350D review and AF Speed

Thanks for an interesting post. What you say pretty much agree with what I said (or try to say) that I've learned so far. My conclusion was that I couldn't see any difference in speed from the test I've been able to make.

When I commented that it seemed 20D was a bit more consistent in some situations, possibly because of more focus points I was talking about ability to focus on a static subject (think I said it was inside a store) and that more focus points icreased the chance of finding a good spot to focus on.

Of course, tracking birds, animals etc. is totally different. I've seen the comment that af will focus with center sensor then lock, and use the other sensors as necessary for tracking. So more points want help initial focus, but might help tracking by having more points to work with. This is only speculation as I haven't tried it.

So, as long as you don't set the camera to use only one point focus you increase the chance that one sensor will lock on to something. You also increase the chance of the wrong sensor being used, but guess that's a worthwhile tradeoff in many situasions. With my old Eos 33 film based body I was impressed by how often it guessed correctly. Can't wait to try the 20D, if it only would stop raining.

As for durability I was only refering the guy in the store, actually one of the better stores where I live in Norway. And he didn't say anything would break. He said that even though the lens mount is metal it's only fastened in plastic. No complete metal frame in the 350. So there's a chance of slight misalignment using the 350 with too heavy lenses. He also claimed to it was info from Canon. He had no reason to exaggerate to sell a 20D as I'd spent about the same amount of money in the store by buying a more expensive lens with the 350. Sounds reasonable to me that a metal body would be slightly stiffer than a plastic one. Still, if it had been a common problem Canon wouldn't have reccommended 350 for use with lenses up to 600mm.

I also wanted better battery life and the 20D viewfinder and like Canon bodies with the second control wheel on the back. On the other hand, I'd prefer the smaller size (and price) of 350. Had to decide, and ended with 20D.

Anyway, thanks for the comment and the impressive shots, both birds and lenses. Wish I could afford one of those!

Svein Eriksen said:

As for durability I was only refering the guy in the store,
actually one of the better stores where I live in Norway. And he
didn't say anything would break. He said that even though the lens
mount is metal it's only fastened in plastic. No complete metal
frame in the 350. So there's a chance of slight misalignment using
the 350 with too heavy lenses. He also claimed to it was info from
Canon.

Click to expand...

This is incorrect, the metal mount is not mounted on plastic but instead to the steel sub-frame.

Svein Eriksen said:

I also wanted better battery life and the 20D viewfinder and like
Canon bodies with the second control wheel on the back. On the
other hand, I'd prefer the smaller size (and price) of 350. Had to
decide, and ended with 20D.

Click to expand...

Amazingly enough, I have been getting more shots per charge out of my XT than I have been getting out of my 20D. I was very concerned about the battery in the XT at first so I have been paying very careful attention to this aspect of the camera. I would have never imagined this would be the case but that has been my experience so far. The battery life between the two has been very close so it is difficult to say if one has a real advantage over the other.

Greg
--



http://www.pbase.com/dadas115/

Those are impressive shots. I love the blur on the wings compared with the sharp body, in No 2 particularly. It's so much more interesting than the "frozen" images that so many bird shooters produce. You must have very steady hands to hold that combination so well!
--
Laurie Strachan

These types of shot are fairly tough to get as the birds' bodies tend to move a lot as they flap their wings. Even if you get the panning motion right, the pictures are often spoiled by the body motion of the bird. It just takes patience and some luck.

Thanks for your kind words,

Greg

--



http://www.pbase.com/dadas115/

Svein Eriksen said:

Thanks for an interesting post. What you say pretty much agree with
what I said (or try to say) that I've learned so far. My conclusion
was that I couldn't see any difference in speed from the test I've
been able to make.

When I commented that it seemed 20D was a bit more consistent in
some situations, possibly because of more focus points I was
talking about ability to focus on a static subject (think I said it
was inside a store) and that more focus points icreased the chance
of finding a good spot to focus on.

