Earthshatteringly Disappointed With 7D

[briansquibb]

In my experience problems like this are either through operator error or camera defect.

Inevitably the camera is blamed - which is why there are so many retruns

I have had a camera returned to me as the IQ was 'cr@p' - from a demonstrably perfect 5D classic. The image shown to me to back it up had little white dots all over it. It was raining ........

Back to this 7D issue. It is clear that the 7D for most people does not show significant noise at 100-200 - certainly not on my 2 7Ds either, I checked.

In other forums the 7D has caught out new users with the changes from the previous xxD and xxxD and Rebel models. There are significant changes to the metering. It is clear that out of the box the 7D is not a high ISO camera - care and pp work has to be applied to get clean images at 1600 and above, but iso100/200 should not be an issue.

So is it the user or the camera? Has the camera been checked under warrantee? Have the factory defaults been used (with perhaps RAW set)? Has a test picture been taken in perfect conditions?

Landscapes are difficult subjects to get perfect due to the high DR needed in strong light.

Settings that would impact noise are:

- Long exposure noise reduction
- High ISO speed noise reduction
- Highlight tone priority (which WILL incease noise in shadow areas if set)

[Orangutan]

You would have to blow up a print at least three fold for it to be large enough for noise to exhibit like it might at 100% on-screen. The simple fact of the matter is, the VAST MAJORITY of viewing contexts require downsizing, usually significant downsizing or increases to resolution density, relative to a 100% crop view on a computer screen. The only time you are actually enlarging any viewing context beyond that scope is with prints larger than 52"x35" (for the 7D anyway), at which point image pixels in print are about the same size as on a screen. Most of the time, enlargements of such scale are done very carefully, with meticulous care around noise and sharpness, so noise is rarely as visible on screen as it is in such a large print.

The only time the levels of noise we have today in current-generation cameras (7D, 60D, 600D, even 5D II) exhibit as a problem is when viewed at 100% crop on a computer. Outside of those who tend to obsess over quality at 100% crop, viewing any photo in that way is impractical and unrealistic. @KeithR's sample photos, while they do appear to be downsized a bit, also demonstrate a REALISTIC viewing context, and thats where any photo's IS should be evaluated...in realistic context.

Jrista, I don't see what part of my post you're responding to or disputing.  It seems to me that we agree that reducing an image masks noise, particularly in light of your your later reply to someone else.  My point was that viewers can't readily compare two sample images that have been produced differently.  If one is a crop at 100%, and the other is a down-sampled, the noise will not be comparable.  Also, if one is taken at 1/5s at f/2 and the other at 1/200s at f/6.3, even if the histograms are similar, the images will not be entirely comparable.  I was specifically responding to  bikersbeard who said the 7D images looked better than those from his 5D2; however, we don't really have enough info to draw that conclusion.

If someone wants to check whether he's got a dud 7D, he/she needs to be careful about the sample images used for comparison.  This is why web sites such as imaging-resource.com (Comparometer) take such care to construct their test environments. 

[jrista]

@Orangutan: It just sounded like you were saying that downsampling for viewing on a screen was effectively different than printing from a noise management standpoint. I guess I was trying to clarify that until your blowing up a print 3-4x or more, both downsamping for viewing on a computer and printing have the same effect...they absorb noise.

[friedmud]

@jrista

Firstly, I know you weren't responding to me directly, but that doesn't stop me from commenting on your posts...

Secondly, you still used the same arguments.

Let me try again.

Your argument basically boils down to: "Prints are generally 4x6 to 13x19 and if they are bigger then your viewers won't be critically close so don't worry about it."

This is a fine, rational explanation... however it doesn't add anything to the conversation.  You could make this argument about EVERYTHING on this site:

Post Title: "1DX rumored"
Your Response:  "Prints are generally 4x6 to 13x19 and if they are bigger then your viewers won't be critically close so don't worry about it."

Post Title: "Should I upgrade from a 450D to a 5DII?"
Your Response:  "Prints are generally 4x6 to 13x19 and if they are bigger then your viewers won't be critically close so don't worry about it."

