!!! SIGMA SD1 - 46Mpx !!!

[LUIS A GUEVARA]

Luis,

I apologize for my serious arithmetic error. My calculation is two dimensional; actually the same as yours. I mistakenly divided by the 1.5 crop factor. What I attempted to do was to calculate the number pixel along the horizontal axis, thus:

Total pixels = Nx*Ny, but Nx=1.5Ny, so it should have been,

Nx^2 = 1.5*15.3MP => Nx= 4800 appr.

but all of this is an aside from my main point. That is that if you look at lens specs from mfgrs. they typically spec sharpness wide open at 30 line pair/mm as a function of radial displacement etc.

They specify Modulation Transfer (MTF) , which is a measure of how a certain Contrast is Transfered by the lens to the Film or Sensor Plane , and it is measured with a traveling slith. Test charts are for Amateur use .

And since all lenses lose sharpness at large distances off axis, the smaller APS-C sensor is less demanding in this regard. It is the difficulty of manufacturing a lens that's sharp over the larger sensors, that drives up the price substantially.

Not True . APS-C cameras use APS-C lenses to save money in Glass and therefore off axis deterioration exist just as in any other lens . What is true is that if you use a lens designed for FULL FRAME , such as a LEICA-R , then the APSC sensor will be using only the so called "SWEET SPOT" . The center of the lens . This is conceptually as IF stopping down to a smaller aperture , to use only the center portion , but USING THE LENS AT FULL APERTURE.

As to measuring lens resolution, it's easier said than done in the field. I'm sure you are familiar with the Imatest Program, for example, which has been devised to do such measurements. The problem is that you always get a kind of convolution of things when you try to analyze the response to various step function targets. So, as an example, if both the lens and sensor had equal resolution on axis, say, The convolved measured resolution would be down by a factor of 1.4 or so. The only way to accurately measure a lens' resolution is to do some sort of scanning slit measurement.

I already answered this point . I assume that you are using the technical word "Field" to refer to the periphery of the lens, same as I am . Serious Bench tests of a lens will be based on MTF , by using a Traveling Slit testing machine , and tests will be performed on all axis as well as Tangentially to measure , among other things , Stigmatism

Convolution of lens MTF and Sensor MTF , is already accounted for in the test bench and it is not really a problem for a Digital Imaging expert or for any Optician that are well versed on Fourier Transforms and the like.

Let me end by asking the question;
Why do you think Sigma states that it has over 40 lenses suitable for working into the SD1?

Sigma has proved to be more deceptive than any other company from Japan. Most of the line up of Sigma lenses was designed for their previous models , that were using sensors that were SMALLER than APS-C. The SD1 sensor is a true APS-C and the current line up of lenses , with the exception of the few remaining Full Frame lenses designed for the Film Cameras , will not perform well unless well stopped down , to eliminate the rays of light coming from the periphery of the lens.

Take a look at the following chart . The purple rectangle, marked Foveon ,indicates the Old Foveon sensor . Compare that with the next step up for APS-C as implemented by Canon and Nikon.

Please observe the big difference that exists with the APS-H Specification that Canon uses . That is what Sigma should use in their next model if they cannot afford to go Full Frame.

View attachment 2203


The short answer to your last question is a no brainer , Sigma wants to sell its new camera and when they stated that the crop factor of the SD1 was only 1.5 instead of the old 1.7 , lots of well informed people posed that question and Sigma avoided the issue by making a statement about how many lenses will FIT the SD1


They can get away with it because even those lenses will not perform well on the larger sensor , still the much improved SD1 camera will outperform the same lens if mounted in any of the previous models.

The good thing is that the camera would perform admirably better if converted to accept a Full Frame lens such as a Leica-R or Zeiss ZF (Nikon-F).

If someone has the money to buy an SD1 that would be the thing to do.

