5D Mark 3 - Please help!

[Cinnamon]

I bought a 5D3 to replace a 5D2 and 7D - and feel like the 5D3 images I'm getting after cropping are noticeably softer than those my 5D2 produced! I've heard that complaint from other people, too (seen quite a number of threads, including here at CR) and have seen various responses to that. I have seen other people post and get helpful feedback, so I'd love to get people's opinions on whether they think the hyperlinked pictures are soft when zooming in (up to 100%)? I first noticed softness when zooming in before 100%, and then decided to compare 5DMkII images side to side and definitely felt in a lot of my 5D3 shots it wasn't as sharp...

FYI, this is NOT a RAW software issue that some other people had - I shoot in JPEG, and notice the perceived softness relative to the 5D2 when zooming in on the LCD screen, too. I have NOT yet done any focus micro-adjustment, but still think the shots are able to be analyzed for detail sharpness when zooming in. I had other shots which appeared softer, which I'm not including: I'd rather show what I think are relatively sharp shots, to see if those have more softness than they should.

People, please give me your feedback on whether there is more softness when zooming in than there should be! That being said, please don't add comments about why you shouldn't "pixel peep" or what could be improved about the composition - these aren't the best pictures as these were shots taken purely for posting here.

I am attaching a link to a sampling of various 5D3 photos I've shot. These are SOOC, and haven't been edited. Since I'm throwing in a variety, I am not listing all the camera settings (although I believe when you save the photo to your computer you can see the EXIF data and make judgments accordingly). The lenses I use are the 24-105L, 70-200 2.8 IS II, and 100mm Macro (non-IS).

5D3 Samples: LINK http://www.glenargov.com/Other/5D3-Test/24800438_Zw5jbH#!i=2029849496&k=73Mshxf

5D2 Samples (with same lenses):
http://www.glenargov.com/Other/5D-2-Samples/24804618_8rnbGd

Please get back to me ASAP as I need to make a decision whether to keep the 5D3 or return it. Thank you in advance!

 

[Cinnamon]

To clarify - by "soft" I don't mean an issue of sharpness, I'm talking about a lack of detail when zoomed in. Again, these are large JPEGs, not converted RAWs. I generally shoot at smaller apertures for a less shallow depth of field, so while these might not be the best photos, I'm not concerned about issues about focus - I'm concerned about detail from crops.

While I'm not including 5D2 shots in the gallery for comparison, I'm hoping someone commenting would have experience with a 5D2 to compare to. Also, I'm aware at High ISOs the 5D3's noise reduction results in a lack of detail - yet I felt I noticed softness even at low ISOs.

Please give your feedback! Again, not the picture zoomed out, but when zoomed in up to 100% - that's what I'm curious about. Thank you!

 

[wickidwombat]

sounds like the ham fisted noise reduction possibly try turning it down or off maybe provide some examples?

 

[Cinnamon]

Thanks for your feedback! The NR is on its default of 'Standard' - since I think the Noise Reduction really kicks in at high ISOs, and I had seen a lack of sharpness at low ISOs too, it might be something else.

If people think there is no detail softness when zoomed in, it could be my imagination - but again, comparing my 5D3 to 5D2 side by side at 100%, I really felt as if the 5D2 had more detail...

 

[TAF]

The photos you've put up don't seem bad to me.

It does look to me like there is perhaps some haze in the air in some of the city shots, but that is not surprising...maybe the top down bus shot has the beginnings of some flare (were you using your lens hood?).

But without anything to compare to, I can't draw the conclusion that there is anything wrong here.

Could you perhaps do a side by side with your 5D3 and another camera using the same lens so we can compare apples to apples?

 

[Richard Lane]

I think the shots look fine! It just looks like you're dealing with some harsh lighting and haze. A CPL may be useful for some of the building shots, in order to improve saturation and contrast and cut back on haze. The shots in the shade with the horse and cars look better to me. You should try to shoot earlier in the morning or find that golden hour later in the day.

The 5DIII isn't going to be that much sharper than the 5DII.

