why do computer scientists say 1KB=1024 bytes?!!

You don't have a need to know how that part of the computing biz is working? It is a pretty bad bias that you're incorporating into your computing usage. I stayed with AOL just to observe what most uneducated users were experiencing. In that way I can extrapolate computing needs and usage 5-10 years from now.

/BAH

Somebody threw a crank into the works.

/BAH

Yup. DUNGEOn. That has also been resurrected. Another one, where I made absolutely no headway (IOW, I played for fifteen minutes and gave up) was HAUNT.

Now that brings back memories about my workshop. The Coke was a second and third generation phenomena.

I didn't get far in that game either. OS stuff was much more intriguing.

/BAH

This happens to be a newsgroup law of computing. Any post correcting spelling will contain another misspelled word.

Are you sure there is no u? Doesn't pneumonic...or I see what my gray matter did. Typically, my brain damage can't see the word in my dictionary even though I know I must be staring at it. You would never believe that I was The One to tell everybody else how to spell when we were spec'ing.

Thanks for the correction. :-)

/BAH

It's time to reexamine my assumptions. As I wrote that post, it struck me that I was making a similar assumption. However, I've been collecting knowledge since I was a child. I didn't meet a science fiction book until I was

1. My memory could be scrambled and the story influenced me.

/BAH

And if you're naked, boys can count to 2,097,151. ;-P

Eh?

My aversion to Intel and Maxtor is through personal experience and overwhelming ancedotal evidence from other people. Not because of some ignorant bias, like what you seem to be implying.

It *isn't* overhead though!

I'm trying to be real clear about this. I am not confused about what overhead is, nor am I confused about hidden files. I am also not saying you are confused, it is just that you are totally correct about the wrong thing! :P

All true, but irrelevant to the situation at hand. :)

My O/S measures and addresses everything in bytes, kilobytes, megabytes, gigabytes.

All of these quantities, as you know, are multiples of 1024.

The drives are sold, though, using the pseudo-SI measurments of 'mebbibytes' and other fantastically annoying measurments that completely defines the phrase "close, but not quite".

One 'mebbibyte' or however the f*ck you spell it is 1x10^6 bytes, not

This is all explained in the drive's fine print, but still annoys the piss out of me because I always feel slightly cheated.

In my understanding, modern drives have a little bit of space reserved [but unrelated to the marketed total] set aside for taking the place of the occasional bad cluster.

...

Speaking of being "right" about the "wrong" thing, you've got it exactly back-asswards.

A "kibibyte" is 1024 bytes, and so on. The ones with the "b" in the middle are the binary ones, multiples of 1024 decimal. (10000000000 binary). The marketing ploy is to take a drive that actually has 279 * 1,073,741,824 (decimal) bytes of storage, and say it has 300 * 1,000,000,000 (decimal) bytes. So when they sell you a "300 MeGaByte" drive, they're selling you a "279 MeBiByte" drive.

Non-geeks don't spot it, and, like audiophools, spend more money per byte than geeks do. ;-P

Cheers! Rich

Try Baf's Guide to the IF Archive:

Not only are people still playing old interactive fiction games, they're still writing new ones. The IFComp attracts dozens of entries each year:

-- Ben

Ah, the difference between men and women.

Do you think the development of GUI was driven by men's never ending desire to get better pinups? If women were in charge, would we still be using TTY interfaces? Not that there's anything wrong with that, I am currently working on making yet another port (3rd or 4th) of the 25 year old Apple ][ text adventure game THE WONDERFUL WORLD OF EAMON to Python so I can still play it without mucking about with an emulator.

Of course, I'm somewhat nostalgic having written my first program with BASIC on a Teletype. It was interesting watching the Teletype s..l..o..w..l..y print out the results of your MIRV warhead cluster detonating over the computer's cities and waiting for it's retaliation. It was so slow you could add up the numbers in your head before the totals were printed to see who won that exchange (and whether 3 independently targeted 1 megaton warheads were better in the long run than single 5-10 megaton warheads). But the novelty of that wore off quickly.

The basis of the internet. Almost all RFCs use octet instead of byte:

RFC791 - IP RFC1332 - PPP RFC1731 - IMAP RFC1962 - SOCKS RFC2251 - LDAP RFC2279 - UTF-8

etc...

Nothing mainly because tender documents and specifications come in printed form on thick bundles of paper. But it makes the client's lawyers happy.

Also all of IEC and ITU documents and specifications.

IEC 60870-5-101 protocol specification for example does not contain the word 'byte'. All data objects are defined as multiples of octets in size.

Nope. Not overflow. Parity check bit.

/BAH

Reread what I wrote. You ensure that your bias remain ignorant. Intel is contantly working on fixing bugs, aspects, and new architectures or twiddles of the current architecture. If you don't play with it occasionally, you won't know if what bit you badly has been fixed or still exists. If you expect any of your work to be used or transferred to or through an Intel architecture platform, then it is wise and frugal to check out the competition every once in a while.

I can't think of any computing that won't be put on or thru an Intel these days.

/BAH

The

going

from

use

I'm correct about the wrong thing very often. You seem to be a tad confused below.

So? Substitute bytes for blocks.

They aren't using any SI. And the numbers have always been that close, but not quite.

You should try to think in bits not bytes. A byte is a variable length, which is increasing over time as more bit patterns in the standards are needed.

Do you always feel slightly cheated when you buy a car and the estimated mileage/gallon doesn't exactly match.

Think about how you would keep track of bad spots on the disk. Then think about how you, the operating system, would prevent itself from writing to that _physical_ spot.

When you go through this design process, solving all of the holes in each iteration of your design, you will probably produce an equivalent of what the olde bit gods designed.

In some things, the old-fashioned way is the only way. The reason for this is that you cannot change physical properties on a whim.

/BAH

Yup. The most pleasurable thing about today's games is the mapping part while solving the puzzles. The thinking and approaches used to do this activity is very similar to operating system work. She should have cut out the maze parts, and let you work at it.

Working at the maze for a couple of days seems a tad long. What kinds of approaches did you use? Or were you simply stuck using the brute force method without making paper notes?

/BAH

Yup. Now I have to warn you that I keep confusing the requirements between DUNGEOn and HAUNT. One of them needs MDL (pronounced muddle). Some blessed soul has done the work to get that available.

I'm hittemal challenged. I can't give a precise address. The way most people are getting it installed is by first installing the hardware emulator, then the operating system (this part of the game can take lots of enjoyable time), then the game.

It would because it took up huge amounts of a large PDP-10 system resources. It came patched so that it could only be run non-prime time...unless you were the wizard and knew the password :-).

Last spring (I think) some guy resurrected that one and couldn't get it to work until I told him to SET TTY TTY. He had a fine time because he also tried to debug it using DDT which automatically caused a SET TTY TTY. I always think of Heisenburg whenever that happened to us.

Good lord! I'd hire him.

Oh, it was very good pre-training.

Boo-hoo. Nobody ever made a male pinup. It was a couple years later (after that printer paper phase) that Burt Reynolds was the centerfold. It caused a huge sensation in my group which was 100% female at the time. Even then they wouldn't let me look and I was too cheap to buy a magazine.

/BAH

I was astonished the other day, having been following this thread, I checked to see what processor my $300.00 e-bay computer has; it seems slow, and I thought I'd blame Intel, but imagine my surprise:

$ cat /proc/cpuinfo processor : 0 vendor_id : AuthenticAMD cpu family : 6 model : 4 model name : AMD Athlon(tm) Processor stepping : 4 cpu MHz : 996.487 cache size : 256 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 1 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov pat pse36 mmx fxsr syscall mmxext 3dnowext 3dnow bogomips : 1985.74

Maybe a fxxxing gigaherts isn't fast enough for today's bloatware? I remember installing Slackware 3, back in the late 1990's, including X, and it was just really lean and clean and quick - really snappy response.

On a Cyrix 6X86-P150.

What happened? Howcome everything is so blasted slow that you need a processor ten to a hundred times faster than we used to have, just to type a message to a newsgroup?

(in case you're wondering, a "bogomip" is a "bogus MIPS" - it's just an ordinary timing loop that doesn't account for caching or arithmetic or anything.)

Thanks, Rich