|
Categories: all aviation bicycle gadgets misc motorcycle theater Thu, 29 Jul 2010Feel free to skip this one, it's kind of heady, and may not make a lot of sense. So, think about how a normal consumer transaction goes. I make a widget, and it costs me $5 to produce. Let's say I spend another $5 on packaging, distribution, marketing, etc. My total cost to put a widget in a customer's hand is $10. I charge them $3 on top of that to have some profit, total $13 charge to the customer. They pay their $13, and go away happy with their widget. Don't dwell on the amounts here, just get that process in your head: cost, product, price, overhead, etc. The end of that process is that I'm out my $5 product, and the $5 I spent on marketing and overhead, but I'm up by $13, so I've made $3 at the end of the day. Pretty straightforward. Do that a lot, and that's the basis of most trades. Theater, however, is an entirely different beast, as occurred to me the other day. In theater, you spend various resources to put together a product (a show) -- time, money, reused materials, etc. Rent costs a certain amount. If you assume a fixed-length run of a show, the product cost (the cost to put on the show) is pretty much fixed. However, unlike many other transactions, the customer doesn't diminish your supply of the product. Whether you have three paying audience members, or a full house, one night of a show costs the same, and uses up the same amount of your product. You end up with the same amount of product left whether you cancelled for lack of sales, or completely sold out. So income is, effectively, completely disconnected from cost. If you spend $100 on a show that is for whatever reason a hit, your profit would be enormous. If you blow $10 grand on a show that's a complete flop, you're out $10 grand. And through all that, the amount of product you have never changes -- your show is always (within the limits of what fate doles out in the form of actor performances, tech successes/failures, etc.) the same. Time is the only thing that diminishes the amount of product you have. I don't really have a point here, it's just such a weird business to be in. The normal rules don't apply. Posted at 16:22 permanent link category: /theater Mon, 26 Jul 2010After the questionable focus tests, I decided that the only real way to test was to take the lens out and use it. So, I wandered around Fremont and took pictures, being sure in each case that I was holding still, the subject wasn't moving, and the focus was locking on a reasonable thing. The end result was that at a variety of focal lengths, and a variety of subject distances, the lens focused pretty well 95% of the time. There were a few missed shots, particularly closer up. Basically, it's Good Enough. I'm disappointed that this is the result of spending $900 on a top-of-the-line lens, but I'm not sure what else to do about it. In the realistic light of day, no lens gets perfect focus every time, because the camera can't always get good focus. Obviously this one is well enough adjusted to work most of the time. So I'll keep plugging away with this one, and if it comes up obviously deficient in some other situation (such as the much lower-light environment of the theater), I'll sell it on to someone for whom it'll be the right thing, and look into other choices. In the mean time, it's nice to have my "good lens" back, and not be limited to primes (which take beautiful pictures, of course, so long as you want that specific focal length). Posted at 15:05 permanent link category: /misc Wed, 21 Jul 2010I sat down with a notepad and did some ordered and logical testing with my 24-70 last night. With no adjustment in the camera, the lens was focusing about right in most cases. This is good and bad. When I say "about right," that means that in something like 20% of cases, the focus was pretty unacceptably soft. When I dialed in +20 on the AF adjustment, the results were less predictable, with the focus sometimes right on (what?) and sometimes dramatically back-focused, like I expected. I didn't collect statistics last night, but I'll post a gallery of the focus shots later (I need to redo them with better light, to ensure that low light wasn't contributing to errors, although the lens has to shoot in low light in practice), and come up with some numbers. The bottom line is that right now I'm not comfortable with this lens's ability to focus correctly. This will probably end up with me renting that Canon 24-70 f/2.8L lens from Glazers again, to compare. The Canon L lens is pretty much the gold standard, so it'll be a good control result. Posted at 14:33 permanent link category: /misc Tue, 20 Jul 2010I got the lens onto the camera last night, and I'm honestly a bit confused about what I discovered. The best I can say is that the linearity of the focusing data seems questionable. With the full (+20 toward distance) adjustment dialed in, my 70mm at 1.5 feet test looked alright. Not ideal, but close enough. However, with that same adjustment dialed in, shots further away were dramatically back-focused (what you'd expect with this adjustment in place, and a correctly-functioning lens). With other tests, it looked like I was getting in focus consistently on the second shot, but not the first, and not by the huge difference I found in some other shots. I didn't have time to sit down and approach the problem systematically last night, but the answer I have right now is that I have no clue what's going on. I don't trust the lens to focus correctly. Once I get some rational test results that hit the situations I mostly find myself in, I'll have a clearer picture. I was too tired last night and fiddling with too many variables at once to have anything conclusive to say right now. More to come. I just need to approach the problem rationally and I should end up with a workable solution. Clearly the lens is getting close, and my messing with the AF adjustment in the camera was a lot of the confusion. Posted at 11:41 permanent link category: /misc Mon, 19 Jul 2010A quick recap of our story so far. Back in February, I took the plunge and bought myself