Accurately comparing FF vs APS-C sensor performance? An open discussion.

Started Apr 8, 2013 | Discussions thread
OP VirtualMirage Veteran Member • Posts: 3,956
Re: You misunderstand

OldClicker wrote:

One more try.

Look, I am really not trying to fight with you.  If it is coming across that way, I apologize.  But I will not deny that we have been displaying some strong minded debating.  Hopefully, the end result will be us walking away knowing more than we did walking in.

I am doing my best to really understand how your views impact the accuracy of my observations through the use of these DoF calculators.

So now you are flipping your stance and admitting that DoF and its related variables are based on mathematical formulas?

No, your backwards.  Reality is NOT based on the model.

I am not saying that reality is based on a model, unless you live in the Matrix.  Models try to interpret reality, some more precise than others.  Maybe my sentence was poor wording, still relying on the assumption that we know what is being implied.

Welcome to the world of math and science!  It doesn't make it any less accepted or valid.

I'm a chemical engineer.  My education and career were based on mathematical modeling of real world processes.  That's all engineering is.

So you are not denying this, then?

Example, proofs of what?  You are saying that pixel size/density has no effect because your model, that totally ignores it, says so.

I was saying that pixel size/density only affects how much detail/definition of the DoF that the optics and format size are creating is perceived in the final output image, but it doesn't effect the amount of DoF that is being created.

Change the pixel density and you might be able to perceive a finer detail of the range of DoF in the final output image but, it doesn't change the size of DoF itself.  The size of the DoF is being created by the optics and the format size, not the pixel size.

The CoC (Circle of Confusion) does not change with pixel size, it changes with format/sensor size.

Assumptions aren't wrong - all models make assumptions.  It is only the misuse of the model that is wrong.  I believe the DoF calculators assume something like 20/20 vision at 1 meter for an 8x10" print at 300 DPI.  If these are not true, then the model is being misused and may be wrong.

So now I am to assume all variables that can change from person to person, including eyesight?  Why would I create a model for someone who is myopic or hyperopic?  Wouldn't they be correcting their vision to be as close to 20/20 as possible?

Even if that variable changed, it would be the same amount of change for both cameras.  And if that is the case, wouldn't the amount of difference in DoF for both formats still be the same?  I can see how it wouldn't be if you had someone with different vision viewing one and someone else viewing the other.  But this observation is to assume it is the same person using both cameras.  Their vision isn't going to change just because they are using a different camera or format size.

This is why I chose to use a standard model to determine depth of field, one that is accepted by many and across the industry.  I am not going to make up my own.  If I did, I would be getting scrutinized even more than I am now.

Yes, many mathematical models usually assume the tests are being carried out in a vacuum, a perfect world.  We know that in real life our scenario wouldn't match this, but the results are usually very accurate for everyday use with exception to the extreme situations or when you are trying to get to a extremely accurate numbers (ie, to several decimal places) that would only need to be registered in a lab environment.  We are not working in a lab, we are not working in the extremes or on another planet, and thus this would not apply.

No, models are not based on "tests being carried out in a vacuum".  The best model for any process is the simplest model that gives the desired result.

But yes, many are based in a so to speak vacuum.  The example with determining DoF is based on perfect optics, perfect light, perfect distance from lens to imaging format, etc..  Any slight imperfection here may change the end result, albeit so slight it may not register.  But the model is not factoring in imperfections.  So wouldn't that be considered modeling in a vacuum?  Or am I using the wrong term here?

Been nice.  Over and out.

Well, it's a shame we couldn't find a way to come to an equal understanding here.  Just when you started to make some sense, too.  Take care!

-- hide signature --


 VirtualMirage's gear list:VirtualMirage's gear list
Sony RX100 Sony a77 II Tokina AT-X Pro 11-16mm f/2.8 DX Tamron SP AF 90mm F/2.8 Di Macro Tamron SP 70-200mm F/2.8 Di VC USD +18 more
Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow