Canon 70-200mm f/2.8 IS II - Is my lens soft? Photos included

[jd_calloway]

<p>Hello all, I purchased the Canon 70-200mm f/2.8 IS II last week. I've used the lens a few times now with my Canon 5D Mark II and I'm underwhelmed with the sharpness of my photos. They just don't seem tack sharp to me. Unfortunately, I don't own (nor have experience with) another 70-200 lens, so I can't personally compare the photos from this newer model with anything else. However, my gut is telling me if the lens' sharpness isn't making me say 'wow,' then something is wrong - either with the way I'm using it, or with the equipment I'm using.<br>

My Question: Is my 70-200mm not as sharp as it should be? Is it the lens or something I'm doing? I've included some photos below. Thank you for the help. I appreciate it tremendously.<br>

Note: Sorry for not posting images directly. Tried to post them using the 'Insert/Edit Image' button in the post creation page, but that didn't work, and I searched all over site for a how-to guide on posting photos, but didn't find anything.<br>

URL:

f/2.8<br>

70.0mm<br>

1/1000<br>

ISO: 50<br>

IS: OFF</p>

<p>URL:

f/2.8<br>

70.0mm<br>

1/1000<br>

ISO: 50<br>

IS: ON</p>

<p>URL:

f/2.8<br>

70.0mm<br>

1/1000<br>

ISO: 50<br>

IS: ON</p>

<p>URL:

f/8.0<br>

70.0mm<br>

1/250<br>

ISO: 100<br>

IS: ON</p>

<p>URL:

f/8.0<br>

70.0mm<br>

1/250<br>

ISO: 100<br>

IS: ON</p>

<p>URL:

f/8.0<br>

70.0mm<br>

1/200<br>

ISO: 200<br>

IS: ON</p>

[dcstep]

<p>To post here, from Flickr, select the Arrow pointing right on Flickr, then Embed, then select a size no longer than 800p on the long side, Copy that. Next, come here, Select "HTML" here, Paste, then delete everything between <script and script> Yes, it's a pain in the butt that this sight doesn't support BBCode and you have to modify the HTML, but you quickly get used to it.</p>

<p>As for your lens, your demonstration subjects don't lend themselves to easy study of sharpness. I'd suggest starting with something like a dollar-bill, taped to the wall. Mount the camera/lens on your tripod, use Liveview for focus, turn off IS and use remote release or the timer. As you've done, shoot wide open, f/4, f/8, f/16 and see how the shots compare in sharpness. Shoot from around 18-feet away. Those test shots should be really super sharp. If they're not, then try other lenses with the same subject to see if it's lens or body.</p>

<p>If your tripod mounted images were sharp with Liveview, next try focusing using the viewfinder. Use One-Shot mode for all the tests. (AI Servo can cause drift between shots). If those are not as sharp, then you need to micro-adjust the camera to the lens.</p>

<p>Your camera body has excellent high-ISO performance, so don't be afraid to raise it high enough to get shutter speeds as high as 1/400-sec. when hand holding. The IS is superb and with good technique, you should be able to hand hold below 1/100-sec. at 200mm, but that does require experience and excellent technique. </p>

<p> </p>

[AzDavid]

<p>I wouldn't panic. Everything takes practice ... as my Canon 100-400 mm reminds me every time I use it. Take some time. Read up on focus/sharpness and try different things. What do images look like at 200mm? Try shooting different subjects ... like people ... and different times of day. Have you tried manual focus? Tripod? Shooting landscapes and other things at f/8 or higher can improve depth-of-field and overall sharpness (unlike your first image shot at f/2.8). And always keep the IS turned on unless you've got a good reason to turn it off. I don't think you have enough data yet to decide the lens is flawed. </p>

[jd_calloway]