Post Title: "Will the 5DIII have 54MP?"
Your Response:  "Prints are generally 4x6 to 13x19 and if they are bigger then your viewers won't be critically close so don't worry about it."

Post Title: "My 18-85 seems a bit soft"
Your Response:  "Prints are generally 4x6 to 13x19 and if they are bigger then your viewers won't be critically close so don't worry about it."

Post Title: "This new 70-200 seems to be off a bit on the focus"
Your Response:  "Prints are generally 4x6 to 13x19 and if they are bigger then your viewers won't be critically close so don't worry about it."

Past Post Title: "Should I upgrade from a 20D to a 40D?"
Your Response:  "Prints are generally 4x6 to 13x19 and if they are bigger then your viewers won't be critically close so don't worry about it."

Future Post Title: "Is an 80D better than a 70D?"
Your Response:  "Prints are generally 4x6 to 13x19 and if they are bigger then your viewers won't be critically close so don't worry about it."

Future Post Title: "120MP 5D9 Rumored!"
Your Response:  "Prints are generally 4x6 to 13x19 and if they are bigger then your viewers won't be critically close so don't worry about it."


Yes, you have a very well reasoned mathematical basis behind your posts: but so what!  It's not adding anything to the conversation about whether or not the 7D has quite a bit of low ISO noise.  In essence, your posts are just adding "noise" to the conversation!

Whatever our motivations are for being interested in low ISO noise on the 7D... just let us talk about it without slapping us in the face with your dogma on printing / viewing sizes.

I mean, seriously... it's like you want this banner at the top of Canonrumors whole website: "Prints are generally 4x6 to 13x19 and if they are bigger then your viewers won't be critically close so don't worry about it."

To me, Canonrumors is all about pointing out existing weaknesses in Canon's cameras and fantasizing about the future of these machines.  If we can't have a real conversation about things people are actually observing in their existing cameras, why do these forums even exist?



NOW - back on topic.  I shot a bunch of tree silhouettes last night with my 7D on ISO 100 and am very happy with the output.  Unfortunately, I only have a tiny loner Macbook Air with iPhoto on it right now so it is hard to get a sense of the noise in the photos... but after a bit of sharpening and a little bit of denoise in iPhoto I feel like I got some really nice shots.

The more I use this camera the more I'm enjoying my time with it.

(Please excuse the quality of the upload... iPhoto is fairly limited, I'll give these shots a proper workup when I get back home)

[jrista]

I'm sorry, but now your putting words into my mouth, and I don't like that. Since you've made this a personal attack, here we go. I've argued consistently in THIS THREAD because I think YOU ARE WRONG about the 7D noise. I'm not the only one who thinks that. Several attempts have been made to explain why your evaluation of noise in comparison to your 450D is wrong, however are ignored with just as much DOGMA as I may present in my arguments that noise on a computer screen at or near 100% crop is IRRELEVANT!!! I think it can be DEMONSTRATED that you are WRONG, both visually (KeithR did a pretty damn good job of that...BTW...CLICK for 100% views of his shots), and mathematically. I think you are evaluating the IQ of the 7D incorrectly in an unrealistic context, and complaining about a factor that does not matter in the real world.

I'm done with this thread now, as its pointless to argue with people who refuse to embrace a little bit of real-world objectivity. I'm glad you like your 7D now, and I hope you enjoy more time in the future complaining about its noise.

[briansquibb]

I still dont think we have established whether it is the operator or the camera that is at fault.

We know that the 7D gives excellent results at low ISO - is it just this particular body that is at fault. Should it be returned to Canon?

Are the settings incorrect and therefore causing noise

Lack of information is making everyone double guess

Lets have some solid facts

[Orangutan]

I still dont think we have established whether it is the operator or the camera that is at fault.


Lets have some solid facts

Agreed.

 â€œData! Data! Data!...I can’t make bricks without clay.”  (Sherlock Holmes)

 

[Isaac]

I still dont think we have established whether it is the operator or the camera that is at fault.


