Lens MTF detail (Page 4): Again you are making it up. It isn't a "correlation". It's what the lens design produces (absent odd sample variations). If the MTF shows low contrast and low...

Apaflo Loc: Anchorage, Alaska

burkphoto wrote:

Not. But numbers folks always think they can "see" or predict results with their figures.

There is, most often, a strong correlation. But correlation is not causation. As my economics professor used to say, "Our best assumptions are always just that --- theoretical assumptions, which is to say they are quite possibly wrong, or they are based on the wrong things. Nonetheless, they are often useful."

Again you are making it up. It isn't a "correlation". It's what the lens design produces (absent odd sample variations). If the MTF shows low contrast and low resolution at the edges that is exactly what you get every single time without fail. If astigmatism is indicated, that is what the lens produces every single time.

If the MTF for a particular lens is very good and you don't get matching results it either needs repair or your technique needs adjustment.

The idea of causation as opposed to correlation is a total misunderstanding of what MTF is.

aflundi Loc: Albuquerque, NM

Apaflo wrote:

... An "impressive looking" MTF is not what is in demand, and an accurate MTF is.

You want an accurate MTF. So do I. That's not the same thing as market demand. If you and I determined demand, AA filters would still be standard.

I'm not asking what you or I want in an MTF. That's *obviously* a full-spectrum-based MTF.

What I'm asking is 'what are they doing?'

Apaflo Loc: Anchorage, Alaska

aflundi wrote:

... What I'm asking is 'what are they doing?'

Producing useful data about their products. Got it yet?

aflundi Loc: Albuquerque, NM

Apaflo wrote:

Producing useful data about their products. Got it yet?

So then you know for sure that all lens manufacturers do full-spectrum calculations or measurements, and those MTFs are what they show publicly?

Architect1776 Loc: In my mind

Apaflo wrote:

Actually it make one Hell of a lot of difference to a knowledgeable photographer when choosing which lenses to buy in the first place! It shows how sharp a lense is across the sensor and in the corners. It shows how contrasty a lens is too, and how much astigmatism there is.

If you buy kit lenses and find them "tack sharp" then MTF charts won't mean much. If you want the highest IQ possible you need to carefully compare alternative lenses.

Apaflo Loc: Anchorage, Alaska

aflundi wrote:

So then you know for sure that all lens manufacturers do full-spectrum calculations or measurements, and those MTFs are what they show publicly?

That is the only way to sell their products. That is what they do except in rare circumstances when it otherwise makes sense and then they would be very specific. That might be true with optics made for special purposes such as astrophotography with monochrome CCDs.

aflundi Loc: Albuquerque, NM

Apaflo wrote:

That is the only way to sell their products. That is what they do except in rare circumstances when it otherwise makes sense and then they would be very specific. That might be true with optics made for special purposes such as astrophotography with monochrome CCDs.

Ok, I'll let it go. I'd like to ask one last favor though. I know you (apaflo) have built up a large "library" of links and other references related to the subject. For the purpose of further education, if you don't mind, do you have a link, book, or other reference that shows the procedure used to calculate an MTF. I ask because I'd like to know if there is a numeric trick involved to be able to make the calculation over a continuous spectrum more tractable.

Apaflo Loc: Anchorage, Alaska

aflundi wrote:

Ok, I'll let it go. I'd like to ask one last favor though. I know you (apaflo) have built up a large "library" of links and other references related to the subject. For the purpose of further education, if you don't mind, do you have a link, book, or other reference that shows the procedure used to calculate an MTF. I ask because I'd like to know if there is a numeric trick involved to be able to make the calculation over a continuous spectrum more tractable.

I actually do not have such a list. (I am traveling away from home and posting with a cell phone anyway.)

There was a link posted to a Nikon website that is good. Also Google lensrental.com for good articles. The ulimate though, if you really want to spend big bucks and do your own is to search on Norman Koren and his Imatest software. Start with the wikipedia page for Imatest.

aflundi Loc: Albuquerque, NM

Apaflo wrote:

There was a link posted to a Nikon website that is good. Also Google lensrental.com for good articles. The ulimate though, if you really want to spend big bucks and do your own is to search on Norman Koren and his Imatest software. Start with the wikipedia page for Imatest.

I've long been familiar with those and many others like them. I'm looking for technical details that none of those mention. But, thanks anyway.

Apaflo Loc: Anchorage, Alaska

aflundi wrote:

I've long been familiar with those and many others like them. I'm looking for technical details that none of those mention. But, thanks anyway.

Read Koren's doumentation for Imatest. It answers all of your questions.

Dan De Lion Loc: Montana

aflundi wrote:

Ok, you've stated what we've all assumed all along... (show quote)

-----

Aflundi – you’ve changed your post from a quest for information to one of a search for conspiracies. If you make false assumptions then anything is possible. Your assumption is that false MTF curves would go unnoticed and uncorrected by the market. Let’s just say that Zeiss advertised MTFs produced without the red end of the spectrum. That they did that in order to misrepresent the performance of their lenses. First, reviews would note that. Second, the informed purchaser would recognize that performance was not up to what the MTF curve indicated. Third, other manufactures would use Zeiss’ lies to their competitive advantage. The result would be a cheapening of the brand and a loss of sales.

I am familiar with Nikon’s, Hassy’s, and some Sigma MTFs. I find them accurate representations of actual lens performance. A good example is Sigma’s new 24-70 art lens ( https://www.sigmaphoto.com/24-70mm-f2-8-dg-os-hsm-a ). Those MTF’s show a brand new Art lens with just so-so edge and corner performance compared to Nikon’s VR 24-70 ( http://www.nikonusa.com/en/Nikon-Products/Product/Camera-Lenses/AF-S-NIKKOR-24-70mm-f%252F2.8E-ED-VR.html#lightbox:/Photography/20052_MTF_01_en.jpg ). If Sigma was going to “cook” the MTFs, you wouldn’t see that kind of chart. Likewise Nikon’s MTFs accurately show the relative strengths and weaknesses of their lenses.

-----

Sinewsworn Loc: Port Orchard, WA

bamfordr wrote:

And, of course, there is LSMFT

Not"Lucky Strike means fine tobacco?"

Could be "Loose straps means flappy tits."