What went wrong.. was it me?

PilotBrad · June 10, 2010

Last week I rented a Canon EF 100-400 L to take with me on a short trip to our friend's vacation house in the mountains. I have a 7D and I've been looking at buying a telephoto lens for it, and this rental was to serve as a test-drive. Now that I am home I am beginning to process all of my images, and I am quite disappointed at the shots I came home with.

Here are some examples of the RAW files processed in LR3 and viewed/cropped at 100% (1:1). These are just a few examples of several hundred images I took and I would say about 99% of them are fuzzy.

I'll go into the details of each shot below, but I am interested to get your opinions.... Was my technique wrong? Am I expecting too much? Or did I did I get a bad copy of this lens?

1) ISO400 (edit: not 100), f5.6, 1/100s, 400mm, spot focus on eyes, tripod and remote shutter used, subject about 50-60' away. Subject motionless, even the tree the bird is sitting in is fuzzy.

2) ISO100, f6.3, 1/800s, 400mm, handheld, spot focus on head, subject about 100' away, subject moving slowly.

3) ISO200, f6.3, 1/1250s, 400mm, handheld, spot focus on head, subject about 25' away, subject motionless.

So please school me a bit... where did I go wrong? I've seen many amazingly sharp images from this body/lens combo, but there is no way any of my images even come close. Or am I just expecting too much?

I know I could have micro-adjusted the focus of the lens, but I didn't have time; I picked up the lens and hit the road.

Art Scott · June 10, 2010

is the lens stabilized?? If so ....were you waiting for stabilization to set??
It takes a second or 3 for stabilization to set on most cams.....
I have stabilized lenses but prefer to shoot off a tripod for my own peace of mind.

PilotBrad · June 10, 2010

Thanks for bringing that up Art... IIRC I had IS off for shot #1 because of the tripod, but on for #2 and #3.

But yes, I was waiting for IS, when used, to set.

SimpsonBrothers · June 10, 2010

Did all of the shots turn out blurry?
Just checking to see if you rented a defective lens.

pathfinder · June 10, 2010

They are a bit soft - so the question is this your camera or the lens you rented. Did the view seem crisp through the viewfinder? Does your 7D seem to focus other lenses satisfactorily?

Did you turn IS OFF? Are you sure? On a tripod, the IS for the 100-400 definitely needs to be turned off.

The EOS 100-400 has never been a favorite of mine, but these images are not acceptable.

i would like to see the full image also, not a crop.

I have not had a lens that looked this bad, even when I found it needed a +10 micro adjust with my 50D. SO I doubt that really is the issue.

Foques · June 10, 2010

I wonder what would happen if you would open aperture as wide as possible, and get a higher shutter speed.

nevermind.. i'm an idiot.

Just read parameters for the handheld shots...

PilotBrad · June 10, 2010

SimpsonBrothers wrote: »

Did all of the shots turn out blurry?
Just checking to see if you rented a defective lens.

Most are blurry yes, but even the better shots aren't as sharp as I had hoped.

pathfinder wrote: »

They are out of focus - so the question is this your camera or the lens you rented. Did the view seem crisp through the viewfinder? Does your 7D seem to focus other lenses satisfactorily? Did you turn IS OFF? On a tripod, the IS for the 100-400 definitely needs to be turned off.

The EOS 100-400 has never been a favorite of mine, but these images are not acceptable. Are these images you linked at 100% or something different?

Yes, they seemed crisp thought the VF, and the camera had no issues finding focus quickly. I don't seem to have this issue with other lenses, but this is my first experience with this lens and one of this focal length. As an example, here are some shots I took just a few weeks ago with my 17-55mm f2.8.... sharp I think.

IS was off for the first shot (#1 above) as a tripod and remote shutter was used. I am not 100% sure, but I do remember having an ah-ha! moment as I remembered to turn it off.

Yes, they are all at 100% (or 1:1 in LR3). I took screenshots of them as they displayed in LR3 and saved them as JPEGs through PS (quality 100). They appear almost exactly as I see them in LR.

I'll throw up some of the full sized images when I am home.

Thanks.

