|
Week of 12 July 1999 Sunday, 18 July 1999 10:50 A (mostly) daily journal of the trials, tribulations, and random observations of Robert Bruce Thompson, a writer of computer books. |
|
|
Monday, 12 July 1999 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week] I've been reading a fascinating book, The Wealth and Poverty of Nations: Why Some Are So Rich and Some So Poor by David S. Landes. When I checked reviews on Amazon.com, I found that there were 45 of them, and the book overall got only three stars. That's skewed, though, by the Politically Correct brigade, who have come down on this book with their usual unreasoned shrillness. "We don't like it, so it can't be right" is always their argument, and that's the case here. I don't read much non-computer-related non-fiction--probably only one or two books a month--but I made an exception in this case. This is a thick book, but it's well worth your while to read. Pick it up at the library. Better still, buy a copy of your own. It's available in hardback and trade paperback (like a computer book). If you want to buy a copy, you can help support this site by ordering it from Amazon.com. Click here to buy a hardback copy, or here to buy a paperback copy. * * * * * I just noticed a subtle change on the Northern Light search engine yesterday. They used to provide 25 hits/page--not as good as hotbot.com and snap.com, which allow you to specify up to 100 hits/page, but still better than search engines like AltaVista which provide only ten hits/page. Now, Northern Light has gone to 10 hits/page. I send email to them protesting this change, and suggesting that they go the HotBot/Snap route by defaulting to 10, but allowing up to 100. Presumably, this change was made in pursuit of inflating page-read counts. As anyone knows who's visited Tom's Hardware, this artificial inflation of page counts has gotten completely out of hand. On Tom's Hardware, a "page" is often just a paragraph or two. That's one reason I don't bother to visit Tom's Hardware much any more. That and that fact that the content is seldom updated and often just plain wrong. * * * * * Don't forget the Season Finale of Buffy the Vampire Slayer tomorrow night. We're having a Buffy party for it. * * * * * This from ROBERT RUDZKI [rasterho@pacbell.net]:
Well, 60 Minutes has no credibility as far as I'm concerned. I haven't watched it in twenty years. I wouldn't even take their word for it that these children had died, let alone anything they claim about the particulars. These are the people who record an interview and then come back and re-record Mike Wallace et alia asking different questions than they asked during the interview and then patch in the responses made by their victims. That's not news. It's entertainment of a particularly disgusting sort. I have never used Abit motherboards, but they certainly have a good reputation. Generally when a system runs properly under Win9X and blows up under NT, the problem is memory. NT is much more rigorous about memory than Win9X. If you really load up the system under Win9X (so as to occupy all physical memory), you may find that problems occur there, too. The problem may be the memory itself, but it also may be a power problem. The first thing I'd suspect is the DIMMs. A lot of memory represented as "PC100" (or even so labeled) ain't. One often sees "100 MHz" memory sold as PC100 (which actually must be 125 MHz or better). One also often sees "10 ns" memory sold as PC100, when actual PC100 is 6 ns or 7 ns. I use Micron/Crucial memory in every system I build, and I've never had a problem with it. I can't say the same for commodity memory, some of which I keep around for testing purposes. The only thing that makes me think perhaps it isn't the memory is the fact that you're having problems with the FSB running at 66 MHz. Nearly any SDRAM DIMM sold today should function properly on a 66 MHz FSB. The next suspect is the power supply. Cheap power supplies are often run at (or even above) their component ratings, and can't provide adequate well-regulated power to the memory. Memory likes higher voltage (within spec, of course) and cooler temperatures. If your power supply isn't providing sufficient ripple-free voltage to the memory, that may be the problem. If heat is the cause, you may find that the system runs properly for a few minutes or a few hours and then becomes unstable. Even then, though, it may be the power supply rather than the memory itself that's causing the problem. I've seen cheap power supplies that provided reasonably clean power when they were first turned on. When they heated up, they no longer delivered usable power. Again, that's why I use PC Power & Cooling power supplies when I build a system.
|
|
|
Tuesday, 13 July 1999 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week] Here's some good news. PC Hardware in a Nutshell (that book I'm still writing) is apparently selling very well over at Barnes & Noble. I did a quick search for all books beginning "pc hardware". That search found eight matches, and my book is number four of eight. And numbers five through eight aren't straw men, either. They're all showing as "In Stock: Ships within 24 hours." Not bad for a book that isn't finished yet. I guess that may mean that book will eventually sell well. * * * * * This exchange between Paul Robichaux [paul@robichaux.net] and me started yesterday morning. (For the record, I should emphasize that Paul and I are co-authoring a book, are both on Microsoft's W2K beta list--and so are both entitled to this product--and do not casually pirate software)
Thanks. Barbara just found mine at the front door a few minutes ago. I was about to mail you to ask if you needed a copy. As usual, I have packing fluff all over me. "Tear here" it says. So I tear. Problem is, the damn glue still seals the envelope flap. So I rip the damned envelope open and fluffy stuff goes everywhere.
