytes to cause a sector read. Service the interrupt. It didn't get any lower-level than that.
We specifically avoided the Linux code at the time since we didn't want to GPL our code or use their implementation.
Writing my own low-level device driver for accessing hard-drives was pretty cool. Before long, I had written a bunch of the simple UNIX command-tools for DOS -- ls, rm, cat, cp. Boot out the DOS handler, read the raw FAT data off the HD, format it, and interpret it.
*sigh* Anyway, this is apropos to nothing. Just waxing nostalgic about a university project 20 odd years ago. It's all been downhill from then.
Lost at C:>. Found at C.