Dan's Data letters #81
Publication date: 29 Dec 2003.Last modified 03-Dec-2011.
Why for PC yodels?
I'm using the Net through a LAN-shared 56K modem connection. The problem is, the host computer with the modem is just sending packets out and I have no idea what it's sending. It's a real pain in the if I'm trying to play games because I get huge pings and can't move. I'm not sure if it's a virus or what. I've tried looking in Task Manager on the sharer PC for anything suspicious but found nothing.
Mr. pRedatoR
Answer:
Well, Mr P, it probably is a virus, trojan or worm, assuming the Net-connected
PC isn't just sharing drives with the universe or the subject of
script kiddie
attack, which is unlikely since you're on dial-up.
A recent probable culprit in this situations is the Blaster worm. If your sharer machine is running Win2000 or WinXP and hasn't been updated lately, it could have been Blasted. Run a Blaster removal tool on it like this one, and then go to Windows Update and suck down the whole list of security updates. Which will take about a zillion years, but such is life.
I like to recommend Trend Micro's free on-line virus scanner in situations like this, but it ain't fast to load over dial-up, and it'll be incredibly slow to load on a computer that's currently yammering away to the universe at the top of its lungs for whatever reason. Try the anti-Blaster tool first.
Tuning in
I saw this review of what seemed to be a pretty nifty TV tuner recently, and I thought I'd like to get me one of those. Then I realised that I did, in fact, need to connect it to a TV antenna.
Can I go to the store and get myself an antenna to work with this card? More importantly, what sort of antenna is best for indoor use (surrounded by brick, bottom level, next to a window if you need to know)?
Dominic
Answer:
Yes, a plain old indoor antenna will work, but for that tuner card you'll
need an antenna whose lead terminates in an F connector, as opposed to the
lousy PAL connector (aka "Belling-Lee
connector" or just "TV coaxial connector") which is still used for most
TV and FM radio RF connections. F connectors screw on and have much better
performance than the cheap and nasty PAL connectors, but you won't find
them on pretty much any indoor antenna. No problem, though; you can just
buy a bare F plug and graft it on, if you're feeling handy, or you can buy
an adapter.
Jaycar, for instance, have a 1.5 metre F-to-TV-plug lead (stock code WV7384) for $AU6.50. I presume one or both ends will be the wrong gender for your application, though, because of the basic rule of the universe which states that TV antenna cables are always the wrong gender, usually at both ends.
As regards what indoor antennas are best - they all suck, I'm afraid.
The problem for indoor antennas is that walls and conduits and water pipes and all the other stuff that makes up a building create a complex and horrible RF environment, full of occlusions and reflections and God knows what else. This is the source of the well known Antenna Dance, where you leap all over the room trying to find a spot for the antenna that works properly for a decent number of channels. And then, upon discovering that it's two feet down from the ceiling, three feet out from the back wall, pointing nor-norwest, you have to try to find a way to mount the damn antenna there.
There is nothing that can be done about this. You can buy yourself a cheap, small outdoor antenna if you like, or a fancy compact amplified unit meant for use on caravans or boats, but if you use that antenna indoors you're just going to have more gain on the same garbage. An amplified antenna may help if you're in a fringe reception zone, but if you're not then pretty much any old thing will do.
Hence: Get yourself a cheapo indoor antenna from a discount store. If it's got some kind of UHF-receiving thingy in between the two telescopic rods it uses for VHF, then it'll probably be about as good as anything you'll find. Then try to find a place to put it where it'll work.
Sensitivity training
I am very new at photography as a hobby, but after your latest article on photography, I have become more interested in the details. I currently own a Canon PowerShot S30, which I really do love and enjoy using. It has a manual mode which allows me to set the F-stop and shutter speed.
While experimenting with how different settings affect exposure, I noticed in my camera's menu there was an ISO setting ranging from 50 to 800. Now, my dad is a huge film photography hobbyist (or once was), and he explained to me that on regular film, one of the things that the ISO (or ASA?) number effects is the "graininess" of the picture. Since digital cameras obviously don't have molecular film processes, I was wondering how this works with a digital camera.
I experimented with a shot at each ISO extreme, one shot at 50 and one at 800, and the 800 shot was much noisier. Does this mean I should always shoot with a lower ISO setting?
