Can someone who is hands-on verify some 40D points please?
1.The white paper says (speaking of "live-view" shooting):
"Short lag time captures near-instantaneous movements."
The "near" has raised my eyebrows. Can someone give an opinion about the lag after the button is pressed, with this situation in mind - The camera /tele is tripod mounted/pre-focused on the edge of the nest with the mommy-bird approaching with food for waiting fledglings. Remote release in hand.
The target "moment" is the instant of touchdown, wings still spread, head forward with food towards gaping baby beaks. "Wait,...wait,....NOW!"
Is the live-view lag time suitable for this sort of timing, or is it annoyingly "late", unless somehow magically "pre-timed"?
2. Can someone give the actual on-screen pixel dimensions of the image when using remote live view with a computer?
3. Is my understanding correct in that a "burst" of rapid shots (or a sequence of slower shots) may be fired using live-view, with the mirror remaining up throughout?
4. How great a concern is "heating" during MLU live-view while waiting for "the moment" as described above?
I posted this as a separate topic but didn't get enough response for me to determine whether or not my sample of the 40D needs to be serviced. Hopefully, owners of both the 30D (or even the 20D) and 40D can perform the same test and report back. Basically it concerns the AI Servo focusing of the 40D vs the 30D in low light (around 2 ev). To quote from the original thread:
"I was trying the 40D and the 70-200IS F2.8 in really low light and had the camera set to AI Servo (from a bird shoot). It wouldn't lock focus at all - to my surprise and chagrin, since I thought my 30D usually had no problems in this type of light. So I put lens on the 30D and sure enough it had no problem acquiring focus. Same results with the 24-105L.
I then switched to one-shot mode and the 40D seems better but not by much, I thought.
The light, in a dimly lit room, was such that: ISO 200, F2.8 at about 1 sec., a fairly typical flash situation say, in a low lit club.
I wonder whether this is a 40D 'problem' or perhaps just my sample."
Heat is a problem with live view and it is really not suited for action type shots, more for macro and landscape work, on a colder day heat may not be a problem, on a hot day especially in the sun it definitely will. On a monitor I'm not sure how large but I am assuming as large as the actual image if you have a monitor that big. I'm not sure about #3, I'm hoping someone else will answer that, I'm assuming it will, since I did use live view for 3 bracketed shots, I'll try it when I get home and tell ya what it does.
Al
Edited by Allen_Maestas on Sep 20, 2007 at 08:22 AM GMT
Sep 20, 2007 at 10:05 AM
Allen Maestas Offline • • • • • • Upload & Sell: Off
I posted this as a separate topic but didn't get enough response for me to determine whether or not my sample of the 40D needs to be serviced. Hopefully, owners of both the 30D (or even the 20D) and 40D can perform the same test and report back. Basically it concerns the AI Servo focusing of the 40D vs the 30D in low light (around 2 ev). To quote from the original thread:
"I was trying the 40D and the 70-200IS F2.8 in really low light and had the camera set to AI Servo (from a bird shoot). It wouldn't lock focus at all - to my surprise and chagrin, since I thought my 30D usually had no problems in this type of light. So I put lens on the 30D and sure enough it had no problem acquiring focus. Same results with the 24-105L.
I then switched to one-shot mode and the 40D seems better but not by much, I thought.
The light, in a dimly lit room, was such that: ISO 200, F2.8 at about 1 sec., a fairly typical flash situation say, in a low lit club.
I wonder whether this is a 40D 'problem' or perhaps just my sample."
No idea man, I could not tell you why the 30D would do it and not the 40D? Make sure everything is the same, focus point, ISO, etc... and if the 30D will do it and the 40D won't then maybe your 40D is bad? I sold my 30D so I can't do a comparison, but I'll try what you described, and let you know, although I don't have that exact lens I do have a prime 2.8 I can use.
mschecan wrote:
Everything was the same. Subject, AI Servo, type of metering(spot, evaluative, etc).
ISO, Av, Tv, etc..
I appreciate your reply.
Thanks
Yeah, unless someone beats me to it, I'll try it tonight and see what I get. Maybe it was just a glitch? Have you tried it again to see if it still does it?
Yes, tried it several times over the last few days.
Anxiously awaiting the resultrs of your low light experiment.
Marq
Allen_Maestas wrote:
Yeah, unless someone beats me to it, I'll try it tonight and see what I get. Maybe it was just a glitch? Have you tried it again to see if it still does it?
