The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

anyone got a Phase 110mm Schneider LS lens yet?

Jack

Sr. Administrator
Staff member
You know, Jack, in talking with you---this is exactly the same type of rationalization I tried to use myself before I bought the Mamiya 150mm D.

You poo-poo'd that. And you were the one who convinced me to buy one more lens...

So I'm here to tell you that you need to add both the 110mm LS and 55mm LS to your 80mm LS. You wouldn't want to break up a set would you? :grin:
I stand by my reco for you and the 150 D -- actually for ANYBODY who shoots people and or events regularly, that 150 D is simply amazing...
 

kdphotography

Well-known member
I stand by my reco for you and the 150 D -- actually for ANYBODY who shoots people and or events regularly, that 150 D is simply amazing...
Jack, your recommendations have always been pretty much "spot-on" ---but then again, that assessment may be slightly biased, as all gear sluts think pretty much alike... :D
 
S

SCHWARZZEIT

Guest
I realized I should have included a 100% crop from the f2.8 focus plane for reference. Here is an unsharpened 800 pixel crop from a different image of the same door where I focused on the door of the old structure:

Jack, are you sure this is unsharpened, not even to some degree during raw conversion?

On screen it looks to me like there was some degree of sharpening applied. The micro contrast really pops out.

I can see some random white and black single pixel artifacts in the crop. I don't think they would show in any form on a print but it looks distracting on screen. Are those artifacts coming from the raw conversion?

-Dominique
 

Jack

Sr. Administrator
Staff member
It was capture sharpened at my normal setting in C1 only --- 160%/0.6/2.0... This in and of itself creates no artifacts, so anything you're seeing there is the result of jpegging (I jpeged this at 10 in CS). The increased micro-contrast is actually fairly obvious when you compare this lens directly to the 80 D and one of the reasons I decided to keep this over the D.

Here is the same crop processed for max detail. Note that in the wood "sign" above the door, you can see some minor artifacting -- this again is only visible in the jpeg, not in the tiff. This lens is "cut your eyeballs" sharp -- again, this is wide open at f2.8:



And for further reference, here is the full frame -- as you can see, it's unlikely even the artifacts above would show in a print:

 

faneuil

Member
Got my 110!

well.. a manual focus Sekor C lens for $104!
First impressions are, well, impressive!
What is expected price of the new 110mm lens?
Always love MF portraits in the 110-120mm range.
The new 150mm 2.8 is mightily impressive but may not be my cup of tea.

Eric
 

Jack

Sr. Administrator
Staff member
Got my 110!

well.. a manual focus Sekor C lens for $104!
First impressions are, well, impressive!
What is expected price of the new 110mm lens?
Always love MF portraits in the 110-120mm range.
The new 150mm 2.8 is mightily impressive but may not be my cup of tea.

Eric
We want PICTURES!!!

:ROTFL::ROTFL::ROTFL:
 
S

SCHWARZZEIT

Guest
Thanks for the demonstration, Jack. The version that you processed for detail is obviously much sharper than the other crop that had only the pre-sharpening applied. But now there are many single black and white pixels that pop out like sparkle. AFAIK jpeg compression may introduce blocking and banding artifacts. I haven't seen something like this from jpeg compression. But if those sparkling pixels do not appear in the tif then there's no other explanation.
Just to make sure we talk about the same thing here, the odd pixels I'm referring to are also obvious in one of the crops from your WR S + Rodenstock 40mm review where almost every bolt of the train has one white pixel on top and even more so in the bush background:



It looks like the sharpening software is creating some artificial highlights to increase the micro-contrast in certain areas. This may be quite effective to increase the sharpness impression in a print but it looks artificial to me on the monitor. But then of course the appearance of sharpness itself is a matter of personal taste.

Sorry for going a little OT with this.

The 80mm LS is indeed impressively sharp, especially considering that this is f/2.8.
How is the corner performance wide open?
Does is it get noticeably better in the center when stopping down?

-Dominique
 

Jack

Sr. Administrator
Staff member
Good points Dominique,

I think we tend to process for our normal output medium, and mine is print. So many of my jpegs will appear over-sharpened at 100% screen resolution. The sparklies are one characteristic for sure. I think my issue with the train shot was I applied my normal workflow to the raws, and that 40HR is so freaking sharp that it ends up over-done for 100% web, and probably for print too. Chalk it up to learning curve with a piece of equipment I had access to for 48 hours only. On the other hand, the train car was freshly waxed and those rivets are lit by direct sun, as are the waxy leaves on the bush in the background, so at least some of those are likely to be true speculars...

Back to the 80LS.

Edge sharpness is very good wide open, but it is not anything close to what the center is. I'll try and post a comparable crop from that above frame if there is something near the corners near the focus plane. If not, I'll find another f2.8 and post the corners. The extreme corners go soft on my P65+ at f2.8.

Honestly, I think it's already resolving better than the sensor at the center wide open, so stopping down doesn't add much there other than DoF. However, contrast pops up a little and that can add the appearance of being sharper. Again, I can try to post some crops if I can find some that are relevant.
 
S

SCHWARZZEIT

Guest
Ok, knowing that the crops were processed for printing explains the slightly oversharpened appearance on screen.
In my own workflow with drum scanned film I do not apply any sharpening to the master file. Every image gets individually sharpened after resizing according to the output medium. It also depends on the film type, magnification ratio, contrast range (all those variables that affect the grain pattern of the final image) and of course the content of the image itself. Some images can take more sharpening while others may not need any sharpening at all. There's no batch processing routine, except maybe for very small size web output.

With that kind of contrast as shown in your crop there's no doubt that the 80 LS easily outresolves the sensor wide open in the center. It's more important how much contrast is transferred within the range of the sensor. That's what you need to resolve very low contrast detail.

An f/2.8 crop from the edge of the frame would be interesting if there's anything in focus. I think the corner performance of a lens, especially at wider apertures, is where the quality of a lens design really shows aside from the price tag ;-)

Do you think the overall performance is on par with the Digaron-S and Digitar lenses of comparable focal lengths?

-Dominique
 

cunim

Well-known member
Good points Dominique,
Honestly, I think it's already resolving better than the sensor at the center wide open, so stopping down doesn't add much there other than DoF. However, contrast pops up a little and that can add the appearance of being sharper. Again, I can try to post some crops if I can find some that are relevant.
Jack, this is somewhat OT but interesting. How does a lens resolve better than a sensor? I often see this comment with respect to photographic lenses. The MTF of the system is the product of the MTFs of the discrete components, so better performance in any one component will yield better performance from the system.

That is, it will until we reach some actual limiting factor such as diffraction. What is going on in this case?
 
S

SCHWARZZEIT

Guest
Jack, this is somewhat OT but interesting. How does a lens resolve better than a sensor? I often see this comment with respect to photographic lenses. The MTF of the system is the product of the MTFs of the discrete components, so better performance in any one component will yield better performance from the system.

That is, it will until we reach some actual limiting factor such as diffraction. What is going on in this case?
In theory diffraction is the only limiting factor when it comes to the resolving power of a lens. But as there is no such thing as a perfect lens, a lens designer has to deal with all sorts of aberrations, keeping them in check because each aberration reduces contrast. As you figured with the cascading of the MTFs of every link in the imaging chain, in the end contrast is all that really matters because you need some minimum threshold contrast to resolve fine detail. Stopping down decreases the effects of the residual aberrations and thus increases contrast until diffraction drops your contrast even more.

The resolving power of a sensor is ultimately limited by the Nyquist frequency which is determined by the sensor's pixel pitch. A 6 micron sensor for instance cannot resolve more than 83 lp/mm. So a lens that resolves more than that at decent contrast outresolves the sensor. It means that you could resolve finer detail with a sensor of higher pixel density or with film but that's another topic.

-Dominique
 

Jack

Sr. Administrator
Staff member
In my own workflow with drum scanned film I do not apply any sharpening to the master file.
Mine too, but when I processed the above crop, I was actually posting from the field and had cooked the files on my laptop using my 'standard' set of settings --- IOW it was not an optimally processed crop.

Do you think the overall performance is on par with the Digaron-S and Digitar lenses of comparable focal lengths?
I think the Digiron HR's and latest Schneider's are still a notch or maybe half a step above this. But this and the latest D lenses from Mamiya and Phase are really pushing that distinction. While the differences are there, you need side-by-side pixel-level comparisons to see them, and thus they won't be realized differences until one prints very large. Factor in the use/convenience advantage and IMHO the scales tip back.
 