Peter
Answer:
If you can afford the larger aperture and/or longer shutter speed then yes,
you should use a lower ISO setting. The native
sensitivity of most, if not all, digital sensors is still around ISO
100; higher ISO settings are just amplifying the signal coming out of the
sensor, and possibly also using cunning noise reduction techniques. Generally
speaking, this means that higher digicam ISO settings give you a noisier
image, just as film with a high ISO ratings, gives you a grainier result.
I talk about this in a few places on my site, most recently the second page of my big photo tutorial.
Drive decisions
I am assembling a new system including an Abit IC7-MAX3 (native SATA) and two Seagate 160Gb SATA drives. My question concerns the next, much faster version of SATA. I have researched and cannot find any info on what it will take to upgrade when the new version of SATA is available. Will it be possible to upgrade recent hardware, or will we have to purchase new motherboard and disk drives?
Paul
Answer:
The next SATA speed bump will give more peak bandwidth, yes, but it won't
actually be faster in real world use.
The current SATA spec has an alleged theoretical ceiling bandwidth of 1.5 gigabits per second, which the Serial ATA Working Group for some reason call 150 megabytes per second, reversing the usual storage company policy of making things sound bigger and faster than they are. 1.5 gigabits is actually about 179 megabytes. Never mind that, though, because considerable overhead stands between the theoretical bandwidth of all ATA interfaces and the amount of user data that can be pumped through them.
And never mind THAT, in turn, because SATA only allows you to connect one drive per channel, and ATA drives with sustained read speeds above 60 megabytes per second are still pretty much nonexistent. Roughly half of current "SATA-150" bandwidth is therefore never used for anything but brief bursts of data to and from drive caches. Past experience with the UDMA-33/66/133 speed steps has shown that the real world impact of bandwidth boosts that only affect cache data is very close to zero.
SATA-300 will be along at some point, and we're assured that SATA-600 will follow that. Drives are going to have to get a lot faster for even SATA-300 to be perceptibly faster, though.
Everything should be backward and forward compatible. Old drives should work with new controllers and old controllers should work with new drives, although only at the fastest speed that both drive and controller support.
So don't worry about being left behind, and don't hold your breath for big speed boosts from new gear, either.
'Phone power
Just read your review on "Luke", and not surprisingly it comes with a cheap and crappy set of headphones. If you replaced the 'phones from hell with a nice set of Sennheisers or Koss(es), or even an $AU80 pair of Sonys, would the drain in Luke's AAA cell be noticeably higher? Surely there is some cost to the benefit of better sound?
Rob
Answer:
Yes, it might make a small difference to the battery life, if you switched
to lower efficiency 'phones and therefore had to set the volume higher.
Generally speaking, though, there's unlikely to be any significant difference.
There's still an issue here, though. Most big headphones are very efficient if you measure output at a standard distance from the transducer, and they're certainly more efficient than the cheapo-'phones that come with many portable music gadgets. But different headphones sit at different distances from the ear. Well outside, for normal over-ear 'phones, or nestled into the outer ear, for regular earbuds, or poking right in there in the case of the more recent "canalphone" designs. Canalphones, of necessity, have very small transducers that aren't terribly efficient, but they're coupled so well to the eardrum that their effective volume-per-milliwatt is very high.
If you can tolerate the earplug feel of canalphones. then they're clearly the best option for most mobile listening these days, particularly since they block external noise very well, too.
You don't have to spend a fortune to get a decent set of canalphones. Sony's "Fontopia" MDR-EX71 set lists for $US49.99; they're not as good as the basic Etymotic ER-6 'phones (list price $US149, street price more like $US130), but they'll do.
Energizer versus Acme
With all your battery operated toys (flashlights, tanks, etc) I'm sure you burn through batteries by the pantload.
While in the store last night looking for new batteries for my Maglite, I noticed "good ones" (Duracells, energizers) were about 5 bucks for the whiz-bang models, and $4.50 for the "regulars", in the 2 pack. 4 packs were 6 to 8 bucks.
Right next to them were the "store brand", at $1.79 for the 2 pack.
How many dollars are in a name, and how many in quality, I wondered? Being a Cheap Bastard, and figuring the quality couldn't be that much different, I got the store brand. Will I be in the dark the next time the power goes out? I'm sure there's some difference, in something that sucks the batteries pretty hard, but in a flashlight?