Luis
SIGMA CUM LAUDE
Please click here to visit our new Forum
--
http://photo.net/photos/Luis-A-Guevara
http://www.pbase.com/luis_a_guevara/galleries
http://gallery.leica-users.org/v/LUIS+A+GUEVARA/
http://www.summiluxart.com/
http://www.sigmacumlaude.com/

 

[JOHNBLUE3]

Luis,

It is clear that a Leica-R lens is sufficient to show the SD1 in all its glory. No one is debating that. What's not clear is that it is a necessary requirement. Again, the smaller sensors, places any applied lens, more within its sweet sport than if same lens were working into a larger sensor. As they often say, a picture is worth a thousand words. All one has to do is download any number of SD1 Raw files, taken with ordinary Sigma DG quality lenses of varios sorts, to see the extrordinary performance possible with this much less expensive lens arrangement.
Again, I know there is anger and disapointment focused on Sigma on account of the high MSRP of the SD1, but I don't think that issue rises to the level where they deserve to be called a dishonest firm. I own a lot of their products and have always found them to be of great value. Even now, as we debate, the SD1 is busily snapping photos, the quality of which we have never seen in DSLR. And all this with one or more of Sigma's more than 40 compatible lenses.
Finally, may I please request that you attach a plot of an MTF of a given lens firom any DSLR manufacturer. Here I mean MTF as a function of Spatial frequency. One that shows that the FWHM, say, is so many line pairs per mm. Regards.

 

[afshalders]

For me, the only problem with the SD1 (and Sigma) is that they have some very good lenses but to find a really good sample is something like a russian roulette game. I used several samples of a particular lens model (18-50 2.8 EX DC) and every single one gave me bad results, with very shar centers and poor edges even at F5.6 and F8 apertures, and this is unacceptable for a $450 lens. I also tried some other lenses with my EOS 5D and every single sample showed some front or back focus problem. To be honest the only lenses that worked fine, at least for me, were the 70/2.8 , 105/2.8 and 150/2.8 macros.

Sigma have the know how to make good stuff but they suffer for a very bad quality control. When I got my SD15, the sensor came with dust from factory !

I also don't trust in magazines or maker's tests and charts. Magazine charts are based in ONE lens sample and any person who studied statistics will know that an one sample test worths absolutely nothing. Maker's MTF charts are usually the THEORECTICAL curves, not the MEAN CURVES FROM A BATCH. Also worths nothing. The only maker I know that makes real MTF charts using the correct method is Zeiss. ALL the others show us the theorectical curves.

There are many other very important issues.

- Taste is a personal matter. For example, I love my 1950 85/2 Zeiss Sonnar for portraits and this lens is not very good wide open. But it's fantastic at f8 for portrait use. Some softness can be very useful and welcome.

- Photography is a very subjective matter. Involves a lot of psycological and physiological things.

What bothers me is to pay the equivalent of a good used car for something that I would not trust in terms of quality control (both cameras and lenses) and you all know that Sigma's quality control sucks.

When you buy something from Leica/Zeiss/Mamiya/Hassel you know that you will have the best in terms of quality and care. That's the point for me.

It's almost the same thing with cars. Ford and Toyota makes some very good stuff, but C'mon... don't dare to compare them with Mercedes. BMW or Porsche. I had lots of japanese cars, but none of them compares with my Mercedes.

Even comparing between japanese makers, the Sigma optics is still the worse option in terms of QC/Cost ratio (for the higher grade optics). It's almost impossible to compare any Sigma lens with Canon's L series or Nikon's Pro series.

I think it's ok to put a $800 Canon L prime on a EOS 5D-II, but not to do the same with that dread 30/1.4 from Sigma on the SD1 (even SD15) body.


Please forgime my rusted English.

Cheers !

