Canon 35mm f/2 IS vs Sigma 35mm f/1.4 Art bokeh comparison

kotmj wrote:

Peter Karbe of Leica said in an interview on the 50/2 M APO that what they tried to achieve with it that was different from most manufacturers is that the MTF needs to fall off steeply away from the plane of focus. That is, the object in focus has great macro and microcontrast -- very high and level MTF curves -- but that the objects infront of and behind the plane of focus have intentionally bad MTF curves. The out-of-focus areas have low contrast and resolution. This difference in contrast between the object in focus and those OOF is what makes for the pop.

He further said the Japanese would never think of doing this.

Leica has had different philosophies in this regard over the decades. Mandler had a piece published where he stated that in what he called "pictorial" photography, even the OOF areas should have high contrast.

Maybe Sigma caught on to Karbe's thinking, or arrived at it on their own.

...the MTF curves measure resolution on the plane at the focal point that runs perpendicular to the line of sight.  If the MTF curves show lower resolution from the center, this could mean the lens sharpness drops off as you describe above (which is a good thing, in my opinion, for central composition wide open) or it could indicate strong field curvature.

Either way, the MTF curves should flatten out as one stops down, becoming sharper across the frame and at top form by f/4 (f/5.6).  The Sigma 35 / 1.4A is a wonderful example of such a curve, although one might argue that lower resolution at the wider apertures in the periphery would have been more desirable.  The 35 / 2 IS, however, would have benefited from greater sharpness in the center at the wider apertures, and, like the Sigma, perhaps a bit lower periphery resolution at the wider apertures.  However, the strong vignetting of the 35 / 2 IS at the wider apertures may serve to give a similar effect.

So, to recap:  an ideal situation with regards to shallow DOF, sharpness, and pleasing bokeh would be near peak sharpness in the central portion of the frame right from wide open, with resolution falling off from the central portion of the frame to the corners as a function of the lens sharpness rather than strong field curvature.  As one stops down, the corner performance steadily improves until it is at top form by f/4-f/5.6 and close to the center performance.

No, no, no. Peter Karbe was talking of MTF outside the plane of focus. Not your father's MTF. He was talking about the contrast and resolution of the bokeh, in particular, how both decrease very rapidly just in front of and behind the plane of focus. This is MTF of areas never published, ever.

He was aiming for standard focal length lenses that shoots like large aperture telephotos.

kotmj wrote:

kotmj wrote:

Peter Karbe of Leica said in an interview on the 50/2 M APO that what they tried to achieve with it that was different from most manufacturers is that the MTF needs to fall off steeply away from the plane of focus. That is, the object in focus has great macro and microcontrast -- very high and level MTF curves -- but that the objects infront of and behind the plane of focus have intentionally bad MTF curves. The out-of-focus areas have low contrast and resolution. This difference in contrast between the object in focus and those OOF is what makes for the pop.

He further said the Japanese would never think of doing this.

Leica has had different philosophies in this regard over the decades. Mandler had a piece published where he stated that in what he called "pictorial" photography, even the OOF areas should have high contrast.

Maybe Sigma caught on to Karbe's thinking, or arrived at it on their own.

...the MTF curves measure resolution on the plane at the focal point that runs perpendicular to the line of sight. If the MTF curves show lower resolution from the center, this could mean the lens sharpness drops off as you describe above (which is a good thing, in my opinion, for central composition wide open) or it could indicate strong field curvature.

Either way, the MTF curves should flatten out as one stops down, becoming sharper across the frame and at top form by f/4 (f/5.6). The Sigma 35 / 1.4A is a wonderful example of such a curve, although one might argue that lower resolution at the wider apertures in the periphery would have been more desirable. The 35 / 2 IS, however, would have benefited from greater sharpness in the center at the wider apertures, and, like the Sigma, perhaps a bit lower periphery resolution at the wider apertures. However, the strong vignetting of the 35 / 2 IS at the wider apertures may serve to give a similar effect.

