lidesun wrote:
I just bought a brand new one after 6-7 bad experiences, the price was too good to refuse the deal
Serial # 56xxxx, datecode UW0106, should be the latest production...
Hmmmm. Hope springs eternal I guess.
I'm still in the "wait and see camp". I thought we were out of the woods when I purchased my 1D3 right after the BDs were released. As many of us discussed before, I'm still thinking the 1D3 may be a short lived model. If Canon popped a new FF version of the sensor in there, I'd be all over it like white on rice.
One thing i noticed right away is that after the microadjustment, the AF performs very well, even @ f/1.2, very considency acurate focus, which was the big trouble back to all the Mark IIIs i had, and i don't shoot sports, and use oneshot mode 95%, so i think i am fine with it...
jamesf99 wrote:
Hmmmm. Hope springs eternal I guess.
I'm still in the "wait and see camp". I thought we were out of the woods when I purchased my 1D3 right after the BDs were released. As many of us discussed before, I'm still thinking the 1D3 may be a short lived model. If Canon popped a new FF version of the sensor in there, I'd be all over it like white on rice.
Good luck with #7.
are you serious? do you enjoy spending thousands of dollars to be a beta tester? If i were a canon shooter it would be Months before i would think of ever buying the "latest and greatest". But i am from Texas and the old saying goes a long way with us there: 'fool me Once , shame on you . Fool me Twice, shame on Me.'
Seems to work pretty well . For me anyway
jmcfadden wrote:
are you serious? do you enjoy spending thousands of dollars to be a beta tester? If i were a canon shooter it would be Months before i would think of ever buying the "latest and greatest". But i am from Texas and the old saying goes a long way with us there: 'fool me Once , shame on you . Fool me Twice, shame on Me.'
Seems to work pretty well . For me anyway
Actually, I think the statement now goes something like this...
'fool me Once , shame on you . Fool me Twice... [blank stare, butt scratch at the podium] uh, hmmm, uh, uh, uh,,uhmmmm, oh yeah, won't get fooled again". (from a "notable Texan )
OK, beta testing sucks. I've been an official MS beta tester for 22 years now so I'm used to being screwed. MS sells their product to QA and sits back waiting to hear the screams.
Actually yes, I'm serious, to a point. There's no doubt in my mind that Canon knows how to make an extremely high quality FF sensor. From my own experience, the 1D3 is a fantastic camera (AF blunder aside) with one major drawback - the cropped sensor. Boy did Nikon get that right...
I have no doubt that Canon will not make the same AF blunder again and will fix the problem before releasing another camera. If I could get a high performance 1Ds2 (i.e. new 16MP FF sensor) released with all the new features of the 1D3/1Ds3 sold at a reasonable $4k-$5k I'd order two and elbow others out of the way to get in line.
12MP too small. 21MP too large. 16MP. mmmmm, just right
I just cannot see them putting out 2 1D series Full Frame bodies..with the only real difference being megapixels and perhaps FPS.
BTW: I am from Texas too...Not that it matters.
With technology races, the consumer benefits from quick development, but also suffers the "beta tester" syndrome. It's the price we pay when companies are eager to out do each other the quickest with QA being the first to be cut.
As long as there is a line at the door to purchase new products, this will remain the case IMO.
And I thought the "Won't get fooled again" quote was from a notable Brit - Roger Daltrey
apdieb wrote:
I would love to agree with you, but the fact that all my Canon bodies perform great without any adjustment makes me think there is more to it than that. This includes my Mark III PRIOR to sending it to Canon.
I adjusted mine with that chart you mention, as well as the new one that was posted here recently (the LCD moire pattern)...But what worked for me (at least for the 1st lens tonight) was just shooting a bunch, making adjustment, shooting a bunch more, adjust, and so on. When I got home it was obvious which adjustment made the most dramatic difference. For me, (at least on my 70-200F2.8 IS) it was +15 microadjustment.
We should not have to be making all these adjustments...If all my other bodies (including Nikon) performed great with a set of lenses, the Mark III should as well. I am not even referring to AI-SERVO specifically either.. One-shot was way off equally....Show more →
Not to dispute the issues anyone is having here with the Mark III, as I also have two imperfect copies with the first currently at Canon for the "fix"... However, my experience regarding AF accuracy is that every 1D body I've owned (six so far starting with the original 1D) has at some point had AF accuracy issues - i.e. general front or back focusing, or more specific problems with certain lenses. Quite often it was right out of the box (it's the first thing I test with a new camera). Therefore I expect this to happen (perhaps Canon has conditioned me over a couple decades to expect this!!). Looking back to the film days, I had the same problems with EOS-1V bodies, a 3 and an original 1.... are we forgetting about the endless front/back focus rants on every photo forum in existence?