Krstofer
Answer:
Actually, many flashlights do work their batteries quite hard. Generally
speaking, the bigger a supermarket-grade incandescent-bulb flashlight is
then the less punishment its batteries will be taking (D cells can deliver
a lot more current than AAs without breaking a sweat), but most flashlights
will give you better value with alkalines than with carbon-zinc cells, and
will run for about the same time with freshly charged NiMH cells as they
will with alkalines, despite the alkalines having rather higher low-draw
nominal capacity.
Generally speaking, off-brand alkalines are considerably better value than Duracells or Energizers. Ultra-cheap batteries may leak, but I can't recall any cheapies doing that to me for some years now. Dirt cheap batteries may also have been on the shelf a lot longer than the fast-selling brand name cells at the supermarket. If they're reasonably fresh, though, super-cheap alkalines are a more sensible choice.
We shouldn't forget good old carbon-zinc non-alkaline cells, either. In low drain devices (little radios, remote controls, wireless doorbells...), they'll give you about half the capacity of alkalines, so if you can get them for less than half the price and don't mind changing batteries more often (for remote controls and such, this still means "only every few years"...), they're worth buying too. In higher drain devices, alkalines are better value.
I'm not sure about the value equation for the relatively rarely seen "zinc chloride" extra-heavy-duty non-alkalines (actually, these are just like regular "carbon zinc" cells, except they've got little to no ammonium chloride in their electrolyte). They're considerably cheaper than alkalines for the larger sizes at least, but don't cost a lot less for AA and AAA sizes.
Eveready have a handy battery chemistry comparison chart here.
If your standby emergency This One Has To Work flashlight takes AA cells, there's a lot to be said for lithium AA cells. They've got higher terminal voltage, though, so they can eat bulbs in non-LED flashlights. CMG's two-AA-cell Reactor (reviewed here) is an excellent choice for this task; not too pricey, quite tough, and perfectly happy with lithium AAs.
Three points define a plane
I recently bought one of these Everglide Giganta Optical mousemats, and am very satisfied with it, except for one niggly little problem. The upper right corner is lifted up about 2-3mm from the surface of my desk... hardly earth shattering, but there's a noticeable seesaw effect when I move over it. Any suggestions on bending it back so that it's flat on the table (I've tried hair dryers and a good bending over my knee to little avail)?
Lucas
Answer:
I don't know of anything tailor-made. But the old LP de-ruffling strategy
might work - leave it in the hot sun between two sheets of glass, and let
it stay there until well after dusk.
Very brief mooning
The full moon is an event that last only an instant - there is an official instant at which it is full. It is not officially full at times other than that instant.
But I want to know if the surface that we see is fully illuminated for a period of time, and not just an instant. If so, for how long?
On newsgroups I've talked to two real (as real as one can hope in a newsgroup, anyway) astronomers, who did not know this, nor know how to find out.
Charles
Answer:
Well, yes, the moon really is completely full only for an instant, if you
define "Officially Full" as "that moment when the light from the moon reaches
its maximum for this lunation (lunar month)".
If you define fullness as "100% illuminated", though, with no portion of the shadowed area visible, then the period of fullness is actually significant. This is because the sun illuminates marginally more than a hemisphere of the moon, because the sun's got non-zero angular size in the sky. If the sun were a point source at infinite distance then it would illuminate exactly a hemisphere of the moon and the earth, but it isn't. Also, we on earth see marginally less than a hemisphere of the moon, for the same reason that you can't see a whole hemisphere of the earth from the top of a mountain.
Since we're looking at a bit less than half of an object that's a bit more than half illuminated, then even a picky definition of fullness gives a more-than-an-instant time span.
Fullness is more normally defined as "looks like it's 100% illuminated even if you're looking at it through good binoculars", though. By that definition, the moon's full for at least several hours each cycle. It'll be within a small fraction of a per cent of full for more than ten hours.
Meanwhile, in Scrabble-Land...
I was just idly playing with an anagram generator and I thought you might get a chuckle from some of the ones that came up for "Daniel Rutter" (I used your name instead of "Dan's Data" because there were more possibilities).
Unread Litter
A Lurid Netterand at the risk of insulting you (but I think you'll see the humour)
I, eternal turd
Keep up the great site.
Derek
Answer:
If you include my middle name, though, I can be "a lurid Internet
bra".
I think there's something in that for all of us, don't you?