I think I'd be inclined to go with the Intel BI440ZX. Although I haven't tried running NT5 on it yet, I'm sure there won't be any problem. The BI440ZX is a relatively inexpensive Socket 370 board with embedded SoundBlaster audio. Add a sub-$100 Celeron PPGA CPU, a 128 MB DIMM, an inexpensive video card, and you're good to go. * * * * * This from Werth, Timothy [timothy.werth@eds.com]:
I haven't played with the Adobe software much yet, and I don't know if the .pdd format is readable by anything else. As far as choice of format, I think that mostly depends on what you plan to do with the image. I use .gif for screenshots and similar images that have a limited number of colors. For posting to the web, I use JPEG, although at a lower resolution and higher compression than you're apparently using. (For the web, I typically scan a 4X6" photograph in 16.7 million colors at 150 dpi and get a resulting JPEG that's about 25K to 50K.) I've used TIFF in the past for screen captures. I believe that it uses lossless compression. We'll see what the experts have to say. * * * * * This followup from Werth, Timothy [timothy.werth@eds.com]:
Yes, that sounds like a good way to preserve your family photographs, with a couple of caveats. Monochrome prints, if processed properly, are inherently at least an order of magnitude more stable than color prints, and probably two orders. Particularly color prints that use dye-creation methods (which is most of them) versus the dye-destruction method used by Cibachromes. Ideally, a monochrome print contains only paper, gelatin, and finely divided silver, all of which are reasonably archivally stable. Certainly more so than any currently available means of storing computer data, including recordable CDs. With data storage, there's not just the issue of media stability, but also changing standards. Just try, for example, to read a DC600 tape or a 5.25" floppy written only five or eight years ago. Chances are, you don't have a drive that can read them. CDs, of course, are much more likely to be readable for longer than that, but in twenty years will you have a drive that can read those CDs? Who knows. From what little I know (or think I know), TIFF uses lossless compression and would be well-suited to your purpose. Unless the subtle color variations in monochrome prints (e.g. sepia or selenium toning) are important to you, I suggest that you scan them as gray scale rather than color. The images should be substantially smaller and no detail will be lost. * * * * * And, as I was sending that last, this showed up from Werth, Timothy [timothy.werth@eds.com]:
And then this:
Well, if what you want is long-term archival stability, it's tough to beat what you already have. Off the top of my head (it's been a while...) Standard silver-gelatin prints are pretty stable, if they were processed properly to begin with, and particularly if they were toned using selenium (warm black) toner or gold (cold black) toner, which many commercially produced prints were. Stored properly (cool, reasonable humidity, basically conditions that people are comfortable in) and not exposed to light, a silver-gelatin print made in the 60's or earlier (before resin-coated papers) should last at least one hundred years. Glass-plate negatives are probably even more stable, as are film negatives made on cellulose acetate film base. Early film was made on a cellulose nitrate base (=nitrocellulose=guncotten) and are very unstable. They generate nitrogen oxide fumes which accelerate further decomposition. Many an old film can has been opened that contained just a pile of reeking ashes. Cellulose nitrate film stock was widely used through the 1930's, and continued to be used in some countries well into the 1960's. If you have any of those lying around, it's probably a good idea to have them professionally duped and then dispose of them safely by burning them, small amounts at a time. If an piece of film from that era doesn't mention "acetate" on the edge, chances are good that it's nitrate-based. If it was me, I'd do a scan for computer storage and distribution purposes, and then put the originals back into safe storage. You can probably find all kinds of resources on the net concerning archival storage of photographs. Various companies used to sell archivally safe sleeves, print storage boxes, etc., and I'm sure that many are still around. * * * * * This from [fremen@pobox.com]:
Thanks for the confirmation. I figured it couldn't be just me, but it's nice to know that for sure. * * * * * I got the following message from Northern Light in response to my message protesting their change from 25 results to only 10 results per page:
So it appears that the inconvenience is temporary.