Jack

Sr. Administrator
Staff member
Here is another frame from the 80 LS at f2.8 that I had some corner detail at least near the PoF. This is not a perfect example as the camera is angled downward slightly, we do not have a planar subject but rather a series of "stuff" in and out of the focus plane in these regions. But I think it will hopefully give some insight to your main question on corner performance as well as some insights to Bokeh. These are not process sharpened, only the basic capture sharpening as per the earlier image.

Here is the full image:



Now Upper Left corner, Lower Left corner and Bottom Center crops:







Bottom line is I think this is pretty commendable wide-open corner performance from any lens in front of a P65+ sensor. The more I use this lens, the more I like it.
 
S

SCHWARZZEIT

Guest
Thanks for showing these, Jack. I find that subtle glow near the edge really beautiful.
BTW, I like how the different greens are rendered in this image.

-Dominique
 

Jack

Sr. Administrator
Staff member
Thanks for showing these, Jack. I find that subtle glow near the edge really beautiful.
BTW, I like how the different greens are rendered in this image.

-Dominique
Yes, I like that soft glow too -- it is oddly pleasing, especially at the corners. Also, it seems to work in harmony with the somewhat busy, but soft far-focus bokeh.

The green tonal rendition is a property of the Dalsa sensor I really appreciate. FWIW, I processed this image at neutral WB, and the greens pop even more if you warm the WB up a few points (which I normally do for landscape type images). I don't have any logical explanation as to why the response in greens is different from the Kodak sensors, since I would assume it's profile-dependent. But whatever the reason, it is there and I like it too ;)

Jeeze, I'm starting to sound like the cameraholic version of a wine snob :rolleyes:
 

cunim

Well-known member
The resolving power of a sensor is ultimately limited by the Nyquist frequency which is determined by the sensor's pixel pitch. A 6 micron sensor for instance cannot resolve more than 83 lp/mm. So a lens that resolves more than that at decent contrast outresolves the sensor. It means that you could resolve finer detail with a sensor of higher pixel density or with film but that's another topic.

-Dominique
I think I understand. If I might paraphrase - in this mixed digital/analog system, detector aliasing is our ultimate limit. To use the detector at its theoretical limit, we need to ensure lens resolution is better than pixel pitch. I think it would need to be much better than twice as high (e.g. 90% mtf at 83 lp), as the lens should be giving us good contrast at the twice the Nyquist limit of the detector.

However, this is a special case in that we are not usually imaging Nyquist-limited data. At any spatial frequency above the Nyquist limit a lens does not outresolve the sensor and the multiplicative MTF relationship maintains.

Interesting and instructive as to why the new digital lenses perform so well on any sensor pitch. Except under a very limited set of conditions a lens cannot ouresolve a sensor. Or have I missed something?
 

Jack

Sr. Administrator
Staff member
Except under a very limited set of conditions a lens cannot ouresolve a sensor. Or have I missed something?
A lens can outresolve a sensor. If you have a sensor that has 200 pixels per linear mm, then before Nyquist it can theoretically "resolve" 100 line-pairs per mm. If your lens now can resolve 120 LPmm, then it "outresolves" the sensor. Nyquist takes the actual resolving ability down to about 67% theoretical, or in this case, 67 LPmm. So all we'd really need in this example is a lens capable of resolving anything better than 67 LPmm and it will be outresolving the sensor.

What is true, is that we won't be able to tell how much better the lens is if we continue to test with that sensor. All we can know is the lens is at least as good as the sensor, or better, but not how much better. OTOH, if the image in the above example only resolves 40 LPmm, then we know the lens is the limiting factor since our sensor can resolve up to 67.

Finally, we do not need a sensor or film to determine a lens' resolving power -- we can alternatively derive it by testing it on an optical bench where we view the projected (aerial) image of the test target through a loupe. Extreme optics are tested this way since some can resolve well beyond what even the finest grain film can render.

Make better sense now?
 

Guy Mancuso

Administrator, Instructor
Actually Phase is trying to get me one for testing now. So hopefully soon i can run it through the mill and report fully on it.
 
Top