Though the mosquitoes and bad weather have kept me close to home so far, I am loving my new 40D. Live view. mode 2, remote switch, 5x = wonderful for extremely long lenses. Tried it with a 2x on my Sigmonster. Great. Not only can I wait for the right time to take the pix, and yes, almost instant since the mirror is already up, but it is very quiet. And, I can see when the lens/camera has stopped shaking, how much the wind is affecting things, etc. Yes, you can take more than 1 shot since the mirror stays up. Though I do have to press the + button to get my 5x back. Mommy bird landing on the nest would be a good example. I have a setup in my back yard where I am 25 feet from the birds. My 20D does scare away many birds. Not the 40D (even at 6.5 fps.) And, the mode 2 is so much quieter than that.
Great camera, can't be much more please (well if it had ECF I'd be even more pleased). WAs surprised at the kinda cheap battery door, plastic hinge and all. Took about 1 gram of force to pull it off. It can see guys losing it in the field during battery changes.
"I was trying the 40D and the 70-200IS F2.8 in really low light and had the camera set to AI Servo (from a bird shoot). It wouldn't lock focus at all - to my surprise and chagrin, since I thought my 30D usually had no problems in this type of light. So I put lens on the 30D and sure enough it had no problem acquiring focus. Same results with the 24-105L.
I then switched to one-shot mode and the 40D seems better but not by much, I thought.
The light, in a dimly lit room, was such that: ISO 200, F2.8 at about 1 sec., a fairly typical flash situation say, in a low lit club. ...Show more →
AI Servo is for moving targets, I doubt if even the best IS would handle a 1 second exposure on a moving target. At 1 second, your target must be stationary, which calls for one-shot AF.
The 40D is a good low light focusing camera.
This test was shot with an EF-S 17-85 at 85mm using only the light from a 4w night-light.
2 seconds @ 5.6 at ISO 100 is EV4 which is pretty low light. Single-shot AF is good enough for me at this level.
I shot my grandson's first soccer game earlier this week using AI Servo, EF 70-300 IS, shooting at ISO 1600 and 1/400th in Tv mode (typically EV10-EV12). Out of 84 frames, I only had 1 OOF shot, and it was clearly pilot error. In other words, I haven't seen any issues with AI Servo under normal use.
Fred
Sep 20, 2007 at 07:18 PM
Allen Maestas Offline • • • • • • Upload & Sell: Off
Yeah, I agree with Fred, tried seeing how AI Servo would do tonight inside with a Tokina 100mm 2.8 macro, which is not a fast focusing lens, and it did great. Got down to the point where there was almost no light before it finally stopped focusing. I then switched it over to AI Focus and it continued to focus until it was practically pitch black, just a small amount of ambient light coming through the windows, no inside lights on. I was kind of amazed. Much better than the 30D ever did. I did notice as it started to not pickup focus on AI Servo, that if I switched from a single focus point to multiple focus points and visa versa that I could sometimes get it to focus, but I mean at that point you would be using a flash anyway, so why not use a pre-flash to pickup focus? I mean trying to take a shot of a moving subject at 1 second or slower even with IS it's going to be blurry, even with flash you are going to have some blur, for a single shot anyway. I really think birders and sports shooters are going to love this camera, especially when considering the price.
Thanks for the test results. I agree that one has to make do with what works, i.e., use one-shot in very dark situations. However, my 30D works in AI Servo mode in lower light than the 40D and am wondering whether this is an indication of a not quite optimized AI Servo mode on my 40D.
Also, in your example, you may have gotten focus because the localized lighted portion of your example may have effectively more EV than the that of the averaged scene.
Thanks for taking the time for the experiment. Your key point was that you thought the 40D performed much better than the 30D. Mine doesn't by quite a bit. And my test subjects was nowhere near approaching pitch black. So I suspect that the AI Servo in my 40D is not quite up to specs. I like to shoot BIF and need all the help that I can get. I see a trip to the Canon service center in the near future for me.
Thanks again.
Marq.
Allen_Maestas wrote:
Yeah, I agree with Fred, tried seeing how AI Servo would do tonight inside with a Tokina 100mm 2.8 macro, which is not a fast focusing lens, and it did great. Got down to the point where there was almost no light before it finally stopped focusing. I then switched it over to AI Focus and it continued to focus until it was practically pitch black, just a small amount of ambient light coming through the windows, no inside lights on. I was kind of amazed. Much better than the 30D ever did. I did notice as it started to not pickup focus on AI Servo, that if I switched from a single focus point to multiple focus points and visa versa that I could sometimes get it to focus, but I mean at that point you would be using a flash anyway, so why not use a pre-flash to pickup focus? I mean trying to take a shot of a moving subject at 1 second or slower even with IS it's going to be blurry, even with flash you are going to have some blur, for a single shot anyway. I really think birders and sports shooters are going to love this camera, especially when considering the price.
mschecan wrote:
Also, in your example, you may have gotten focus because the localized lighted portion of your example may have effectively more EV than the that of the averaged scene.