|
|
|
Wednesday, 14 July 1999 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week] The cool, wet weather persists here. We've had several days of rainy weather totaling perhaps 4 inches (10 cm), and temperatures in the upper 50s to upper 60s (15C to 20C). Very refreshing for July. It was an abrupt change. We'd had several days with highs in the upper 90s (37C). Then one day the high was 64F (18C), and it's been that way ever since. No transition. Just hot to cool. * * * * * Barbara is off on errands all day today, but she told me before she left that it looks as though we're going to be having Riley, a rescue Border Collie, as a temporary visitor starting this weekend. She said when she got involved with BC Rescue that the one thing she wouldn't do was foster, but I suspected then that she'd change her mind about that. When it comes down to choosing between taking a dog temporarily or letting it go to the gas chamber, I knew which one she'd pick. Our only real concern is how the visitor will interact with our older Border Collie, Kerry. He's an old dog, about 85 in people years, and is used to being boss dog. We're concerned that a visitor may challenge him and Kerry will get hurt. The other thing, of course, is that a dog rescued from a shelter may harbor disease. I told Barbara this morning that we need to talk to Sue Stephens, our vet, about what needs to be done in the way of more frequent inoculations against parvo and so on. Bob Shar, one of Barbara's former co-workers at the library, generously donated a wire cage that's large enough to hold a baby elephant. She's set it up in the unfinished part of the basement and plans to keep visiting dogs there. The dog that's arriving Friday comes from a home rather than from a shelter, however, so it's likely to be healthy and up-to-date on its shots. That means we'll probably try letting it stay in the house with us and our dogs. If they get along, fine. If not, it's the cage for Riley. * * * * * Nearly all the stuff has now arrived for the Colossus: The Forbin Project, so I'll probably be building that machine this coming weekend, although work on it may well slip over into the following weekend. Colossus (not its final name--I'll probably name it gordon after NASCAR's Jeff Gordon just to annoy my wife) will be built around a PC Power & Cooling Full Tower case, and use an EPoX KP6-BS dual-CPU motherboard, two Intel Pentium III/550 processors, 256 MB of Crucial memory, a Matrox G400 video card, an Intel PRO/100+ fast Ethernet adapter, a Turtle Beach sound card, an Adaptec 2940U2W SCSI host adapter, a Seagate Cheetah U2W 18 GB hard drive, a Plextor U2W 40X CD-ROM drive, a Plextor 8/2/20 SCSI CD burner, and an OnStream ADR50 50 GB SCSI tape drive. The best of everything, in other words. I'd originally planned to use the Tecmar Travan NS20 SCSI tape drive in Colossus, but I'll probably use the OnStream instead, because I may well install a Seagate Barracuda 50 GB drive in Colossus as well as the 18 GB Cheetah. If I do that, I'll want the 50 GB capacity of the OnStream tape drive. I realize that the money-no-object Colossus isn't practical for most people, so I'm also going to build a second performance system, but on a more reasonable budget. This box will be built around a PC Power & Cooling Personal Mid-Tower case, and use an Intel SR440BX "Sun River" motherboard (which has embedded nVIDIA RIVA TNT graphics and Creative sound), an Intel Pentium III/450 processor, 128 MB of Crucial memory, an Intel PRO/100+ fast Ethernet adapter, an Adaptec 2930 SCSI host adapter, a Seagate Barracuda 18 GB hard drive, a Plextor 40X CD-ROM drive, and the Tecmar Travan NS20 SCSI tape drive. It will be interesting to see how these two systems compare. * * * * * This from Werth, Timothy [timothy.werth@eds.com]:
Well, don't take it too much as gospel. One of the things I studied at Rochester Institute of Technology (RIT, or "Kodak U", and actually located in South Henrietta, but that would never do as an acronym...) was archival stability, but that's been nearly 25 years ago, and I'm sure that things have changed since then. Obviously, dye stability is going to be an issue with inkjet prints. I don't know how stable they are, but my guess is that you'll see noticeable degradation after a couple of years at most if the prints are exposed to light. It'd be worthwhile doing two copies of one and putting the control copy in a box so that you could compare them after a few months. Dye stability is one of the reasons I did a lot of work with dye-transfer prints, and later with Cibachromes. With standard color prints, the dyes are actually created during processing, which limits your choices. With dye-transfer, the dyes can be chosen based on factors like archival stability, and with Cibachrome (a dye destruction process), the dyes are already present in the unprocessed emulsion and are destroyed as necessary during processing. So either dye-transfer or Cibachrome prints are probably close to an order of magnitude more stable than ordinary c-prints. Sadly, I understand that Kodak no longer manufactures the dye-transfer matrix film or the dyes themselves, so that process is effectively dead. It took all day sometimes to turn out a good dye-transfer print, but the results were gorgeous (and reasonably permanent.) And once you'd produced the matrices, knocking off additional copies was reasonably fast. Incidentally, if you have any slides that you need to scan, check into the HP slide attachment. It came standard with my 6200C, was a $20 option for the 5200C, and I assume is also an option for your 4200. Basically, all it is is a little triangular plastic thingee that has a position to mount the slide and a mirror to direct light through it. I haven't tried it yet, but it appears that it should work fine. A mounted 35mm slide has a viewable area of a bit less than 1" X 1.5", so scanning at 600 dpi would give you a roughly 600 X 900 image--certainly not photo quality, but nothing to sneeze at. * * * * * This exchange with Keith Soltys [ksoltys@solect.com]:
I vaguely recalled the Tamron adapters. I think I actually owned a Tamron lens at some point. I still may. It's been a while since I dug through that closet. I wasn't pleased by their web site, however. When I went to http://www.tamron.com, the splash screen that came up said:
They sound like they're *bragging* about it. Geez. As it happens, I have Javascript disabled in IE5, so I fired up Navigator. It didn't help much though. I couldn't find a search function on the site, and nowhere could I find a list of what adapters were available. But Tamron may indeed provide a solution for someone who has time to research it.