So, to recap: an ideal situation with regards to shallow DOF, sharpness, and pleasing bokeh would be near peak sharpness in the central portion of the frame right from wide open, with resolution falling off from the central portion of the frame to the corners as a function of the lens sharpness rather than strong field curvature. As one stops down, the corner performance steadily improves until it is at top form by f/4-f/5.6 and close to the center performance.

No, no, no. Peter Karbe was talking of MTF outside the plane of focus. Not your father's MTF.

He was talking about the contrast and resolution of the bokeh, in particular, how both decrease very rapidly just in front of and behind the plane of focus. This is MTF of areas never published, ever.

OK, sure -- my bad for addressing a different point.

He was aiming for standard focal length lenses that shoots like large aperture telephotos.

Yes -- I get that, and do not disagree.  Thanks much for correcting me!

BAK wrote:

Your shots are the best examples I've seen showing the difference in contrast between lenses.

I note the Canon's background is brighter and richer. Red plant on the right edge, and the orange whatever is significantly different in the pix from the two lenses.

The Canon is a great lens, don't get me wrong - especially at f/4. I think the colours are more pleasing to my eye at the same WB. Check out the pale spruce top on the guitar. It has ever so slightly a warmer look to it. Nothing that's so drastic that can't be tweaked in post for the Sigma, of course but both lenses were set to the same manual WB for the shots to keep tests consistent.

So perhaps the lower contrast is what makes the foreground "pop."

It could very well be, or a combination of factors I guess.

Anyway, thanks for your efforts.

Thanks for your kind words, and you are most welcome.

Cheers,

Mark.

Great Bustard wrote: To be honest, I think we are all picking nits, here -- it's not the equipment that's holding people back from getting great photos.

Amen to that. I guess what I didn't disclose as to why I was looking at the Sigma in the first place over the Canon is that as I'm shifting my main focus to cinematography and so I figured that the extra stop of light from f/2 to f/1.4 could be a real asset for me in dimly lit locations whilst maintaining a low ISO, the "better" (in my opinion) bokeh is a bonus which I thought I'd share as it'd be interesting to photographers.

On the back of this Sigma tests I've done (these being just bokeh tests) I've decided to keep it because I think the extra stop will genuinely be useful now that I know it's still usably sharp wide open with less vignetting than the Canon has at f/2. I've also ordered the Sigma 24mm f/1.4 A and the Sigma 50mm f/1.4 A for those reasons and the reason I never went for the Canon 35L II is because I can get all three for only a little bit more than the 35L II on its own. It's a no-brainer really.

My other concern with the Sigma when I ordered it was how well it would cope with manual focusing with such a short throw (I think it's around 100º) but as it turns out it's actually ok as the focus ring is very smooth and precise with zero slack - which is more than could be said of my recently sold 17-40L! ;-))

I could post some 24mm and 50mm comparisons in the same vein as this one if anyone's interested but alas my comparison lenses are now limited. My other half has a nifty fifty, a 24-105 f/4 L (so could do 24 and 50 on that one) and a 16-35 f/4 L (24mm). Not the best choice of lenses to compare against really as aside from the nifty fifty they're all f/4, but if anyone's interested, let me know.

Cheers,

Mark.

Great comparisons, thank you. Its strange Ive looked at hundreds of images on flickr from each of these lenses and I always prefer the bokeh from the Canon when viewing web images. But in side by side comparisons that I have seen including this one the Sigma is always the clear winner...weird. I do believe that bokeh cannot be judged at 100%, It can only be judged when viewing an overall image which is why your samples work so well. I am really surprised that the Sigma shows so much more bokeh at the same apertures. However, I have always noticed that the Sigma produces double lines in bokeh areas which has put me off it. I will definitely have to revisit this comparison when I finally do upgrade my original 35 f2 which does not have great bokeh but with which I can get great images btw.