Marq,
Good thought, I set it up again using the same focal point with spot metering instead of evaluative and obtained the same setting of 2 seconds @ 5.6 with ISO 100, so I still think the correct EV is about 4. Also, in both tests I used the center focus point (which I nearly always use) which, if I understand the metering system correcly, gives metering weight to the selected focus point when using evaluative.
So you obtain AI Servo focus with the 40D at 4EV. My tests were I guessed at about 2 EV (ISO 200, 2 sec at 2.8) when the 40D couldn't lock focus whereas the 30D had no probem. Canon specs claims AF sensitivity down to -0.5 EV. AL in another post could acquire focus in almost pitch darkness. This must be approaching that bottom limit.
Marq.
fredkinder wrote:
Marq,
Good thought, I set it up again using the same focal point with spot metering instead of evaluative and obtained the same setting of 2 seconds @ 5.6 with ISO 100, so I still think the correct EV is about 4. Also, in both tests I used the center focus point (which I nearly always use) which, if I understand the metering system correcly, gives metering weight to the selected focus point when using evaluative.
I'm not sure if this has been covered before, but the AWB on my 40D seems to be totally off taking pics outside, compared to my 400D. It records it way too blue. Setting the the WB to shade or overcast records well though.
Anyone else having this problem??
Sep 21, 2007 at 04:26 AM
Allen Maestas Offline • • • • • • Upload & Sell: Off
Infrasound wrote:
I'm not sure if this has been covered before, but the AWB on my 40D seems to be totally off taking pics outside, compared to my 400D. It records it way too blue. Setting the the WB to shade or overcast records well though.
Anyone else having this problem??
AWB is excellent for me, I shoot nothing but RAWS, so changing WB is a snap! I've found that colors, contrast etc... are all fantastic. I think you are just used to your 400D which is probably a bit warmer in WB than it should be.
Al
Sep 21, 2007 at 07:09 AM
Allen Maestas Offline • • • • • • Upload & Sell: Off
Good thought, I set it up again using the same focal point with spot metering instead of evaluative and obtained the same setting of 2 seconds @ 5.6 with ISO 100, so I still think the correct EV is about 4. Also, in both tests I used the center focus point (which I nearly always use) which, if I understand the metering system correcly, gives metering weight to the selected focus point when using evaluative.
Fred
I also used spot metering but bumped my iso up to 800 otherwise I was looking at 2-3 second exposures even at f2.8. Before you go sending in your camera I would try another lens and make sure it isn't your lens. Also try turning IS off. Plus remember that you are not going to see or hear the focus lights in AI Servo, but you should hear the lens. I also had the custom function (can't remember what it is now, I'm at work) for disabling/enabling continues focus search (where the camera continuely tries to focus even when it can't get focus), I had that disabled. So anyway, yeah, try some different things, and try another lens before you send it in.
I did try all that you suggested before I posted this thread. Lenses: 24-105L, 17-55IS, 70-200 2.8IS, and I did have the focus search set to on, the deafult C.Fn III-1-0. Also remember that I was comparing with my 30D, so the lens problem would affect both cameras.
Thanks again for your time,
Marq
Allen_Maestas wrote:
I also used spot metering but bumped my iso up to 800 otherwise I was looking at 2-3 second exposures even at f2.8. Before you go sending in your camera I would try another lens and make sure it isn't your lens. Also try turning IS off. Plus remember that you are not going to see or hear the focus lights in AI Servo, but you should hear the lens. I also had the custom function (can't remember what it is now, I'm at work) for disabling/enabling continues focus search (where the camera continuely tries to focus even when it can't get focus), I had that disabled. So anyway, yeah, try some different things, and try another lens before you send it in.
mschecan wrote:
So you obtain AI Servo focus with the 40D at 4EV. My tests were I guessed at about 2 EV (ISO 200, 2 sec at 2.8) when the 40D couldn't lock focus whereas the 30D had no probem. Canon specs claims AF sensitivity down to -0.5 EV. AL in another post could acquire focus in almost pitch darkness. This must be approaching that bottom limit.
Correct, I had no problem focusing at EV4 levels using one-shot, AI Focus or AI Servo (although AI Servo does not confirm focus). I saw no difference in performance with one shot or AI Focus - both gave instant confirmation, with no hunting; AI Servo seemed as fast.
I don't doubt the Canon specs on AF sensitivity based on my limited testing and a couple weeks of use. The 40D is significantly better in low light performance than my XT (which I always thought was pretty good, but the XT did stumble some in low light focusing).
I don't believe it is reasonable to expect AI Servo to yield good quality results at EV4 and below - it's simply the wrong tool for the job (IMHO).