Yes, it does work fine with Javascript enabled. And you're right that the easiest way would be to contact a Tamron dealer. I'm still amazed, I don't know why, when I find useless web sites like this one. The worst example was the Generac site. I visited there before I bought my backup generator. They'd gone to all the trouble to put up a web site, but instead of putting up specs, tips, and so on about their products, they put up a survey form asking whether they should put up information about their products. Duh. I just visited the Generac site, and find that they've sold their portable generator line to a separate company, Generac Portable Products Inc. There's an 800 number and an email address (info@gppi.com) that you can use to get information, but no web site at all. These people are clueless. * * * * * This from Chuck Waggoner [waggoner@gis.net]:
Hmm. I don't think it's mouse creep. It happens on too many different systems, and is completely unpredictable. I think it's a bug in NT and the screen saver itself. I haven't listened to shortwave in years, but I should probably start again. Back in the late 60's, I had an old WWII-era Hallicrafters receiver that someone gave me. It was built like a tank, and I wish I still had it. I have no idea what happened to it. Probably lost in a move. I doubt that Hallicrafters even still exists, but they certainly made fine equipment. I don't doubt that people exist who call themselves vampires, or even that they may bite people in the throat and drink blood. There are deranged people all over. I don't even doubt that driving stakes through their hearts would kill them, although I'd bet money that they wouldn't disappear in a puff of dust. But writing a book about them seems a bit outre. I sincerely hope that Buffy is having more than a casual impact among our young. If I had pre-teen to teenage children, I'd do everything I could to encourage them to watch it and think about it. Buffy is an allegory and a morality tale. It delivers exactly the kinds of lessons that children need to be taught, viz. Good and Evil are real things. Good may be ugly and Evil beautiful, so one must judge by actions rather than appearance. Good people may do bad things, and Evil people may do good things, but they always return to type. Evil triumphs if Good men do nothing. The triumph of Good over Evil is by no means assured, and Good people suffer and die fighting Evil. And so on. In watching the season finale episode last night, it was immediately clear why WB decided to delay running it. It had students arming themselves and packing the school with explosives, and a final shootout at the school. Superficially similar, in other words, to Littleton. But the key difference was that it was Good guys defending themselves against Evil rather than allowing themselves to be victimized by it. * * * * * Late Morning - This from Tom Syroid [tsyroid@home.com]:
Yep. I think I found the problem. One of the reader mail messages had a sig that used a solid underline full screen width. I got rid of that, which should solve the problem.
|
|
|
Thursday, 15 July 1999 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week] Another x86 CPU manufacturer bytes the dust. The Register ran a story this morning saying that the IDT WinChip is no more. With Cyrix gone, that leaves only Intel and AMD as mainstream x86 producers. And the future of AMD is by no means certain. I am a big fan of Intel. They make the best processors available, and I wouldn't even consider using a non-Intel CPU on any of my production systems. But the presence of AMD, Cyrix, IDT, and other x86-clone makers has kept Intel honest. Without that competition, Intel processors would not be as fast or as cheap as they are now. I hope that AMD does well with the K6-III and the K7/Athlon, not because I would consider buying or using one myself, but because those alternative processors keep the pressure on Intel to continue innovating and cutting prices. * * * * * Riley the Border Collie should be showing up here shortly. Carolina Border Collie Rescue has already placed Riley. In fact, James Davis, the man who will be Riley's new human, was the one who drove to Raleigh last night to pick him up. But Mr. Davis was already scheduled to leave on a trip tomorrow, and Riley's old owners were not willing to keep him any longer. So Mr. Davis picked him up last night and kept him overnight. Now he's dropping Riley with us to stay until he returns from his trip. Dogs don't much like change, and this is a lot of change for any dog. We're planning to keep him with us in the house, and will cage him only if things get unworkable. * * * * * This from Mathew A Fuller [mjpi7maf@fs1.ce.umist.ac.uk]:
Thanks for the kind words. The concept of hydrostatic shock is not entirely a myth, but hydrostatic shock becomes a factor only at very high impact speeds--above 2,500 ft/s (762 m/s) or thereabouts--well beyond anything practical for a pistol bullet. A rifle bullet traveling at or above that speed has greater stopping power than its momentum, cross-sectional area, and other factors (e.g. bullet construction) account for. The concept of hydrostatic shock as applied to pistol bullets probably got started because there are some people who see things as they wish they were rather than as they really are. In fact, by long experience, it is demonstrable that a pistol bullet must be larger than 40 caliber and 200+ grains to be an effective stopper against a determined man. Also by long experience, hitting someone multiple times with a lesser caliber doesn't accomplish much. As a response to the first hit, the body goes into crisis mode--pumping adrenalin and shutting down its response to external shock. Hitting someone two or three times (or even twenty times) with a minor caliber like a 9 mm or a .357 has very little more likelihood of stopping him than does hitting him once (unless, of course, subsequent rounds hit a nerve center or other critical area). Until the late 19th