The simple fact is this is not a new issue. Agreed, it would be nice if there was better consistency in production but these are complex devices. At the risk of throwing out analogies... consider high performance vehicles... don't they require more maintenance and attention? Likewise with high end cameras, photographers need to be very aware of any performance degradation through constant monitoring...
This is why I'm thankful that Canon has included the microfocus adjustment feature because it means I can finally fine tune problematic lenses rather than sending in the kit on an annual basis for calibration and hoping that it doesn't suddenly go out of whack months down the road. I do hope though that it's not an excuse to ship sloppily assembled/calibrated cameras. BTW, it's also on the Nikon D3... so you don't think some Nikon users also have AF calibration issues?
As for why a Mark III would come back after the fix with worse calibration... we're only guessing. Maybe the "known good lens" that they use to calibrate AF is not representative of many high speed Canon lenses in circulation?
Here's a crazy suggestion from back in the Mark II and Mark II N days: Try reinstalling the firmware, even if you're already on the latest, and see if that makes any difference in AF precision. Some II and II N users claimed this worked for them when AF would suddenly drift off and they didn't want to send their cameras in for calibration. I can't guarantee it will work...
I had noticed differences in sharpness/detail with certain color areas between the two versions. Looking closer at the RGB channels, and R&B in particular, revealed that there was indeed some sort of sharpening/softening going on. In 3.0.1 the red channel was a touch softer than in 3.2 and the blue channel considerably sharper (for certain color areas with detail) than in 3.2... I also went so far as to take blue channels from 3.0.1 and paste them over v.3.2 blue channels, sharpening certain details even more. But on the whole, I seem to prefer the results from 3.2. It should be noted that I had sharpening in DPP set to zero.
Therefore, yes, I would guess that Canon is applying some sharpening to DPP conversions separately from user controlled sharpening settings.
One thing I forgot to mention that might be of interest to some is that I found during AF micro-calibration, the optimal settings I get depend on the subject distance. For the record, the lenses I calibrated were the 600/4 IS, 400/2.8 II, 300/2.8 and 70-200/4 IS. I normally shoot with these at subject distances just this side of infinity, so you might think calibration is a waste of time, but such was not my experience. I found that I had to perform my calibration at target distances that are typical for my shooting, which isn't as easy as it might seem, considering that my typical subject distance on the 600 is roughly 300 meters. I found that if I calibrated these long lenses with a target that was much closer to me than my typical subject, the results I got in the field were less than acceptably sharp. When I calibrate at the desired distance, my actual results are a lot better.
Also, when I do calibration, I shoot a minimum of four shots at each setting, simply because the Mark III AF bounces around so much (meaning I often get inconsistent results). I find that if I shoot a lot of shots and then average them, I can come pretty close to what works out to be the "best" setting.
Just thought I'd pass that along in case anyone finds it interesting or helpful.
stating the obvious: it shouldn't be so difficult to get AF to work on a camera. it should just work. if we have to think about it so much, it's a waste of time. the camera's a piece of junk.
simonella_viru wrote:
stating the obvious: it shouldn't be so difficult to get AF to work on a camera. it should just work.
Unfortunately very simplistic. Also unfortunately AF is a continuum from really bad to really good. All AF will fail at some time or another, and I guess I'm lucky to have a 1DIII at the better end of the scale (90% keepers, and when it doesn't, it's usually my fault...)
jaclarkaus wrote:
Unfortunately very simplistic. Also unfortunately AF is a continuum from really bad to really good. All AF will fail at some time or another, and I guess I'm lucky to have a 1DIII at the better end of the scale (90% keepers, and when it doesn't, it's usually my fault...)
don't take my statement too literally.
the brakes in your car simply work. you don't question them, you don't have to tweak them every day for different weather, different temperatures, different light levels etc. they simply just work. sure, your brakes won't work forever and they don't always stop you in 50m. but virtually always they'll stop your car and keep you safe.
we only understand a small sliver of what's going on with the mk III and that's likely the way it'll stay. if canon had a stable, reliable system (and manufacturing/ QA process) that wasn't on a cun1t's hair edge of instability, we wouldn't all be having this discussion for the past 9 months or so. it's obviously an unstable (not controlled) system that is sensitive to _________ (?). some got lucky cuz the constellations aligned when their camera was being built. others, like myself, got total kife.
lidesun, you're the most persistent and patient man on earth. i have no idea how you've put up with so many mk III bodies (7). i gave up after 1 and now i'm stuck with it.
