Bought 24-70 f/2.8 II -Soft, owners, please advice

[absente]

Hey there,

so I went on and bought a new 24-70 f/ 2.8 II. The strange thing in China (where I currently live) is that they don't let you test the lens before you buy it - nowhere. So anyway, I picked one finally after searching the whole day for a good price - and to be quite honest, I am very disappointed.

Here are two crops, 100%, everything is on neutral, no sharpening.
First one is AF-Live mode where I zoomed in on the 24 using 10x magnification. Second shot is using the normal way.
Lightning is not so good, but since on a tripod and with remote control I don't see any problem using slow shutter speeds.
I used AV mode - strange thing is that life-view gives me a different meter reading.

Any advice from 24-70 owners will be highly appreciated.

PS: I am shooting with a 5D3, newest firmware.

Using live-view, 1/4 f/2.8 ISO50 24mm

lifeview1

Using normal focusing, 0.3 f/2.8 ISO50 24mm

lifeview2

Both pictures are shot in jpeg, L mode. I usually don't shoot RAW so thats why I keep testing like this.
I don't see any differences in those two pictures except the exposure, which means that AFMA can't help me - Do I have a "bad copy" or can I fix this somehow?

 

[sturdiva]

Did you have mirror lockup turned on? With such a long exposure, even on a tripod, you can get camera shake.

Better bet is to try this with better lighting, to remove that variable.

 

[Random Orbits]

Did you have mirror lockup turned on? With such a long exposure, even on a tripod, you can get camera shake.

Better bet is to try this with better lighting, to remove that variable.

+1 on repeating the test in better light. But also do a back-to-back test with your 24-105 at 24mm in the same light with IS on and off. The 24-70 II should be better than the 24-105. The IS vs. non-IS will give you some insight as to how steady your tripod is.

 

[absente]

Alright, I am now repeating the test in better light and in M mode to keep the exposure the same. I will enable mirror lockup in PF too, and set the timer to 10 seconds. I hope this will be enough.

PS: I don't have a 24-105 to compare it with - the only thing (right now) I do have is an old 50/1.8, but they seem to perform the same at 50mm.

I will post a little more test-shots, this time. Hold on (and thanks for your support).

 

[Zv]

First of all you need to calibrate it with either dot tune method or Reikan Focal. If you can spend that much on a lens you can afford the software to calibrate it.

Secondly, and after calibration you should take it outdoors on a day with decent amount of light place it on a tripod and snap a few different types of shots at different apertures. Not just f/2.8. You should also fire some shots handheld but with a fast shutter speed. Aim it at something that has a lot of fine detail like a cracked wall.

I dunno how close you were to that box but I bet it was near MFD. That would give a very shallow dof making much of the scene soft.

And you should test it in RAW since jpeg adds sharpening and NR. Switch in camera NR to off for both high ISO and long exp when testing.

Repeat test with 24-105 and compare in LR or other processing app. If the 24-70 shots still look softer than the 24-105 ones in similar conditions you should return it or have Canon look at it.

Just read that you don't have the 24-105 so compare to the 50 1.8. @ f/2.8

 

[neuroanatomist]

If you have a flash, you can avoid a slow shutter by shooting with flash. Also, if you were near MFD the 24-70 II isn't at it's best when focusing close.

 

[Random Orbits]

If you have a flash, you can avoid a slow shutter by shooting with flash. Also, if you were near MFD the 24-70 II isn't at it's best when focusing close.

+1. If your 24-70 II doesn't beat the 50 f/1.8 at f/2.8 in good light then something is wrong.

 

[rmrc]

same opinion as Neuro, you should do the test away from the MFD

 

[absente]

Okay, doing new tests now - is 3 meters enough? Pictures follow.

PS: Somehow while seeing the picture in live view and later comparing it with the final image - live view seems much more in focus (at 10x) before taking the shot.

 

[BoneDoc]

Best bet is to do it with sunshine helping you. Just shoot outside, and you'll be able to see the difference at 100% magnification.

One thing I notice once I got my 24-70 MkII is how much camera shake I induce into the photograph, even at 1/100 . The lens is that sharp!

 

[absente]

some new shots. (PS, the 1m distance is actually 1,5 meters)


vk1


vk2


vk4


vk3


vk5


vk6

 

[absente]

I will test it outside tomorrow and in much better conditions - it's 1AM currently here, so the only place I can test it is at home.

To be quite honest, my 50/1.8 doest do much better at 2.8, but I was really expecting more from this lens. The MFD is about 0,4m so at 1,5m it should deliver good results, especially in the center where I frame the test-shots.

Something thats unclear to me is also why I can't get the same "focus" like on the live preview (10x) on the final picture.

I guess I gotta test it tomorrow in real world conditions and see how it performs. Yet, I still would like to hear if even those crappy testshots are in the normal range in terms of focus (in similar conditions)

 

[absente]

2V6A3204_crop


2V6A3205_crop


2V6A3206crop

 

[Mt Spokane Photography]

There is such a thing as a defective lens, you should be able to get a sharp image.

 

[Canihaspicture]

You don't have a cheap filter on that lens do you??? The pic looks blurry to the point were it doesn't even look like missed focus, it looks like a dirty lens.

 

[absente]

You don't have a cheap filter on that lens do you??? The pic looks blurry to the point were it doesn't even look like missed focus, it looks like a dirty lens.

Of course not - no filter at all. I used my flash for the bacardi shots, the light here is not very good so perhaps it didn't focus quite well - yet i repeated those shots a couple of times.

Just as a notice - in-body sharpening is completely switched off.

 

[absente]

There is such a thing as a defective lens, you should be able to get a sharp image.

I am not saying it's defective, and as you can see, I am trying to get a sharp image by getting suggestions from you and other forum members.

 

[Canihaspicture]

You might consider having sharpening turned on and a little contrast if you plan to shoot jpg and not post process... a little goes a long way.

This image is pretty sharp for the resolution in my opinion after minor post processing. (obviously it destroyed your text though haha)

 

[absente]

You might consider having sharpening turned on and a little contrast if you plan to shoot jpg and not post process... a little goes a long way.

This image is pretty sharp for the resolution in my opinion after minor post processing. (obviously it destroyed your text though haha)

That pic has no PP whatsoever. All the samples are straight from the camera, everything to 0.
But I agree, in camera sharpening solves the problem if I push it to 5-6-7, yet I wouldn't say its the perfect solution.

Maybe it's my mistake, I've read so many happy reviews about the sharpness of this lens, so I wanted to give it a try without in-body-sharpening turned on. Yet at 2-3 meters, a 2300$ lens should be sharp, without resampling the file (e.g. apply sharpening) - or am i wrong?

 

[Canihaspicture]

In all of my experience I've never had a camera with an anti-aliasing filter and any lens produce a perfectly sharp image at 100% with no post.

Sharpening images is an art and if every picture came out of a camera perfectly sharp then sharpening wouldn't be such a popular topic.

Your choices are sharpen in camera or plan on pp for every image (no matter what lens).

Obviously if you aren't displaying at 100% then it won't matter as much. Granted, stopped down to f/8 or f/11 with a good flash and it'll be pretty darn sharp.