century, all that was well-recognized by everyone. The British at the time were using the .455 Webley, a serious man-stopper, but were concerned about the weight of ammunition and the fact that the recoil of the .455 made it difficult to shoot. They replaced it with the .38 Enfield, which turned out to be a disaster. On more than one occasion, the British cavalry arrived at the site of a massacre only to find a dead officer holding an empty .38 Enfield, surrounded by empty .38 cartridges and dead attackers riddled with .38 holes. The .38 Enfield *killed* people just fine, but they took too long to die. In the mean time, they were busy hacking away at the officers who had shot them. In a defensive pistol, what you want is stopping power versus killing power, which are two very different things. Ask any emergency room physician if he was going to be shot in the torso whether he'd prefer to be shot by a .22 or a heavy caliber handgun. He'll tell you he'd pick the heavy caliber handgun every time. It punches big holes and stops someone from doing what they're doing, but those holes are often not fatal. The .22, on the other hand, punches little tiny holes and rattles around inside the abdominal or chest cavity, ripping up lots of other stuff. But it has no stopping power whatever. The unfortunate fact is that, in order to be effective, a defensive handgun must be something on the very close order of a .45 ACP or a .44 Special. Men who buy their wives or girlfriends a "lady's gun" for protection aren't doing them any favors. The size and physical capabilities of the shooter have nothing whatever to do with what's required to stop an attacker. I saw to it that my wife trained with a .44 Special and a .45 ACP. And even those are not really adequate. They're the best imperfect alternative when portability is a factor. I've told Barbara many times that if she should ever have to shoot to defend her life, pick up the 12 gauge riot gun and keep shooting the SOB until he looks like hamburger. And even that's not guaranteed to work. The unfortunate fact is that there is no hand-held weapon that is guaranteed 100% effective against a determined attacker. Incidents have been reliably reported where men have continued fighting effectively for quite some time after receiving multiple central body hits from 50 caliber and 20 mm machine guns. But the best you can do is the best you can do, and in practical terms the best you can realistically do is a .45 ACP Colt 1911 pistol. * * * * * This from Anna Coates [acoates@bellsouth.net]:
You probably can't. If it's done legally (by court order), the phone company does it all at the central office, and it's completely undetectable. If an individual has tapped your phone, you also won't be able to tell unless he's incompetent. Those little boxes with the LEDs on them are entirely useless unless the tapper is incredibly clueless. The only real way to tell (and that's not certain) is to trace wires to see if there are any that don't belong. If there are extra wires present, it may not be obvious to casual inspection, even to someone who knows telephone systems well. And those wires won't necessarily be close to your home or office. It's quite possible to connect to your pair anywhere between your demarc and the central office. To make matters even harder, a tap does not require a hard-wired connection. It's quite possible to use an inductive pickup to tap a line without ever physically connecting to that line. I've seen some very cunningly concealed inductive pickups, including one that appeared to be (and functioned as) as standard surface mount jack. The short answer is that if you suspect you're being wiretapped by the police or another government agency, there's not much you can do other than avoiding saying anything on the phone that you don't want recorded. If you suspect you're being wiretapped by a private party, use a different phone to call the phone company and report your suspicions. * * * * * This from Bo Leuf [bo@leuf.com]:
Thanks. I probably should have added, "much though many wish otherwise, violence is the only argument that is effective against Evil. Those who attempt to deal with Evil by peaceful means always fail, and ultimately drag others down with them. Turning the other cheek to Evil accomplishes nothing but getting both cheeks slapped." * * * * * This from Bo Leuf [bo@leuf.com]:
Yes, the OpenGL screensavers are one of the nastiest little things that Microsoft foists on people. Unless you run a video card that has OpenGL acceleration in hardware, the OpenGL screensavers don't just want part of your CPU, they want all of it. That's bad enough on a Workstation box, but at least most Workstation boxes aren't doing anything when the screensaver kicks in. What I find incredible is that Microsoft includes the OpenGL screensavers with Server. I've spoken with several people who made the mistake of enabling an OpenGL screensaver on their network servers (which almost never have OpenGL video cards installed) and then wondered why server response dropped to nothing. Duh. * * * * * This from Tom Genereaux [entropy@lawrence.ks.us]:
Thanks. * * * * * Here's a long one from ROBERT RUDZKI [rrasterho@pacbell.net]:
Thanks for the long post. I wish I had time to respond in detail, but I have so much I have to get done today that I simply don't have time. One comment, though. I think you got it exactly backwards with regard to civil suits. The problem isn't the juries. They're pretty much handcuffed. The problem is the lawyers and judges. We'd be much better off getting rid of the lawyers and judges and keeping the juries than we would doing the converse. And trial by jury is an important right. The last thing I want is the government deciding anything at all for me. I'd sooner trust a jury of my peers. And if your suggestion came to pass, how long would it be before they decided to get rid of those inconvenient juries for criminal trials as well? We'd be back to the star chamber in no time.
