Bryan Carnathan has posted test shots:

http://www.the-digital-picture.com/Reviews/ISO-12233-Sample-Crops.aspx?Lens=487

From this it appears the 17 isn't quite as impressive as the 24. Furthermore a negative shift seems to result in softer images than a positive one.

FWIW, my experience has been markedly different as I find it equally pin sharp in either direction of shift.

Center sharpness unshifted is also not as impressive as the 24. Once again, this hasn't been my experience with the lens. I'm curious what other 17 TS-E owners think.

That center softness at f/8 just doesn't look right. Reminds me of the first 24/1.4 II test. Did they forget to turn off NR again?

alundeb wrote:
That center softness at f/8 just doesn't look right. Reminds me of the first 24/1.4 II test. Did they forget to turn off NR again?


Yea that was my thought at first as well but two things:

1. It doesn't explain the poorer corner performance
2. It doesn't explain why the corners are softer when shifted to one side vs. the other

Oh and I also figured they'd be be more careful with RAW settings after the 24L f/1.4 II test.

QA issues I guess?

ILOVECANONL wrote:
QA issues I guess?

Man if thats true, that really blows... seriously on a freakin' $2.5k lens...

thedigitalbean wrote:
[2. It doesn't explain why the corners are softer when shifted to one side vs. the other



Strange also that the +12mm shift, edge crops have virtually no light fall-off.
The +/-6 mm shift crops are much more symmetrical in that respect.
Hov can there be less light fall-off at +12mm than at +6mm?

Edit: This appears only in the f/4 crops.

ILOVECANONL wrote:
QA issues I guess?

I hope you're wrong. Let's wait for the test itself.

Happy shooting,
Yakim.

Could it be as simple as the "corner" section of the chart is at the bottom and therefore the negative shift puts it at the very edge of the image circle and the positive shift puts it more toward the center?

steeeven wrote:
Could it be as simple as the "corner" section of the chart is at the bottom and therefore the negative shift puts it at the very edge of the image circle and the positive shift puts it more toward the center?

If I understand you correctly, I don't think so as I think they reset the camera position such that the chart precisely covers the captured image.

alundeb, good eye on the vignetting difference between 6 and 12. I can't fathom how there could be less corner vignetting at 12 vs. 6.

Sadly I can't offer any quantitative information since I don't shoot test charts with new lenses, rather opting to just go out and use them to get a feel for them.

I just wonder how realistic a test chart test is, with what is likely a very short working distance and the inevitable difficulty in ensuring everything is perfectly lined up, compared to the real world normal working distance results we've already seen from a number of photographers.

I'd be truly surprised if the 17 is equal to the new 24, however it still looks good compared to the 16-35, 14, etc.

thedigitalbean wrote:
steeeven wrote:
Could it be as simple as the "corner" section of the chart is at the bottom and therefore the negative shift puts it at the very edge of the image circle and the positive shift puts it more toward the center?

If I understand you correctly, I don't think so as I think they reset the camera position such that the chart precisely covers the captured image.

Intuitively, I think steeeven is into something.

I have both the 17 and the 24mm, and in my case I would say that the 17mm is easily as good as the the 24mm. It is mind-blowingly sharp...