BadlyDrawnBoy, don't feel bad, according to the Sigma reps I spoke to they've had tons of reports of greatly satisfied customers with this 50/1.4 We're obviously the distinct minority who need our camera bodies calibrated. </sarcasm>
I'm still expecting a .CR2 file from Nick at CRK in Perth, he says he has 3 .CR2 files that are properly focused using AF beyond 10' on a 1Ds MarkII so that'll be interesting... It won't make me buy this lens again, but it will show that it's a programming error and not a design/build fault.
I will say I have gotten images focused just fine with the lens over 10feet..its just it was few and far between and never know when it would get it right
I just got a call from work saying Nick dropped in the CR2 files... But my mate (photographer) who called said it sounded a little suspicious as Nick said he focuses then recomposes for every shot... We'll see if ZoomBrowser EX thinks the "in-focus" part is actually in focus. Heading off to work now.
lorriman wrote:
Are any Nikon and other non-Canon users having these focusing problems?
I get the impression it is just Canons, which, from what I understand about Canon's secretiveness and patents, makes some sense.
As I understand it is the lens available only in EOS mount for now. Nikon, Sony, Sigma and Pentax will follow later.
To my understanding, not own experience, most Sigma lenses focuses correctly, don't they? The 30/1.4 is a victim of bad QC and many seem to need a recalibration. Are there other, general, Sigma focusing problems as well?
I got a few CR2 files from Nick, but the ones focused at *around* 10'... are usually at f/2.8 or smaller (couple at f/1.6) and are very soft IMO - he hasn't included much in the background to tell if they're back-focused. He focused and recomposed every shot after locking focus, I don't know if Zoombrowser will identify if AF was locked elsewhere using centre point then recomposed... it doesn't seem so.
He seems to think it works perfectly on his 1Ds MarkII but who knows, without seeing concise methodical testing it's very hard to tell from these CR2's.
thrice wrote:
I got a few CR2 files from Nick, but the ones focused at *around* 10'... are usually at f/2.8 or smaller (couple at f/1.6) and are very soft IMO - he hasn't included much in the background to tell if they're back-focused. He focused and recomposed every shot after locking focus, (...)
So, do we have a single confirmed case at all of a Siggy50 with working AF in all distance ranges?
After seeing some of the samples of this lens out on the web, I went out and picked up a copy earlier today. However, I wound up with similar AF problems, so I thought I'd add another data point for those following this thread.
After my first real-life pictures turned out to have missed the focus by a wide margin, I set up the standard test- a large, flat, high contrast object with detail in front and behind the target. At 8', the focus plane was about a foot in front of the AF target. Since DOF at 50/1.4/8' is only ~5", this is obviously a major problem. IIRC, Canon's target variance with the 5D's center sensor is 1/3 DOF, so this is ~6x that.
In practical terms, this meant that with a seated portrait autofocus wasn't just on the wrong eye, it was all the way out to the subject's knee.
I've made arrangements to return this one and try the store's other copy, so hopefully that one will turn out better- I'll let you know. Of course, I probably wouldn't have bothered with the testing if the lens had worked as expected, so you can take this report with a grain of selection bias salt...
+2 lenses for Canon mount with serious front focusing problems. I just received a replacement lens from Beach Camera and it front focuses just like the first one did. If not for carefully keeping files separate I would not be able to tell the two apart. So Monday I call them again, send in the 2nd and ask for a 3rd.
FYI, I used both lenses on two separate 40D bodies, one of which just came back from recalibration at Canon Service in Irvine CA two weeks ago. I used that freshly calibrated body to verify the other seems to be working fine.
Does anyone else here think that Sigma has a bunch of mis-calibrated Canon bodies they use to test lenses
Yeah, Sigma QC is less than perfect. I had a Sigma 30 f/1.4 for a while, but it frontfocused horrendously. I have had success with the 18-125 (which I apparently had a very good copy), the 70-300 APO, and my current favorite Sigma lens, the 150 f/2.8 Macro (which is awesome, and spot on, AF wise). I also have a 24-60 f/2.8, but it too front focuses, so it needs to make the trip to Sigma for calibration (haven't sent it in yet.)
So, for me, I'm 3/5 for properly AFing lenses with Sigma.
OTOH I don't hear much about Tamron or Tokina requiring calibration? These brands need to support FF more, they're shooting themselves in the foot as it's only a matter of time until demand and decreasing manufacturing costs make big silicon sensors the best option.
It is possible however that the users who typically buy those brands don't test as carefully?
I've never had an issue with either of my two Tokinas, nor my one Tamron. I think enough people own the 17-50 that any issues there would come to light. Tokina's most popular for their ultra-wide lenses, which are much more forgiving on small focus errors, so I doubt even if there were errors, if they'd even be noticeable most of the time.
It is possible however that the users who typically buy those brands don't test as carefully?
Yes, definitely a large majority of people in general will fit that category. Many budget conscious buyers are also less demanding about the results obtained with their "economy product", they expect the lower cost product must be somehow inferior and theorize that it must be "just the way it is...".
For every detail oriented pixel-peeper there are probably 10-50 people who will not even recognize they have a lens or camera that is under-performing. Hence the reports from Sigma that they have many happy users of this new lens. They probably do, so far. More likely, there is no communication channel telling sales reps exactly how many are going back in for service and how many returns there are.
We're obviously the distinct minority who need our camera bodies calibrated. </sarcasm>