<p>Thank you for the responses<br /> David Hoye, I did shoot at f/8.0 to test whether there'd be an improvement in sharpness. The photos above suggest a slight improvement in sharpness at f/8.0 as opposed to 2.8. However, both apertures still underwhelm. I've shot landscapes, people, and objects in the week that I've had the lens. Except for some close up shots taken midday when the lens was pointed straight down (leaves on the ground at about 3-4 feet from lens) which seemed close to tack sharp, all other shots seemed, while not exactly soft, not exactly super sharp, either.<br /> David Stephens, thank you very much for taking the time to explain all of that. I just performed the test and uploaded the first batch of results. I used the 70-200 f/2.8 IS II along with my 5D mark II. As instructed, I used a tripod, shot from about 18 feet away using a remote release, and IS was off. Unfortunately, again, I don't have any experience with the 70-200 so I can't really judge whether my lens is functioning as it should be, so any feedback would be really appreciated.</p>

<p><a title="_MG_9130" href=" data-flickr-embed="true"><img src="https://c6.staticflickr.com/6/5324/30607698445_ed069fdcd6_c.jpg" alt="_MG_9130" width="800" height="433" /></a><br /> URL: /> 1/50<br /> f/2.8<br /> 200.0mm<br /> ISO: 1600</p>

<p><a title="_MG_9132" href=" data-flickr-embed="true"><img src="https://c5.staticflickr.com/6/5676/30307691300_b90d1ef5ff_c.jpg" alt="_MG_9132" width="800" height="410" /></a><br /> URL: /> 1/50<br /> f/3.2<br /> 200.0<br /> ISO: 1600</p>

<p><a title="_MG_9133" href=" data-flickr-embed="true"><img src="https://c7.staticflickr.com/9/8618/30571262366_354e3d71fd_c.jpg" alt="_MG_9133" width="800" height="409" /></a><br /> URL: /> 1/50<br /> f/3.5<br /> 200.0<br /> ISO: 1600</p>

<p><a title="_MG_9134" href=" data-flickr-embed="true"><img src="https://c8.staticflickr.com/6/5741/30607698215_2aced68792_c.jpg" alt="_MG_9134" width="800" height="422" /></a><br /> URL: /> 1/50<br /> f/4.0<br /> 200.0mm<br /> ISO: 1600</p>

<p><a title="_MG_9128" href=" data-flickr-embed="true"><img src="https://c1.staticflickr.com/6/5471/30571262496_be8640c554_c.jpg" alt="_MG_9128" width="800" height="380" /></a><br /> URL: /> 1/30<br /> f/4.5<br /> 200.0mm<br /> ISO: 2000</p>

<p><a title="_MG_9137" href="https://www.flickr.com/photos/144446874@N07/30607698325/in/datetaken/" data-flickr-embed="true"><img src="https://c6.staticflickr.com/6/5685/30607698325_01798e9fde_c.jpg" alt="_MG_9137" width="800" height="402" /></a><br /> URL: https://flic.kr/p/NCGsmD<br /> 1/50<br /> f/5.0<br /> 200.0mm<br /> ISO: 1600</p>

[kerry_grim]

Have they been sharpened in post processing?

 

Did you check your focus accuracy?

[dcstep]

<p>LOL, perhaps I should have said, "Crisp, new dollar-bill." Also, the upload to Flickr should be full-resolution. I would expect a file that's several MB and yours are under 1MB. One final thing, shoot in Manual mode so that the relative exposure doesn't change, in a brighter room. ISO 1000 at f/2.8 only yielded 1/50-sec. That's a pretty dark room. There's nothing wrong with supplemental light near the subject. Flash would be okay, but you would have trouble synching at high shutter speed. </p>

<p>If your files are small because you needed huge crops to fill the frame, then move closer, to crop as little as possible.</p>

[jd_calloway]

Kerry,

 

No, I haven't sharpened the photos at all in post processing. I didn't brighten a few up, though. Also, I have checked for

focus accuracy and micro adjusted. I used manual focus in live view for the tests.

