Friday, February 2nd, 2007
I believe we’ve fixed the shading problem reported last Fall — which practically zeroes my PV array’s output at 5:00 PM, with one hour of valuable peak-period sunlight remaining in the day.
The photo from September is difficult to make sense of. The trees on the left side of the image are tall Eucalyptus on the edge of the property, a few hundred feet from the roof. The lighter green foliage on the right 60% of the picture is attached to three birch trees planted adjacent to the house. It’s impossible to tell from the picture, but the sun is behind both.
The arborist returned this week to execute the plan from last Fall: trim the tops of the distant Eucalyptus and the nearby birch trees. We didn’t have to take any trees down, and fortunately for my breakeven date we only had to trim one Eucalyptus — those trees are tall.
The lower image shows what looks like success — a great big hole in the treeline for the sun to shine through. We’ll know next Fall whether we’re still generating electricity after 5:00 PM.
Tags: solar, pv, shading
posted to channel: Solar Blog
updated: 2007-02-05 14:51:01
Wednesday, January 31st, 2007
What does it cost to play music on a mid-fi stereo?
Electricity consumption for a 65 watt-per-channel integrated amp plus a pair of bookshelf speakers, as measured with my Kill-a-Watt:
| Mode | Watts |
| off (aka “phantom load”) | 0 |
| idle | 19 |
| playing | 20 |
The speakers are Polk Audio RT-25i.
Power consumption rose a bit with volume, presumably because the amp draws more power as it works harder. The “playing” value in the table above was measured at a moderately loud volume; in a bigger room, the amp could be made to work harder without deafening the guy with his hand on the volume knob.
It’s nice to see that this amp isn’t burning power when it’s turned off. The same can’t be said for all the components of my stereo; see measurments for the Polk Audio subwoofer.
Tags: kill-a-watt, a500, rt-25i
posted to channel: Conservation
updated: 2007-02-06 05:57:03
Tuesday, January 30th, 2007
I’ve never seen anyone play a drum machine like this.
Here’s another video that shows Haynes’ hands in close-up.
(Chuck, thanks for the link.)
Tags: hr-16, david haynes
posted to channel: Music
updated: 2007-01-31 06:00:28
Monday, January 29th, 2007
Looking into backup media options this morning, I came to the conclusion that I guess everyone else had come to already: tape drives are impractical for backups, because disk drives have gotten so inexpensive.
Why do I say tape drives are impractical? Because I can’t find a tape drive for less than $500. Here’s 36/72 GB internal DAT drive for $554
Seems to me if tapes were still a viable backup mechanism, the drives would be fast, have multiple interfaces, sell for about $99, and be more reliable than, say, the two DAT drives I used to use that died after a year apiece. Look at the example of inkjet printers: Amazon sells the Epson Stylus C88+ InkJet printer for $69.95. Sure, they make their money on the ink and paper. The people with tape drives have to buy media, too.
Iomega, famous for its Zip drive, has a “removable hard drive” product called the REV. It seems to offer the best of tape and fixed-disk worlds: pay for the complex mechanism just once, buy as much storage as you need, swap media around for offsite backups, etc. Its disks are likely more reliable than tapes, and offer faster (random) access to data.
The one problem of tape-media solutions that REV hasn’t yet addressed is cost. The drive costs over $300 (amazon: Firewire REV for Macs, $333; USB 2.0 REV for Win/Mac, $327), and a 4-pack of 35GB disks costs $180
GraniteDigital makes an interesting hybrid solution: firewire enclosures with hot-swappable drive bays. The unit is portable. Additional “media,” in the form of hard drives, can be purchased in various sizes and unlimited quantities. The device of course offers random access to files, is maintenance free, and supports offsite backups — although I’d guess hard drives are less robust, meaning more sensitive to physical trauma, than tapes or even REV cartridges. Best of all, this solution benefits from the aggressive pricing in the disk-drive market.
Here’s a table that rounds up my options. The items are somewhat arbitrarily configured, in terms of storage space; the three options with removable media would show a lower $/GB price if I’d buy more media, but I wouldn’t want to spend more money to get started.
| Mechanism | Configuration | Total $ | $/GB | Specs, purchase |
| Standalone USB2 disk drives | 2 500GB | $330 | $0.33 | Western Digital 500 |
| Internal SCSI DAT | drive + 5 DAT-72 tapes | $653 | $3.63 | Certance CD 72 Internal |
| Iomega REV | drive + 4 35GB disks | $510 | $2.91 | Firewire, or USB 2.0; media |
| FIREVue hotswap drive bay | enclosure + 2 160GB disks | $410 | $1.28 | enclosure, drives |
If nothing else, this investigation suggests that setting up a mirrored RAID for every workstation is an important first step in data longevity, as that halves the risk of a bad disk destroying days, weeks, or months of files. But a mirrored RAID won’t insulate anyone from an app that goes haywire and corrupts its database, or an accidental deletion, or a virus attack or other exploit.
