[Maxer]

I think you should prove your point by testing your theory. Go take a picture of a room then render it in Maxwell using all the same camera settings and then present it to us. Right now you’re just making unsubstantiated claims without any proof.

[giacob]

i dont need any proof to demonstrate the fact that maxell is biased... how can a render that dont let the light pass trhough glasses be defined unbiased.. defining it so is simply ridicoulus
the calculation method is unbiased, perhaps, but the risults are more than biased

[iandavis]

Mihai,

Actually your eye can not see that much of a dynamic range... granted it's 1000:1 or so... which kills photography and video being only a few hundred to one...

No, your brain compensates for your eyesight to such an extent that our crappy little lenses seem almost magical. We only see a degree or so but since we move our point of focus around so quickly our brain builds a picture of the world around us so convincingly that we imagine we see everything in focus, where in fact our focus range is (generously) the size of the moon.

As for contrast and light levels... In a dark room during the day we quickly change our 'exposure' when we glance out the window, however, if you pay careful attention to the room around you while doing so you may see the details of the room suddenly darken. No, unfortunately, though the eye/brain visual system is remarkable it actually isn't too much better then maxwell for dynamic range... in fact an HDRI contains more information then we can see AT ONE TIME, though we can see millions of distinct shades of colour we still can't effectively see light levels spanning several THOUSAND times between dark and light simultaniously.

With Maxwell, generally people are reproducing what camera's can see, though it is much less of a range then we can comfortably see it is done for the same reason CG includes motion blur. Weakness in our own visual system and that of photography convinces us that this 'render' is either a photograph, or reality. Since we don't often see 'reality' outside of reality we should base our renders on the limitations of photography.

In order to match the light levels of outdoors at noon you would need several 1500W spots and various reflectors etc.

In a round-about way this answers the poster's original complaint about maxwell being biased. Maxwell is lighting the scene purely based on the behaviour of light and physics. Our brains don't see the world realistically, we see interpretively. We we lightbalance interiors instantly and since we don't notice the light shift when we look outside, we assume the light levels to be similar... they are not. Outside is often thousands of times brighter then inside. If WE saw the world unbiasedly WE would see the scene identically to what maxwell produces. It's not maxwell that's biased... it's US. So... in order to create a scene that we think is 'realistic' we in fact have to cheat.

I don't mean to be harsh, but the whole thread is based on something which is just plain wrong.

Cheers all.

[Mihai]

i dont need any proof to demonstrate the fact that maxell is biased... how can a render that dont let the light pass trhough glasses be defined umbiased.. defining it so is simply ridicoulus

unbiased renderer != complete reality

Rendering complete reality is not possible but the methods used can get very close.

Unbiased rendering simply means that using the same settings, you will get identical renders while a biased one will be slightly different because it guesses some values to speed up rendering.

Like iandavis said you are judging an image based on what your eyes see which is kind of ridiculous. Don't compare it with professional photographs either because I think you will find they use a lot of extra light just to make the photo ressemble what our eyes (or brains) would see.

[giacob]

jandavis instead of repeating your confused an poor lectures on the issue u better read well what i wrote.. i was taking of a room with shutters more than half down and lightened just from indirect light at noon
...u are right ....a so called unbiased render that dont let the light pass trhough glasses is quite far from beeing real
...... anyway everybody is free to bealive to flying horses!!

[giacob]

"Unbiased rendering simply means that using the same settings, you will get identical renders "
i really dont think so....

[Mihai]

i really dont think so....

Why not?

[Maxer]

giacob you do need to prove your statement otherwise everyone here will just assume your a pissed off Vray user looking to stir up some sh*t. If you believe Maxwell is biased then give us some proof. Also just because there are some bugs currently in the system doesn’t mean that the whole thing doesn’t work.

[giacob]

giacob you do need to prove your statement otherwise everyone here will just assume your a pissed off Vray user looking to stir up some sh*t. If you believe Maxwell is biased then give us some proof. Also just because there are some bugs currently in the system doesn’t mean that the whole thing doesn’t work.

u are right i am a malicious secret agent of vray producers making the dirty work here... but the enormous bug .. repeat ENORMOUS ...of maxwell help me a lot
a propo maxwell would like to be unbiased but is still surely not...o sorry i was wrong ..i can see a n horse flying.. yes u are right maxwell is unbiased

[giacob]

so mihai if i would make a render engine in which using the same settings, you will get identical renders, for instances all blacks and red and yellow u will have another umbiased render engine... i'll try
i forgot u was the one who tried to demonstrate that constant could also means three time in a millennium

[Maxer]

a propo maxwell would like to be unbiased but is still surely not...o sorry i was wrong ..i can see a n horse flying.. yes u are right maxwell is unbiased

I have no idea what you’re saying, calm down and type a little slower.

[Bige]

uhm... giacob my tiny little student room (10m2) has on one side a giant window along the whole lenght. I have shutters in front of the window. And still my room can get pretty dark with difus light. Only with direct sunlight things get light. Besides that if you take a picture with a normal camera you will see nothing becouse the contrast is to high. Try it with your camera, you can not take good pictures with light comming towards you.

[Micha]

My expirience: if you like to get a lighting like the human eye it see (you can see indoor and outddor at the same time), than you should try a very low burn value (0 ... 0.1) and adjust the contrast with the gamma and the brightness with ISO/shutter. I'm very impressed by the burn option. I have used it in my image here:
http://www.maxwellrender.com/forum/view ... 92&start=0
Without a low burn it was impossible to get a bright lighting of the room and no lost in the bright areas of the window.

For me is the burn value a great option that Maxwell user ignore for a to long time.

[otacon]

Doesn't the fact that you can create renders that look fake mean maxwell is biased? Not trying to argue, just wondering....or am i missing the definition of unbiased?

[aitraaz]

i'm assuming unbiased in terms of calculation is meaning something like no use of an stochastic/random calculations therefore the solution always converges to one solution without any computational shortcuts...as far as this leading to a 'fake' or 'non-fake' solution that's a whole other issue...