Of course, tracking birds, animals etc. is totally different. I've
seen the comment that af will focus with center sensor then lock,
and use the other sensors as necessary for tracking. So more points
want help initial focus, but might help tracking by having more
points to work with. This is only speculation as I haven't tried it.

So, as long as you don't set the camera to use only one point focus
you increase the chance that one sensor will lock on to something.
You also increase the chance of the wrong sensor being used, but
guess that's a worthwhile tradeoff in many situasions. With my old
Eos 33 film based body I was impressed by how often it guessed
correctly. Can't wait to try the 20D, if it only would stop raining.

Click to expand...

Interesting how different our needs are. The only time I use the multiple focus points on my XT is when I let my wife take snapshots. I bought a DSLR for the selective DOF, and the last thing I would ever want is for my camera to decide which of the 7 or 9 (!) points of my image will be in focus. Before AF cameras, I never randomly focused on everything in the frame. Now, I always use only one AF point at a time, but maybe I'm strange.

Sal

nonsense. You will see a difference in AF speed with the 20D, even without a f2.8 lens. With a f2.8 lens, the AF is then even MORE precise. But for capturing action, the AF on the 20D is much better than the 10D I had, which was better than my old D60/D30. I could catch action shots easily with my 70-200 F4L lens on my 20D, that my 10D (with the same lens) would have a bit of a struggle doing.

HappySnappy said:

But it's already been said that the only way you wil see a
difference is with a 20D AND a lens at F2.8 or better.

Click to expand...

--
http://www.garylambert.us/gallery
http://www.onemodelplace.com/jahfakin

really? they compared the AF between a 350D and 20D?

link?

i really would love to see an on-body AF assist light. that strobe flash thing is about as effective as rowing upstream w/a plastic spoon.

River Side said:

just use the same lens and give us figures.. we'll estimate if it's
going to be slower with a different lens.. at least we'll get to
know the ballpark estimates..

imagingresource gives these timing figures.

--
* * * * * * * * * * * * * *
3 0 0 D + 1 8 - 5 5 m m
5 0 m m f / 1 . 4 U S M
7 5 - 3 0 0 m m U S M I S

* * * * * * * * * * * * * *

Click to expand...

It was just surprising to me that the shutter lag on the decent point and shoot cameras these days is equal to that of a dSLR, especially since they talk about it in every single P&S review, but rarely mention it in the dSLR reviews.

I disagree that .1 seconds is not a lot or not noticable, the bottom line is the picture isn't taken when you push the button, and you have to account for that.

If you half press for focus and exposure lock, I don't see why there would by any shutter lag, yet alone a tenth of a second, on an SLR. What does the camera have to do between pushing the button and releasing the shutter? It certainly doesn't take 100 ms (or even 1 ms, or even 1/1000th of a ms) for the shutter trigger to physically communicate with the shutter that it should open.

I'm sure there's an explanation -- like I said, it's surprising to me, and disappointing, that the shutter lag on these cameras isn't non-existent (or at least nominal).

John.

The mirrir must flip up.

Do you really hold your camera like that even with big lenses attached? Or is that only for a demo that its possible?

--
Deepak

http://www.images.rnhinfo.com/

OCJohn said:

It was just surprising to me that the shutter lag on the decent
point and shoot cameras these days is equal to that of a dSLR,
especially since they talk about it in every single P&S review, but
rarely mention it in the dSLR reviews.

I disagree that .1 seconds is not a lot or not noticable, the
bottom line is the picture isn't taken when you push the button,
and you have to account for that.

If you half press for focus and exposure lock, I don't see why
there would by any shutter lag, yet alone a tenth of a second, on
an SLR. What does the camera have to do between pushing the button
and releasing the shutter? It certainly doesn't take 100 ms (or
even 1 ms, or even 1/1000th of a ms) for the shutter trigger to
physically communicate with the shutter that it should open.

I'm sure there's an explanation -- like I said, it's surprising to
me, and disappointing, that the shutter lag on these cameras isn't
non-existent (or at least nominal).

Click to expand...

