7D Mk II Image Quality: Is it as bad as DxoMark says?

[goodguy]

Continuing my post from yesteray

 

[hamlet]

I think it's a hit and miss with most cameras. My d3200 handled underexposed areas better than my d7100 in my experience. Banding Noise creeps in fast.

 

[goodguy]

From looking at Tony's second more thorough video I learn one thing which I already implemented, if you really want to get good low light performance go full frame

 

[runnah]

Dx0 is a bit biased towards nikon.

 

[snerd]

I would be buying the 7DII not for low light, but for daytime sports and wildlife. General daylight outdoor stuff. Perfect compliment to a 5DIII, I think.

 

[goodguy]

Dx0 is a bit biased towards nikon.

I doubt that very much, I think even Tony after correcting the NR on the software saw his initial assessment of the low light performance were incorrect.
Canon got very close to the average APS-C sensor in the market today when it comes to low light but its definitely not leading the market and that without taking into account the other parameters where is still far behind like DR.
Overall its a good sensor and the 7D II is a good camera for what it was designed to do but Canon needs to work a bit harder to reach the leading sensors in the market.
BTW the Nikon bias you suspect DXO has are not really Nikon sensors they are Sony sensors.

 

[mcap1972]

Is it true Canon is using Sony sensors for this camera?

 

[goodguy]

Is it true Canon is using Sony sensors for this camera?

Nope, its Canon's

 

[MarshallG]

Here's the question I have about DxOmark: Where are the photographs? Where is the difference that my eyes can see? They never show your eyes how the 87 camera produces a visibly better result than the 85, but people buy based on their numbers anyway.

Dpreview.com has an outstanding Studio Comparison Tool, which lets you see the side-by-side differences between four cameras at a time. And your eyes will tell you a different story than DxO. See for yourself.

 

[Derrel]

Look at the images from the New York City Marathon, shot with the new 7D-II. OMG....open shade and small bits of sunlight = unrecoverably blown-out highlights. WEAK color saturation. Look at on-line bird photography...again....weak color, narrow dynamic range. The 7D-II has been optimized for shooting FAST, at up to 10 frames per second, with a deep buffer, with a smallish file size (20.2 MP, as opposed to say....a 36 MP Nikon) on a 1.6x, smallish crop-body size frame, and to come in at $1799 at intro, with the world's highest-spec'd AF system.

Again, LOOK at the pictures on Flickr. You can SEE the difference between 7D-II images and those made by a Canon 6D or 5D-III...the 7D-III has limitations in image technical quality, but it's designed to, "Get the pictures, FAST, and IN-FOCUS, across long shooting sequences, and at elevated ISO values if needed."

It's pretty ridiculous to try and infer that the 7D-III has image quality that surpasses cameras with bigger sensors, with higher MP count, and better sensor technology. If you want a 7D-II for what you want to shoot, then get one. But please, stop trying to argue with what is obvious: this camera's sensor is nowhere near what state of the art 2014 is. This camera is not designed for the highest technical image quality: it's designed to shoot a LOT of in-focus frames FAST, across a deeeeep buffer, cheaply, not at $6999, but at $1799.

 

[TCampbell]

Look at the images from the New York City Marathon, shot with the new 7D-II. OMG....open shade and small bits of sunlight = unrecoverably blown-out highlights. WEAK color saturation. Look at on-line bird photography...again....weak color, narrow dynamic range.

When you think about the science behind how a color sensor works, there's really no such thing as "weak" color. This is ENTIRELY a processing choice. Technically color exists only in our heads. Light has "wavelengths" and it's our brain that translates those wavelengths to color. The sensors themselves register signal. Since most color sensors employ a Bayer mask, they register the "green" signal, "red" signal, and "blue" signal and all of this is de-bayered to create a resulting "color" for that pixel. The process behind this is entirely in software ... there's not much to the chip design.

If you like stronger color saturation... just dial it up.

 

[Derrel]

HERE is an official sample from the 7D Mark II, shot in natural, full-spectrum daylight--not under spectrally deficient light at some stadium. Shot at 3,200 ISO with the very expensive Canon 200-400mm f/4 L IS USM lens that costs $11,799 at B&H Photo.

http://canon-premium.webcdn.stream.ne.jp/www09/canon-premium/eosd/samples/eos7dmk2/downloads/07.jpg

Color saturation? LOW.

Detail rendition? LOW. Shockingly lacking on fine detail.

This image looks like s**+. It's pathetic. THIS is daylight, 3,200 ISO with a nearly $12,000 Canon uber-lens? No amount of work can save this crap image.

And sorry, but there ***is*** such a thing as weak color. Are you seriously so brand-loyal that you are going to argue that all color is equally "strong" at each ISO level, or that there are no differences in degrees of color saturation? HILARIOUS, and flat-out incorrect.

Seriously, are you trying to go there in public? You need to stop drinking the Canon Kool-Aid.

 

[runnah]

So people are getting upset that a camera that is billed as a blazing fast sports oriented camera isn't as good as one that is designed for studio work and costs almost twice as much?

 

[Derrel]

Here's a good example of the 7D Mark II's lack of detail. I would call this, "mush-level fine detail".

An "Official Sample", hosted by Canon. Low noise? ehhh.

"Weak color" too; see how the color is almost absent? One can "crank the saturation to MAX," as TCampbell suggested one do, and this will still look like crap.

http://canon-premium.webcdn.stream.ne.jp/www09/canon-premium/eosd/samples/eos7dmk2/downloads/06.jpg

And we have people here deriding DxO Mark's test results from this sensor, trying to discredit DxO mark's rigorous tests, which are applied the same way for all cameras, in a million-dollar testing lab? Download the two photos from Canon and actually ZOOM in on them! And then try your hardest to mentally make the DxO mark score go wayyyyyyyy up!

 

[snerd]

Derrel, we get it. You hate this camera. But, dude, what's with the literal hate? Every thread, you come in swinging! You okay?