My MKIII's CR2 files open up in PS's ACR 4.1 (after new update/download note the new 4.1 ACR features!) and it formats my 16Gb SanDisk Extreme III to 15.2 Gb. Some data rich RAW files are 14.5Mb and are, typically, bigger than my 5D's RAW files. Maybe that's the extra 2 bits of data?
I just bought a spare battery at a local camera store for $170 + tax-Ouch! Li-Ion is more expensive than Ni-MH and, since Li-Ion battery chemistry is very touchy, I wouldn't recommend 2nd party batteries (remember those laptop fires?).
I did some non scientific tests with the lens AF Micro Calibrate function. It seems that the AF calibration varies with distance, F-stop and zoomed FL (if a zoom). The wide lenses seem to need the feature the most and are the hardest to check. I will do more field tests under actual shooting conditions. Difficult since the zoomed 10X LCD viewfinder is hardly adequate to check critical focus and it takes some time to swap the card out and check images on a computer screen. Absent hard data it seems best to leave it set at "0."
I am developing some impressions of ISO vs IQ but want to be more certain. Part of this is the nature of the MkIII noise (it seems patterned/processed) and the different IQ of the sensor which seems to take more/different sharpening. I am certain that the News/Press folks will be very happy with ISO 6400. I'm happy with the quick set ISO/in viewfinder, viewfinder IQ, LCD, fast(er) AF and the high quality of the camera construction.
When using dual record to both flash cards I find that erasing pictures happens only to the SD card-not the CF. Anyone have some ideas about this?
SoundHound wrote:
I did some non scientific tests with the lens AF Micro Calibrate function. It seems that the AF calibration varies with distance, F-stop and zoomed FL (if a zoom).
Which makes sense given that the 1/8 single side DOF of each step takes into account each of those variables you mentioned to determine a physical amount. If one has a zoom that is off by varying amounts over the focal range, sending it in is still the option as there is no provision for making focal length dependent changes to a zoom.
SoundHound wrote:
Absent hard data it seems best to leave it set at "0."
That is what many owners have found so far; Canon didn't intend this feature as some calibration akin to formatting a memory card that one must do before taking pictures.
dcmiller wrote:
I think because the numbers an estimate, and it's estimating the worst-case scenario. Does the number change to 24 when ISO is set to 3200 and no safety shift??
Yep, safety shift is disabled, ISO set to 3200, indicated burst shots are 24. Seems like the indicated burst speed is worst case only.
Still, I was out today with my 1D3, set myself up in a hide where kingfishers are, the light is terrible, too much shadow, set the ISO to 3200 to give me a shutter speed of 1250s so if the bird does show up I will have the speed. In the field next door the farmer was cutting grass and straight after the tractor goes past a pheasant lands, I grab 2 shots before it flies off. Total grab shot hence the silly shutter speed! Still, I have to say, I am very very impressed with the ISO 3200, it really is useable. Far far better than my N at 1600. Anyway, for your comments, the original RAW file is .here 15.8mb file so downloading it is probably better.
Other things I noticed today:
- With liveview you can only use MF. Not a feature I would use so it doesn't bother me;
- Must get into the habit of turning the on switch 2 clicks so the main control dial on the back works. Still in 1D2 mentality where the 2nd click gives the beep!;
- My menu feature is brilliant. I chose my 5 most used menu items and grouped them under my menu. About the closest we are going to get to an MLU button;
- Since today is the first time I shot with it, not properly I admit, it do have the sneaking suspicion that my 500/4 IS focuses even faster with this camera. Perhaps the 1D3 drives the USM harder I am not sure, but it just felt like the lens was snapping, and it feels like a snap into focus even faster;
- Is it me or in 10fps mode, is the shutter much louder than the N?
You will all laugh but the only weird thing I have found about the 1D3 so far is this:
When I am setup behind the N for long periods of time looking through the viewfinder, sometimes I would see beads of condensation just underneath and to the left of the on/off switch, obviously the moisture from my breathe. I never really paid any attention to it. On the 3 though, the 4 small holes for the mike are directly where the droplets form. Nothing to worry about I imagine but it made me laugh!
If a Canon tech in Japan reads this and mutters under his breathe 'damn the English guy is right' and suddenly on new bodies the position of the mike holes are relocated slightly, remember guys, you heard it here first! _poke_
Fabulous bit of kit, I really am impressed and despite the Ł3K GBP price tag (NA people no comment please!) I have to say it really does feel excellent value.
It should be called High ISO speed "color" noise reduction.
Like you, I don't see any noticeable difference in the RAW files when High ISO speed noise reduction is ON.
It does work on JPG files though. However it mainly eliminates chroma noise (Which can be eliminated very easily in Photoshop without any loss of detail). It does not seem affect luma noise to a noticeable level.
Because I mainly shoot in RAW mode, I will be leaving this CF OFF.
I notice that too. But luma noise adds character. I'm glad they didn't further slow down the camera with luminosity noise reduction.
I believe the outstanding question is : When shooting raw only, does the camera do NR on the embedded jpg?
Photoshop can easily reduce color noise, but I believe the point of camera NR is to produce a finished file, or a file that will only require sharpening for printing.
It's also there for fair comparison to other brands by reviewers.
dcmiller wrote:
I notice that too. But luma noise adds character. I'm glad they didn't further slow down the camera with luminosity noise reduction.
I agree with that. I also like the "grain look" ...However there is a fine line where you can reduce luma noise and still keep the character and detail on your image...the type of luminosity noise reduction that would give you a "one stop" reduction at most.
I believe the outstanding question is : When shooting raw only, does the camera do NR on the embedded jpg?
I remember that BB allows you to get the embedded JPG. However it is a "preview" jpg and it's highly compressed.