|
|
|
Friday, 16 July 1999 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week] I didn't get much accomplished for the last couple of days. I had a testbed box that had Windows NT 4 installed on a 4 GB partition on a 4.3 GB drive. I forgot when I installed the OS that I also wanted to run Win98 SE and W2K Professional on that testbed. It's a long, sad story, but basically I can no longer get anything other than Win98 SE to boot and run on that system. Attempting to install any flavor of NT generates a blue-screen crash right at the end of Setup. I've taken the disk down to bare metal with the Western Digital low-level format utility, and I still have the problem. I'm going to swap in a new hard disk and see what happens. Nothing is ever easy... * * * * * This from ROBERT RUDZKI [rrasterho@pacbell.net]:
Yes, the .357 is a minor caliber regardless of loading. The "Marshall and Sanow" data is worthless, simply because they picked and chose only the data they wanted to use, and then manipulated it without regard for proper statistical analysis methods. For example, they list one particular .357 round as providing 97% one-shot stops, when by their own data it actually provides 37% one-shot stops. See this page for some details, or simply search the web for "+marshall +sanow". My late friend Mel Tappan, in his 1979 book Survival Guns, provided a pretty reasonable ad hoc method for calculating stopping power. He didn't represent it as scientific, but observed that it fit observed results very closely. Basically, Tappan used momentum multiplied by bullet surface area and adjusted for bullet efficiency. He called it the WAVE method, where [W]eight is the bullet weight in grains, [A]rea is the surface area of the bullet in square inches, [V] is the velocity, and [E]fficiency is an semi-arbitrary figure assigned to bullet efficiency, where standard .45 ACP round-nose hardball is 1.0, and the typical .357 125 grain semi-wadcutter bullet is 1.25. He multiplied those numbers together and then divided by 1,000 to give a nice round working number. On that basis, comparing your 125 grain .357 at 1,400 fps to my 230 grain .45 ACP hardball at 850 fps or my .240 grain 44 Special semi-wadcutter at 850 fps gives the following: .357 Mag: (125 * 0.1001 * 1400 * 1.25) / 1000 = 21.90 .45 ACP: (230 * 0.1605 * 850 * 1.00) / 1000 = 31.37 .44 Special = (240 * 0.1445 * 850 * 1.25) / 1000 = 36.86 Tappan observed that, based on actual experience over thousands of recorded shootings, any round that exceeded about 30 on this scale would be a 95%+ effective one-shot stopper, and that anything that calculated out significantly less than 30 would be effective in about the percentage ratio relative to 30 at 95%. That puts your .357 round at about 69% effective, which is on the close order of observed behavior. Actually, it will be less than that, because you're probably not going to get anything near 1,400 fps out of that round. Fired from a typical defensive revolver with a 4" barrel, you'll probably see something more like 1,200 fps, which takes your figure down to 18.77 and your effectiveness to perhaps 59%.
|
|
|
Saturday, 17 July 1999 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week] Well, after doing a lot of research, I've joined the digital camera revolution. I decided on the Olympus D-400 Zoom. I spent the better part of a day reading web reviews and digital camera discussion forums. With the exception of one aberrational review in PC Magazine, the Olympus D-400 Zoom got very high praise indeed, including the kind of reviews that really count—those written by users. But perhaps the most influential element in my decision was my friend and co-author Jerry Pournelle’s advice: "The Olympus D-400 Zoom is the camera you want. It's just about perfect. I have a bunch of digital cameras, and the D-400 Zoom is the one I actually use." He should know. He's been using digital cameras for years, and has used a lot of them. When FedEx showed up yesterday morning with the camera, I of course immediately dropped what I was doing and started playing with it. I’ll have much more to say about the D-400 Zoom after I’ve had some in-depth experience with it, but for now I’ll offer my first impressions. The first thing that struck me is the camera’s size. Seeing product photos of the D-400 Zoom didn’t prepare me for how small it would be. In appearance and size, the D-400 Zoom closely resembles a small point-and-shoot 35 mm camera. Although small, the camera appears solidly built. The D-400 Zoom is indeed, as some have described it, a “shirt-pocket” camera, although to avoid dropping it if nothing else, I’d limit that carrying method to something like a flannel shirt with a buttoning flap. The camera is light enough to carry all day long without even noticing that it’s there. If I was pleasantly surprised by the camera’s small size and weight, I was truly shocked by the size of the included 8 MB SmartMedia Card. This thing is literally the size of a large postage stamp, and not much thicker. When I read during my preliminary research that a SmartMedia card weighed 480 milligrams—less than an extra-strength aspirin tablet—I thought it was a typo. It wasn’t. What’s truly incredible to me is that this tiny card stores about 40 1280X960 images in HQ mode or 120+ 640X480 images in SQ mode—both of which use standard compression—or 18+ 1280X960 images in SHQ mode, which uses minimum compression. Perhaps most intriguing is that you can store two images in uncompressed 1280X960 mode. Uncompressed mode uses .TIFF format, versus the .JPG used by the other modes. There are also 16 MB and 32 MB SmartMedia cards available, which store a correspondingly larger number of images. Of course, I did some test shooting just to see how the D-400 Zoom did. I created an Olympus D-400 Zoom Sample Images page that contains images that I shot while playing around. These are by no means the best that this camera is capable of. I'm still learning how to use it, and looking forward to that process. Much more later after I've had a chance to do more than just play with the camera.