So, it’s still necessary to make backups. But I have to say, I don’t love any of these options. Maybe I’m back to Amazon S3 after all, e.g. via JungleDisk. Is there really any difference between renting storage from Amazon and buying fresh media every few months? I guess that’s an analysis for another day.
Tags: backups, rev, dat, raid
posted to channel: Personal
updated: 2007-01-30 22:05:59
Tuesday, January 23rd, 2007
I was a sophomore in High School in 1982, at the height of Erno Rubik’s slick-fingered stranglehold on Western culture. His eponymous puzzle was everywhere, including my book bag.
The teacher of my Introductory Physical Sciences class, a bookish old Jesuit who wore tinted eyeglasses and whistled when he spoke, had already earned major points with the class by conducting a wood-distillation experiment in which one team of distracted amateur scientists had inadvertently shot a flaming section of tongue depressor across the room with a startling bang. He put himself into my hall of fame by taking a Friday off of the usual curriculum to teach us all how to solve Rubik’s Cube.
He distributed a one-page solution, which he’d typed up and mimeographed. (Check out the purple ink! If you’re my age, you’ll remember the sharp chemical smell, and that damp flaccid feeling that contradicts every sense of the phrase hot off the press. But, forgive an old man, I digress.)
I know not whether his work was original. Two solutions had been published by then, according to Wikipedia, although I would not be surprised if my professor had spent a few long evenings in the J.S. residence, peering through dark glasses at the bright cube, working out his own algorithms.
His steps are presented longhand, e.g. “front clockwise.” My pencilled shorthand is a variation of what I learned tonight has come to be called “cube notation.” Whether this shorthand was my invention or my teacher’s, I don’t remember; the official cube notation, introduced (acc. to Wikipedia) by David Singmaster in his 1980 publication Notes on Rubik’s Magic Cube, uses a prime mark rather than the “-1” I used to indicate counter-clockwise rotation. Had Singmaster’s book been the source of my teacher’s solution, I imagine he would have presented the notation as well.
Then, too, there’s the enigmatic description of the 12-step “Rubik’s Maneuver,” which Google finds only six references to — all of which are in Japanese. Had that phrase appeared in one of the published solutions, one would expect it to have propagated into the community, quotes intact. But no.
My personal best time was on the order of 65 seconds, according to the sweep second hand on the oven clock, before which I would perch on the kitchen stool for a few minutes of what passed for speed-cubing before the M*A*S*H reruns started. This was of course for the 3x3x3 cube; the 4x4x4 was a travesty of human puzzle-solving ability, as far as I was concerned. (Yes, I have seen the videos of Frank Morris solving a 5x5x5 in 1 minute 46 seconds, but clearly he’s not human.).
I took my cube apart during that quest for speed, and swabbed the guts with Vaseline — yet another Cube-born instance of simultaneous invention. But I didn’t spend a lot of time trying to improve on the basic algorithms, aside from one maneuver I discovered by manipulating a solved cube — it did a nifty job of orienting and permuting a fairly common combination of final edge pieces in fewer moves than prescribed by the ditto sheet. Unfortunately, the mnemonic for this maneuver escapes me.
The other formulas I can recite to this day, although I had no recollection of how to apply them until I found the original solution sheet (pictured above) during a garage-purging exercise. Under it was my original, much worn and beloved circa-1981 Rubik’s Cube. My immediate thought was to work through the solution again in silent tribute to the best day of my Jesuit education.
Unfortunately, 20 years’ worth of neglect and Vaseline residue had taken a toll; the adhesive on several of the Cube’s colored decals had been eaten away.
Two great resources for Cube aficionados: Wikipedia’s Rubik’s Cube article, and this amazing algorithm playback applet, which apparently takes any Cube notation as a URL argument, reverses it out of a solved cube, and then plays it back. Drag the cube to rotate it in three dimensions — amazing!
Tags: rubik's cube
posted to channel: Personal
updated: 2007-01-25 16:43:19