Photoshop can easily reduce color noise, but I believe the point of camera NR is to produce a finished file, or a file that will only require sharpening for printing.
It's also there for fair comparison to other brands by reviewers.
Agree. But if it's only color noise in exchange for a bursting hit, I would prefer to reduce that in a Photoshop batch.
Unfortunately, Micro Focus Adjustment (MFA) seems to affect/be needed with a zoom lens I already sent to Canon Irvine (who replaced a handful of parts). I am concerned that MFA is highlighting focus issues that have been lurking below the radar (until now?).
I am hoping that I can approach 5D 12.7Mp sensor IQ with the MKIII @ ISO 3200 by tighter framing (less cropping) with a (F2.8) zoom instead of my usual primes (set @ F2.0). That way I can take advantage of the MKIII's much better focusing and bigger buffer for action dancers in low light.
My preliminary RAW tests (MKIII @ ISO 3200+ verses 5D @ ISO 1600) indicate much more luminance NR is needed coupled with PS' "Smart Sharpen" (I seldom use Smart Sharpen with 5D files) to approach my typical IQ standard.
Alistair101 wrote:
- Is it me or in 10fps mode, is the shutter much louder than the N?
It's definitely a different sound from the 1D Mark II and II N. Sounds more robust IMO. Reminds me of the Nikon F5 to a degree. The Mark II N in direct comparison without lenses attached sounds more "tinny" - a somewhat higher pitched sound. While I would guess the overall sound volume is roughly the same, the Mark III's deeper sound may in some situations be less intrusive. With lenses attached, the N retains more of a "clattering" sound while the III in comparison is smoother, deeper and a bit less irritating.
ben_is_in wrote: Nill Toulme wrote:
Which CF is new?
Nill
~~
On the 1DIII, C.Fn III -2 is AI Servo tracking sensitivity like C.Fn 20 on the 1DII. C.Fn III -4, as posted above, seems to be new.
It's new, in that it allows the user to set an absolute distinction. Either it sticks with the obstructed subject or it immediately focuses on the new, closer one. On the previous 1D models, the AF was designed to always initially ignore momentary obstructions by closer objects. The delay before it would focus on something closer was set in CF 20 as Ben pointed out.
Interestingly on the Mark III, CF III-2 still allows setting the AI Servo tracking sensitivity, which seems to be the same as CF-20 in the Mark II... so maybe CF III-4 at 0 will only hold focus on the obstructed subject for the time specified by CF III-2? Maybe it's time to read the manual...
rscheffler wrote:
Interestingly on the Mark III, CF III-2 still allows setting the AI Servo tracking sensitivity, which seems to be the same as CF-20 in the Mark II... so maybe CF III-4 at 0 will only hold focus on the obstructed subject for the time specified by CF III-2? Maybe it's time to read the manual...
From what I can tell:
C. Fn. III-2 sets the delay time prior to changing the subject of the tracking.
C. Fn. III-4 sets whether the camera will continue to track whatever is in the center point or if it will follow the original subject outside the set AF point.
rscheffler wrote:
It's definitely a different sound from the 1D Mark II and II N. Sounds more robust IMO. Reminds me of the Nikon F5 to a degree. The Mark II N in direct comparison without lenses attached sounds more "tinny" - a somewhat higher pitched sound. While I would guess the overall sound volume is roughly the same, the Mark III's deeper sound may in some situations be less intrusive. With lenses attached, the N retains more of a "clattering" sound while the III in comparison is smoother, deeper and a bit less irritating.
That is a great description. I could actually hear the sound while reading!
Interestingly on the Mark III, CF III-2 still allows setting the AI Servo tracking sensitivity, which seems to be the same as CF-20 in the Mark II... so maybe CF III-4 at 0 will only hold focus on the obstructed subject for the time specified by CF III-2? Maybe it's time to read the manual...
Perhaps now we can set the tracking to low and set CF III-4 at 1. That would allow AI Servo to track the subject while holding focus even when outside the set AF point. Only testing will tell...Is Ben getting a mk III?
It's a noisy bugger for sure...just tested it next to my 5D and there is no comparison, the 1D III is about three times as loud to my ears. Not good for wildlife, but then again other than that it's perfect for wildlife.
john660 wrote:
It's a noisy bugger for sure...just tested it next to my 5D and there is no comparison, the 1D III is about three times as loud to my ears. Not good for wildlife, but then again other than that it's perfect for wildlife.
I have just done as scientific a test as possible. 1D3 and 1D2N side by side, no lens, same settings. They sound exactly the same 'volume' just the pitch of the 3 is different, higher. Short of testing with a sound meter (anyone got a recording studio that can measure the actual noise dbs?) this is the best I can do.
Nill Toulme wrote:
How do their respective "silent modes" compare?
The N doesn't have a silent mode that I know of. With the 3 in silent single shot mode you still get the snap as the mirror rises and a sort of slow 'schlaapp' as it drops, almost like the mechanism has some kind of damping.
I shot some cats with the camera in silent mode. While they were spooked by normal single shot drive, Silent was just ignored. My human subjects didn't hear it going off either.
Today I was testing the M3 with the 580 EX II inside. I was in the TV mode. I was stunned to find that my shutter speed went from 1/250 to 5". This happened several times. I contacted Canon and found that the CN FnI # 8 is set for the safety to adjust shutter speed if not enough light. This is the default.
I had a scare like that once when my 20D came back from Canon servicing. Not sure that this is supposed to be the default setting, though. I'd expect the default setting to be 0: Disable.
...However there is a fine line where you can reduce luma noise and still keep the character and detail on your image...the type of luminosity noise reduction that would give you a "one stop" reduction at most.