I like the control and flexibility I have with my own web site, but (alas) I have been unable to do a couple of things here (like deal with the spammed comments, upgrade to a new version, and to get Ping.fm to write to this blog).

I have been scanning old B&W photos from my Canon F1, and by and large, these pix are not as sharp as I remember!  For example, here is a 100% crop  of a pretty good B&W image from 1993:

You can see the grain of the ASA 400 film here pretty clearly, but the image is still pretty nice.  But it is not very sharp.  Compare this to a picture I took at the market this weekend:

This comparison favors the digital camera, in my (humble) opinion.

2. The various races on the Jumbotron. I think we all realize that the race itself is fake (they must have 10 or 20 of these races to choose from), but that sorta doesn’t matter because the winner does not matter. You can take your ticket for that Dunkin’ Donuts “Cuppa Coffee” and get a free cup of coffee, even if it doesn’t win. How are they going to know? Check the UPC on those cards–they are ALL THE SAME.

3. Cheers from the Jumbotron. Some/many of the cheers that are led from the Jumbotron have crowd noise already built in–they pump extra crowd noise through the PA system to make it seem like there is more noise than there really is.

Everything else at The United Center is real, though.

Random, Somewhat Related Comments About Bulls Games

• I wish they would offer one or two Bulls games where they turn off all the extra audio/visual crap: No Jumbotron, no loud music, no stupid, contrived races during timeouts, no scrolling and flashing ads on the ring-scoreboard.  In my childhood, all basketball games were like this.  I went to a West Aurora basketball game on Friday, and it was like this.
• How about the occasional National Anthem where the featured singer is the crowd! (They have the words on the Jumbotron–good; everyone should know the words, but it is good to have a reminder.  But, guys, the last sentence of the National Anthem is a question, and should end in a question mark.)
• I’d like to see something sold for under $1–how about a 6-ounce Coke. I bet they’d sell a ton of these!

On Saturday, I ran 3.8 miles in 48 minutes, and it felt great!  On Monday, I did a one-hour workout in the gym–also great.

I now know the benefits of steroids on excercise!  And my thumb is better, too.

First impressions

• Super fast!  I accidentally took 3 pictures when I pressed the shutter the first time.  Sweet!
• Very quiet!  I’d say (remember, I’m standing in Best Buy) that the shutter/mirror makes half the noise of the XSi.
• Sturdy as can be!
• The reviews are fantastic (as you might expect).

All that, and it is a real bargain: See the Best Buy Web Site.  (Just kidding–it is close to 3X more expensive than the XSi!!)

 I have had my Canon EOS XSi (known in Europe as the 450D) for one year now, and I’d like to give you all an update of my experience with it.

Equipment

• Canon EOD XSi
• Canon EF-S 18-55 mm f3.5-5.6 IS
• Canon EF-S 55-250 mm f 4-5.6 IS
• Canon EF 50mm f1.8
• Tokina 11-16 mm f2.8 (purchased 6 weeks ago)

The Newly Discovered Good Parts

I bought the 11-16 mm Tokina lens, and this lens seems to be very, very good.  The constant f2.8 is way cool, and the 11mm wide angle is stunning!  So, the “kit lens is not too wide” gripe is mollified.

I discovered how to rent lenses: lensrentals.com.  I rented the amazing Canon 85mm f1.2 L lens recently.  This lens issweet. But it was less than spectacular for me, probably because of my inability to focus manually effectively (see below).

RAW mode, coupled with the bundled Canon software, gives excellent control over the final image quality.  I took some pictures by candle light at Channukkahhh and processed the images to stunning results.  The biggest piece of that is the color balance control (which, I have now discovered, all of the image tools I had been using already have).

The Problems

In my previous post, I mentioned some shortcomings of this equipment.  In summary, I said that this is still a great choice, but there are a few little problems.

These problems are still there, and (I think) I feel a bit more strongly about them.  My biggest problem is my difficulty in getting truly sharp pictures.  By this, I mean that when I look at any image I take at 100% (that is, with enough magnification to see individual pixels), I almost always see evidence of camera motion.

One piece, mentioned previously, is the lightness of the camera.  The motion of the mirror seems to be shaking the camera in a way that neither image stabilization nor holding it steady can mitigate.  Even shooting on a tripod is problematic!

Another piece of this is that it is impossible to focus this camera manually.  This is a real problem for me, having always focussed cameras by hand, up to now.  There is no chance of focusing through the lens with any accuracy because there are no focusing aids available.  (I just discovered this focusing screen, for $105.  I might have to buy this little toy.)  The Live View focusing is enhanced by a 10X magnification view, but there is so much image motion when you touch the lens, and the focusing ring on each of my lenses is so light and so sensitive, this method is pretty much useless.

I am learning about Lens Microcalibration,  which my camera does not have.  Maybe this is part of my lack-of-sharpness problem?

I am beginning to suspect that my 55-250 zoom’s IS is faulty.  I have done a simple test on this lens and published the results here as a PDF document.  I took 4 pictures: At 55mm and at 250mm; with and without stabilization turned on.  I fixed the shutter speed at 1/125 second, which I might be able to hold steady without stabilization, but WITH stabilization should be very sharp.  (I actually took 8 pictures (two at each setting) and selected the sharper on in each case.)  I concentrated on holding the camera very steady.  Look at the results for yourself, but I do not see any sharp pictures, and there is only a small difference between stabilization on and off.  Here is a very sharp image taken with this lens, but not by me.

The eye cup on the viewfinder keeps falling off.  It clicks onto the eyepiece, but only barely.  The action of putting the camera into my backpack is often enough to jar it loose.

Tentative Conclusions

My experience with Canon’s in-the-lens Image Stabilization is less than impressive.

Maybe I should have gotten the Canon 40D, or the 50D.  The price of the XSi is still a BIG advantage.  And some of the professional reviews say that there is very little difference in image quality between the XSi and (in particular) the 50D.  These two reasons were why I did not seriously consider a ?0D.  But the professional-level controls (especially the easily interchangeable focusing screens, and the availability of focus microcalibration), and the heavier body are looking pretty good to me right now.

I am tempted to rent the 70-200mm f2.8-L IS lens, if only to see if my suspicion about my big zoom are justified.

Right now, the LHC is the particle accelerator with the highest particle beam energy (1.18 TeV), surpassing the Tevatron (0.98 TeV) by a hair.

But “power” is voltage *TIMES* current, and the Tevatron had them beat by a mile in that measure (at least for now).

The powers of these beams are proportional to:

LHC : 1.18E12 volts * 5E9 particles (~5E21)
Tevatron: 0.98E12 volts * 1E14 (or so) particles (~1E26)

Let me count the ways!

1. Header files. Why do we need header files??
2. Object files. The only way to know if everything is compiled right is to recompile everything. This is related to #1, I’m pretty sure.
3. Memory management–what a PITA!! I know, I know–doing your own memory management is much more powerful. I DON’T CARE! Garbage collection is (IMHO) much more conducive to GETTING THE PROGRAM COMPLETED!
4. No inner classes
5. What’s up with all the double colons? NOT READABLE!
6. Every method has to begin with “ClassName::”. Can you say “Redundant”?
7. The default visibility for any subroutine is GLOBAL. Global variables are EVIL!
8. You can’t test a class by simply adding a “main” method (see 7: You don’t have a concept of “ClassName::main”)

Of course, you can do everything in C++. I hate that argument: you can do everything with ANY programming language (Fortran, Basic, Ada, assembly language, even the Turing Machine).