I understand the merits of micro adjustment and understand that there will be some inconsistency. What I do not understand is why on every other body all of the lenses either perform superbly or very close too it. On the Mark III, there is drastic calibration necessary and to varying degrees based on lens. That is the part that bothers me and defies logic IMHO.
It wasn't like that with my (2) 20Ds, (2) 5Ds, Mark IIN, or my Mark III prior to the submirror stopper replacement. I am going to continue to calibrate each lens as I get a chance on real life shooting (not paying gigs) and see where I end up. I am also going to take the advice and reflash the firmware (assuming it will accept it). I also need to see how it does at varying distances as was mentioned. I believe I have a similar issue...When miscalibrated, my upclose shots seemed "ok", where subjects at a good distance were extremely front focused. After calibration at the long end, it "seems" like both are acceptable. I reserve final judgment until all are tested.
Thanks again for everyone's opinion. I find them all valuable.
I finally got Canon to give me a copy of the service work they did after the sub mirror fix. They list:
Problem - Auto Focus Back Focus
Condition - Always
Cause - Misadjusted
Service - Electrical Adjustment
Location - AF Assembly
It is still screwed up and in trying to use the micro-adjustment to correct I am seeing the same thing as others. That being that adjustment for one distance can cause a worsening at other distances. Since what they did in the shop and what I can do in the field are both electrical adjustments they are almost certainly related in some manner and this has all sorts of interesting potential ramifications.
PS. The Canon tech person I talked to today acknowledged that they were hearing rumors of an impending second fix from staffers who know people in Canon Japan. He also stressed that Canon USA has not said word one about it and everyone is in the dark about what might be going on. He claimed they found out about it from RG as well which I sensed didn't leave them very prepared for the phone calls and questions. They are very much following the party line and I am now convinced they are working from a script as the wording from three different people has been almost identical and very carefully phrased.
apdieb wrote:
I also need to see how it does at varying distances as was mentioned. I believe I have a similar issue...When miscalibrated, my upclose shots seemed "ok", where subjects at a good distance were extremely front focused. After calibration at the long end, it "seems" like both are acceptable. I reserve final judgment until all are tested.
Thanks again for everyone's opinion. I find them all valuable.
Andrew
I wonder about the AF microadjustment methodology too, like the logic in making adjustments based on a target that is quite close to the camera. Granted, with certain lenses, that is typically how I work, such as the 50 & 85 for people photos... but definitely not typical for some other lenses.
According to pg. 169 of the manual:
"It is best to make the [AF micro]adjustment at the actual place to be photographed. This will make the adjustment more precise."
I also wonder about light quality. Like, would it be wise to calibrate indoors under tungsten lighting when the camera will be used primarily outdoors?
Today I received the Mark III I sent to Canon for the "fix." It was already dark by the time I opened the box, so have only been testing it inside. I tried the computer monitor AF test chart (linked somewhere here) for the first time and so far have mixed feelings about it (tried both the crosshairs and rings versions). When I got the f/1.2 lenses to focus consistently on it, I found they both back focused on "real world" objects around the house. But the slower the lens, the less of an issue it was (kind of obvious). An 85 1.8 seems to be more consistent shot to shot based on focus set with the chart and the 70-200 f/4 so far seems rock solid in One Shot mode. It's the 1.2 lenses that were initially giving me some concern as they were not hitting the exact point consistently, particularly the 50 1.2. However, after some further "real world" shots around the house, the impression is that the camera *seems* to be better in One Shot than before sending it in - so far. Focus consistency with everything except the 1.2 lenses seems to be very, very good. Even the 1.2 lenses behave better than before the fix (it used to be I could hit the focus button numerous times before an exposure and each time the lens would refocus - not so much the case today). With this camera in particular I also had headaches with the 16-35 often being way out shot to shot. So far it looks promising with that lens too.
For the most part I'm not dialing in much, if any, microadjustment. The only significant offender so far at -7 is the 85 1.8, but I already knew it tended to back focus with everything. Other lenses are at zero or between -3 and +2. Even with such small adjustments, I only really notice a difference with the 1.2 lenses wide open and there the shot to shot AF inconsistency may be a greater factor.
simonella_viru wrote:
don't take my statement too literally.
the brakes in your car simply work. you don't question them, you don't have to tweak them every day for different weather, different temperatures, different light levels etc. they simply just work. sure, your brakes won't work forever and they don't always stop you in 50m. but virtually always they'll stop your car and keep you safe.
Tell that to a Formula 1 driver. Your brake analogy is the same as saying how the 1D3 is fine if as long as you print 4x6 and nothing larger. I'm pretty sure even if you have OOF issues, if you're not printing larger than 4x6, they'll look acceptably sharp.
). i gave up after 1 and now i'm stuck with it.