Stock Notice: Canon RF 15-35mm f/2.8L IS USM

Canon Rumors Guy

Canon EOS 40D
CR Pro
Jul 20, 2010
10,779
3,158
Canada
www.canonrumors.com
Looking at the image quality over at the digital picture, it doesn't seem to be quite up to the image quality of the EF16-35 f2.8L III. I'm a little disappointed, as I would have expected the image quality to be as good or better ... and there is no image stabilization. My EF 16-35 f4L is heavily used and stabilization often makes a difference in outdoor photography. I'm surprised there is no stabilization on this new RF lens. Hopefully, the next R-series camera has IBIS.
 
Last edited:
Upvote 0
Looking at the image quality over at the digital picture, it doesn't seem to be quite up to the image quality of the EF16-35 f2.8L III. I'm a little disappointed, as I would have expected the image quality to be as good or better ... and there is no image stabilization. My EF 16-35 f4L is heavily used and stabilization often makes a difference in outdoor photography. I'm surprised there is no stabilization on this new RF lens. Hopefully, the next R-series camera has IBIS.
If you’re using their lens comparison tool, you have to keep in mind the camera the images were shot on. The EF 16-35 was tested on the 5DsR, while the RF 15-35 (IS, btw) was tested on the EOS R. Not the same sensor at all.
 
  • Like
Reactions: 1 user
Upvote 0

Ozarker

Love, joy, and peace to all of good will.
CR Pro
Jan 28, 2015
5,933
4,336
The Ozarks
If you’re using their lens comparison tool, you have to keep in mind the camera the images were shot on. The EF 16-35 was tested on the 5DsR, while the RF 15-35 (IS, btw) was tested on the EOS R. Not the same sensor at all.
They should both get tested on the R. Strange that digital picture would make such a comparison on those separate cameras. Recycled review on the EF model, I guess.
 
  • Like
Reactions: 1 user
Upvote 0
They should both get tested on the R. Strange that digital picture would make such a comparison on those separate cameras. Recycled review on the EF model, I guess.

They let you choose which camera to use in combination with the lens for the comparison, however, they don't necessarily have all camera and lens combinations in their archive. If they were to run through the tests with every single camera and every single lens for each release, it would take them an extraordinary amount of time.

In this case, the EF lens was actually never tested on the 5Div (which would be the same sensor and AA filter as the EOS R), so unfortunately they don't have a useful tool for comparison at the moment.

It's important to note, that The Digital Picture did not themselves indicate this was an appropriate comparison to make. KirkD simply selected these settings using their available comparison tool on the website, and came to a misleading conclusion.
 
  • Like
Reactions: 1 user
Upvote 0

Ozarker

Love, joy, and peace to all of good will.
CR Pro
Jan 28, 2015
5,933
4,336
The Ozarks
They let you choose which camera to use in combination with the lens for the comparison, however, they don't necessarily have all camera and lens combinations in their archive. If they were to run through the tests with every single camera and every single lens for each release, it would take them an extraordinary amount of time.

In this case, the EF lens was actually never tested on the 5Div (which would be the same sensor and AA filter as the EOS R), so unfortunately they don't have a useful tool for comparison at the moment.

It's important to note, that The Digital Picture did not themselves indicate this was an appropriate comparison to make. KirkD simply selected these settings using their available comparison tool on the website, and came to a misleading conclusion.
Very true.
 
Upvote 0
But the EF lens was on a 50 Mp sensor and the RF on a 30 Mp sensor. This means that if the RF was tested on the 50Mp sensor it would look even worse. So an apples to apples comparison makes the RF version less optically impressive.

It doesn't quite work that way, and if you extrapolate out you'll quickly see why. In other words, would an image taken on a 10Mp sensor look the best? How about a 1Mp sensor?

The reality is that perceived sharpness is a combined result of sensor resolution (the higher the Mp count, the more individual data points to build the image from) and the resolving power of the lens (the more precisely the optics can focus light, the more distinct the reading from each photo-site on the sensor will be).

And in the case of these two cameras, there is also the issue of the Anti-Aliasing filter on the EOS R. AA filters very slightly blur an image in a way that is designed to reduce moire. The 5DsR has no AA filter (that's why is has the "R" in the name, for increased Resolution).

The disadvantages of higher Mp sensors (given the same sensor size) is in the additional noise that results from having a smaller photo-site per pixel, as well as a higher data rate to handle which can slow down shooting. But at low ISOs, such as in the test conditions used by TDP, the 5DsR simply has a superior sensor to the EOS R, and will produce much more detailed images.
 