 

[JOHNBLUE3]

Hi Luis,

It sounds like you have great confidence in brands. My wife is like that, she'll pay any price for a certain designer's dress. She knows little or nothing about what's happening on the clothing factory floor; those sweat shops in China and India etc. That stamp out fabrics and clothing to all commercial takers.
Your confidence in these brands is comparable to that, Leica/Zeiss/Mamiya/Hassel. We all know of their cachet, we know their products are exorbitantly expensive. We drool. But as an old Optical designer, myself, I know full well, that their designers don't know anything that I don't already know. I know, too, that in these modern times when one designs a complex lens on his computer, the model is extremely precise and one can easily use computer based raytracing techniques to very accurately model the MTF or anything else of interst. Even before the fact. Let me digress a moment to tell you a personal story.
Here in the US, we have Costco stores. Costco is a discount warehouse type opeeration which sells bulk items of all sorts. I buy bottled water from them. They carry two brands, their own Kirkland brand and the more widely advertized Arrow brand which most Americans are very familiar with. The Arrow brand water is sold at a substantially higher price than the costco Kirkland brand. Yet if one stands and watch customers choose water, there are many man y folks who will choose the Arrow almost subliminally. The actual facts about the two products is that the Costco water is substantially purer, has less sodium chloride and so forth and is lower priced. There are paradoxes here, I am always puzzled by how folks can be at once cynical and at the same time beholden to brands.
Now back to my earlier discussion of raytracing, as you, perhaps know, One can do many millions of simulations on a give lens arrangement, sufficient to map out completely its spatial frequency response. And to do that with Statistical Robustness.
Regarding Sigma, as far as I know, they are the only DSLR manufacturer that provides detailed lens profile characterization data to Adobe for built in use in CS5/LR CameraRaw etc. Further this little company makes lenses of all sorts for operation will most of the major DSLR manufacturers. With all of this experience in lens manufacturing, it's baffling to me why people are too dismissive; too quick to denigrate their products.
They are often caught between a rock and a hard place. For example, they make that perfect lens, 200-500mm/ F2.8 APO EX DG Telephoto that sells for $38000. To my knowledge, nobody makes a lens of that quality. yet because of their lack of cachet, a lot of people dismiss that product without review. Brands are cool, but I know full well, that the Engineers at Sigma know perfectly well whether their own in house lenses are appropriate to the SD1.
Regarding cars, I own a couple of Lexuses, one with over 250,000 miles on it and still running. It's quite luxurious with all systems still working after more than 15 years; all that with nothing more than routine maintenance. I admit it doesn't have the Mercedes cachet but I am hard put to find anything else it's missing. Regards.

 

[afshalders]

Hey, I'm not Luis !!! =)

I don't care about brands. I care about results and quality control. I have many cheap and excellent optics here besides two leica lenses, like the Fujinon 50/1.4 EBC and a really good (believe me) Industar 61 L/Z and other ones. What about my SD15 coming with sensor dust from factory ?

What I don't like is to pay $450 ina crappy lens like the Sigma 18-50 2.8 EX DC, Had to return 3 samples and still not happy with them. Tamrons and Tokinas have a much better QC.

I also have some cheap Chinon branded lenses that u can find from less than $100 at ebay, but those ones were made by Tomioka !

Sigma CAN do great stuff. The problem again is reliability and QC. Maybe I can pay $10k for a MF but not for the SD1 sorry. And I'm a big Foveon fan.

 

[JOHNBLUE3]

Hey, I'm not Luis !!! =)

I don't care about brands. I care about results and quality control. I have many cheap and excellent optics here besides two leica lenses, like the Fujinon 50/1.4 EBC and a really good (believe me) Industar 61 L/Z and other ones. What about my SD15 coming with sensor dust from factory ?

What I don't like is to pay $450 ina crappy lens like the Sigma 18-50 2.8 EX DC, Had to return 3 samples and still not happy with them. Tamrons and Tokinas have a much better QC.

I also have some cheap Chinon branded lenses that u can find from less than $100 at ebay, but those ones were made by Tomioka !

Sigma CAN do great stuff. The problem again is reliability and QC. Maybe I can pay $10k for a MF but not for the SD1 sorry. And I'm a big Foveon fan.