Lets have some solid facts

Agreed.

 â€œData! Data! Data!...I can’t make bricks without clay.”  (Sherlock Holmes)

 

+1 we need some info before judging what the problem is.

[Maui5150]

I still dont think we have established whether it is the operator or the camera that is at fault.


Lets have some solid facts

Agreed.

 â€œData! Data! Data!...I can’t make bricks without clay.”  (Sherlock Holmes)

 

Since when did Star Trek the Next Generation encounter Sherlock Holmes??? 

And wasn't it Bones McCoy "I'm a Doctor... Not a brick layer"

All this confusing and mis-matching sci-fi has me befuddled!

[neuroanatomist]

Since when did Star Trek the Next Generation encounter Sherlock Holmes??? 
All this confusing and mis-matching sci-fi has me befuddled!

In the holodeck, of course, when Commander Data was Sherlock Holmes, and Moriarty attempted to take over the Enterprise. 

[Orangutan]

“Data! Data! Data!...I can’t make bricks without clay.”  (Sherlock Holmes)

Since when did Star Trek the Next Generation encounter Sherlock Holmes??? 

Google tells me this quote is from Before Roddenberry:

http://en.wikiquote.org/wiki/Sherlock_Holmes#The_Adventure_of_the_Copper_Beeches

I observed that he sat frequently for half and hour on end, with knitted brows and an abstracted air, but he swept the matter away with a wave of his hand when I mentioned it. “Data! Data! Data!” he cried impatiently. “I can’t make bricks without clay.” And yet he would always wind up by muttering again that no sister of his should ever have accepted such a situation.

[awinphoto]

I brought my 5d2 with me this weekend to take pics of my kids with santa and playing around on train rides and a "fun center" style thing...  I think I should start a new thread called Earthshatteringly Disappointed with 5d Mark II...  Holy crap the AF focus was horrendous even in so so light... my lord... granted when the camera did nail focus it was magical what it could do, but you almost needed to fire a few off focus shots to get the in-focus goodie... I think i'll keep the 7D strictly for everyday shots and 5d for studio... holy crap.   

[thepancakeman]

“Data! Data! Data!...I can’t make bricks without clay.”  (Sherlock Holmes)

 

Since when did Star Trek the Next Generation encounter Sherlock Holmes??? 

And wasn't it Bones McCoy "I'm a Doctor... Not a brick layer"

All this confusing and mis-matching sci-fi has me befuddled!

+1.  Almost any Star Trek reference is going to get an applause from me.    

And actually ST:TNG did have encounters with Mr. Holmes: http://en.memory-alpha.org/wiki/Sherlock_Holmes

[neuroanatomist]

I brought my 5d2 with me this weekend to take pics of my kids with santa and playing around on train rides and a "fun center" style thing...  I think I should start a new thread called Earthshatteringly Disappointed with 5d Mark II...  Holy crap the AF focus was horrendous even in so so light... my lord... granted when the camera did nail focus it was magical what it could do, but you almost needed to fire a few off focus shots to get the in-focus goodie... I think i'll keep the 7D strictly for everyday shots and 5d for studio... holy crap.

<sarcasm>It must be operator error.  Briansquib shoots motorsports with his 5DII and the AF performs flawlessly, tracking fast-moving carts around curves, even moving toward and and away from the camera.  If you can't even shoot a running kid, don't blame the camera.</sarcasm>

Yes, I know from experience that the 5DII's AF is pretty bad with anything moving.  That's the main reason I plan to get a 1D X as soon as they're available.

[Orangutan]

xxxx shoots motorsports with his 5DII....Yes, I know from experience that the 5DII's AF is pretty bad with anything moving.  That's the main reason I plan to get a 1D X as soon as they're available.

Kids are particularly difficult.  Even though moving cars are faster, they're predictable and have nice straight lines to focus on.  Moving kids, on the other hand, are little more than blur to begin with: not much to lock focus on.