Compare 5D2 vs 5D3 vs 7D dark noise shots, 100% crops online w +4 EV shadow push

[ew20]

No offense, but a camera company could make a sensor that would win this test quite handily, and suck when it comes to real-life shooting.

Under real conditions, the noise performance of the 5D3 has surpassed the other two bodies, AINEC. I'm getting images at 12,800 that aren't just usable, but good. I guess it's fun to test the limits of equipment, but tests like these should in no way impact whether you buy this camera or not, unless your work requires routinely pushing your exposures ridiculous amounts in post.

 

[Aglet]

Re: Compare 5D2 vs 5D3 vs 7D dark noise shots, 100% crops online w +4 EV shadow

Thanks for the comparison. It clearly shows my main gripe with the 7D: Lots of shadow noise even at ISO 100. It's almost impossible to get something useful out of the darker areas of an image because of all the noise. From my experience, this wasn't as much of a problem with older Canon APS-C Sensors.

That's what ground my work to a halt, all the gross banding in 100, 200 and 400 ISO on the 7D.
I'll be putting up more pages with more comparisons, including 40/50/60/450/1000D bodies, none of which have the 8 pixel wide stripes seen on the 7D, likely from the dual-channel sensor readout electronics mismatch.

 

[Aglet]

Re: Compare 5D2 vs 5D3 vs 7D dark noise shots, 100% crops online w +4 EV shadow

That's a really interesting test. Is there any reason to shoot with a lens attached? Why not with just the body cap on? Also, in ACR, did you zero the brightness and contrast sliders or did you leave them at the default 50/20?

I left the lens caps on so as not to bother swapping lenses on other people's cameras and risk adding dust to their sensors (another gripe for another time ).

Yes, thanks for confirming; ACR brightness and contrast were left at default 50/20 or 50/25, whichever it is, i can't remember and I'm not in my studio to check for another 8 hours.

 

[Aglet]

Re: Compare 5D2 vs 5D3 vs 7D dark noise shots, 100% crops online w +4 EV shadow

Hi Aglet, it'd be great if you took crops of an area corresponding 5% of the image to get relative noise performance of the whole sensor, not on pixel level. Of course if you didn't do that in your test

I've got something like it coming on per-camera sample pages later, I hope you'll find it useful.

Same files used for the noise crops are scaled 1/8 linearly so you can get an overall noise map of sorts.
some cameras generate a lot of plaid and one 5D2 was rather uneven.

 

[briansquibb]

Why use ACR when you have the RAW converter from Canon that is able to do justice to the RAW files?

 

[sach100]

No, it's not. It's just that the completely random noise in both shots happens to match up perfectly. It's like that one monkey with a typewriter among an infinite number of them that just happens to bang out the complete works of Shakespeare.

Actually it is, they're both "7Diso3200c.jpg", there is no "5d3iso3200c.jpg"

Silly me, did I forget the <sarcasm> tags?

Well Spotted sir..

 

[Aglet]

Re: Compare 5D2 vs 5D3 vs 7D dark noise shots, 100% crops online w +4 EV shadow

Somewhat related so I'm not starting a new thread.

http://www.dpreview.com/previews/canoneos5dmarkiii/12

I can't save it in the link but I switched the Sony in the lower right to the canon 7D. That makes it a comparison between the D800 - 5DIII, 5DIII and 7D (all for RAW shooting).

I gotta be honest, as far as this DPReview test alone goes, I can't see any real-world difference between any of the cameras. There is nothing about any of the minute differences at the same settings (doesn't really matter which settings either) and for real world usage they all seem nearly the same.

for the most part they ARE very close and very impressive overall.
Previously I posted a topic that describes where to look to see shadow noise of these cameras without having to push any shadows at all.

www.canonrumors.com/forum/index.php?topic=5101.msg99797#msg99797

This is pixel-level noise you won't see in small prints and even in larger prints at effective 75ppi it'd still be difficult to notice unless the noise shows a pattern and is on a large smooth-tone area of image with no other subject texture to hide it.

 

[Aglet]

Re: Compare 5D2 vs 5D3 vs 7D dark noise shots, 100% crops online w +4 EV shadow

I appreciate the the test an the effort. I believe the 5DIII 3200 ISO shot is the same as the 7D.

It tripped me out a little. I couldn't figure out why 3200 was so bad on the 5DIII the I looked at the 7D.

DOH! Nice catch. Thank-you. I'll fix that tonite.
I was trying hard not to make a mistake while exhausted and with a head cold ( X-6 )but I guess it still crept in there.
If you want to find the correct file before then it is

http://a2bart.com/techimg/dknz/5d3iso3200c.jpg

all individual files at this time still include EXIF data

 

[Aglet]

Re: Compare 5D2 vs 5D3 vs 7D dark noise shots, 100% crops online w +4 EV shadow

Why use ACR when you have the RAW converter from Canon that is able to do justice to the RAW files?

I like and use DPP for much of my work, it's a great and underrated bit of software.
I don't have the version that supports the 5D3 yet and I also wanted to level the field as much as possible by using the same workflow with other brands like Nikon, Panasonic, Pentax, Sony, etc.

 

[Alker]

Nice test, but I prefer some real world examples.

These Lens Cap + 4 images ...well what does it say ?

 

[Alker]

Some more real world images review........

Canon 5D MK III vs Nikon D800 with Nathan Elson

 

[Aglet]

Re: Compare 5D2 vs 5D3 vs 7D dark noise shots, 100% crops online w +4 EV shadow

Nice test, but I prefer some real world examples.

These Lens Cap + 4 images ...well what does it say ?

Like the topic intro, this is just a teaser.
Real world examples will be prepped and uploaded later.
Thanks to my 7D, i have a few to choose from.

what these images tell you can be quite useful if you have to push any of your raw files to recover dark area/shadow data.