Yours is an anecdotal story, regarding the 18-50mm lens. I own one of those and it has been a workhorse lens for me. The internet can be quite a dangerous place. One user can make deleterious statment(not generally true), then millions will later repeat the same claim. Some of us own quite a few Sigma lenses and have never had a problem with any of them. There are folks using them to good purpose as we speak, to take glorious photos with the SD1. BTW, have you seen and analyzed any SD1 Raw files yet?

 

[afshalders]

Yours is an anecdotal story, regarding the 18-50mm lens. I own one of those and it has been a workhorse lens for me. The internet can be quite a dangerous place. One user can make deleterious statment(not generally true), then millions will later repeat the same claim. Some of us own quite a few Sigma lenses and have never had a problem with any of them. There are folks using them to good purpose as we speak, to take glorious photos with the SD1. BTW, have you seen and analyzed any SD1 Raw files yet?

I got 3 from B&H and all the 3 ones were crap. The fourth one is better but still not good. I have 3 Sigma lenses, the 3 macros 70/105 and 150 and i said that they are great lenses.

What's your problem ? Anedoctal ? Do you even know me to said this ? Are you telling me that I'm a liar and just repeating what some schmuck said ? Are you serious ?

 

[afshalders]

And here is a sample from the "good" 18-50 at 28mm F=7.1. No wind.

http://farm6.static.flickr.com/5039/5901979682_79d065d4ca_o.jpg

Borders are CRAP !

 

[LUIS A GUEVARA]

Luis,

It is clear that a Leica-R lens is sufficient to show the SD1 in all its glory. No one is debating that. What's not clear is that it is a necessary requirement. Again, the smaller sensors, places any applied lens, more within its sweet sport than if same lens were working into a larger sensor. As they often say, a picture is worth a thousand words. All one has to do is download any number of SD1 Raw files, taken with ordinary Sigma DG quality lenses of varios sorts, to see the extrordinary performance possible with this much less expensive lens arrangement.
Again, I know there is anger and disapointment focused on Sigma on account of the high MSRP of the SD1, but I don't think that issue rises to the level where they deserve to be called a dishonest firm. I own a lot of their products and have always found them to be of great value. Even now, as we debate, the SD1 is busily snapping photos, the quality of which we have never seen in DSLR. And all this with one or more of Sigma's more than 40 compatible lenses.
Finally, may I please request that you attach a plot of an MTF of a given lens firom any DSLR manufacturer. Here I mean MTF as a function of Spatial frequency. One that shows that the FWHM, say, is so many line pairs per mm. Regards.

Sure , here is the Leica=R 19mm f2.8 Elmarit:


And this is what Erwin Putts from leica has to say about it:

At full aperture the Elmarit-R 19mm f/2.8 has a high contrast
image with crisp definition of very fine detail over a
large part of the image area. The extreme corners are a bit
soft, but when using slides this edge area will be covered
by the slide mounts. The sagittal and tangential lines are
very close, indicating absence of coma and astigmatism.
Very fine detail is being represented by the 20 Lp/mm line
and we can see that even at 2.8 the contrast of 60% is held
till an image height of 15mm, giving an image circle of
30mm diameter

View attachment 2204

View attachment 2205

If you dont like what Leica says and prefer Japanese lenses here are some Canon lenses by Luminous Landscape:
View attachment 2206

Luis
SIGMA CUM LAUDE
Please click here to visit our new Forum
--
http://photo.net/photos/Luis-A-Guevara
http://www.pbase.com/luis_a_guevara/galleries
http://gallery.leica-users.org/v/LUIS+A+GUEVARA/
http://www.summiluxart.com/
http://www.sigmacumlaude.com/

 

[LUIS A GUEVARA]