|
|
|
Sunday, 18 July 1999 [Last Week] [Monday] [Tuesday] [Wednesday] [Thursday] [Friday] [Saturday] [Sunday] [Next Week] I've added several images to the Olympus D-400 Zoom Sample Images page, including one SHQ macro shot. I'm amazed at the level of detail it provides. I didn't think a digital camera could do this well on illustration shots, but the D-400 Zoom took it in stride. This is one great little camera. Incidentally, I noticed before adding the new photos that the estimated download time for the page at 28.8 was already at 4 minutes, so I thumbnailed everything. I also realized that I hadn't gotten into any detail about how I transferred the images from the camera to the PC. The camera comes with a serial cable, which I have not used. Instead, I used the FlashPath floppy disk adapter, which looks a lot like a floppy disk and fits into a floppy drive. There's a slot on the side of the FlashPath that you slip the SmartMedia card into. You insert the device into the floppy disk drive, log to A: and copy the images directly from the FlashPath to wherever you want them. The FlashPath does require drivers, and three separate disks are provided with drivers for Windows 95/98, Windows NT 4, and the Mac. I've tried the first two, and both work fine. Just as an estimate, files appear to copy about 10 or so times faster than they would from an actual floppy disk. Copying a dozen or so 640X480 SQ images takes only a few seconds. And speaking of 640X480 images, to see just how good 640X480 images can look, check out my friend Steve Tucker's site. He's posted some very nice images he took at 14,000+ feet at the summit of Mount Evans, near Denver. He even has an MPEG video clip posted. Pretty impressive. * * * * * This from Daniel C. Bowman [DanBowman@worldnet.att.net], subject "You costing me money!":
Well, depending on how much film you shoot, this could actually end up saving you money. Barbara tells me that a 24-exposure roll runs $10 or $12 counting film and processing, or roughly 50 cents a shot. The D-400 Zoom puts 20 to 150 images on a $14 8 MB SmartMedia card, which presumably can be re-used at least 100 times, and probably much more (I'm just arbitrarily assuming 100 uses, guessing that the things must eventually wear out.) Even using very pessimistic numbers, that's maybe half a penny per shot for the card. Batteries would certainly be an expense item if you used standard alkalines, but a set of four NiMH batteries with charger runs about $40. It's rated for at least 300 full charges (although I've been told that 500 or even 1,000 is not that unusual), and Pournelle says he gets 100 shots per charge. Call it 30,000 shots for $40, and that counts the charger which you don't throw away. Call it another tenth of a cent per shot for batteries. If all you do is post pictures on the web, there aren't any additional costs for the D-400 Zoom. With the regular pictures, you either have to buy a scanner or pay to have photodisks made, which runs another 25 cents or so per photo. If you want prints from your D-400 images, you need an inkjet printer capable of printing photos. Those are perhaps $200, and the per photo consumables costs are close to that of a standard print. Still, you typically don't print anything near every shot on a roll, so the actual print costs are typically going to be lower. Actually, what I expect to see in the very near future is photo printing kiosks that contain a fast, high-quality printer. You'll stick your SmartMedia card or CompactFlash card into the slot, display the contents, punch in the numbers of the ones you want printed, how many you want, and how big you want them, put your credit card in the slot, and get your prints out the other end. When that happens, you'll see the death of film and of drugstore processing. * * * * * This followup from Daniel C. Bowman [DanBowman@worldnet.att.net]:
Alkaline batteries. Ugh. I plan to carry a set, just for emergencies, but I can't imagine using them routinely in a digital camera. I'm sure that Energizer, Duracell, and Rayovac would love you to do that. I've heard reports that some digital cameras provide as little as 10 to 20 shots per set of alkalines if flash, zoom, and other power consuming features are used heavily. The D-400 Zoom isn't nearly that hungry by all accounts, but I'm sure that even it would get expensive to run on alkalines. I'll probably burn a set or two of alkalines just to see what really happens, but I'm not really expecting much out of them. Still, you can buy AA alkalines anywhere, so it's nice to know that if you are stuck with dead NiMH batteries, a broken charger, or no power to recharge with, you can continue to shoot with the D-400 Zoom. That's an option you don't have with cameras that don't accept standard AA alkalines. If I really care about a photograph, I'll still shoot it on 35mm Kodachrome or 120, because they still resolve at least a full order of magnitude more detail linearly (and two orders areally) than the best available digital cameras. But for stuff that I'm going to put up on the web, family snapshots, product shots for my books, and just about everything else, I'll go with the D-400 Zoom. Who knows? In a couple of years, we may see consumer-grade digital cameras with full interchangeable lenses, 10KX15K resolution, and multi-gigabyte memory cards. At that point, I'd probably give up on film entirely, and I suspect I won't be alone. But until then, this D-400 Zoom will suit me just fine. * * * * * This followup from Daniel C. Bowman [DanBowman@worldnet.att.net]:
Absolutely. And the thing that's easy to forget is that although they'll be better and cheaper later, they're Good Enough now. But we all buy PCs fully aware that what $1,000 buys us today, $250 will buy us in a year or two. Heck, I have an old Gateway 386 sitting five feet from me. I paid $3,000 for it in 1991, and I don't begrudge a cent of that cost. I got my money's worth out of it. And the same thing is true of digital cameras now. What you pay $600 for today may seem kind of primitive a year from now, but who cares? In the mean time, you'll have the use of it, and it'll still be doing the same job it always did even when better and cheaper cameras are available. And the job it's doing now is perfectly adequate for what I want--family snapshots, stuff to put up on the web, and illustration shots. * * * * * This from Tom Church [tom48@ecqual.net]:
Geez, I have no idea, except that it's pretty fast. Boris Becker has one of the truly overpowering serves in tennis, on a par with those of Ellsworth Vines, Pancho Gonzales, Michael Sangster, and Roscoe Tanner. Part of the problem is, where do you measure the speed? A tennis ball is ballistically inefficient, both in shape and surface. So where you measure speed is critical. Years go, I remember that the Guinness Book of World Records listed Michael Sangster as having the record: 154 MPH off his racket and 108 MPH as it crossed the net. My guess is that Becker at his best is at or above that. * * * * * This from Mike Boyle [mboyle@toltbbs.com]:
I'll have to look into that. Right now, I have no means of printing these images other than my monochrome laser printer. I'd planned to look at standard inkjets, but I may have to check this out as well. Thanks. * * * * * This from bdenman [bdenman@FTC-I.NET]:
No, I don't have any special power or AC going to my office. There are two standard 15-amp circuits, and the vent is under my desk. In the winter time, I close the air vent because the machines keep the room plenty warm. In the summer time, I keep the air vent wide open, and it seems to be adequate, although the room is noticeably warmer than elsewhere in the house. As far as cases, those are a couple of PC Power & Cooling cases you spotted, one full tower and one mini-tower. I don't think that PPC cases and power supplies are overkill for any application. As you may have noticed, I use one for my main workstation, and wouldn't consider using anything else. Like everything else in computers, the case and power supply segment is very competitive. You get exactly what you pay for. Perhaps I should say, you get no more than you pay for. If you want to economize on case, that's one thing. But don't go with an inexpensive power supply. I've seen too many power supply related problems over the years. What I always tell people who balk at the cost of a PPC unit is this: "Okay, if you must save money, buy a cheaper case. But at least put a PC Power & Cooling power supply in it." I have a couple of Antec cases here for testing. Although they're not as good as the PPC units, they're very nice units--nicer than the Enlight and InWin units I've seen--and I'll probably build a couple of testbed systems around them. But check out the power supply specs. Antec has better specs than Enlight, but both pale next to the PPC specs. * * * * * This from Daniel C. Bowman [DanBowman@worldnet.att.net]:
I'm also using X-10. I wish a reasonable alternative existed. I had high hopes for Echelon for a while, but they haven't rolled out any consumer-oriented products. But X-10 does for now... * * * * * This from ROBERT RUDZKI [rasterho@pacbell.net]:
I secured full length 2X4 ledger strips to the walls, using 4.5" lag screws driven into the studs. I then screwed 2X2's flush with the top edge of the 2X4's, laid the door flat on top of that assembly, and then drove screws up through the 2X2's into the bottom of the solid door. That sucker is mounted solidly. I can literally jump up and down on it (I weigh about 230 lbs) without making it jiggle. I must say that that's the first time I've heard of a cat killing a PC in that fashion. I assume the cat didn't hit the power supply, or you'd have needed a new cat. Come to that, if a cat did that to one of my PCs I'd probably need a new cat anyway... * * * * * This from Bo Leuf [bo@leuf.com]:
Yes, battery life is definitely something I'll check into. I'm a bit confused about the NiMH batteries. Pournelle says he charges his for seven hours, but my kit said that it reaches full charge for one or two batteries in 80 minutes, or for three or four batteries in 160 minutes. The instructions said the batteries needed to be charged before first use, so I stuck them in the charger and fired it up. They finished ten minutes earlier than they were supposed to, at 150 minutes. The lights on the charger went out, which I assume means that there's no charging going on rather than a trickle charge. But the manual says not to charge them for more than 48 hours (I think), so I'm not sure exactly what the lights mean. Also, the manual says they should be charged two or three times before using them. I'm not sure how to do that, since the charger lights go out, and it doesn't seem to be charging at that point. At any rate, the batteries were warm after a full charge, but putting them back in the charger seemed to do nothing. The lights stayed out, and the batteries hadn't gotten warm after half an hour. Oh, well. * * * * * This from H [hstuck@excite.com]:
I'll probably check into photo printers at some point, although it's a low priority at the moment. Frankly, I'm not entirely sure what the point is of printers that don't require a PC. Surely, anyone who is using a digital camera also has a PC? If not, what are they doing with their images? Presumably printing them out and the deleting them, although that seems foolish given the lack of stability of current inkjet inks. I suspect that a general purpose inkjet printer with the ability to print photos would be a better purchase. * * * * * This from H [hstuck@excite.com]:
Thanks. This doesn't impact me, because I don't run multiple monitors, but I'm sure some of my readers do. * * * * * This from Bo Leuf [bo@leuf.com]. He sent it to me last Thursday, but I just found it at the bottom of my inbox:
Well, regarding your last paragraph, as I've said before, I base my own actions on Kant's Categorical Imperative. Ultimately, it all boils down to personal responsibility. And as far as the question of non-violent and passive, the simple fact is that non-violent means do not work against Evil. Gandhi's non-violent resistance against the British occupiers had some effect because Britain was wrong, but not Evil. The same is true of the success of non-violent civil rights resistance against de jure racism. Most people, including those in power, knew that what was being done was wrong. But Evil is an entirely different beast. Gandhi would have gone to the ovens had he attempted non-violent resistance against the Nazis. When one encounters Evil, the only solution is violence, actual or threatened. That's all Evil understands.
|