I wish I had this kind of info 2 years ago! it would have saved me hundreds of shots with my 7D that I now can't use the way I want because I can't get the quality level I demand. I would have known to use a different body for those shots.


if you shoot for a living or are trying to get a once-in-a-lifetime shot (frankly, they're ALL 1/lifetime shots) then this could save your bacon.

 

[japhoto]

Re: Compare 5D2 vs 5D3 vs 7D dark noise shots, 100% crops online w +4 EV shadow

Thanks for the comparison. It clearly shows my main gripe with the 7D: Lots of shadow noise even at ISO 100. It's almost impossible to get something useful out of the darker areas of an image because of all the noise. From my experience, this wasn't as much of a problem with older Canon APS-C Sensors.

That's what ground my work to a halt, all the gross banding in 100, 200 and 400 ISO on the 7D.
I'll be putting up more pages with more comparisons, including 40/50/60/450/1000D bodies, none of which have the 8 pixel wide stripes seen on the 7D, likely from the dual-channel sensor readout electronics mismatch.

Don't know what causes this, but it's annoying as f***!

The exposure doesn't even have to be "way off", but a little push to bring up the shadows and there it is. The noise itself isn't that bad (not good either), but the banding is an issue for sure.

I'm probably going to move forward (or timewise backwards) and try to find a 1Ds MkII and a 1D MkIII for faster stuff. I live between ISO 100-800 and most of the time 100-400, so I'd like those values to be clean.

 

[Alker]

Re: Compare 5D2 vs 5D3 vs 7D dark noise shots, 100% crops online w +4 EV shadow

Thanks for the comparison. It clearly shows my main gripe with the 7D: Lots of shadow noise even at ISO 100. It's almost impossible to get something useful out of the darker areas of an image because of all the noise. From my experience, this wasn't as much of a problem with older Canon APS-C Sensors.

That's what ground my work to a halt, all the gross banding in 100, 200 and 400 ISO on the 7D.
I'll be putting up more pages with more comparisons, including 40/50/60/450/1000D bodies, none of which have the 8 pixel wide stripes seen on the 7D, likely from the dual-channel sensor readout electronics mismatch.

Don't know what causes this, but it's annoying as f***!

The exposure doesn't even have to be "way off", but a little push to bring up the shadows and there it is. The noise itself isn't that bad (not good either), but the banding is an issue for sure.

I'm probably going to move forward (or timewise backwards) and try to find a 1Ds MkII and a 1D MkIII for faster stuff. I live between ISO 100-800 and most of the time 100-400, so I'd like those values to be clean.

Just to be sure.
The 5D mark III is not giving you clean images at ISO (100-400)
Have you tried it ?
Or will these test images tell you enough ?

 

[japhoto]

Re: Compare 5D2 vs 5D3 vs 7D dark noise shots, 100% crops online w +4 EV shadow

Thanks for the comparison. It clearly shows my main gripe with the 7D: Lots of shadow noise even at ISO 100. It's almost impossible to get something useful out of the darker areas of an image because of all the noise. From my experience, this wasn't as much of a problem with older Canon APS-C Sensors.

That's what ground my work to a halt, all the gross banding in 100, 200 and 400 ISO on the 7D.
I'll be putting up more pages with more comparisons, including 40/50/60/450/1000D bodies, none of which have the 8 pixel wide stripes seen on the 7D, likely from the dual-channel sensor readout electronics mismatch.

Don't know what causes this, but it's annoying as f***!

The exposure doesn't even have to be "way off", but a little push to bring up the shadows and there it is. The noise itself isn't that bad (not good either), but the banding is an issue for sure.

I'm probably going to move forward (or timewise backwards) and try to find a 1Ds MkII and a 1D MkIII for faster stuff. I live between ISO 100-800 and most of the time 100-400, so I'd like those values to be clean.

Just to be sure.
The 5D mark III is not giving you clean images at ISO (100-400)
Have you tried it ?
Or will these test images tell you enough ?

Hey Alker,

No, not talking about the 5D MkIII here. I actually thought that they weren't available here yet, but apparently they have started shipping to customers here in Finland. With a hefty price tag attached to them though (3799€).

The 5DIII is the first 5D camera I could be interested in, but at the moment it's out of my reach at that price. That said, even the 5Dc and 5DII are probably quite good between ISO 100-400.

The thing these test images (actually looked pretty much only the 7D results) "prove" is what I've known all along (poor low ISO performance) and yes, I've done a similar test with my own 7D as well. Also you don't need a test to figure this one out. Just a high contrast scene is enough where you have to underexpose a tad to preserve the highlights and then try bringing the shadows back in post. Not pretty.

I also forgot to add that if the Nikon 5100 results are right and the camera doesn't do aggressive noise reduction in-body, I'm one impressed Canon shooter...

 

[Alker]

@japhoto.

Oke thanks for your reply.

Strange how all the prices in other countries are so different.
Here in the Netherlands the 5D mark III is 3499 euro.

The 5D Mark II will cost you 1700 euro.

 

[japhoto]

@japhoto.

Oke thanks for your reply.

Strange how all the prices in other countries are so different.
Here in the Netherlands the 5D mark III is 3499 euro.

The 5D Mark II will cost you 1700 euro.

I'm not surprised at all that Finnish customers have to pay a premium. Feels like that with pretty much everything around here

The 5D MkII is 1799€ body only at that same store, so not that much different, but I think they are the first ones to carry the MkIII, so they are riding that and making a better profit with the first wave of eager customers.

As an addition I found interesting that the biggest online store values the 5DIII at 3299€ and the Nikon D800 and D800E at 2699,90€. No price difference between the Nikons and a good price with the 5DIII, but I don't think they are going to ship them out any time soon though...