I would think the main reason to upgrade to the 5DIII would be for; more accurate and faster Auto focus, faster frames per second and improved higher ISO.

Rich

 

[Cinnamon]

Glad to hear you don't think they're too bad...yes, I almost always shoot with a lens hood.

Good idea...I'm adding a link to shots with my 5D2 also taken with my 24-105 and 70-200. I know this isn't exactly apples-to-apples, which is partly why I'd love people's feedback on whether they think there is a lack of detail when zooming on my 5D3 samples based on their own experiences with their 5D2. But hopefully it can help make a comparison!

I tried selecting photos that have minimal editing, but a few of these had contrast adjustments...no major cropping (if any at all), so hopefully comparing these zoomed in can help establish a baseline...with that said, I'd love to hear people's thoughts based on their own experiences, too.

Some 5D2 Samples: http://www.glenargov.com/Other/5D-2-Samples/24804618_8rnbGd

Thank you so much everyone who has provided their feedback...I'd love to continue hearing thoughts!

 

[Cinnamon]

I think the shots look fine! It just looks like you're dealing with some harsh lighting and haze. A CPL may be useful for some of the building shots, in order to improve saturation and contrast and cut back on haze. The shots in the shade with the horse and cars look better to me. You should try to shoot earlier in the morning or find that golden hour later in the day.

The 5DIII isn't going to be that much sharper than the 5DII.

I would think the main reason to upgrade to the 5DIII would be for; more accurate and faster Auto focus, faster frames per second and improved higher ISO.

Rich

Hey Rich! Thanks so much for posting feedback, I really appreciate that you and the other posters took the time to give thoughtful responses.

Definitely know what you mean regarding harsh lighting and haze - these are definitely not my best photos lol.

I didn't expect the 5D3 pictures to come out much sharper/have more detail when zoomed in than my 5D2, I just was surprised by what I felt was inferior detail when zooming in. I understand Noise Reduction at High ISOs could be the culprit behind some shots appearing a little 'soft' but I was surprised by low ISO shots looking more pixelated when zooming in than my 5D2 shots.

I definitely LOVE the better features of the 5D3 over the 5D2, I just want to make sure people think the shots don't come out looking too pixelated when zoomed in. I've read online some people say they ended up getting another copy of the 5D3 which was sharper - I'd prefer not to do that, which is why I really appreciate people giving their feedback...

 

[ageha]

5D3 Samples: LINK http://www.glenargov.com/Other/5D3-Test/24800438_Zw5jbH#!i=2029849496&k=73Mshxf

Looks a bit too mushy for a full-frame sensor. Though, I don't think it's a reason to return it. It's still a great camera and I don't think you can get better results from a Canon right now.

 

[Luke]

Gotta say, I feel the same... Your 5D II pics do look a little better...

I upgraded from a T2i to the 5D III, and thought my 5D III jpg's looked soft.
My raws, once 'finessed' and then converted to jpg look gorgeous though... I wonder if there's some simple setting we're missing...

 

[Hugo Fisher]

Read this: http://www.dpreview.com/reviews/canon-eos-5d-mark-iii/29

citation: "Conclusion - Cons

Destructive noise reduction results in mushy JPEGs, even at base ISO"

My solution: shoot in RAW.

 

[revup67]

Have you tried any of the following:

Manual Focus and AF on the same scene tripod preferred. Take off IS when on tripod however. Shoot RAW to get max sharpness just for this test run. leave default settings in DPP
Swapping the 5D2 and 5D3 with same lens on tripod - same scene
doing an AFMA if you have software such as Focal or Lens Align. It is possible the lens is not out of alignment since last used with the 5D

Because the camera is new does not mean it is without fault. I brought out of the box a new 400mm and the AF mechanism had issues, 580 EX II flash 2 months had issues, and 7D 9 months old the AF motor had issues. Sure you can send it back but if you have 30 days to return and are local to Canon it might be worth dropping it off (what I did), they found the issues and all is good. Had camera back in less than 5 days.

