Archive 2010 · Lenses with best micro contrast
p.3 #1Lenses with best micro contrast
It's simple math.
I'm talking about 24x36mm image area magnified to 200x300mm.
please do not confuse with resolving power of sensor. that is not relevant.
edwardkaraa wrote:
Your reasoning doesn't work.
Following your logic, even 40 lpmm will make a total of 2880 lines per frame length, much lower than the resolving power of any current sensor.
We all know that even film could resolve 160 lpmm (slide film) and up to 400 lpmm (black and white).
A full image viewed on the 17" monitor is roughly about 300mm in width. Assuming this image is taken by a FF sensor, it's original size is 24 x 36mm. There is a magnification of 300/36 = 8.3x. So referring to the width (ie 36mm) of the image, 5-10 lp/mm (taking 8.3 lp/mm for convenience of calculation) on 36mm translates to about 1 lp/mm on the image on screen. if you use a ruler to measure the details on the screen where one would call "micro-contrast", the details are definitely finer than 1 lp/mm. This means that micro-contrast cannot be in the range of 5-10 lp/mm on the 24x36mm area.
In simple term, 1 lp/mm means there is one black (dark) line and one white (light) line per mm. Alternatively, one can take the width of the line (of whatever colour) as 0.5mm.
Please do not confuse with pixel as it's not relevant.
I'm talking about 24x36mm image area magnified to 200x300mm.
please do not confuse with resolving power of sensor. that is not relevant.
edwardkaraa wrote:
Your reasoning doesn't work.
Following your logic, even 40 lpmm will make a total of 2880 lines per frame length, much lower than the resolving power of any current sensor.
We all know that even film could resolve 160 lpmm (slide film) and up to 400 lpmm (black and white).
makron wrote
Sorry, after thinking, I have to totally disagree that 5-10 lp/mm is micro contrast. It is simply contrast (macro or global or whatever).A full image viewed on the 17" monitor is roughly about 300mm in width. Assuming this image is taken by a FF sensor, it's original size is 24 x 36mm. There is a magnification of 300/36 = 8.3x. So referring to the width (ie 36mm) of the image, 5-10 lp/mm (taking 8.3 lp/mm for convenience of calculation) on 36mm translates to about 1 lp/mm on the image on screen. if you use a ruler to measure the details on the screen where one would call "micro-contrast", the details are definitely finer than 1 lp/mm. This means that micro-contrast cannot be in the range of 5-10 lp/mm on the 24x36mm area.
In simple term, 1 lp/mm means there is one black (dark) line and one white (light) line per mm. Alternatively, one can take the width of the line (of whatever colour) as 0.5mm.
Please do not confuse with pixel as it's not relevant.
Feb 25, 2010 at 10:32 PM
p.3 #2Lenses with best micro contrast
"Historically, for example, Zeiss was reputed to design their lenses for maximum resolution, while Leica apparently tended to favour maximum contrast. It is design decisions like these that account in part for the differing "looks" of different lens brands." MR
I am not sure if I should laugh or cry.
I am not sure if I should laugh or cry.
Feb 25, 2010 at 10:34 PM
p.3 #3Lenses with best micro contrast
Ahh, here you go. Straight from Zeiss
But what does ‘contrast rendition’ actually mean? We must not forget that when we talk about ‘contrast’ we always mean micro contrast, i.e. structures, which we can just about see or just cannot see with the naked eye, for example on a slide.
He goes on to talk about macro contrast and differentiates it from micro contrast.
But if we photograph a chessboard, for example, so that it fills the format, the contrast between the black and the white squares has nothing to do with this.
MTF measurements say nothing about this macro contrast. They gauge only the correction of the lens, i.e. the small deviations of the light beams, while the macro contrast depends on the veiling glare of thedeviations.
Look at pg 16 in that pdf. Some examples are then shown on page 17 of high and low micro contrast, and low macro contrast.
But what does ‘contrast rendition’ actually mean? We must not forget that when we talk about ‘contrast’ we always mean micro contrast, i.e. structures, which we can just about see or just cannot see with the naked eye, for example on a slide.
He goes on to talk about macro contrast and differentiates it from micro contrast.
But if we photograph a chessboard, for example, so that it fills the format, the contrast between the black and the white squares has nothing to do with this.
