All posts relating to Maxwell Render 1.x
User avatar
By andronikos916
#62276
I do not know how the hell I missed that thread...very helpfull indeed!

thank you guys,
Andronikos :lol:
User avatar
By goncalo
#68866
Hy thomas can you share that scene in other format please, the Mac people can't decompress using .7z

Thanks
User avatar
By Thomas An.
#97161
This is a new test comparing:
  1. Photo
  2. Maxwell alpha 1.1.33
  3. Maxwell RC3
All values/scene/parameters are identical
(using plugin material values).

Image

Maxwell Alpha 1.1.33 (f5.6 and s30, @ 12 bounces)
Image

Maxwell RC3 (f5.6 and s30)
Image

It seems that the RC3 engine behaves quite differently especially in regards to caustics and dielectrics and that the findings in the beginning of this thread might now be obsolete. In other words, the new engine might be using a new mechanism to handle dielectrics.

Caveat:
  • It is possible that the discrepances might be attributed to faulty built in translation of materials between the plugin and the new material editor in RC3. Even though I checked to make sure the ND values transfered properly for the RC3 render, the refractions seem off.
  • Also, I show somewhere that objects partially intersecting with trasmissive objects will produce bad results, which might play a factor here because of the red straw penetrating the liquid
Last edited by Thomas An. on Sat Dec 10, 2005 7:56 pm, edited 3 times in total.
User avatar
By Thomas An.
#97163
Here is the same RC3 image using a slower camera shutter (s5) so as to better observe the faulty refracted images in the liquid (through the top of the liquid and through the side of the glass).

Image
Last edited by Thomas An. on Sat Dec 10, 2005 8:05 pm, edited 2 times in total.
By lllab
#97167
not good....

the alpha beta was sooo close to the photo. the rc looks like a rendering. caustics are much to strong too.

cheers
stefan
User avatar
By psanitra
#97172
We all know how bad current RC engine is . This is another proof of it. :cry:
By JesperW
#97177
Very interesting. I was not around at alpha time, would you say the beta would have been as close to the photo as the alpha? I understand the alpha was slower still than the beta.

(And cred to you for having the energy to do tests with RC3, since NL have said themselves that it's crap. I suspect the tests are still useful to them. Myself I'm busy with other pieces of buggy software :? )
/j
User avatar
By Thomas An.
#97178
JesperW wrote:... would you say the beta would have been as close to the photo as the alpha?


Yes, the beta would be close to the alpha result.
I understand the alpha was slower still than the beta.
Nope. As far as I can tell the alpha was even faster (provided we made judicious use of the bounce control)

As far as I am concerned the Alpha was the best Maxwell ever (and it had working clipmaps and volumentrics) they should have backaged it and sold it as Maxwell Lite....
(And cred to you for having the energy to do tests with RC3, since NL have said themselves that it's crap. I suspect the tests are still useful to them. Myself I'm busy with other pieces of buggy software :? )
/j
Thanks Jasper :)
The only reason for doing this test was to keep a log or progress report and see how RC4 (or 5 or 6) will differ (if at all)
User avatar
By rivoli
#97179
back to glass/water again. thanks for keep testing this thomas, looks we have to start it over again. do you think that refractions are totally off because of rc limitations or it may be the old method that doesn't apply with the new engine?
User avatar
By aitraaz
#97182
Thomas An. wrote:
As far as I am concerned the Alpha was the best Maxwell ever (and it had working clipmaps and volumentrics) they should have backaged it and sold it as Maxwell Lite....
Agree as well, but that photo looks a bit cg'ish :shock:

RC3 caustics seem to be on steroids :shock:
Last edited by aitraaz on Sat Dec 10, 2005 8:56 pm, edited 1 time in total.
User avatar
By rivoli
#97184
Aksel Karcher wrote: Talking about this, does anyone know where to get the Alpha and the according Max-Plugin back from again?
i still have them all backuped in my drive, but i doubt we are allowed to share them (of course not here on the forums).
what do you think tom, are registered users supposed to share old releases?
User avatar
By Thomas An.
#97185
rivoli wrote:back to glass/water again. thanks for keep testing this thomas, looks we have to start it over again. do you think that refractions are totally off because of rc limitations or it may be the old method that doesn't apply with the new engine?
I don't know yet. It could be either or.

It is possible that the new engine uses a different mechanism. So, when I find some time I will have to restart this test (particularly the ray/beam experiments of the front page... but I will not do anything now untill we are at RC200 or something :twisted:
User avatar
By Thomas An.
#97187
aitraaz wrote:but that photo looks a bit cg'ish :shock:
It is a true photo, with no post process other than converting to grayscale.
(This photo was a controlled experiment. If you followed this thread you will see how it was made)
  • 1
  • 10
  • 11
  • 12
  • 13
  • 14
Will there be a Maxwell Render 6 ?

Let's be realistic. What's left of NL is only milk[…]