0.077 - 0.095 second shutter lag is nominal unless you are Superman.

I handle it like that pretty much all the time without any problems.

Greg

--



http://www.pbase.com/dadas115/

I can second that. I had a 10D and then a 20D, and the 20D was usable for action shots (maybe 70% + keepers) as opposed to less than 50% with the 10D.

Well, the two new ones I bought are actually for my kids... the XT did not exist when I bought my own 20D half a year ago. I also have the Mark II and I believe the AF on that one is even better than with the 20D, but I have not measured it precisely.

PeaceFrog said:

PeaceFrog said:

He said that even though the lens
mount is metal it's only fastened in plastic. No complete metal
frame in the 350. So there's a chance of slight misalignment using
the 350 with too heavy lenses. He also claimed to it was info from
Canon.

Click to expand...

This is incorrect, the metal mount is not mounted on plastic but
instead to the steel sub-frame.

Click to expand...

Also thought there were a metal subframe, and was really surprised. The dpreview test seem to confirm that there's not a lot of metal in the 350. Take a look at the construction diagram on this page. http://www.dpreview.com/reviews/canoneos350d/page4.asp

Thanks Sal. Your info is very useful.

Great bird in flight. I am pretty sure the person behind the camera is more important for shots like this. Give me the MkII and I am sure I can't catch flying birds like this!
Thanks for misleading me with your great skill...

Good post, thank you... of course, I already ordered my 20Ds (2 of them, one for each kid) so it's kind of too late to change my mind. Part of the reason is also battery compatibility, so when the 3 of us are in the field, we can help each other there, and when on a long trip, we can take ONE charger for all 3 cameras etc. Also, menu settings will be the same and it will all around be more of a "sharing/learning" experience with the same camera, rather than dad having the better one and the kids don't... rightly or wrongly, if the shots by the kids are lacking due to technique, they'd blame the camera if they had a different one, rather than focus on technique.

That aside, I would second your opinion that it would be hard to get a bird to train to do the same performance in the same light for a series of test shots with different bodies and lenses. Have you tried hearding cats? Kind of the same thing. Or what about bringing all the parents on one of your kids sports teams to agree to practice times? Maybe the bird is not all that hard to train after all.

In regard to your bird shots, like mine, I try to take them when the bird is flying at a steady distance to the sensor, which should present little problem for the AF. My kids like action sports, mostly soccer, where the subject is often moving rapidly and iradically either toward or from the sensor. Hence, AF performance and predictive tracking is critical. I had problems with this with my 10D, which is why I bought a Mark II. My rate of keepers went from less than 50% with the 10D to 90% + with the Mark II, using the same lenses. Big difference, obviously. Since I have a Mark II, I have not used the 20D for action shots, as I don't want to be in experimental mode, when the once in a life time shot were to happen. But in shooting non-soccer stuff, some of it action, I can tell the 20D is much better than the 10D in this regard, but I don't know if it's on par with the Mark II, and/or now the XT.

One would think that purely from a marketing perspective, if Canon has decided to keep 3 product lines, as in the 1 Series, a 10/20D type series, and then a Rebel series, they will need to differentiate their product so that their customers will want to pay more for a 20D type body. As they did with the original Rebel, where they crippled some of the software, simply for product differentiation. Of course, that backfired and, perhaps as a result, they have now decided to discontinue a 20D type in between product line, and hence felt comfortable with putting the same AF sensor in the XT as in the 20D. We shall see.

Based on what we are seeing from both Nikon and Canon, now that their top of the line products produce consistent results that are largely superior to what is needed, they are focusing thier efforts more on having a variety of mass market products to get consumer SLR users to go DSLR. I suppose, the pros and very serious amateurs who decide they need to invest the $5,000 or so in a pro body have already done that, and with the current series of pro bodies out there, the need to upgrade to future models is a study of diminishing returns (how much more than a Mark II does one possibly need???).

Great post, thanks again, now back to training a bird so we can do the subject flying to and fro the sensor...