poynton%vector@Sun.COM (Charles Poynton) (11/06/88)
I propose some nomenclature which might reduce the incidence of confusion about data rates which has surrounded the recent SCSI discussions on comp.sys.next. First, international standards. Upper case "M" is the symbol for mega, a multiplier of 1000000. Little-m is the symbol for milli, 1/1000, and care should be taken not to confuse it with mega. Lower-case "k" is the internationally-standardized symbol for kilo, a multiplier of 1000. Its being lower case is the exception: all other multipliers greater than unity have upper-case symbols, and ALL multipliers less than unity have lower-case symbols. [The Greek symbol micro is for a multiplier of 1/1000000; but the letter "u" can be substituted without confusion.] Lower-case "s" is the symbol for second: upper case unit symbols are reserved for units named after individuals, such as "S" for Mr Siemen, "Hz" for Mr Hertz, or "B" for the unit bel [sic], named after Mr Bell. The full name of a unit itself is NOT capitalized whether derived from a proper name or not. Proper form places a space between the numeric value and the multiplier, no space between the multiplier and the base unit, and no hyphens anywhere. Now, my proposals. I suggest adopting upper-case K for 1024. This does not conflict with any international symbols. The use of this upper-case letter is somewhat mnemonic, in that the 1 K is a little larger than 1 k. [The unit symbol K is for kelvin, the unit of absolute temperature, but this is a unit and not a multiplier symbol so there's no conflict.] I suggest using lower-case "b" for bit, and upper-case "B" for Byte. Neither conflicts with any metric symbols. I use "w" for "word", whose width is context-dependent. [Lower case distinguishes it from the weber, symbol Wb.] For those who measure data communications rates, all but the most esoteric uses of the unit "baud" should more properly use bits per second, symbol "b/s". All conventional data communications rates are related to 1000 [and not to 1024], for example, 9.6 kb/s. Big-M is problematic. My custom is to use M for both 1000000 and 1024*1024, and let context determine the interpretation. Little-m, milli, should certainly not be confused with either of these. Rates are nearly always referenced to decimal mulipliers (for example, SCSI 4.8 MB/s, audio 44.1 Mw/s). Multiples of 1024 are generally restricted to describing capacity. Charles Poynton poynton@sun.com
snoopy@sopwith.UUCP (Snoopy T. Beagle) (11/10/88)
In article <76349@sun.uucp> poynton%vector@Sun.COM (Charles Poynton) writes: |Upper case "M" is the symbol for mega, a multiplier of 1000000. Little-m |is the symbol for milli, 1/1000, and care should be taken not to confuse |it with mega. Now if you can just get the people who label capacitors to comply. _____ /_____\ Snoopy "Look Marge, I found a 47 Mega-Farad capacitor!" /_______\ |___| tektronix!tekecs!sopwith!snoopy |___| sun!nosun!illian!sopwith!snoopy
claudio@forty2.UUCP (Claudio Nieder) (11/10/88)
In article <76349@sun.uucp| poynton%vector@Sun.COM (Charles Poynton) writes: |I propose some nomenclature which might reduce the incidence of confusion |about data rates which has surrounded the recent SCSI discussions on |comp.sys.next. | |First, international standards. | |Upper case "M" is the symbol for mega, a multiplier of 1000000. Little-m |is the symbol for milli, 1/1000, and care should be taken not to confuse |it with mega. | | .. stuff deleted .. | |I suggest adopting upper-case K for 1024. This does not conflict with any | | .. stuff deleted .. | Rates are nearly |always referenced to decimal mulipliers (for example, SCSI 4.8 MB/s, audio |44.1 Mw/s). Multiples of 1024 are generally restricted to describing |capacity. So why not start to forget about those 1024 and 1024*1024 units ?? claudio UUCP: claudio@forty2.uucp BITNET: K538912@CZHRZU1A Mail: Claudio Nieder, Kanalweg 1, CH-8610 Uster
mark@adec23.UUCP (Mark Salyzyn) (11/11/88)
In article <55@sopwith.UUCP>, snoopy@sopwith.UUCP (Snoopy T. Beagle) writes: > Now if you can just get the people who label capacitors to comply. > > _____ > /_____\ Snoopy "Look Marge, I found a 47 Mega-Farad capacitor!" > /_______\ > |___| tektronix!tekecs!sopwith!snoopy > |___| sun!nosun!illian!sopwith!snoopy I don't believe it :-). Most of the capacitor makers label their capacitors with the colour code standard or the numbers associated with the colour code standard. eg yellow blue yellow (or 474) is 470000 uF. The other method used is the European standard of using the muliplier as the point placement. eg. 4n7 is 4700 uF or 4.7 nF. I have seen a few (when space permits) to put the units in their entirety. I have seen non that have used upper case letter (keep in mind that a small upper case `m' (smaller than other letters and numbers) is a LOWER case letter). Does any one know the capacitance of a 60Ah battery, or is it not possible to measure it ?!:-) -- Mark Salyzyn @ mark%adec23@nexus.ca
tgg@otter.hple.hp.com (Tom Gardner) (11/11/88)
FLAME ON While we are discussing unit nomenclature, here is my pet peeve. What is the international standard symbol for seconds? If you think it is "S", then what is the standard symbol for conductance? (hint: the unit is a Siemen.) Right. The symbol for Siemens is "S". The unit for seconds is, wait for it, "s". THAT'S RIGHT, lower case ess. So the next time you see a magazine article which, for example, quotes a RAM speed as 125nS, you may well begin to wonder whether they really understand what they are writing. And then there are the instruments that get it wrong on their front panel. Would you buy a car which had its speedometer calibrated in Hertz? FLAME OFF I feel better now. Tom Gardner -- usual disclaimers