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!

best DB for lens cast

tashley

Subscriber Member
Thanks Guy, much appreciated!

Tim not yet but we specially did shoot the 35mm lens with you in mind. From just what we can tell given the software , hardware today you will lose about 5- 10 percent more than your P65+. But I would not do anything until he has those final results posted. The 28 and 35 look from the previews to be a little limited on movements but I'm being cautious here as well as Phase maybe working on something with C1. I'm a little afraid to commit fully to a yes or no on it without seeing those final results.

I know your sitting there playing russian roulette on this. Totally understand and I don't want you to panic or mislead you but it most likely will be more limited than your P65+

BTW the 35mm is a nice lens compared to my 35D which I have the new version and is pretty good. I know Doug took a few days off for the weekend but should be back in full swing starting today.
 

tashley

Subscriber Member
Tim - I got my back on Friday. So, far I'm pleasantly surprised at how good the LCC's are working to correct the images. I'm only shooting the 43 and I know you are interested in the 35. I really was expecting much worse. Still waiting to see Doug's test.
Good to know Terry, thank you for the feedback!
 

Bob

Administrator
Staff member
At a given level of sensor construction technology the issue is made apparently worse if the pixel pitch is smaller and the sensor is larger and if the filter layer is thicker. In addition some lcc algorithmic adjustment needs to be made for such things as periodic polysilicon conductors that are spaced every so often across the sensor surface.
Most ccd sensor makers are pretty much at the same technology level give or take a little, so it mostly boilds down to the quality of the lcc software algorithms that your raw converter ustilizes and if it has been tweaked for your specific sensor.
All things considered:
Noise is worse as the sensor data is "pushed", so physical center filters needed for the best quality in terms of noise (but worse flare characteristics)
chromatic shifts due to the diffraction gradient effect caused by pixel pitch causes some colors to need more boost than others.
Bottom line, is that given a decent center filter to handle the lens fall-off and good algorithms for cosine law and diffraction color contamination, no wide will have the same noise and DR in the corners as they do in the center.
YMMYV so it is pretty much up to you how much you can take.
-bob

p.s.
there is a universal constant relation of angle of impingement, pixel pitch, filter thickness, and sensor size, that can predict the amount of misbehavior, but it is the quality of the correction and the loss of noise margin an dr that is the real limiting factor.
 

rga

Member
Hi Bob,

I'm not an engineer so I'm trying to apply what you said to how I shoot.

1. If I have a 35mm, using a center filter and LCC will give me better results than LCC alone?

2. If I use left and right shift and then stitch the images together (essentially putting more of the center of the lens toward the edges of the center shot of the image, will it still be better to use a center filter? Is that true whether or not I crop the final stitch?

Thanks for helping me understand,
Nontechnical Bob


At a given level of sensor construction technology the issue is made apparently worse if the pixel pitch is smaller and the sensor is larger and if the filter layer is thicker. In addition some lcc algorithmic adjustment needs to be made for such things as periodic polysilicon conductors that are spaced every so often across the sensor surface.
Most ccd sensor makers are pretty much at the same technology level give or take a little, so it mostly boilds down to the quality of the lcc software algorithms that your raw converter ustilizes and if it has been tweaked for your specific sensor.
All things considered:
Noise is worse as the sensor data is "pushed", so physical center filters needed for the best quality in terms of noise (but worse flare characteristics)
chromatic shifts due to the diffraction gradient effect caused by pixel pitch causes some colors to need more boost than others.
Bottom line, is that given a decent center filter to handle the lens fall-off and good algorithms for cosine law and diffraction color contamination, no wide will have the same noise and DR in the corners as they do in the center.
YMMYV so it is pretty much up to you how much you can take.
-bob

p.s.
there is a universal constant relation of angle of impingement, pixel pitch, filter thickness, and sensor size, that can predict the amount of misbehavior, but it is the quality of the correction and the loss of noise margin an dr that is the real limiting factor.
 

Jack

Sr. Administrator
Staff member
Simple answer is YES -- using a physical CF makes it easier on the LCC file. The associated algorithm does not have to work as hard to correct the entire frame because the physical filter has done a lot of the heavy lifting re falloff.
 