MTF measurements say nothing about this macro contrast. They gauge only the correction of the lens, i.e. the small deviations of the light beams, while the macro contrast depends on the veiling glare of thedeviations.
Look at pg 16 in that pdf. Some examples are then shown on page 17 of high and low micro contrast, and low macro contrast.
Feb 25, 2010 at 10:40 PM
p.3 #4Lenses with best micro contrast
You may change your mind after reading this:
http://www.zeiss.com/C12567A8003B8B6F/EmbedTitelIntern/CLN_30_MTF_en/$File/CLN_MTF_Kurven_EN.pdf
makron wrote:
It's simple math.
I'm talking about 24x36mm image area magnified to 200x300mm.
please do not confuse with resolving power of sensor. that is not relevant.
Following your logic, even 40 lpmm will make a total of 2880 lines per frame length, much lower than the resolving power of any current sensor.
We all know that even film could resolve 160 lpmm (slide film) and up to 400 lpmm (black and white).
A full image viewed on the 17" monitor is roughly about 300mm in width. Assuming this image is taken by a FF sensor, it's original size is 24 x 36mm. There is a magnification of 300/36 = 8.3x. So referring to the width (ie 36mm) of the image, 5-10 lp/mm (taking 8.3 lp/mm for convenience of calculation) on 36mm translates to about 1 lp/mm on the image on screen. if you use a ruler to measure the details on the screen where one would call "micro-contrast", the details are definitely finer than 1 lp/mm. This means that micro-contrast cannot be in the range of 5-10 lp/mm on the 24x36mm area.
In simple term, 1 lp/mm means there is one black (dark) line and one white (light) line per mm. Alternatively, one can take the width of the line (of whatever colour) as 0.5mm.
Please do not confuse with pixel as it's not relevant.
http://www.zeiss.com/C12567A8003B8B6F/EmbedTitelIntern/CLN_30_MTF_en/$File/CLN_MTF_Kurven_EN.pdf
makron wrote:
It's simple math.
I'm talking about 24x36mm image area magnified to 200x300mm.
please do not confuse with resolving power of sensor. that is not relevant.
edwardkaraa wrote
Your reasoning doesn't work.Following your logic, even 40 lpmm will make a total of 2880 lines per frame length, much lower than the resolving power of any current sensor.
We all know that even film could resolve 160 lpmm (slide film) and up to 400 lpmm (black and white).
makron wrote
Sorry, after thinking, I have to totally disagree that 5-10 lp/mm is micro contrast. It is simply contrast (macro or global or whatever).A full image viewed on the 17" monitor is roughly about 300mm in width. Assuming this image is taken by a FF sensor, it's original size is 24 x 36mm. There is a magnification of 300/36 = 8.3x. So referring to the width (ie 36mm) of the image, 5-10 lp/mm (taking 8.3 lp/mm for convenience of calculation) on 36mm translates to about 1 lp/mm on the image on screen. if you use a ruler to measure the details on the screen where one would call "micro-contrast", the details are definitely finer than 1 lp/mm. This means that micro-contrast cannot be in the range of 5-10 lp/mm on the 24x36mm area.
In simple term, 1 lp/mm means there is one black (dark) line and one white (light) line per mm. Alternatively, one can take the width of the line (of whatever colour) as 0.5mm.
Please do not confuse with pixel as it's not relevant.
Feb 25, 2010 at 10:43 PM
p.3 #5Lenses with best micro contrast
I'm sorry i'm a little agitated at this moment (not due to your post).
I think you should read the article you posted carefully. Then read my post carefully. I'll rest my case here.
Thanks for responding.
edwardkaraa wrote:
You may change your mind after reading this:
http://www.zeiss.com/C12567A8003B8B6F/EmbedTitelIntern/CLN_30_MTF_en/$File/CLN_MTF_Kurven_EN.pdf
I'm talking about 24x36mm image area magnified to 200x300mm.
please do not confuse with resolving power of sensor. that is not relevant.
Following your logic, even 40 lpmm will make a total of 2880 lines per frame length, much lower than the resolving power of any current sensor.
We all know that even film could resolve 160 lpmm (slide film) and up to 400 lpmm (black and white).