a very nice lens, a Sigma 24-70mm f/2.8 HSM lens. It's roughly equivalent to the $1500 Canon lens of the same specs, but for $600 less. It was, at the time, the most I'd ever spent on a single piece of camera equipment; that includes the camera it was going on to. Fast forward a few months to June. I was talking with a friend, and we got to talking about autofocus, how it works, and how to test that it was working right. He recommended a test page, and I tried it out. Lo and behold, my new 24-70 lens was front-focusing a bit (the point that was actually in focus was about 10mm in front of where it should have been). So, I contacted Sigma, confirmed that they could fix the problem, and sent it in. A couple weeks later, it comes back. I excitedly fix the lens to the camera, and check it out. Nope, it's still out of focus. Oh wait, once I take out the camera's adjustment (which I'd put in before), it's way out of focus! It looks to my critical but unpracticed eye that they adjusted the focus point by 10mm alright, but in the wrong direction! Ok. Call Sigma back, explain the situation. They're very nice about the whole thing, offering to ship the lens back to them at their own expense. Get it shipped off, and the waiting starts. Almost two weeks after it was shipped off (having requested expedited service, since I'd already had the lens in once before), I got a phone call from them. It was one of their techs, and the (fairly confused) conversation boiled down to them saying, "Our calibration procedure doesn't cover the situation your test shows," and me asking, "So, the lens isn't guaranteed to focus in that situation?" The answer seemed to be that sure, it should, but that's not what we're testing for. (For the nerds in the house, I was testing at 70mm focal length, f/2.8, focused ~1.5 feet from the lens -- which is within its nominal capabilities.) There was a strong undertone of "Why would you do that? It's not a macro lens. You're an idiot, aren't you." Now, nearly two weeks after that phone call (the lens came back to me three weeks to the day after I sent it, so take all my "two weeks" descriptions with a grain of salt), it has arrived at my desk again. The note that came with it says:
IF YOU WANT THE BEST RESULT WE NEED YOUR CAMERA WITH THE LENS. CHECKED ALL IN GOOD WORKING ORDER. That sounds to me like they helpfully left the lens with its even-worse calibration in place. I'm not sure why my camera should be that different from their camera. I know that I saw the exact same error with two very different camera bodies (an XTi and a 7D), so it's hard to imagine that their test 7D could have been so vastly different. Perhaps the most galling thing about the whole experience was that phone call. The tech sounded like he was rolling his eyes and making the yapping-mouth gesture with his hand to the other guys in the room as he talked to me. It's news to me that a lens should be expected to not accurately focus within its design parameters. The fact that it has the problem doesn't make me feel very confident of its ability to accurately shoot the demanding photos I'm taking (low light, moving subjects, etc.). The fact that I had to send the lens back twice with a $900, top-of-their-line piece of equipment is very worrying, particularly if (as it appears) the second trip to the factory resulted in them leaving things exactly as they were after the first trip messed it up. The half-suppressed "You're a complete idiot" vibe is just icing on the cake. Of course, the real test will come tonight, when I can bring lens and camera together. I'll have an update soon with the outcome, but if at all possible, I'm never going to deal with Sigma's warranty service again. The method I've heard described for dealing with this is to buy 3 copies of the lens you want, then send back the two with the worst focus. An expensive way to do it, but it honestly sounds more surefire than going back to the factory for adjustments. Color me seriously unimpressed. Posted at 13:43 permanent link category: /misc Fri, 09 Jul 2010I had a free day today, and decided to take a little ride before it got too hot. Seattle's been dealing with fairly bipolar weather lately, and a week ago, it was in the 50s and raining. Today it topped 90 with perfectly clear skies. I loaded up Google's map page, and started scrolling around. I found a little road near Carnation, called Griffin Creek Road. It looked delightfully twisty and interesting, and I'd never heard of it before. I figured there was a reasonable chance it was a dirt road, but it was worth a trip to find out. I loaded up the Camelbak with ice and water (a fine idea, and it stayed very cold until my return home several hours later), and headed out on the Ninja 250. My usual route for this kind of trip is to take 520 out to Avondale (520 really just turns into Avondale where it terminates), then turn right on Union Hill Road, and follow that through its twists and turns to Snoqualmie Valley Road, which runs up the west side of the Snoqualmie Valley, and eventually leads to some of the best twisties in the area (although they're brief, and relatively high traffic). This time, I continued out to Carnation, and south a bit to pick up Griffin Creek Road. I got there, and it was indeed a dirt road, but with an internal shrug of the shoulders, I headed up, figuring I'd turn around if it got too uncivilized. I finally stopped a few miles up, where the road suddenly narrowed from a wide single lane to a narrow single lane. I didn't have anything to prove, and paused in a convenient pool of shadow to take off some of my gear and spend a few minutes without earplugs in. It was very pretty, although not particularly picure-worthy -- the picture would have been titled, "Trees." Any follow-on shots would have been titled things like "More trees," or "Bike with trees." I turned around, and on a whim, decided to turn south on Carnation-Fall City Road, and make my way to Snoqualmie Falls. I was most of the way there, so why not? I made my way to the falls, and, sweating in my gear among the tourists, took a couple of quick, "I was really there" photos.
