Maxwell Render | Official public forum

Hi all,

Thomas started a very smart initiative in http://www.maxwellrender.com/forum/viewtopic.php?t=8857

I can imagine that people want to comment on some questions, want to support or question them, are wondering what's meant in the first place etc.

To make the original topic as clear as possible I thought it should be a good idea to open this shadow topic. Victor c.s. can skip all the ramblings here, but it can still be a way to discuss some questions. Could well be that it leads to an updated or clarified question or makes clear that an additional question is appropiate.

If it's no succes, it will drop down the topic list and nothing is lost

I'll kick off:

Kabe wrote:Is there an option to deal with cos^4 vignetting, or is the mxi format documented in a way a 3rd party could write something for it?

Summary of the problem:


This is an issue if trying to do matte shots or cubic QuickTime VRs.

I found this post really amazing. It's very in depth, very to the point and very good documented.

What I was wondering: is there really a strict rule concerning vignetting. I always thought that in the real word it's was a matter of quality of lenses. Do you refer to a perfect lens?

They are lenses, like some Zeiss for Arri cameras that have total vignetting compensation. Some Zeiss lenses for Arri have also a total light halo suppression.

Yes it is a super idea with Thomas' thread ... and also this one to keep all the "talk" away from the Q/A.

I suggest that as soon as question is answered in Thomas' thread is is removed from the thread (if possible?) .. so as to constantly keep the thread "clean". This will also help in people not posing the same question again as the thread grows. But migth require someone with administrator rights to do that or ?

Just a thought.

>What I was wondering: is there really a strict rule concerning vignetting.
>I always thought that in the real word it's was a matter of quality of
>lenses. Do you refer to a perfect lens?

optical vignetting happens because of the way the light enters a phsyical lens at wider angles, the aperture would appear smaller because it's at an angle and therefore more oval (amount of light would be reduced), rather than the circular hole if you look at it dead on. But i don't think maxwell models a physical lens because i would guess you would end up with other lens type aberrations.

It's probably natural vignetting which is more a mathematical issue with lens design full stop, even a perfect lens. But again im not sure why maxwell can't compensate - it's not like it's an unknown amount for a given focal length.

perhaps im missing something?

DELETED

max3d wrote:What I was wondering: is there really a strict rule concerning vignetting. I always thought that in the real word it's was a matter of quality of lenses. Do you refer to a perfect lens?

Well, in the real world vignetting is quite a complicated issue. In theory, it it relatively well understood, and I found a great reference here -> http://www.vanwalree.com/optics/vignetting.html#natural

Such artifacts occur in Maxwell, and all I did then was to develop a setup that demonstrates if it was actually cos^4 in Maxwell, which isn't such a major task with some background in physics.

So thanks for your kind words, but we all are standing on the shoulders of giants.

Cheers

Kabe

Then again vignetting in Maxwell could actually be a problem with the render engine giving an illusion of vignetting.

Becco_UK wrote:Then again vignetting in Maxwell could actually be a problem with the render engine giving an illusion of vignetting.

Well, please read the referenced article. If you have a flat diaphragma and an illumination model of the camera - which M~R has -, then natural vignetting is to be expected.
Also have a look at my image: It proofed that the intensity falloff is following the cos^4 law and is not angle dependent. This is an important point because it means that this can be compensated.

Kabe

Back to the original thread topic

max3d wrote:Thomas started a very smart initiative in http://www.maxwellrender.com/forum/viewtopic.php?t=8857

The initiative is smart indeed, and it delivered already a couple of answers.

I'm not so shure about this quota thing. To consider a quota met with 3 questions in a week is pathetic in their situation, exspecially if this becomes a "Tom's answers only" session.

Though I would bite my tongue if the RC would actually hit my door on friday, it is NL which has promised more info numerous times and delivered sparsely.

Kabe

Yes I think so too...perhaps it's best not to keep it so strict, if they feel like answering more questions, please go ahead

I did read the first answers by tom and i am puzzled!!

No fog?? (we were told than no displacement)

I also expect to hear no ies, no texture baking, not this not that.

And the question is

What have they been doing all that months? Just making the new material editor? The 3d env was ready from realflow, so i wonder what they have been doing?

When is see " Feature XX will be provided in versions 1.xx" i am very scared.
It like saying it will be provided in the comming years.

daros wrote:They are lenses, like some Zeiss for Arri cameras that have total vignetting compensation. Some Zeiss lenses for Arri have also a total light halo suppression.

Thanks for the hint. I am not really acquainted with film camera's, my reference were photolenses.

Kabe wrote:

max3d wrote:What I was wondering: is there really a strict rule concerning vignetting. I always thought that in the real word it's was a matter of quality of lenses. Do you refer to a perfect lens?

Well, in the real world vignetting is quite a complicated issue. In theory, it it relatively well understood, and I found a great reference here -> http://www.vanwalree.com/optics/vignetting.html#natural

Such artifacts occur in Maxwell, and all I did then was to develop a setup that demonstrates if it was actually cos^4 in Maxwell, which isn't such a major task with some background in physics.

So thanks for your kind words, but we all are standing on the shoulders of giants.

Cheers

Kabe

Thanks for the link. So vignetting is theoretically a 'natural' thing, but in practice corrected by high end lenses. This raises the question how real we want camera's in Maxwell to behave? Halo's can be fun, but tend to come in unexpected. Chromatic abberation is a pain in the ass. The typical loss of sharpness in corners with wide open cheaper lenses is again a pain in the ass.
Which features of 'real' lenses should be emulated? Flaires can be usefull, bokney is important. What else?

Xlars wrote:Yes it is a super idea with Thomas' thread ... and also this one to keep all the "talk" away from the Q/A.

I suggest that as soon as question is answered in Thomas' thread is is removed from the thread (if possible?) .. so as to constantly keep the thread "clean". This will also help in people not posing the same question again as the thread grows. But migth require someone with administrator rights to do that or ?

Just a thought.

I opened this topic to keep the clutter away of the main thread, but that's not been completely succesfull.

The way Thomas summarizes the questions and their status seems very good. Better than removing the questions as you can always go back and see what the original and more detailed question was.

The maximum quota seems nonsense to me. NL answer what they can. They promised at least three answers a week. That's fine, but I don't see any reason why they wouldn't answer lots of easy ones in a row.