A full image viewed on the 17" monitor is roughly about 300mm in width. Assuming this image is taken by a FF sensor, it's original size is 24 x 36mm. There is a magnification of 300/36 = 8.3x. So referring to the width (ie 36mm) of the image, 5-10 lp/mm (taking 8.3 lp/mm for convenience of calculation) on 36mm translates to about 1 lp/mm on the image on screen. if you use a ruler to measure the details on the screen where one would call "micro-contrast", the details are definitely finer than 1 lp/mm. This means that micro-contrast cannot be in the range of 5-10 lp/mm on the 24x36mm area.
In simple term, 1 lp/mm means there is one black (dark) line and one white (light) line per mm. Alternatively, one can take the width of the line (of whatever colour) as 0.5mm.
Please do not confuse with pixel as it's not relevant.
I think you should read the article you posted carefully. Then read my post carefully. I'll rest my case here.
Thanks for responding.
edwardkaraa wrote:
You may change your mind after reading this:
http://www.zeiss.com/C12567A8003B8B6F/EmbedTitelIntern/CLN_30_MTF_en/$File/CLN_MTF_Kurven_EN.pdf
makron wrote
It's simple math.I'm talking about 24x36mm image area magnified to 200x300mm.
please do not confuse with resolving power of sensor. that is not relevant.
edwardkaraa wrote
Your reasoning doesn't work.Following your logic, even 40 lpmm will make a total of 2880 lines per frame length, much lower than the resolving power of any current sensor.
We all know that even film could resolve 160 lpmm (slide film) and up to 400 lpmm (black and white).
makron wrote
Sorry, after thinking, I have to totally disagree that 5-10 lp/mm is micro contrast. It is simply contrast (macro or global or whatever).A full image viewed on the 17" monitor is roughly about 300mm in width. Assuming this image is taken by a FF sensor, it's original size is 24 x 36mm. There is a magnification of 300/36 = 8.3x. So referring to the width (ie 36mm) of the image, 5-10 lp/mm (taking 8.3 lp/mm for convenience of calculation) on 36mm translates to about 1 lp/mm on the image on screen. if you use a ruler to measure the details on the screen where one would call "micro-contrast", the details are definitely finer than 1 lp/mm. This means that micro-contrast cannot be in the range of 5-10 lp/mm on the 24x36mm area.
In simple term, 1 lp/mm means there is one black (dark) line and one white (light) line per mm. Alternatively, one can take the width of the line (of whatever colour) as 0.5mm.
Please do not confuse with pixel as it's not relevant.
Feb 25, 2010 at 10:53 PM
p.3 #6Lenses with best micro contrast
Edward--having just finished through page 18 of the document you and kidtexas shared, it appears the author asserts that maintaining high contrast through the 20-40 lpmm range is critically important to edge contrast.
Obviously, lower frequency contrast must also be high in order to appreciate the transitions.
Obviously, lower frequency contrast must also be high in order to appreciate the transitions.
Feb 25, 2010 at 11:15 PM
p.3 #7Lenses with best micro contrast
Ok, let me sum it up what this document says regarding this specific subject.
First, it states that when contrast is mentioned, he's specifically talking about micro-contrast, as macro-contrast cannot be judged from MTF. He defines micro-contrast as contrast that cannot be seen by naked eye or on a slide film. So it does mean contrast in micro-detail.
He shows that some lenses can have high micro-contrast but low macro-contrast and vice-versa, they are not interrelated.
He doesn't say it as clearly, but it does seems to me that the 10 lpmm frequency is the one used to judge micro-contrast, while the 40 lpmm is used to judge resolving power. The 40 lpmm figures are only critical for large enlargements.
First, it states that when contrast is mentioned, he's specifically talking about micro-contrast, as macro-contrast cannot be judged from MTF. He defines micro-contrast as contrast that cannot be seen by naked eye or on a slide film. So it does mean contrast in micro-detail.
He shows that some lenses can have high micro-contrast but low macro-contrast and vice-versa, they are not interrelated.
He doesn't say it as clearly, but it does seems to me that the 10 lpmm frequency is the one used to judge micro-contrast, while the 40 lpmm is used to judge resolving power. The 40 lpmm figures are only critical for large enlargements.
Feb 26, 2010 at 05:03 AM
p.3 #8Lenses with best micro contrast
BTW...that article was very good...really helped me understand the real ins and outs of the MTF curve. I would imagine for non-engineers or science people, it would be a tough read, though. 