PilotBrad · June 10, 2010

One thing I forgot to mention is that each of the samples above is just one in a series. For example, I must have taken 30 shots of that hawk, 5 of the geese and 3 of that bird on the pole, tweaking settings along the way, and all of them exhibit the same lack of sharpness.

craig_d · June 10, 2010

One issue I don't see being discussed here is that you are using a hellaciously high-res sensor (18 MP in APS-C format), which at 100% crops will make the most of any softness that is present in the image. Also, at 400mm the EF 100-400mm f/4-5.6L is rather soft until you stop it down to about f/8. Even so, the first shot is softer than I would have expected.

There have been reports of softness in 7D images, with some people claiming the 7D just isn't all that sharp. I haven't been keeping up with the latest news in that area, so I don't know if those reports were debunked (convincingly) or not, but these images remind me of some of the samples that were posted online as proof of the 7D's softness back when the 7D first came out.

Jane B. · June 10, 2010

What about the technic used in making the 100% crops presented here. What is the preferred method? Could that be a factor in how sharp they appear? Found where he said they were screen shots. “Yes, they are all at 100% (or 1:1 in LR3). I took screenshots of them as they displayed in LR3 and saved them as JPEGs through PS (quality 100). They appear almost exactly as I see them in LR.”

PilotBrad · June 10, 2010

Jane B. wrote: »

What about the technic used in making the 100% crops presented here. What is the preferred method? Could that be a factor in how sharp they appear? Found where he said they were screen shots. “Yes, they are all at 100% (or 1:1 in LR3). I took screenshots of them as they displayed in LR3 and saved them as JPEGs through PS (quality 100). They appear almost exactly as I see them in LR.”

IMHO the difference between the RAW images as they are displayed on my PC and the JPEG images presented here is inconsequential. The one change I can see is a small change in color from the ProPhoto color space used by LR3 to the SRGB color space used to output the JPEGs.

As far as I can tell my method of output made no significant difference in their sharpness (I checked before posting), at least not in any amount that would change your opinion of the originals.

But like you said, if anyone has a better way of doing it please let me know.

PilotBrad · June 10, 2010

craig_d wrote: »

There have been reports of softness in 7D images, with some people claiming the 7D just isn't all that sharp. I haven't been keeping up with the latest news in that area, so I don't know if those reports were debunked (convincingly) or not, but these images remind me of some of the samples that were posted online as proof of the 7D's softness back when the 7D first came out.

While it's been tempting, to date I've resisted drinking the "7D is not sharp" kool-aid.

D I read a lot of that stuff before I bought it, and it was certainly a consideration.

I can say that I have never felt that the 7D is soft, especially when I shooting in JPEG. On the other hand, learning to deal with the RAW files has been an experience for me; they are never as sharp as I think they should be but I've been able to output some good JPEGs after sharpening, so I've had to learn that soft RAW files are normal. If you are curious, every picture in my public galleries (link below) has been taken in RAW with my 7D... take a look and tell me if you think they exhibit softness.

I can't remember where but I recall reading something about the pixel-pitch of the 7D contributing to its rumored softness; something about it makes the camera very senstitive to movement. I have no idea if that is true, but it sounded good when I read it.

PilotBrad · June 11, 2010

Ooops!
DELETED: Server error when posting resulted in a dup.

PilotBrad · June 11, 2010

Ooops!
I mistakenly reported image number 1 as ISO 100, when it was 400. That's good new I guess because I was beginning to wonder why I had so much noise and grain at what I thought was ISO 100.

If interested, you can view full sized images here. A little bit of input sharpening was applied when I imported them into LR3, but no NR or output sharpening was used.

EDIT: I've also uploaded another shot of the hawk at ISO 100, which is considerably less grainy. In addition, I've uploaded a shot of a squirrel taken with my 17-55 f2.8 (captured for my 3 yr old daughter who kept calling for the "squallers"). No output sharpening was applied and it is tack-sharp at 100%! So that then begs the question, am I expecting too much from the 100-400? Regardless of whether it was the technique or the equipment, should I expect the same or similar sharpness from it when viewed at 100%?

Dan7312 · June 11, 2010

PilotBrad wrote: »

I've uploaded a shot of a squirrel taken with my 17-55 f2.8 (captured for my 3 yr old daughter who kept calling for the "squallers").

Hmmm, same body one lens sharp the other not. Seems like by process of elimination it is a lens problem.

You could try the newspaper test. Tape a newspaper to the wall. Shoot on a tripod, both at say f8 or so. Get newspaper as close to the same size in the images as possible (you'll have to move the tripod). Enough light so both at about 1/60 to 1/125.

Now you've got a pretty good comparison to tell you whether it is lens or technique.

I don't have the experianced eye that many on smugmug do, but none of those 100-400 shots have the quality I'd expect from that lens.

ziggy53 · June 11, 2010

1) Craig has some observations which most certainly play into the evaluation of this lens performance from these sample images.

2) Since micro-focus calibration was not performed, that is still a potential explanation for "some" of what we see in the sample images.

3) Subjects at great distance are more subject to atmospheric degradation, especially apparent using very long focal lengths and 100 percent magnifications.

4) IS is not a panacea and does not eliminate camera shake. A tripod does not guarantee freedom from camera shake either. User technique, wind, shutter slap and mirror slap can all contribute to shake and resulting blur. Using an exemplary tripod/head (or 2 tripods) and MLU will generally provide best results.

5) The lens may not be an optimal example. Only testing another sample of the lens under the same conditions, preferably at the same time, is going to show a practical comparison.

Combine all of the above elements to understand the variabilities and possible contributors to what the sample images seem to show. Trying to isolate any single contributor may be too simplistic to be useful. Attacking all causes is the best road to ultimate understanding.

pathfinder · June 11, 2010

I agree that the pixel size- pixel pitch places real demands on the resolving abilities of lenses on the 7D. There was a good article on the Luminous-Landscape about the pixel pitch of the 50D ( with a lower pitch than the 7D) being very near the theoretical limits for lenses available for it.

Micro adjustment of the AF, and great care in camera, long glass technique will help optimize what the lens is able to capture. As Ziggy suggested, the best improvement of your lens comes from a good sturdy tripod, MLU, and a remote release, as well as the lowest appropriate ISO. Basics still count.

I just micro adjusted my 7D and my Tamron 200-500 last night, and found it to be a quite nice sharp pairing, so I will be taking it to the Indy Air Show tomorrow. We'll see how correct I am, soon shooting hand held without IS or a tripod

PilotBrad · June 11, 2010

Thank you EVERYONE for your very helpful comments!!!

ziggy53 wrote: »

5) The lens may not be an optimal example. Only testing another sample of the lens under the same conditions, preferably at the same time, is going to show a practical comparison.

That's where I am heading next. I have a pro-photog-friend and I've asked him if I can take a few shots with his 100-400 on my camera. He assures me his is tack sharp on him 5DMKII, so I'll take a few shot and I'll have him to the same and we will see what happens.

pathfinder wrote: »

I just micro adjusted my 7D and my Tamron 200-500 last night, and found it to be a quite nice sharp pairing, so I will be taking it to the Indy Air Show tomorrow. We'll see how correct I am, soon shooting hand held without IS or a tripod

I'd love to see some samples when you have them. One of the reasons I want to get the bottom of this 100-400 thing is that I was planning on buying or renting one to take with me to Oshkosh / AirVenture this summer.

pathfinder · June 11, 2010

I used the Tamron 200-500 without micro adjust here - http://pathfinder.smugmug.com/Airplanes/Terre-Haute-Air-Show-September/3483858_LtjNf#211806892_u4g9L

PilotBrad · June 11, 2010

pathfinder wrote: »

I used the Tamron 200-500 without micro adjust here - http://pathfinder.smugmug.com/Airplanes/Terre-Haute-Air-Show-September/3483858_LtjNf#211806892_u4g9L

I was admiring those earlier today... nice! Have you used that lens much on the 7D? If so, how does it perform?

pathfinder · June 11, 2010

Not yet - but this was from about 1/4 frame of a 40D I have 16 x 20 prints from this frame.

I am fairly agnostic about camera bodies, I tend to use what I have at the time. I will be using the 7D tomorrow though for airplanes in flight.

