aegl@root.co.uk (Tony Luck) (08/25/87)
Instead of making the disks go faster (and then having to fill them with Hydrogen - or run them in vacuum flying the heads magnetically or electrostatically or whatever to avoid turbulence problems) would it not be possible to put more than one set of heads on a drive - then you could reduce rotational delay by only having to wait for the disk to turn to the nearest set of heads. Has anybody ever tried it, or does it make the whole drive ridiculously expensive. Tony Luck
rcd@ico.UUCP (Dick Dunn) (08/27/87)
In article <416@root44.co.uk>, aegl@root.co.uk (Tony Luck) writes: > Instead of making the disks go faster (and then having to fill them > with Hydrogen - or run them in vacuum flying the heads magnetically or > electrostatically or whatever to avoid turbulence problems)... Turbulence is more a problem during takeoff and landing than "in flight" and anyway, current disk heads fly at well below the speed of sound. >...would it > not be possible to put more than one set of heads on a drive - then > you could reduce rotational delay by only having to wait for the > disk to turn to the nearest set of heads. Has anybody ever tried it, > or does it make the whole drive ridiculously expensive. It was done on the IBM 3380--there are two separate actuators. This not only has the potential to reduce rotational latency; it lets you service two requests for the same surface simultaneously. I don't know what part of these capabilities are actually realized by the firmware, tho. It was also done on some fairly old disks ('60's vintage) although I'm not sure whether they had multiple sets of movable heads or only multiple groups of fixed heads. -- Dick Dunn {hao,nbires,cbosgd}!ico!rcd (NOT CSNET!) (303)449-2870 ...Never offend with style when you can offend with substance.