home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!ogicse!hp-cv!sdd.hp.com!cs.utexas.edu!convex!convex!patrick
- From: patrick@convex.COM (Patrick F. McGehearty)
- Newsgroups: comp.benchmarks
- Subject: Re: Disk storage size?
- Message-ID: <1992Aug19.172246.8264@news.eng.convex.com>
- Date: 19 Aug 92 17:22:46 GMT
- Article-I.D.: news.1992Aug19.172246.8264
- References: <0001@umd.edu> <1992Aug19.152525.15741@ornl.gov>
- Sender: usenet@news.eng.convex.com (news access account)
- Reply-To: patrick@convex.COM (Patrick F. McGehearty)
- Organization: Engineering, CONVEX Computer Corp., Richardson, Tx., USA
- Lines: 43
- Originator: patrick@mozart.convex.com
- Nntp-Posting-Host: mozart.convex.com
- X-Disclaimer: This message was written by a user at CONVEX Computer
- Corp. The opinions expressed are those of the user and
- not necessarily those of CONVEX.
-
- >In article <0001@umd.edu>, pc@umd.edu writes:
- >>When people talk about 200MB disk, do they mean 200x1,000,000 byte, or
- >>200x1024x1024 ? When talk about the RAM, people seems use 1k=1024, but
- >>when go to disk, it is metric. Is this always the case?
- >
- ...stuff deleted
- >In article <1992Aug19.152525.15741@ornl.gov> de5@ORNL.GOV (Dave Sill) writes:
- ...stuff deleted
- >
- >In practice, as when calculating transfer rates or repartitioning a disk, I'm
- >careful to calculate in terms of true K, M, and G, but when someone asks me how
- >big a file is or how much space is left on a partition, I'm sloppier and
- >sometimes use K as an abbreviation for "thousand", etc.
- >
- >Bottom line: if it matters, make sure it's clear which meaning is being used.
- >
- Ironic point. From your posting, I can't tell whether you consider
- metric or binary to be "true K, M, and G". Just confirms the
- ambiquity. When I can't tell which type of mega a user means, I
- try to assume it was whatever is most favorable to the user of the term.
-
- I have noticed that almost anything marketing or sales oriented uses metric
- Mega and Giga terms. To do otherwise is to give up 5 or 7% apparent
- relative value (recalling 1024*1024=1,048,576, and 1024**3=1,073,741,824).
- This has not happened in RAMs because they are almost always sold in
- increments which are a power of 2 or a small integer constant times a power
- of two. So most people can tell that 1 Mbyte of RAM is the same as 1.047
- Mbytes of RAM. Disks are not forced to be so regular.
-
- The case of nominal disk space is made even worse by another practice. The
- difference between unformatted and formatted disks is quite large. A nomial
- 1.1 Gbyte disk (unformatted, G=10**9) might convert to 0.91 Gbytes of
- formatted space. The Berkeley Unix fast file system traditionally reserves
- 10% of the space to insure reasonable performance and minimize
- fragmentation. (Most OS's use some disk space for their own purposes, I
- just happen to be most familar with Berkeley Unix). So a user only gets
- 0.82 Gbytes (G=10**9) or 0.75 Gbytes (G=2**30) of real usable disk space,
- considerable less than he might have expected. Very much a buyer beware
- scenario. Of course, if you can buy twice as much for half the price
- next year, it may not hurt as much as it would if disk technology were
- on a slower growth curve.
-
- - Patrick McGehearty
-