Feb 26, 2010 at 05:35 AM
p.3 #9Lenses with best micro contrast
So If you used a really good macro lens that had great resoloving power, and used it to take a picture of a micro brewery and had an enlarged print made, would you then have a macro image of a micro by a macro with micro resolving power? Do I have that straight??
Feb 26, 2010 at 06:11 AM
p.3 #10Lenses with best micro contrast
Docs--I believe so. But for having to read through such conundrums, I rightfully demand a macro sampling of said micros. 
Feb 26, 2010 at 01:24 PM
p.3 #11Lenses with best micro contrast
edwardkaraa wrote:
First, it states that when contrast is mentioned, he's specifically talking about micro-contrast, as macro-contrast cannot be judged from MTF.
Well, the micro-contrast gives you a lower bound for the macro-contrast.
First, it states that when contrast is mentioned, he's specifically talking about micro-contrast, as macro-contrast cannot be judged from MTF.
Well, the micro-contrast gives you a lower bound for the macro-contrast.
Feb 26, 2010 at 01:46 PM
p.3 #12Lenses with best micro contrast
edwardkaraa wrote:
"Historically, for example, Zeiss was reputed to design their lenses for maximum resolution, while Leica apparently tended to favour maximum contrast. It is design decisions like these that account in part for the differing "looks" of different lens brands." MR
I am not sure if I should laugh or cry.
Quoted from Erwin Puts Leica Lens Compendum:
"Leica lenses are optimised for maximum contrast at 20 lp/mm and this will automatically give the best results for the higher spatial frequencies."
"Historically, for example, Zeiss was reputed to design their lenses for maximum resolution, while Leica apparently tended to favour maximum contrast. It is design decisions like these that account in part for the differing "looks" of different lens brands." MR
I am not sure if I should laugh or cry.
Quoted from Erwin Puts Leica Lens Compendum:
"Leica lenses are optimised for maximum contrast at 20 lp/mm and this will automatically give the best results for the higher spatial frequencies."
Feb 26, 2010 at 10:02 PM
p.3 #13Lenses with best micro contrast
makron wrote:
I am not sure if I should laugh or cry.Quoted from Erwin Puts Leica Lens Compendum:
"Leica lenses are optimised for maximum contrast at 20 lp/mm and this will automatically give the best results for the higher spatial frequencies."
20 and 40 lp/mm are already in the realm of resolving power.
Traditionally Leicas are designed for maximum resolving power, while Zeiss are designed for maximum micro contrast.
edwardkaraa wrote
"Historically, for example, Zeiss was reputed to design their lenses for maximum resolution, while Leica apparently tended to favour maximum contrast. It is design decisions like these that account in part for the differing "looks" of different lens brands." MRI am not sure if I should laugh or cry.
"Leica lenses are optimised for maximum contrast at 20 lp/mm and this will automatically give the best results for the higher spatial frequencies."
20 and 40 lp/mm are already in the realm of resolving power.
Traditionally Leicas are designed for maximum resolving power, while Zeiss are designed for maximum micro contrast.
Feb 26, 2010 at 10:08 PM
p.3 #14Lenses with best micro contrast
edwardkaraa wrote:
I am not sure if I should laugh or cry.Quoted from Erwin Puts Leica Lens Compendum:
"Leica lenses are optimised for maximum contrast at 20 lp/mm and this will automatically give the best results for the higher spatial frequencies."20 and 40 lp/mm are already in the realm of resolving power.
Traditionally Leicas are designed for maximum resolving power, while Zeiss are designed for maximum micro contrast.
References?
makron wrote
edwardkaraa wrote
"Historically, for example, Zeiss was reputed to design their lenses for maximum resolution, while Leica apparently tended to favour maximum contrast. It is design decisions like these that account in part for the differing "looks" of different lens brands." MRI am not sure if I should laugh or cry.
"Leica lenses are optimised for maximum contrast at 20 lp/mm and this will automatically give the best results for the higher spatial frequencies."
Traditionally Leicas are designed for maximum resolving power, while Zeiss are designed for maximum micro contrast.
References?