  • Like
Reactions: 1 user
Upvote 0
I can see I wasn't clear enough in how TDP shows their results, in my original comments. They do not use the same image magnification for the different sensors they test. The resolution of a lens has nothing to do with the Mp of the sensor. But on TDP, their results use a higher image magnification for the 50 Mp sensors than they use for lower Mp sensors. So if TDP's image magnification for a 30 Mp sensor shows that the sensor is already outperforming the resolution of a lens, the higher image magnification they use for a 50 Mp sensor will magnify the resolution limitations of that lens. So if the magnification they use for a 30 Mp sensor shows a lack of sharpness, then the higher image magnification they use for a 50 Mp sensor will magnify the lack of sharpness we already see on the 30 Mp magnification they use on TDP. So when I do an image comparison on TDP for the RF and EF versions of the 15-35/16-35 f2.8L lenses, they use two different image magnifications. So if the higher magnification they use for the EF results looks about as sharp as the lower magnification they use for the R results, then I would expect that the EF results should look even sharper if the test image is reduced to the same size as the R results. Having said all this, I did not take into consideration the Anti-aliasing filter on the EOS R mentioned by C. Judge. My concern was that if the 30 Mp sensor on the R is already outperforming the RF lens, then magnifying the image results as they do for the 50 Mp sensor, (so that I could compare the RF lens with the EF lens results) is going to magnify the resolution limitations of the RF lens even more for the simple reason that the 50 Mp test images are magnified more than the 30 Mp test images.
 
Upvote 0
I can see I wasn't clear enough in how TDP shows their results, in my original comments. They do not use the same image magnification for the different sensors they test. The resolution of a lens has nothing to do with the Mp of the sensor. But on TDP, their results use a higher image magnification for the 50 Mp sensors than they use for lower Mp sensors. So if TDP's image magnification for a 30 Mp sensor shows that the sensor is already outperforming the resolution of a lens, the higher image magnification they use for a 50 Mp sensor will magnify the resolution limitations of that lens. So if the magnification they use for a 30 Mp sensor shows a lack of sharpness, then the higher image magnification they use for a 50 Mp sensor will magnify the lack of sharpness we already see on the 30 Mp magnification they use on TDP. So when I do an image comparison on TDP for the RF and EF versions of the 15-35/16-35 f2.8L lenses, they use two different image magnifications. So if the higher magnification they use for the EF results looks about as sharp as the lower magnification they use for the R results, then I would expect that the EF results should look even sharper if the test image is reduced to the same size as the R results. Having said all this, I did not take into consideration the Anti-aliasing filter on the EOS R mentioned by C. Judge. My concern was that if the 30 Mp sensor on the R is already outperforming the RF lens, then magnifying the image results as they do for the 50 Mp sensor, (so that I could compare the RF lens with the EF lens results) is going to magnify the resolution limitations of the RF lens even more for the simple reason that the 50 Mp test images are magnified more than the 30 Mp test images.
I would recommend using TDP’s tool to compare the EF24-70 2.8 ii on the Eos R vs the same lens on the 5DsR. You’ll see very similar results.

Increased sensor resolution does not simply “magnify” the image. If the image consists of black on white lines, any further sensor resolution would definitely show increased detail.
 
  • Like
Reactions: 1 user
Upvote 0
May 11, 2017
1,365
635
I can see I wasn't clear enough in how TDP shows their results, in my original comments. They do not use the same image magnification for the different sensors they test. The resolution of a lens has nothing to do with the Mp of the sensor. But on TDP, their results use a higher image magnification for the 50 Mp sensors than they use for lower Mp sensors. So if TDP's image magnification for a 30 Mp sensor shows that the sensor is already outperforming the resolution of a lens, the higher image magnification they use for a 50 Mp sensor will magnify the resolution limitations of that lens. So if the magnification they use for a 30 Mp sensor shows a lack of sharpness, then the higher image magnification they use for a 50 Mp sensor will magnify the lack of sharpness we already see on the 30 Mp magnification they use on TDP. So when I do an image comparison on TDP for the RF and EF versions of the 15-35/16-35 f2.8L lenses, they use two different image magnifications. So if the higher magnification they use for the EF results looks about as sharp as the lower magnification they use for the R results, then I would expect that the EF results should look even sharper if the test image is reduced to the same size as the R results. Having said all this, I did not take into consideration the Anti-aliasing filter on the EOS R mentioned by C. Judge. My concern was that if the 30 Mp sensor on the R is already outperforming the RF lens, then magnifying the image results as they do for the 50 Mp sensor, (so that I could compare the RF lens with the EF lens results) is going to magnify the resolution limitations of the RF lens even more for the simple reason that the 50 Mp test images are magnified more than the 30 Mp test images.
you have data on the resolution of the R with the RF 15-35 mounted and data on the resolution of the 5Dsr with the EF 16-35 mounted. There is no way from this data that you can compare the resolution of the two lenses because the sensors are different. With the 5Dsr, the 16-35 has higher resolution than it would on the 5DIV, which has a sensor equivalent to the R.
 
  • Like
Reactions: 1 user
Upvote 0