I have always shot single shot AF, but will give AI Servo a chance tomorrow with the 7D using the central AF sensors and see if it does a better job for me.

When I was micro adjusting the 7D with the Tamron 200-500 tethered to my MacBookPro with EOS utility last evening, I thought the images looked quite acceptable, to better than what I was getting with my 50D. We'll know in few days.

kdog · June 11, 2010

Honestly, the larger problem with these shots is that they're all at least a full stop underexposed and very noisy. I'm guessing they started off even worse before they were processed, judging from the noise. There are some esteemed photographers who've contributed to this thread, and maybe I'm a dummy for going out on a limb, but I will say this. The sharpness of these images cannot be evaluated with such poor exposures.

PilotBrad · June 14, 2010

kdog wrote: »

Honestly, the larger problem with these shots is that they're all at least a full stop underexposed and very noisy. I'm guessing they started off even worse before they were processed, judging from the noise. There are some esteemed photographers who've contributed to this thread, and maybe I'm a dummy for going out on a limb, but I will say this. The sharpness of these images cannot be evaluated with such poor exposures.

kdog, thanks for commenting. Did you look at the full sized images I linked to later or just the crops? I'm honestly asking because as I've said, I am trying to learn.

I am not going to disagree with you, if I had to do it again I definately would expose more. However, I will say that of roughly 30 shot of the hawk, many with higher exposures, none are sharp.

As far as the noise goes, I sat down with a friend in his studio this weekend to go over some of these, and it turns out much of the noise was created by the capture/input sharpening preset I was applying during import. What I thought was a light or moderate amount of input sharpening turned out to be higher than I thought, and when combined with underexposed images has been leading to excessive noise. I've subsequently dialed back my sharpening preset and it seems to be helping a lot. In addition, the shots had no sharpening mask applied to them at this point, which when applied has also helped combat the noise.

The one interesting comment he made was that he was surprised to see how quickly noise developed in the images from my 7D when sharpening was applied as compared to those from his 5DmkII, even in areas of blue sky. I'll leave it at that.

Anyway, with the benefit of little more knowledge, I took several test shots with his EF 100-400, and I happy to say that I was able to get MUCH better results this time around. Here is a full sized sample taken handheld from a block away.

After much thought and research I think my issues last week were caused primarily by poor technique, underexposure, and overly agressive input sharpening. Thanks to everyone for helping me figure that out! That being said, the images I got with my friend's lens are all sharper than the ones from last week, so maybe just maybe that particular lens didn't play nice with my camera body.

kdog · June 14, 2010

Yeah, that looks awesome. Keep doing whatever you were doing for that shot, and you'll be fine.

ziggy53 · June 14, 2010

Keep "that" friend. (The one who identified your workflow issues.)

thumb

pathfinder · June 14, 2010

PilotBrad wrote: »

I was admiring those earlier today... nice! Have you used that lens much on the 7D? If so, how does it perform?

I am working up the shots from Sunday here - the air was hot and very, very humid. The show was cancelled due to heavy rain Saturday and the ground was heavily saturated with water, hence the air was far from crisp, but dull and grey at a distance.
http://pathfinder.smugmug.com/Airplanes/Indianapolis-Air-Show-Mt/12547630_hezVS#900261275_Ut25U

I shot the all parked images with a 5DMkii 24-105 f4IS as MRaw images, and the in flight images with the 7D in AI servo mode with my Tamron 200-500 handheld without support, again inMRaw. I used the smaller RAW files to allow me not to have to swap cards during the day long event.

The only processing on these images was done in Lightroom3 with global editing. they have had no pass through PS, although a few might benefit from it.

You don't get to see three P-51s in the air together that often anymore

PilotBrad · June 15, 2010

Very nice, thanks for posting these!

It always amazes me just how big that Skyraider is for a single-engine aircraft, but I guess 2700HP helps a bit.

Stella7d · June 19, 2010

Fabulous shot of the P-51's!!

ChatKat · June 20, 2010

When trying to determine if the error is me or the camera, use one of the auto modes (I use P for this) and see what the camera does without my help. If the photo is decent then I know that it was me. If the image is awful then it's the gear.

What went wrong.. was it me?

Comments