Feb 27, 2010 at 07:29 PM
p.3 #15Lenses with best micro contrast
That Zeiss MTF article is a great service to those with inquiring minds - Hubert Nasse is obviously expert on the subject, though, while very thorough, his analyses stretch comprehension to the limit thanks to the complexity of the concepts involved and their interactions.
I feel what is being said is that all spatial frequencies are very important and should be high, and that the rate of change of MTF is intrinsic to the making of a 'brilliant image'.
On spatial frequencies and contrast (micro), this may help:
"Good MTF values at low spatial frequencies are necessary, but they are no guarantee for brilliant images." p16, How to Read MTF Curves
The clear implication of this and other statements is that *high MTF at high spatial frequencies is of fundamental importance to micro contrast*:
"We must not forget that when we talk about ‘contrast’ we *always mean
micro contrast*, i.e. *structures, which we can just about see or just cannot see with the naked eye*, for example on a slide." p16, How to Read MTF Curves
This also provides me with an opportunity to mention a related issue - that of the significance of structures and their characteristics that are *almost visible*, that lie just beyond our visual cognitive system...I was recently taken to task by member Carsten for comments on this subject. I stick by what I wrote in a comment on several posted images; and also mention here that Ctein (quite an authority IMO) has stated, paraphrasing: that almost visible detail does play a role in human perception of (yes, a poor term) 'sharpness'.
[The waters in the Zeiss article are muddied a little by mention of glare, inherent contrast of structures and lens correction. The analysis would have benefited from an assumption of keeping such factors equal for the purpose of explanation.]
Astered sections are all mine, for emphasis.
I feel what is being said is that all spatial frequencies are very important and should be high, and that the rate of change of MTF is intrinsic to the making of a 'brilliant image'.
On spatial frequencies and contrast (micro), this may help:
"Good MTF values at low spatial frequencies are necessary, but they are no guarantee for brilliant images." p16, How to Read MTF Curves
The clear implication of this and other statements is that *high MTF at high spatial frequencies is of fundamental importance to micro contrast*:
"We must not forget that when we talk about ‘contrast’ we *always mean
micro contrast*, i.e. *structures, which we can just about see or just cannot see with the naked eye*, for example on a slide." p16, How to Read MTF Curves
This also provides me with an opportunity to mention a related issue - that of the significance of structures and their characteristics that are *almost visible*, that lie just beyond our visual cognitive system...I was recently taken to task by member Carsten for comments on this subject. I stick by what I wrote in a comment on several posted images; and also mention here that Ctein (quite an authority IMO) has stated, paraphrasing: that almost visible detail does play a role in human perception of (yes, a poor term) 'sharpness'.
[The waters in the Zeiss article are muddied a little by mention of glare, inherent contrast of structures and lens correction. The analysis would have benefited from an assumption of keeping such factors equal for the purpose of explanation.]
Astered sections are all mine, for emphasis.
Feb 28, 2010 at 12:49 AM
brainiac
Offline
[X]
Details
p.3 #16Lenses with best micro contrast
Cableaddict wrote:
Great link, thanks. -but actually, they don't use the term "micro contrast" at all, unless I missed it.
They use "LOCAL contrast" which I take it is the same basic idea.
I like this term better, because it more clearly defines what it is, by specifying the areas in question. Perhaps we should also adopt this term, and retire the fairly-confusing "micro contrast" designation altogether?
I agree.
We are talking about a contrast-scale response. Large scale contrast is usually called 'contrast', and small-scale contrast is usually called 'sharpness' or 'resolution'. Those two qualities lie on an axis, and lens performance varies along that axis. But we should also remember that it is much more complicated than this, which is why I ignore MTF graphs. For a start, the contrast-scale response can vary by colour, i.e. frequency of light. Issues like LoCA and bokeh deformation over the frame are not always expressed in MTF's, and the ultimate test of a lens is whether you like the pictures it produces. The term 'micro-contrast' is no more specific than 'sharpness'. Performance at one arbitrary scale of image data (e.g. 20 lpmm) doesn't guarantee performance at another.
kidtexas wrote
Here is a link that talks some about microcontrast.They use "LOCAL contrast" which I take it is the same basic idea.
I like this term better, because it more clearly defines what it is, by specifying the areas in question. Perhaps we should also adopt this term, and retire the fairly-confusing "micro contrast" designation altogether?
I agree.
