7D mk II image quality RAW worse than 7D for now

2n10 wrote in post #17234763
Strange none the less that it started with the more recent models if Gabe is right.

Remember the 6D was tested just 9 months before the 70D. I don't think it is a time thing or a fading thing but rather a deliberate choice made by Canon as to how the red channel looks and perhaps it has to do with noise as well. I am just guessing. This is why I think it is going to show up in the 7DmkII production model too. It is the flavor that Canon wants to offer and IMO it is wrong. Time will tell. I don't have $1800 burning a hole in my pocket.

I don't know if it helps but this is how the 70D red cloth looks in ACR, default sharpening, WB unchanged, NR zeroed out.
Just barely but you can see the threads of the red cloth.

Content warning: NOT AN IMAGE

And this the same via DPP, neutral P style, NR 0, sharpness 3

You cannot see the threads here. In "Standard" P style it's even worse.

If you're not happy, switch to Nikon. Canon is not a strong contender for clean images. Great sports cameras, impressive glass lineup, good cinema support. However, their sensors absolutely suck compared to competitors.

To demonstrate the change of the red cloth over the years:
(All RAW, ISO 100, processed in ACR, 0 NR, default sharpening, same for all.)

Nikon D300, 2007

A couple more examples, from Canon, only 4 years apart.

5DII, 2008:

I hope the above examples help, at least to see the progression of the red cloth. Lighting has also changed somewhat over the years as IR stated for a while, BUT the rest of the image looks very similar when you compare an old one to a new one. Mostly it's the textiles that faded over the years, especially, the red leafy cloth.

Great work Gabe. Certainly looks like degradation of the swatch to me based on the timing. I'll know more when my 7D2 arrives, hopefully on Friday, and I find some nice cherry red cars to shoot or some other bright red items in direct sunlight.

I'll keep this in mind when I take photos of red cloth.

gabebalazs wrote in post #17234914
I hope the above examples help, at least to see the progression of the red cloth. Lighting has also changed somewhat over the years as IR stated for a while, BUT the rest of the image looks very similar when you compare an old one to a new one. Mostly it's the textiles that faded over the years, especially, the red leafy cloth.

It is good work and deduction Gabby, it could be fade but wth those jacard type cloths the thread contrast may be strongly affected by lighting changes too.

'I'll keep this in mind when I take photos of red cloth.' HA! Love it

AllenF wrote in post #17234725
OK now apply that same method to the 7DmkII image and then look at and compare ALL of the fabric samples with the 6D sample. Area for area. This will tell you if you tweeked the image to look the same in one place all the while it has messed up the rest of the image. If all the rest is fine then you have shown that the difference I have seen is mostly due to how the scene was exposed and PP can fix it. If on the other hand the image is right in one area and messed up in others then the fault is real and we must wait for more samples from IR of the production 7DmkII. Then we must compare those to the 6D and see how well it compares.

You were able to get them to PP well and I applaud your effort. This is what I was hoping to have happen. Not just speculation but real work with visual results. Thank you again.
Sadly you have some more testing/comparing to do but I bet you can do it in 5 mins time.

I suspect it will not work on the rest of the image as I could not get it to work for me but I did not do an intense effort as it was quickly seen to not work for me. For example darkening the leaf pattern detail to make it show up better caused a loss of the light swirling detail in the black fabric pattern. Also look at the very light white fabric and see if the embossed leaf pattern is over accentuated or blown out. These two will establish the limits to white and black levels. It may be acceptable to give up one or the other,however the 7D does not need to make this compromising choice as it renders them all correctly with no visible loss.

Once again great job!

It's an interesting issue. As noted above, when I compared the IR red leafy cloth sample between the 6D and 7D2 with identical processing, I noticed that the 7D2 cloth looked worse in some areas, but was essentially identical to the 6D cloth in other areas, as shown by the sample I posted above (where the 7D2 is at least as good, if not better, than the 6D sample in terms of leaf detail). That, plus Gabe's series of samples above, convinces me that this is not a camera issue, it's an IR issue with their sample cloth and/or lighting.

My preliminary conclusions, after looking at various samples, including RAWs that I converted myself, are that:
1) The 7D2 has superior image quality to the 7D, although processing plays a big part in how much better, as demonstrated by Teamspeed (the 7D2 noise appears to be finer-grained and is probably more easily cleaned up than 7D noise);
2) The 7D2 has approximately equivalent high-ISO noise to the 1D4;
3) The 7D2 is very close to the 5D2 in high-ISO noise, probably less than half a stop behind if not even closer;
4) The 7D2 is about a stop behind the 5D3 in high-ISO noise (7D2 ISO6400 roughly comparable to 5D3 ISO12800).
Personally, I find this very impressive for a 1.6 crop sensor camera.

stsva wrote in post #17235238
It's an interesting issue. As noted above, when I compared the IR red leafy cloth sample between the 6D and 7D2 with identical processing, I noticed that the 7D2 cloth looked worse in some areas, but was essentially identical to the 6D cloth in other areas, as shown by the sample I posted above (where the 7D2 is at least as good, if not better, than the 6D sample in terms of leaf detail). That, plus Gabe's series of samples above, convinces me that this is not a camera issue, it's an IR issue with their sample cloth and/or lighting.

My preliminary conclusions, after looking at various samples, including RAWs that I converted myself, are that:
1) The 7D2 has superior image quality to the 7D, although processing plays a big part in how much better, as demonstrated by Teamspeed (the 7D2 noise appears to be finer-grained and is probably more easily cleaned up than 7D noise);
2) The 7D2 has approximately equivalent high-ISO noise to the 1D4;
3) The 7D2 is very close to the 5D2 in high-ISO noise, probably less than half a stop behind if not even closer;
4) The 7D2 is about a stop behind the 5D3 in high-ISO noise (7D2 ISO6400 roughly comparable to 5D3 ISO12800).
Personally, I find this very impressive for a 1.6 crop sensor camera.

Your conclusions match mine as well after playing with all the files. I think the 7dii is going to be a very successful camera. It is looking more and more like I will add this body alongside my 1D4. The final image quality is always a result of camera body, lens and PP skill. If any of the 3 are lacking you will not get the best possible result.

butterfly2937 wrote in post #17235341
Your conclusions match mine as well after playing with all the files. I think the 7dii is going to be a very successful camera. It is looking more and more like I will add this body alongside my 1D4. The final image quality is always a result of camera body, lens and PP skill. If any of the 3 are lacking you will not get the best possible result.

So, photography skill is not involved? But when you get your, I'll hope to hear your thought about how it compared to 1D4 in the real world usage, ie: build, ergonomic, AF, tracking and everything that separate the 1D from 1Ds series. It's sad to say, but many review sites now a day are too focus on the final image quality and ignore 90% of what happened before the image hit the sensor.

JimmyG wrote in post #17234859
If you're not happy, switch to Nikon. Canon is not a strong contender for clean images....

I KNEW it.... this entire fake controversy was started by the Nikon marketing department.