Dan's Data letters #88Publication date: 5 Feb 2004.
Last modified 03-Dec-2011.
You wrote "Big nasty secret: There does not exist a colour CRT tube, whether shadow mask or aperture grille, TV or computer monitor, that can clearly display all 1920 horizontal pixels of full-res HDTV."
Oh yes there is (but only just) - the Sony GDM-FW900.
Sony quote a 0.23mm to 0.27mm "variable" stripe pitch for this tube. I didn't know what that meant, but Cezar knows more about aperture grille CRTs than me, and clued me in to the fact that big Trinitrons commonly have a smaller pitch in the middle of the screen than out at the edges.
Assume that the average horizontal stripe pitch is 0.25mm. Also, forget about adding the usual 15% to the aperture grille pitch to get an approximation of the equivalent shadow mask pitch (aperture grilles are sharper horizontally than vertically); we're only talking horizontal resolution, here.
According to its PDF brochure, the FW900 has a peculiar 16:10 aspect ratio - but then they quote a 18.98" horizontal, 12.13" vertical viewable area, which is more like 16:11.9. These numbers plugged into Pythagoras' Theorem match the 22.5" quoted diagonal, so let's go with them. 18.98 inches is 482mm; 482mm of 0.25mm stripe triads is 1928 triads.
Now, using the usual rule of thumb that says 1.25 dot triads are enough for a reasonably clearly displayed pixel, the acceptably clear horizontal resolution of this CRT would appear to be... [drum roll] around 1540 pixels. Around 1680 pixels if you only cared about the smaller-pitch area in the middle of the screen and could tolerate more fuzziness around the edges.
Using the 0.25mm pitch figure, to wring 1920 horizontal pixels out of this tube, you'd have only about one triad per pixel - or about 1.09 triads, using the 0.23mm figure. This is better than the clarity a lot of people put up with (when viewing 1600 by 1200 or more on cheap 19 inch monitors, for instance), but it's still not really "clear", if you ask me.
The 1.25-times rule-of-thumb divisor figure is what gives you the shadow mask sharpness shown here. Some people find even this unacceptable (it looks worse in close-up than from normal viewing distances...), and insist on a larger divisor. Personally, I think that just gives you sharper jaggies. Nowhere is it written, after all, that you have to be able to clearly see individual pixels; you can see 'em on LCDs and many people seem to enjoy doing so, but I question the wisdom of this.
The FW900's "recommended" resolution is 1920 by 1200, but recommended CRT resolutions are, routinely, above the clear display capacity of a tube. Sony do this as much as anyone else does. Often, recommended resolutions seem to be whatever the highest resolution is that the monitor can display at an 85Hz or better refresh rate.
All this isn't to say that there aren't lots of big CRTs that look great running at two megapixel and higher resolution, but close examination will reveal that there just aren't enough phosphor splodges for each pixel to have even three of its own, much less three times 1.25.
The picture is sharp and there is nothing obviously wrong with it, except for moire. In most resolutions there's a pattern of dots over everything. The only resolution I haven't seen this occur at is 1600 by 1200 (which is unusable anyway because the refresh rate is maxed at 70). I have tried different refresh rates at different resolutions, and the monitors own crummy moire controls for vertical and horizontal blurring. This results in the screen looking much worse, but no moire artifacts!
My question is, is there any other way to reduce or eliminate this annoying condition? Or is this a result of the design of the monitor itself? My cheap 17 inch Mag Innovision 720v never had any issues with moire even at its maximum resolution of 1280 x 960 (in fact, I didn't know what moire was until I got this monitor).
It's possible that you've got a monitor that exhibits moire problems as a normal feature. It's more likely, however, that the particular monitor you've got has had a hard life.
High quality, expensive CRT monitors may not show moire on any image (LCDs never do). Mid-range monitors, like the Samsung 955DF I'm using at the moment, commonly show moire sometimes, but not always; mid-greys and single-pixel checkerboards tend to be the culprits. You can often reduce the moire by tweaking the monitor's settings, but one kind of moire often gets better as another gets worse. It's pointless to try to make the monitor the equal of one that cost three times as much.
Dirt cheap monitors may show moire all the time, but it's not actually common. A monitor that's been dropped and had its shadow mask or aperture grille squished out into an odd shape may have the problem constantly, though. That could be what's up with your KDS.
(A friend of mine has some Apple monitors that were rescued from the dump. They all have moire issues, but you can't beat that price!)
If you can't fix it with the monitor's own controls, then you probably can't fix it at all. Learn to live with $30 worth of 19 inch CRT.
You say that "There's no cheap liquid ("Cheap", here, means "under $AU150 a litre") that's anything like as good as water at moving heat."
What about olive oil? It is not as corrosive as water, incredibly good at transferring heat - probably better then water - but also I do not think it conducts electricity at all (well, hardly at all).
Now, this is all just my personal observations over time, as I have no data to back it up. Would this not be a good solution? You might need a stronger pump, but eh!
Yes, olive oil's non-corrosive, but neither is water with ordinary radiator additive in it. Olive oil will go rancid over time, especially if it's warm, and it's about a hundred times as viscous as water; you'd need a special very high-power pump to move it.
And it's not as good at moving heat as water. I don't have numbers for olive oil in particular, but as far as I know there's no cheap "oily" liquid that comes within a mile of water.
Substances like mineral oil (which won't go rancid) are used for cooling in some situations, but usually because their freezing and boiling temperatures are much lower and higher, respectively, than water's.
You're right about olive oil being non-conductive, as well, but neither is clean water. Water with radiator additive in it may not be negligibly conductive, but it won't cause galvanic corrosion, which is the main reason why you want non-conductive coolant.