I like the Sigma 24A it seems to have better bokeh than the 35A and its a nice size/weight.

Tiosabas wrote:

Great comparisons, thank you.

You are most welcome.

I like the Sigma 24A it seems to have better bokeh than the 35A and its a nice size/weight.

Could that perhaps be because with the 24mm, one will be closer to the subject due to the wider angle of view, and hence create "more" bokeh, or do you mean that it creates better bokeh if both lenses were taking a picture of the same subject from the same distance?

As it happens, I have 2 x Sigma 24mm Art lenses here but I've not used them. I've only checked them for decentering (they're both decentred) as they arrived in haviliy damaged boxes. One was a replacement for the last which arrived damaged(!). Still, that's what you get from buying lenses from an online store which started out selling books...

Hopefully I shall have yet another replacement tomorrow and it won't be damaged. Fingers crossed!

Cheers,

Mark.

Sorry - I don't have access to a Siggy 35mm Art; but here is my Canon 35mm f/2 IS just about as close as I can get with it (OOC JPEG):

I've chosen a reference subject that I hope is familiar to you! The size & weight of the lens are perfect on my 6D for carrying around. It isn't a "bokeh king" but it is sharp, sharp, sharp, and the IS works really well. It is my most-used lens - hardly ever comes off the camera.

J

I think the following is an interesting observation about the bokeh of the Canon 35mm f2 IS, and can be of value to this discussion. I've been thinking about buying that lens myself, but after seeing this I'm reconsidering my options. I think the Tamron 35mm f1.8 VC looks like the most promising lens right now.

https://flatworldsedgephotography.com/2015/11/15/corner-bokeh-quality-of-canon-35mm-f2-0-is-vs-canon-35mm-f1-4-ii/

Alex78 wrote:

I think the following is an interesting observation about the bokeh of the Canon 35mm f2 IS, and can be of value to this discussion. I've been thinking about buying that lens myself, but after seeing this I'm reconsidering my options. I think the Tamron 35mm f1.8 VC looks like the most promising lens right now.

https://flatworldsedgephotography.com/2015/11/15/corner-bokeh-quality-of-canon-35mm-f2-0-is-vs-canon-35mm-f1-4-ii/

Bokeh comparisons of different scenes are all but useless, really, since the bokeh is very dependent on the scene itself.

SarahBK wrote:

Alex78 wrote:

I think the following is an interesting observation about the bokeh of the Canon 35mm f2 IS, and can be of value to this discussion. I've been thinking about buying that lens myself, but after seeing this I'm reconsidering my options. I think the Tamron 35mm f1.8 VC looks like the most promising lens right now.

https://flatworldsedgephotography.com/2015/11/15/corner-bokeh-quality-of-canon-35mm-f2-0-is-vs-canon-35mm-f1-4-ii/

That is indeed quite distracting.

that review of 35 f2.0 IS is not fair and pretty foolish if you ask me! i have used the best canon and other top brand lenses for background bokeh--they turn out horrid when there are random tree branches and foliage! so, i don't take that unfair review seriously at all!!!

1Dx4me wrote:

SarahBK wrote:

Alex78 wrote:

I think the following is an interesting observation about the bokeh of the Canon 35mm f2 IS, and can be of value to this discussion. I've been thinking about buying that lens myself, but after seeing this I'm reconsidering my options. I think the Tamron 35mm f1.8 VC looks like the most promising lens right now.

https://flatworldsedgephotography.com/2015/11/15/corner-bokeh-quality-of-canon-35mm-f2-0-is-vs-canon-35mm-f1-4-ii/

That is indeed quite distracting.

that review of 35 f2.0 IS is not fair and pretty foolish if you ask me! i have used the best canon and other top brand lenses for background bokeh--they turn out horrid when there are random tree branches and foliage!

This is a big generalization. After all, that page has samples of good bokeh with trees and branches, as well.