I made my way back to the bike, and back down the hill, the way I'd come. I stopped at the delightfully deceptive "$1.00 sweet cherries" booth ($1 buys you a miniscule amount of cherries, although you can certainly get those few for a dollar). I rolled on. As I was coming back along highway 203, I was reminded of something Jeniffer and I had experienced and discussed in our trip around the North Cascade loop last month: no one knows how to pass any more. So, the way it works is this: on any two-lane road (ie, one lane each direction) where passing is permitted, well, passing is permitted. Generally speaking, to do it legally, the vehicle in front of you should be going less than the speed limit, and you shouldn't exceed 15 MPH over the speed limit in your overtaking maneuver. In Washington, at least, it's illegal for any vehicle to detain 5 or more vehicles behind it, and it's required to pull off the road to let the other guys go past. I suspect other states have similar rules, and it's certainly a common-sense idea. Anyway, it was interesting to me how atrophied the skill is. Everyone who drives on a multi-lane freeway is used to simply having a second lane available for overtaking. Someone in front of you not going fast enough? Pull into the next lane, and overtake them at your leisure. On a two-lane road, it's not that easy, of course. First, you have to determine if it's safe to pass. That is, is there oncoming traffic? Do you have a dashed line? (If there's a solid yellow line in your lane, passing is prohibited, usually for a very good reason.) Is the guy in front of you behaving predictably? Sometimes the best course of action is to back off, or pull over for a few minutes to let them get ahead. If things are safe, turn on your turn signal, gun the motor, and jam past the slowpoke. Signal back into your lane, and slow down to something approximating the speed limit for the benefit of the local constabulary. It's really easy, particularly if you're on a motorcycle. Of course, not all motorcycles are created the same. The Ninja 250, while it embodies many fine attributes, will never be mistaken for a powerhouse in the modern pantheon of motorcycles. With the 250, you have to plan your strategy with a bit more care -- find somewhere with good visibility, a useable gap between cars ahead of you if you can't pass the whole line in one go, etc. It's actually a pretty interesting challenge. On our trip on Highway 20 last month, Jeniffer was riding her BMW F650GS (a confusingly-named 800cc vertical twin), and I was on my Ninja 250 (a rationally-named 250cc vertical twin). Her bike produces something like 80 HP, mine screams along with maybe 28, for a not terribly substantial difference in weight. It was with interest, therefore, that I would pass people, and wait a surprisingly long time for Jeniffer to catch up, despite rolling through some prime passing areas. When I asked her later what was happening, she said she's just not comfortable passing, and deferred to my decade-plus of experience. This is by no means an attempt to rag on Jeniffer or her riding skills, it's just interesting to me that we have such different approaches to it. But similar to both my ride today, and our ride last month, was the fact that very few other drivers showed any inclination to pass. We'd come upon these long trains of cars unhappily guttering along behind a slow vehicle, and I'd pass them in one or two leaps as conditions warranted, and Jeniffer would catch up to me a bit later. Therefore, I'm urging you, fellow drivers, break free of your multi-lane roadway habits, and actually take a chance on your next road trip. If you're driving a modern car, you have so much power available to you in most cases that passing is little more than a trifle, to be accomplished with as much difficulty as most people execute the pouring of a cup of coffee. You can do it. You can break free of the mental chains. And you can get around that slug ahead of you. Posted at 18:22 permanent link category: /motorcycle Categories: all aviation gadgets misc motorcycle theater Written by Ian Johnston. Software is Blosxom. Questions? Please mail me at reaper at obairlann dot net. |