[JRCrowe]

<p>I have found that that lens can out resolve the sensor on the 5D MKII. If you can try it out on one of the newer Canon cameras with more megapixels I think you could see the difference.</p><div></div>

[JRCrowe]

<p>Here is a 700 pixel crop out of the above photo. This was shot handheld so may have turned out sharper with a tripod.</p><div></div>

[Spearhead]

<p>If you want to know if the lens is sharp, put the camera on a tripod and shoot a good high contrast test subject. Try it with autofocus and manual focus without moving the camera. If the MF ones are sharper, you have an AF problem, look into calibrating the lens. If they are fine, then shoot the same target handheld with a shutter speed fast enough to minimize any issues. If those are softer than the tripod shots, then you have three things that could be causing the problem: 1. technique, 2. settings, 3. malfunctioning IS. </p>

[jd_calloway]

John Crowe,

 

Thanks for uploading those photos. Without hesitation I can say my lens has not shot photos as sharp as those -

especially handheld. That's the kind of sharpness I expected when purchasing the 70-200 and my instincts told me

something was wrong when I didn't see it. I mean look at the dollar bill photos above. That was on a tripod and, for me,

those shots don't come anywhere close to the sharpness of your handheld photos.

 

Fortunately, the lens is still within the return period, so I'll definitely take it back. I dont know if I'll be exchanging it for a

new unit or not, but if I do, I'll update the thread for future references.

[JRCrowe]

<p>This was shot with the 70-200 with the 5D MK IV with the Canon 2x extender at Yosemite last Wednesday. I used a monopod. I never had real good results with the 5D MKII and the extender, but am having better results with the MK IV. Oh and do what Jeff and David said to do.</p><div></div>

[JRCrowe]

<p>700 pixel crop</p><div></div>

[Spearhead]

<blockquote>

<p><br />Fortunately, the lens is still within the return period, so I'll definitely take it back.</p>

</blockquote>

<p><br />Why not try some scientific tests first? It's amazing how many lenses get returned because of the problems I cited. Test it correctly.</p>

[Tim_Lookingbill]

<p>Keep in mind all the samples posted here of outrageously sharp results of tiny objects in the distance were taken in direct full spectrum sunlight. The EXIF data in your first sample shows no flash was used. What lighting did you use?</p>

<p>AutoFocus metering requires a lot of contrast (provided by bright full spectrum light) to detect edges especially shooting wide open and if you're not using enough full spectrum light, autofocusing can be hit or miss.</p>

<p>Do what Jeff and David suggest.</p>

[mark_pierlot]

<p>I don't understand why one should have to subject a lens to rigorous testing for sharpness if it's not producing sharp results under normal shooting conditions. I've had two lenses, the 35/1.4 L and 24-70/2.8 L, that were disappointingly soft until I replaced then with different tack sharp "copies." Nothing in my technique had changed, only the lenses with which I was shooting had.</p>

<p>I love Canon gear, but certainly won't be an apologist for the company when it comes to their selling of demonstrably substandard L-class lenses.</p>

[JRCrowe]

<p>I downloaded the full sized version of your Queens Museum photo and applied my standard settings in Adobe camera raw, then did a 700 pixel crop. It appears you lens is as sharp as it should be. I can read the museum hours stenciled to the window. it seems the limiting factor is the camera sensor in this case. I have seen complaints of how the 5D MK II sometimes renders foliage and have noticed it myself so that may not be the best subject for a test.</p><div></div>

[Spearhead]

<blockquote>

<p>I don't understand why one should have to subject a lens to rigorous testing for sharpness</p>

</blockquote>

<p> <br>

There have been numerous reports here of people who get two or three lenses before they figure out it's technique that is causing the problem. And this helps drive up prices for everyone, including people who do test their lens rigorously to make sure it's the lens that is the problem.</p>

[dcstep]

<p>What Jeff said. It's particularly true with long focal lengths. </p>

[peter_rowe]

<p>Hi JD,<br>

Sorry I am late to the party. You did good on the dollar bill test but for one thing; you need to fill the viewfinder with the dollar bill. In other words, get as close as needed so that the dollar bill takes up most of the viewfinder (15 feet maybe?). Also, ISO 1600 is a bit high and has the potential to soften images so since you are on a tripod take the shot at ISO 100, f8, with mirror lockup and timer, put some descend light on it and post a 100% portion of the shot here without any post-processing (no resizing). This will show if there's a problem with your lens, it's highly unlikely believe it or not.</p>