 

[M.ST]

Even if you turn off the noise reduction the 5D Mark III add a noise reduction to the images and you lost details.

If you compare ISO 3200 and ISO 6400 images no notice, that with ISO 6400 you lost additional details.

JPG-quality is definitely better if you shoot with the 5D Mark II.

Shoot in RAW and develop the images in DPP or Photoshop CS6.

I report this to Canon four month ago and Canon is not able to optimize the image quality with a firmware update until today. It´s a shame, because we all had payed a lot of money for the 5D Mark III.

 

[keithfullermusic]

I don't mean this in an insulting way, but why are you shooting jpegs on that sweet camera?

I agree, that they should still look better than the ones on the mark ii, but still. You paid all that money for that camera, so why don't you get the most out of it.

I will say that I initially had my sd card backup in jpegs, but when I compared the raw to the jpegs the jpegs looked terrible compared to the raws, so now I just have it backup to raw also.

 

[Cinnamon]

5D3 Samples: LINK http://www.glenargov.com/Other/5D3-Test/24800438_Zw5jbH#!i=2029849496&k=73Mshxf

Looks a bit too mushy for a full-frame sensor. Though, I don't think it's a reason to return it. It's still a great camera and I don't think you can get better results from a Canon right now.

Thanks for your response! So you think that while the results are mushy, that's to be expected from a 5D3, and not an issue with mine in particular???

Thanks again

 

[Cinnamon]

Gotta say, I feel the same... Your 5D II pics do look a little better...

I upgraded from a T2i to the 5D III, and thought my 5D III jpg's looked soft.
My raws, once 'finessed' and then converted to jpg look gorgeous though... I wonder if there's some simple setting we're missing...

Hey Luke, thanks for your response. I'm glad to hear I'm not the only one who thinks the 5D2 results look better! I don't want to sound like a lunatic, as if I'm pixel-peeping needlessly and imagining issues that aren't there...but at the same time, if I drop $3,500 on a camera, I want to make sure it gives the results I paid for

I'm interested to hear you thought your JPGs looked soft, too - if you think that coming from a crop-sensor body like the T2i (a great camera, I got many awesome shots from mine up until I upgraded to the 5DMk2), then it makes me feel a little better that I'm not the only one noticing it (even though I upgraded from a full-frame). While I'd prefer not to have 'mushy' JPGs, it's much more reassuring to hear it's a universal thing and not an issue with mine in particular.

And yes, I'd love it if it's just a setting we're missing that we can fix. While I admit that I don't do 100% cropping regularly, I do like to have the option to crop when necessary, and it is a little disappointing to see that crops on the 5D3 don't seem to have as much detail as those on my 5D2. Don't get me wrong, I love my 5D3's high ISO performance, autofocus, etc, but it seems annoying that to have more detail in my shots I have to turn down the noise-reduction - NR was part of why I was enamored with the 5D3 to begin with!

 

[Cinnamon]

Read this: http://www.dpreview.com/reviews/canon-eos-5d-mark-iii/29

citation: "Conclusion - Cons

Destructive noise reduction results in mushy JPEGs, even at base ISO"

My solution: shoot in RAW.

Thanks for posting this link, Hugo! Before buying the 5D3 I read a ton of reviews/analyses, but I guess I was selectively looking for the good stuff, because I overlooked that part even though I do think I read the DPReview analysis on the 5D3...

While I'm not crazy about mushy results, I am relieved to see it might be an issue with the 5D3 more generally, and not my copy in particular...

Shooting in RAW is something I might consider, but for my work I generally need to shoot in JPEG. I shoot for an agency that requires uploaded images be in JPEG format, so even if I do shoot RAW I'd have to go through the process of converting them, which is more time-consuming for work which needs to be uploaded relatively quickly. That being said, I might try shooting in RAW+JPEG to dual cards (now I shoot with 2 cards, but in JPEG to both with one as a backup of the other).

Thank you!

