Nagered hard drive;'(..
in [Computer]
|
From: John Rumm on 25 Jun 2010 04:31 On 25/06/2010 07:34, Roland Perry wrote: > In message <kLadnTeW2aZIfb7RnZ2dnUVZ8rednZ2d(a)brightview.co.uk>, at > 23:48:40 on Thu, 24 Jun 2010, John Rumm <see.my.signature(a)nowhere.null> > remarked: >>> RLL Compression? Pray tell me more. >> >> Run length limiting. Using a RLL controller gained an extra 50% >> capacity, but at the expense of a reduced signal to noise and grater >> risk of unrecoverable read error. > > Other way round, surely? It was a coding method to make sure you never > had too many of the same polarity bits "in a row", which risks > unreadability. What makes a magnetic storage medium work is *changes* in > polarity. It possibly aided clock recovery by a small margin (but remember the data were MFM encoded anyway, which mostly eliminated the problem), but the boost in areal density is what lowered the reliability. Hence why many drives were not certified as compatible with RLL controllers even though the hardware interface was identical. There was a spell when folks would swap out controllers and "gain" extra storage capacity - but that came at a price of lower reliability. -- Cheers, John. /=================================================================\ | Internode Ltd - http://www.internode.co.uk | |-----------------------------------------------------------------| | John Rumm - john(at)internode(dot)co(dot)uk | \=================================================================/
From: baron on 25 Jun 2010 04:45 Roland Perry Inscribed thus: > In message <5b2dnXjjK-cQ6r7RnZ2dnUVZ8vGdnZ2d(a)brightview.co.uk>, at > 16:19:28 on Thu, 24 Jun 2010, John Rumm > <see.my.signature(a)nowhere.null> remarked: >>IIRC I bought my first HDD about '87. A "huge" 42MB seagate. > > By around 1981 I was the UK distributor for the Micropolis range of > drives, which were originally in the same form factor as an 8" floppy > drive. The most capacious was 33MB, and cost about the same a small > family car. > > One of my customers was the BBC newsroom, who bought one to store > digitised images to project behind the newsreader's head - to replace > the infamously unreliable slide projector they used to have. In those > days it was difficult to find people who thought they needed that much > storage (outside of a classic mainframe scenario). Couldn't resist, :-) I still have my first ever hard disk drive, purchased in 1983/4. It has four 5.25" platters and a linear stepper driven head. A whopping 5Mb. Well in was in those days ! It still functions, though its more of a showpiece nowadays. -- Best Regards: Baron.
From: baron on 25 Jun 2010 04:55 Andy Champ Inscribed thus: > dennis(a)home wrote: >> >> Most drives rotated quite slowly and it took a lot longer to read the >> data from a single drive than an array, the latency was the same. >> > > I'd argue that drives rotate more slowly now. > > No, really. They've gone from 3600 to 10000 RPM - but the capacity > has grown several orders of magnitude, so the speed is much lower in > proportion. > > Andy I would argue that that the speed has increased markedly simply because the storage density has improved dramatically. 15K spindle speeds and 360Gb plus on a single platter. Add to that on board two way cache, on some drives 32Mb, means that the bottleneck is moving back to how fast the mainboard circuits can handle the data stream. -- Best Regards: Baron.
From: baron on 25 Jun 2010 05:17 dennis(a)home Inscribed thus: > > > "Roland Perry" <roland(a)perry.co.uk> wrote in message > news:xiItDKI5iEJMFA2Y(a)perry.co.uk... > >> Somewhere I have an early "PROM", you programmed it by soldering >> diodes in, and the foot-square PCB probably has a couple of dozen >> bytes capacity. > > I had a piece of PROM where you sewed wires through the ferrite rings > to program it. I lost it years ago which is a shame as people couldn't > believe it was a bit of computer. A similar "core" based memory was used in a "Seeburg" Jukebox to store record selections in the late fifties early sixties. -- Best Regards: Baron.
From: Andy Dingley on 25 Jun 2010 05:48
On 24 June, 23:36, "Jeff Strickland" <crwlrj...(a)yahoo.com> wrote: > I remember drives that were sensitve to changes in orientation. I had a rack server a month or two back that was sensitive to whether it was the upper or lower slot in the rack. As always, it turned out to be thermal. Some muppet had lost an internal air deflector from the other machine, so it was running with a hot spot on the top of its case. Place that at the top and it was OK. Place that underneath and the drives in the server above got cooked. |