ghuff wrote:
I don't know how long Canon can continue to test and / or ignore the issue with it seeming to be so widespread. Hopefully they will announce a direction soon.
Hopefully. I wrote the dealer and asked them if they had heard anything from Canon. If not, the lens is going back their way.
Just curious, what is your date code and first 3 digits of your serial #?
Canon is really getting themselves into a public relations nightmare.
In recent memory, the 70-300 had a droopy barrel in portrait mode, the 24-105 had flare problems and now the 50 has major backfocusing.
All three of these issues are probably not manufacturing defects. Rather these are design problems.
This is a lot to screw up! I think that Canon could lose customers to Nikon over this. I don't see people selling their Canon kits because of this, but when buying new, you might think twice about Canon.
CMOS wrote:
Canon is really getting themselves into a public relations nightmare.
In recent memory, the 70-300 had a droopy barrel in portrait mode, the 24-105 had flare problems and now the 50 has major backfocusing.
All three of these issues are probably not manufacturing defects. Rather these are design problems.
This is a lot to screw up! I think that Canon could lose customers to Nikon over this. I don't see people selling their Canon kits because of this, but when buying new, you might think twice about Canon.
.
It really makes you wonder why they just keep shipping out the new 50L, with all the documented problems. Sounds to me like poor business practice in the realm of public relations. If I was the CEO of Canon, I would have been all over this this thing, from the start. Maybe we should have a mass exodus to Nikon or Leica or something and let Canon play with their own lenses, till the lights go out.
cactusclay wrote:
It really makes you wonder why they just keep shipping out the new 50L, with all the documented problems. Sounds to me like poor business practice in the realm of public relations.
Exactly. A few weeks ago, the 50L was out of stock at all the major retailers. At that point, I was calling it that Canon was holding off on new shipments while they sorted out this mess. Nope. Back in stock everywhere now.
This type of crummy behavior by Canon could impact initial sales rates going forward. Canon is probably going to introduce some new lenses at PMA 2007 and people are going to hold off for a while to see if it works.
I think since they are distributing lenses it is more of a firmware issue. If it was a design problem they would stop. It wouldn't make any sense to keep producing lenses if you knew you would have to recall them. It also wouldn't make sense to have a recall right before PMA, maybe they are only releasing a slow trickle of lenses and waiting until later this summer to say anything? All I know is I love my 50 and even if there is a minor niggle with the lens, Canon will make it right.
ward1066 wrote:
I think since they are distributing lenses it is more of a firmware issue. If it was a design problem they would stop.
I missed part of the technical discussion in the thread above, but I don't see how this could be a firmware issue. The lens is focused properly at f/1.2. Stopping down with aperture blades causes it to backfocus. What can be changed in firmware?
Either way, all these lenses would have to be sent back to Canon for the firmware/mechanical update.
Remember the surprise delay in availability of the lens? My guess is that there was some other problem with the lens that cropped up shortly before shipping. So they delayed shipping to fix that problem by making some other change in the design. But they didn't have time to test out that fix. Unfortunately, that late fix caused this problem we are seeing now.
Oops.
As described in this thread, the backfocus only happens when autofocusing, and at least in the 1DII cameras, it does it very consistently vs. aperture setting. This sounds like something which a firmware change can fix, e.g. by adding a look-up table for each aperture in the firmware, and making the focusing algorithm to correct the original focus calculations by a value fetched from a look-up table.
Koivulehto wrote:
As described in this thread, the backfocus only happens when autofocusing, and at least in the 1DII cameras, it does it very consistently vs. aperture setting. This sounds like something which a firmware change can fix, e.g. by adding a look-up table for each aperture in the firmware, and making the focusing algorithm to correct the original focus calculations by a value fetched from a look-up table.
If the firmware fix works by adjusting a lookup table so that your selected aperture will be in focus properly, what happens when you look through the viewfinder at f/1.2 (the way it usually works unless DOF is depressed)? Your subject would be out of focus!
A firmware fix would render the lens nearly impossible to use. You would see the (for example) right eye out of focus in the viewfinder at f/1.2 but you would have to know that the right eye will be in focus in the photo when the lens stops down to f/4 to actually take the photo. You'd have to chimp every shot!
Am I missing something here about how the firmware fix would work?
CMOS: I don't know how precisely you are able to see through the viewfinder, but I know I don't even try to look that precisely when I am using AF. I point the AF spot where I want to and expect the AF to do its job. The MF is another business, since there I am in full control.
I would expect the firmware fix to compensate the backfocus using the aperture specific table and measured distance seen by the selected AF spot. If the lens is not able to give accurate enough distance info to the camera body, then this principle won't work. By all means, there might be other flaws as well in this principle.
Yes, I think your missing the point. The firmware would have the lens focus properly at 1.2 and then focus properly at 1.4, 1.8, 2.0, 2.8, 4.0, etc. and every where in between. After all, that's what it does on every lens... there is always a look up table to which the camera refers to calculate how far to spin the focus motor to achieve focus from wherever it's starting. Or, maybe I'm missing your point...?
Hmmmm, there are aperture positions, and lens positions, with overtravel available in the lens movement on both ends... a look-up table pairs up the correct lens positions and aperture positions. Seems it should work, to me.
flatdraft wrote:
the 50L is mechanically perfect. it's the firmware that needs help.
Yep... agreed...but I think Canon's people need help in admitting they have an issue. I still stand by Canon on this, but enough is enough. PMA or no PMA.... the darn lens has issues, and has been on the market for almost 3 months in this condition. What does it take to get someone to stand up, be a good corporate citizen and admit....we have a problem?
Even if they don't have an answer yet...they need to acknowledge the issue....
OK...I'm cranky tonight. But this is part of the same discussion I was having in early December with Canon service after seeing multiple copies that were doing the same thing....
CMOS wrote:
Exactly. A few weeks ago, the 50L was out of stock at all the major retailers. At that point, I was calling it that Canon was holding off on new shipments while they sorted out this mess. Nope. Back in stock everywhere now.
This type of crummy behavior by Canon could impact initial sales rates going forward. Canon is probably going to introduce some new lenses at PMA 2007 and people are going to hold off for a while to see if it works.
Edited by CMOS on Feb 21, 2007 at 12:47 PM GMT
Yep. In stock everywhere..From $1399 and falling fast
From day one of the release I was never a believer of this lens. And that sucks 'cause I wanted one bad :0(
DavidP wrote:
Hot dang! The white paper clarifies what this "AF micro-adjustment" is about.
It's a way the PHOTOGRAPHER can adjust the AF to correct for front and back focusing. No more trips to Canon for this. You can even do this for up to 20 individual lenses!
Nirvana!
I must be dreaming!
So maybe the new 50L's fix is a 1D Mark III? Haha.