 

[Cinnamon]

Have you tried any of the following:

Manual Focus and AF on the same scene tripod preferred. Take off IS when on tripod however. Shoot RAW to get max sharpness just for this test run. leave default settings in DPP
Swapping the 5D2 and 5D3 with same lens on tripod - same scene
doing an AFMA if you have software such as Focal or Lens Align. It is possible the lens is not out of alignment since last used with the 5D

Because the camera is new does not mean it is without fault. I brought out of the box a new 400mm and the AF mechanism had issues, 580 EX II flash 2 months had issues, and 7D 9 months old the AF motor had issues. Sure you can send it back but if you have 30 days to return and are local to Canon it might be worth dropping it off (what I did), they found the issues and all is good. Had camera back in less than 5 days.

Thanks for replying. I have not had the time to thoroughly test out the 5D2 vs the 5D3. I have taken a number of shots with the 5D3, which is why I noticed the mushiness, but I haven't taken the time necessary to properly compare the 5D2 with the same lens from the same spot with the 5D3. While I've never done a microadjustment before, it's something I might try to see if it makes a difference...but again, I wasn't concerned about issues of focus, but detail. So even in out-of-focus areas, I expected to see more detail when zooming in than I did see. I've just been really busy to do a proper controlled experiment, which is why I genuinely appreciate everyone chiming in with their thoughts based on their own experience...

Thanks for your suggestion of sending it to Canon. It seems from some other comments that it might not even be necessary, that the 5D3 simply produces JPEGs without as much detail as the 5D2?

 

[Cinnamon]

Even if you turn off the noise reduction the 5D Mark III add a noise reduction to the images and you lost details.

If you compare ISO 3200 and ISO 6400 images no notice, that with ISO 6400 you lost additional details.

JPG-quality is definitely better if you shoot with the 5D Mark II.

Shoot in RAW and develop the images in DPP or Photoshop CS6.

I report this to Canon four month ago and Canon is not able to optimize the image quality with a firmware update until today. It´s a shame, because we all had payed a lot of money for the 5D Mark III.

Thank you for your post! I'm glad to hear that more and more it seems like it's not my 5D3 in particular that has soft JPEGs. On the one hand, it's a relief to see I wasn't just imagining stuff when I thought my 5D2 produced more detailed JPEGs...at the same time, I'm disappointed (like you) that Canon hasn't addressed this issue yet! I hope they have a firmware release at some point in the near future to fix this...thank you once again.

 

[Cinnamon]

I don't mean this in an insulting way, but why are you shooting jpegs on that sweet camera?

I agree, that they should still look better than the ones on the mark ii, but still. You paid all that money for that camera, so why don't you get the most out of it.

I will say that I initially had my sd card backup in jpegs, but when I compared the raw to the jpegs the jpegs looked terrible compared to the raws, so now I just have it backup to raw also.

Thanks for your comments. No offense taken - the reason I shoot JPEGs is because most of the professional work I do is for an agency where the photos need to be uploaded in JPEGs, and there needs to be quick turnaround time. Therefore I like to shoot and submit with the least editing possible (frequently submit results SOOC). For example, I shot the Tony Awards here in NYC and filled up so many cards with thousands of shots, through which I just selected what I thought were the best, and uploaded those. While it would have been nice to tweak each shot's settings in RAW, then convert to JPEG, I wanted to upload as quickly as possible so they'll sell (a lot of times they sell within minutes of the event). Perhaps I'm mistaken about shooting in RAW, but it seems like a very time-intensive process to edit RAW shots?

When you say "I agree, that they should still look better than the ones on the mark ii"...do you mean that the 5D3's JPEGs should generally look better than the 5D2 but that they generally don't across all 5D3's, or that my 5D3 sample images should in fact look better than the 5D2 samples I posted and therefore there might be an issue with my 5D3? I would much prefer that it's a 5D3 issue and not an issue with the copy I got, and from some of the other posts it seems like that might in fact be the case?...

Thanks again for your post, and for everyone's help. I really appreciate you all taking your time to provide your thoughts...