We are talking about a contrast-scale response. Large scale contrast is usually called 'contrast', and small-scale contrast is usually called 'sharpness' or 'resolution'. Those two qualities lie on an axis, and lens performance varies along that axis. But we should also remember that it is much more complicated than this, which is why I ignore MTF graphs. For a start, the contrast-scale response can vary by colour, i.e. frequency of light. Issues like LoCA and bokeh deformation over the frame are not always expressed in MTF's, and the ultimate test of a lens is whether you like the pictures it produces. The term 'micro-contrast' is no more specific than 'sharpness'. Performance at one arbitrary scale of image data (e.g. 20 lpmm) doesn't guarantee performance at another.
Feb 28, 2010 at 06:42 AM
p.3 #17Lenses with best micro contrast
brainiac wrote:
I agree.
We are talking about a contrast-scale response. Large scale contrast is usually called 'contrast', and small-scale contrast is usually called 'sharpness' or 'resolution'.
I don't agree. Units are different; resolution is a spatial frequency whereas contrast is a dimensionless quantity. Contrast is the modulation transfer at a given spatial frequency,
resolution is the highest spatial frequency at which a given (subject to discussion) contrast value is achieved.
I agree.
We are talking about a contrast-scale response. Large scale contrast is usually called 'contrast', and small-scale contrast is usually called 'sharpness' or 'resolution'.
I don't agree. Units are different; resolution is a spatial frequency whereas contrast is a dimensionless quantity. Contrast is the modulation transfer at a given spatial frequency,
resolution is the highest spatial frequency at which a given (subject to discussion) contrast value is achieved.
Feb 28, 2010 at 07:39 AM
p.3 #18Lenses with best micro contrast
Hi Jman13,
Which Zeiss 85/2.8 was used to take that flower shot?
I'm looking for a good lens on that range to use on my 5D, and that seems to be great!
Saw few on eBay for pentax, rollei and Hasselblad but I'm not sure if they are all the same.
I appreciate advices.
Thanks
Sam
Which Zeiss 85/2.8 was used to take that flower shot?
I'm looking for a good lens on that range to use on my 5D, and that seems to be great!
Saw few on eBay for pentax, rollei and Hasselblad but I'm not sure if they are all the same.
I appreciate advices.
Thanks
Sam
Mar 07, 2010 at 08:16 AM
p.3 #19Lenses with best micro contrast
SamRobles wrote:
Hi Jman13,
Which Zeiss 85/2.8 was used to take that flower shot?
I'm looking for a good lens on that range to use on my 5D, and that seems to be great!
Saw few on eBay for pentax, rollei and Hasselblad but I'm not sure if they are all the same.
I appreciate advices.
Thanks
Sam
It would be one of these: Contax 85m f/2.8 Sonnar (Contax made the lenses under the Zeiss name).
http://www.keh.com/Product-Details/1/CX06009010382N/CX06/FE.aspx
They used to run more like $200-250, but they seem to be creeping up in price.
Jim
Hi Jman13,
Which Zeiss 85/2.8 was used to take that flower shot?
I'm looking for a good lens on that range to use on my 5D, and that seems to be great!
Saw few on eBay for pentax, rollei and Hasselblad but I'm not sure if they are all the same.
I appreciate advices.
Thanks
Sam
It would be one of these: Contax 85m f/2.8 Sonnar (Contax made the lenses under the Zeiss name).
http://www.keh.com/Product-Details/1/CX06009010382N/CX06/FE.aspx
They used to run more like $200-250, but they seem to be creeping up in price.
Jim
Mar 07, 2010 at 08:55 AM
p.3 #20Lenses with best micro contrast
Thanks Grenache!
The only one I saw on ebay was around $500. I saw a Zeiss Sonnar 2,8/85mm f Rolleiflex SL35, do you think they are about the same or the Contax is different?
Also, I have a Leica Summicron 90mm f/2, and I was wondering if it's better for portraits than the Zeiss Sonnar 85/2.8.
Sam
The only one I saw on ebay was around $500. I saw a Zeiss Sonnar 2,8/85mm f Rolleiflex SL35, do you think they are about the same or the Contax is different?
Also, I have a Leica Summicron 90mm f/2, and I was wondering if it's better for portraits than the Zeiss Sonnar 85/2.8.
Sam
Mar 07, 2010 at 05:41 PM