Should this new Sigma 50mm 1.4 go back for retuning

So I just got a new Sigma 50mm f/1.4 for my 5Dmk2 and am noticing that it just doesn't look as good as I was expecting. Am I asking too much of this lens or am should it go back it for repair (I've had it for 2 days..). All shots tripod mounted and remote shutter released.

Test Shot:

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

HTTP response: 404 | MIME changed to 'text/html' | Byte size: ZERO

in the full view, it appears as though the focus is on the wire to the front left side of the photo, did you have the lens wide open causing extremely shallow DOF?

yes the lens in the full view was shot @ 1.4, though focused to the central focusing point (which is to what all the above photos have been cropped to @ 100%)

sjafari wrote in post #8508885
yes the lens in the full view was shot @ 1.4, though focused to the central focusing point (which is to what all the above photos have been cropped to @ 100%)

If you are saying that the 100% are centre crops, and the focus point used was in the centre, than yes there is a slight front focus issue here.

The green/magenta colours are longitudinal CAs, which are common with very fast lenses that have good bokeh. They show up in high contrast transitions. The magenta/red always appears in the OOF zone in front of the focus point, and the green in the OOF zone behind the focus point. It disappears at smaller apertures.

The neat hing of course is that you can exactly see where the lens focused: the area between the two colours where the image appears normal. IOW, this lens shows slight frontfocus, not backfocus.

I reckon the LoCAs are about average for the Sigma, as they are similar to the two I tested. This is common with fast Canon lenses, too, BTW. Personally, I think it is quite charming.

The only software I am aware of that is capable of partly fixing this is DxO Optics, provided it supports the lens. Unlike "standard" lateral CA this is not completely fixable at all times. Maybe DPP can fix it these days, but likely not for the Sigma.

Kind regards, Wim

Were the shots taken indoor w/ minimal amount of light? AF might not be so accurate. Second, the shot might be too closed. Looks like the object might be less than 2 ft away. Try a littel further away. The MFD is around 18-24" if I remember correctly.

You need to separate two issues:
1) AF issue. Should take at f/8
2) softness wide open. Take at f/1.4, 2, 4 and 8

Need lots of light for both test shots so you take it at ISO100, >1/200. Prefer tripod than hand held. Take shots at both 3ft and then 10ft and then 20ft.

I own 5D MK2 and Sigma 50mm 1.1
a few suggestions.

1. use live view, x10, switch to manual focus, dont rely on the AF confirmation beep. use the lcd screen to MF. shot and compare with your AF results.

2. I experienced almost the same problems and yours. My sigma 50mm 1.4 was front focus at close distance, +20 cannot bring it back. back focus while target is far away, MA -20 doesn't help much. I realy like the optic of this lens. I paid $10 sent it to Sigma NY for calibration. It came back less than a week. Now it foucs almost perfect at all distances. No MA needed.

So if you want to give it a try, send it to Sigma. Return and exchange is always a bad idea.

My new Sigmalux is perfect.....perfectly front focusing like so many other copies out there.

I see no reason to test with a tripod. With real life handheld testing you should clearly see if you have a lens with issues.

I tested my Sigmalux at the "store". I had a gut feeling it was a tad soft wideopen. When I brought it home I discovered it was clearly front focusing. Taking photos of my 3mth old I found the eyelash detail extremely soft wideopen. Taking photos of my 2yr old I found the same issues as her mouth became in focus as I was focusing on her eyes.

I found that taking pics of a keyboard is a real quick method of testing for front/back focus at mfd.

At f/1.8 to f/2 my photos were acceptably sharper and by f/2.5 the lens was extremely sharp. My Canon f/1.4 is by far solid in the AF department and my current Sigma is extremely sketchy at all distances.

Well after doing a bit more testing and some research, I decided to send the lens back to the store for a refund.. although purchased new ~3 weeks ago, the S/N on the lens was 2xxx, so a very production run of the lens. I will be purchasing the lens from a dealer where i expect they would go through more volume and thus have a newer stock (something like Adorama or B&H).. hopefully the second one does the trick.

thanks for the input everyone!

I can't tell from the picture on FF, but just keep in mind something I flog around all the time.

MFD for this lens is 17.7". You need to have at least that from the sensor to the object in focus.

You probably know that already, but I see a ton of posts/charts where it's obvious they aren't that distance away from the subject.

Alan U says "I see no reason to test with a tripod. With real life handheld testing you should clearly see if you have a lens with issues."

The reason for "testing" on a tripod is to eliminate any chance of outside movement. Granted, in the real world you will not be using the tripod for every shot, however, if you want to do a true test, you must eliminate the possibility of movement as a result of hand holding the camera. Especially if you are shooting close range wide open when the dof is only inches.

sjafari wrote in post #8570388
hopefully the second one does the trick.

thanks for the input everyone!

Hope your's does, mine didn't. ...

DarenM wrote in post #8574742
Alan U says "I see no reason to test with a tripod. With real life handheld testing you should clearly see if you have a lens with issues."

The reason for "testing" on a tripod is to eliminate any chance of outside movement. Granted, in the real world you will not be using the tripod for every shot, however, if you want to do a true test, you must eliminate the possibility of movement as a result of hand holding the camera. Especially if you are shooting close range wide open when the dof is only inches.

I use a flash when doing the "hand held test" to stop motion. I'll typically use a human subject since its easier to pixel peep eye lashes. Taking photos of stationary objects is a tad harder for me to guage sharpness.

I'll also take a picture of a handful of pens (in a cup) and aim at a specific pen. The click type "chrome" clickers will show if there's significant CA's.

The keyboard "tool" is a very quick test that'll show front/back focus.

I guess if your very particular you can start using a tripod after the quick preliminary tests have been performed. It seems people add user error more so when using test charts.

My 2nd copy of the Sigma 50 was wonderful... failed with flying colours. Problem with Sigma Canada is that it takes aprox 1 mth ++ to get a lens calibrated. When I took a photo of my 3mth old daughter her eyes were a complete blurr wideopen. I swapped the sigma with my Canon 50 f/1.4 and her eyes became detailed and sharp.

"if" I buy a Sigma 50mm again (then that'll be my 3rd copy) I'll get it calibrated and wait a month.

yes