I just got dual monitors set up and I'm damn proud of the big time baller status that they confer upon me. However, since these two monitors are not the same brand or model I've been having some difficulty getting the two to display the same image in an identical manner, no matter how much I fiddle with the brightness and contrast. I figure a good monitor calibration tool would do the trick, but I've yet to find a really great one.
I have Adobe Photoshop, but its bundled Adobe Gamma is quite inadequate. I was therefore wondering if you knew of a really good calibration tool, preferable one that's free.
I don't know what monitor calibration tools will help you, here. The Spyder calibrator I reviewed the other day will be no good to you until its software supports more than one screen - not that many people aren't perfectly happy with one screen for colour-critical work and another one for toolbars or whatever else where calibration isn't important.
One of the other calibrators I mention in that review may suit you, but there certainly aren't any free options that'll work. Software-only calibration solutions can't really give you a top-quality result; there needs to be objective colour measurement at some point, and that requires hardware.
Still, the Spyder or one of the other calibrators I mention in that review might well be suitable, after a software revision or two.
I know you're a busy bloke but can you answer a couple of questions for a PI who is too dumbfounded by modern technology?
Will the Mustek DV3000 give suitable video quality to hold up in court, and does it have a recordable date/time?
Regarding standing up in court - maybe. All sorts of lousy images have been used successfully as evidence in court. Whether DV3000 video would cut it depends on what it's of, the context in which it's presented and how much of the case hinges on it.
These little Aiptek/Mustek/GrandVision/whatever cameras have excellent audio quality (well, by integrated-microphone voice recorder standards), but low resolution, low video frame rate, and a pretty wide angle lens. So they're generally useful, but not if (for instance) you need to recognise someone in the distance whose face ends up covered by only 12 pixels, or catch someone quickly handing a small object to someone else.
I can think of lots of situations where one of these things would be a great covert camera for investigation purposes, but it wouldn't be much use for many others - low-light work, for instance.
As far as I know none of the Aiptek/Mustek-alike cameras, including the DV3000, have a real time clock. All of the DV3000's files think they were made on the first of January, 2001.
I've enjoyed reading your articles and though you might be interested in these.
From the site: "Vibrant-coloured acrylic housed in an anodised alloy casing, the amazing GlowRing X2 uses state-of-the-art technology (gaseous tritium sealed inside internally phosphor-coated borosilicate glass vials, to be exact!)."
They sounded pretty interesting to me, but the Web site says that they don't ship outside of the UK. Have you seen anything similar available here in Australia?
The GlowRing X2 is an update of an earlier, smaller version, which used the same glow-tubes that showed up in Traser's Teknolite flashlights, which I reviewed here.
I haven't seen anyone selling them in Australia recently, though someone might well stock them. There used to be an eBay dealer selling them locally, but not at the moment.
Traser.com.au exists, but only seems to sell the Traser watches.
Google doesn't convert this yet
I am currently building an arcade machine and am following plans from the USA. The plans say to get 22 Gauge wire.
I have called a few electronics places here in Australia and still haven't got a straight answer about the size of wire I need to get! What metric size should I be asking for? The only thing I know about the US gauges is the higher the number the smaller the wire!
American Wire Gauge (AWG) can be converted to metric sizes pretty easily. Here's a page on the subject.
My sister and I have been arguing about whether the human body is mostly made up of water. She says it is, I say it isn't. I was sure I was right and so I jumped on the computer to prove it. However, it seems I'm wrong. Most sites agree with her.
Please, can you save me from the embarrassment of being wrong, or at least come up with something so confusing that she'll give up and leave me to sneak away?
Nope. She's right, you're wrong.
The water content of a human body varies, but if it's under 50%, you'll probably be feeling pretty darn thirsty.
This can be easily verified by weighing someone, killing them, drying their body thoroughly, then weighing them again.
Just curious, do people really give you money for no reason? I am asking because I had heard about a woman who had a Web site where she asked people to send her money to help her pay off her credit cards. It seemed odd to me that people would send her money for that. How come she is the one that became so famous? Was there something different about her site? How did she get so well known?
But that all being said, I was wondering how to set up a site were people could send me money. There is a legitimate cause around this and I would lay that out on my site and just let people decide if they want to donate or not. The other thing is, how would people find my site so they could send me something?
Yes, people do donate money to me. I still have to steal VCRs to support my cat habit, but every little bit helps.
People don't really send me money for no reason, of course; they give me money because they like my site and want to help me out. But they certainly don't have to.
There had been "e-panhandlers" before Karyn, but she's the most famous. Most of them didn't get diddly, of course.
There wasn't anything very "different" about Karyn's site, except for its sheer unmitigated gall.
So why was she the one that made it? It's a crapshoot, really; you're more likely to achieve this sort of success if you're the first person to think of something, but Karyn wasn't that original, and most of the people who become flavour of the month on the Internet for doing Some Damn Thing do so for the same reason that some band that's released four albums over the last ten years suddenly has a number 1 hit, then flops back into obscurity. No good reason at all, in other words. Popularity breeds popularity, in any environment where you've got an oversupply of stuff for a very large market, like music or goofy Web sites.
Setting up a site, however, is easy enough. Find yourself a cheap hosting service (free hosting probably won't cut it for a site that you expect to get any significant traffic at all, but $US10-a-month hosting for a lightweight almost-all-text site will probably be adequate, unless you succeed beyond your wildest dreams), tap away at the HTML editor of your choice, and stick that sucker up.
Getting traffic to your site is, as you may have guessed, the tricky bit. Post on relevant newsgroups and Web-boards, tell relevant Weblog-operators, produce press releases...
Spam isn't going to get you anywhere, but if you actually can make a good case for your cause, and (importantly) demonstrate that you're on the level, you ought to be able to get some money.