Wayne Fox

Workshop Member
Hi Bob,

I'm not an engineer so I'm trying to apply what you said to how I shoot.

1. If I have a 35mm, using a center filter and LCC will give me better results than LCC alone?


Nontechnical Bob
Another reason for the center filter is so the sensor can handle the dynamic range of the scene. Not an issue on many lenses but I find the 35mm schneider has a two stop difference between center and corners. Without the filter, there are scenes within the dynamic range of the sensor but with the added two stops push it too far. Personally I dislike my 35mm very much ... but I think that's reason why. Hoping the center filter will make it much more useful to me.
 

GrahamWelland

Subscriber & Workshop Member
Wayne,

I'd love to hear more about how you get on with your 35XL. I just picked up one of these myself, without the center filter, and I've been seriously considering getting one to even out the light fall off prior to LCC post processing. I'm sure that for a crop sensor (44x33mm) it's much less of a consideration than a P65+ or IQ180/160 etc but I am tempted to get the optical filter just in case.
 
Last edited:

tashley

Subscriber Member
I've been using the 35xl on a p65+ for about a year now without cf and it has been fine, though looking possibly a little stretched with shifts over 10mm, but still pretty good. It has very little distortion and, at f16, gives good focus to infinity from about 3m if set to focus at 5m, though that may vary per sample. For me it is a lens with no character whatsoever, a very good thing for certain kinds of work. It simply reproduces reality - as long as you LCC it.
 

rga

Member
I've not had the same issues as Wayne, though eventually I'd like to get a center filter. I've done several panos with the 35 doing 10mm shifts to either side.

This one was pointed into the sun:
http://www.rgaphoto.com/benches/content/Baylands_Bench_In_Memory_Of_Panorama_MASTER_large.html

This was at mid-day on a beach in Florida:
http://www.rgaphoto.com/FL/content/2011_05_27_Jacksonville_Big_TalbotSnag_1_BW_Pano_large.html

As was this:
http://www.rgaphoto.com/FL/content/...e_Pano_Cropped_BW_Big_TalbotSnag_2_large.html

However I do firmly believe that getting exposure right at the time of capture is much better than correcting in post processing.

I am concerned about the loss of 2 stops, the possibility of vignetting when putting a polarizer in front of the center filter, and the high cost of the center filter. So for now I'll have to put it off. It does seem like a good idea.

Bob

Wayne,

I'd love to hear more about how you get on with your 35XL. I just picked up one of these myself, without the center filter, and I've been seriously considering getting one to even out the light fall off prior to LCC post processing. I'm sure that for a crop sensor (44x33mm) it's much less of a consideration than a P65+ or IQ180/160 etc but I am tempted to get the optical fiter just in case.
 

Bob

Administrator
Staff member
Hi Bob,

I'm not an engineer so I'm trying to apply what you said to how I shoot.

1. If I have a 35mm, using a center filter and LCC will give me better results than LCC alone?

2. If I use left and right shift and then stitch the images together (essentially putting more of the center of the lens toward the edges of the center shot of the image, will it still be better to use a center filter? Is that true whether or not I crop the final stitch?

Thanks for helping me understand,
Nontechnical Bob
Yes, whatever can be corrected by center filter does not need a mathematical fix.
and yes to Nr2
-technical bob
 

jlm

Workshop Member
anytime you shift to the edges of the image circle you will pick up more extreme casts and vignetting.
 

cng

New member
The 35mm XL needs a CF and attempting to shoot without one is an exercise in masochism. I even have a CF on order for my 43mm XL, even though some believe that a CF is not as essential for the 43mm as it is for the 35mm.

I don't understand anyone who questions the need for a CF where one exists for a particular lens. CF's are notoriously difficult to manufacture, so they are not produced just for the sake of it. If Schneider/Rodenstock offer a CF then the default position should be to assume that it is necessary to obtain the best results from a given lens.

Also, the specs for the 35mm XL's CF specifically state +2 compensation, so why the surprise that the lens has a 2-stop falloff at the edges?

Two common complaints I hear about using CF's is the light loss and cost. The light loss can admittedly make life difficult in some situations, but the cost of a CF is negligible relative to the lens itself.
 

rga

Member
I honestly would love to have one and will get one. Unfortunately I leave for a workshop Thursday morning and will not have it for that. Sure wish I could get one in time...
Bob

The 35mm XL needs a CF and attempting to shoot without one is an exercise in masochism. I even have a CF on order for my 43mm XL, even though some believe that a CF is not as essential for the 43mm as it is for the 35mm.

I don't understand anyone who questions the need for a CF where one exists for a particular lens. CF's are notoriously difficult to manufacture, so they are not produced just for the sake of it. If Schneider/Rodenstock offer a CF then the default position should be to assume that it is necessary to obtain the best results from a given lens.

Also, the specs for the 35mm XL's CF specifically state +2 compensation, so why the surprise that the lens has a 2-stop falloff at the edges?

Two common complaints I hear about using CF's is the light loss and cost. The light loss can admittedly make life difficult in some situations, but the cost of a CF is negligible relative to the lens itself.
 

dchew

Well-known member
I got the following note from Schneider a few days ago regarding a CF for the 43mm:

"Hi Dave,
I heard back from our photo product manager in Germany and he informed me that calculations have been made for glass Center Filters for the new 28mm, 43mm and 60mm Apo-Digitar lenses and these should be in production soon. As soon as we have part numbers and pricing, we will post the information on our website."

Ciao,
Dave
 

cng

New member
calculations have been made for glass Center Filters for the new 28mm, 43mm and 60mm Apo-Digitar lenses and these should be in production soon.
Just adding to this, I got an email from Cambo about a week ago: "Center Filters for the 28XL as well as the 43XL will become available within the next few weeks, so meaning in July ready for sale."

It's great that Schneider is planning to produce CF's for the 28mm and 60mm as well.
 

rhsu

New member
Also from the list above to include Sinar 86H (48mp) (multishots) w/ Dalsa new generation micro lenses - allowing greater angle than the 1st gen micro lenses - 86H specially designed for tech-camera.
 

Woody Campbell

Workshop Member
I've been using the 35xl on a p65+ for about a year now without cf and it has been fine, though looking possibly a little stretched with shifts over 10mm, but still pretty good. It has very little distortion and, at f16, gives good focus to infinity from about 3m if set to focus at 5m, though that may vary per sample. For me it is a lens with no character whatsoever, a very good thing for certain kinds of work. It simply reproduces reality - as long as you LCC it.
Thanks to Lance Shad at Digital Transitions I had the use of a loaner IQ 180 over the weekend. I tested three Schneider lenses on my Alpa Max: the 35 xl and the 47 xl (both in Alpa dress) and the 72 L. I plan on posting samples whenever I can get an hour of free time to assemble them.

My observations on the 35 xl can be summarized as follows: The LCC correction files look like they have some fairly steep gradients but the LCC correction appears to work well. Ambient light has a strong effect on the LCC correction so it's a good idea to take one with each new situation - I found this out when I tried to use a shade LCC to correct a daylight image and the result was terrible. Properly corrected the unshifted images look very good, espically if you back off the vignetting correction - it will take a lot more experience for me to have a view on whether the LCC corrections for this lens are using up capacity to make further image adjustments but overall I was pleased.

There is very little scope for shifts on the IQ 180 with this lens. Maybe 7mm max on the short axis, and none on the long axis. If you look at the lens's mtf curves the sweet spot for the lens barely covers the IQ 180's 67mm diagonal. This is consistent with Jack's observations that the really wides seem to hit the zone of blue death at about the same point that they run out of resolution. As to the 35 that's ok with me because I'm actually using this lens on my Alpa TC as a digital SWC - I really didn't expect to get any shiifts out of this focal length.

The issue that I'm not clear on is whether the IQ 180 oversamples this lens - in other words the issue is whether the lens delivers enough resolution to meet the demands of an 80 meg sensor. It's fine with my 60 meg Blad. My experience varied on this subject but I did get a few images of outstanding clarity. If I end up owning an IQ 180 I'll sort this out - variables include the fact that this lens's perfromance peaks sharply at f 11 (not all of my images were at f11); based on the mtf curves the lens may be optimized for closer distances; Lance's Alpa adapter may or may not be optimally shimmed (which Alpa says is important for wides because of their short focus throw); and I need a bit more experience with focus masking to be sure that I'm getting it right.
 
Top