Hi Luis,

It sounds like you have great confidence in brands. My wife is like that, she'll pay any price for a certain designer's dress. She knows little or nothing about what's happening on the clothing factory floor; those sweat shops in China and India etc. That stamp out fabrics and clothing to all commercial takers.
Your confidence in these brands is comparable to that, Leica/Zeiss/Mamiya/Hassel. We all know of their cachet, we know their products are exorbitantly expensive. We drool. But as an old Optical designer, myself, I know full well, that their designers don't know anything that I don't already know. I know, too, that in these modern times when one designs a complex lens on his computer, the model is extremely precise and one can easily use computer based raytracing techniques to very accurately model the MTF or anything else of interst. Even before the fact. Let me digress a moment to tell you a personal story.
Here in the US, we have Costco stores. Costco is a discount warehouse type opeeration which sells bulk items of all sorts. I buy bottled water from them. They carry two brands, their own Kirkland brand and the more widely advertized Arrow brand which most Americans are very familiar with. The Arrow brand water is sold at a substantially higher price than the costco Kirkland brand. Yet if one stands and watch customers choose water, there are many man y folks who will choose the Arrow almost subliminally. The actual facts about the two products is that the Costco water is substantially purer, has less sodium chloride and so forth and is lower priced. There are paradoxes here, I am always puzzled by how folks can be at once cynical and at the same time beholden to brands.
Now back to my earlier discussion of raytracing, as you, perhaps know, One can do many millions of simulations on a give lens arrangement, sufficient to map out completely its spatial frequency response. And to do that with Statistical Robustness.
Regarding Sigma, as far as I know, they are the only DSLR manufacturer that provides detailed lens profile characterization data to Adobe for built in use in CS5/LR CameraRaw etc. Further this little company makes lenses of all sorts for operation will most of the major DSLR manufacturers. With all of this experience in lens manufacturing, it's baffling to me why people are too dismissive; too quick to denigrate their products.
They are often caught between a rock and a hard place. For example, they make that perfect lens, 200-500mm/ F2.8 APO EX DG Telephoto that sells for $38000. To my knowledge, nobody makes a lens of that quality. yet because of their lack of cachet, a lot of people dismiss that product without review. Brands are cool, but I know full well, that the Engineers at Sigma know perfectly well whether their own in house lenses are appropriate to the SD1.
Regarding cars, I own a couple of Lexuses, one with over 250,000 miles on it and still running. It's quite luxurious with all systems still working after more than 15 years; all that with nothing more than routine maintenance. I admit it doesn't have the Mercedes cachet but I am hard put to find anything else it's missing. Regards.

Leica does not use Ray Tracing in their designs , Their Concept of a lens is as a conduit for Optical ENERGY , not rays of light ( the ancient newtonian idea ) ,and their aim is to eliminate or minimize any obstructions to its flow.

To that effect Professor Helmut Marx, created the famous COMO ( Correcting, Optimizing and Minimizing by means of Orthogonalizing) program.

This program was very soon installed in the factory of Leitz in Midlands (Ontario) to be used by Walter Mandler. Today this program (or licensed variations of it ) is now the staple of all Optical manufacturers.

As early as 1953, Leitz was the first photographic firm in the world which made use of a computer to help in the designing of the optical calculum of a lens. The driving force behind this revolutionary method that allowed the opticians to save thousands of manual calculi through some years to make the best lenses .

Computers are specially suitable to find feasible design routes, beginning from introduced parameters related to possible types of glass used, production costs, refractive indexes of them, capacity to be grinded or moulded and a huge list of specifications.


Leica makes use of the tandem MTF (Modulation Transfer Function) + OTF (Optical Transfer Function) as very valuable and necessary tools for the measurings by the designers, but that isn´t everything l.

Leica gives tremendous significance to the feedback received from professional photographers with a lot of years of experience who have made hundreds of thousands or millions of photographs to fine tune the optical performance of each new lens it designs before putting it into the market, and frequently , abandons designs which have rendered very good outcomes in MTF curves , because they do not make good images. MTF is just a tool .

Luis
SIGMA CUM LAUDE
Please click here to visit our new Forum
--
http://photo.net/photos/Luis-A-Guevara
http://www.pbase.com/luis_a_guevara/galleries
http://gallery.leica-users.org/v/LUIS+A+GUEVARA/
http://www.summiluxart.com/
http://www.sigmacumlaude.com/