vinegar bottle

[s0real]

here's a bottle i rendered today.
hope you like it.

[heroes5]

i like it,the price? :P

[J_Bug]

Wow that looks Sooo real

The liquid is great but kinda like oil though. Did you find the IOR of vinegar somewhere or were you just going by look?

--Jon

[s0real]

had a bottle in front of me...

i agree it does look a bit 'thick' though.

[dbuchhofer]

My only comment would be that the label looks like it is part of the bottle.. a small clue that you could add would mabye be to have the label have a smallll thickness, or lightened color along the edges. or mabye having a backside of the label slightly visible around one of the edges..

[Matt_3D]

Looks very nice

I agree with Jon regarding the thickness of the liquid, but other than that it looks awesome.

-Matt

[zuliban]

looks very nice but i thnik the caustic is too strong, the scene almost have no lights or shadows to have that casutic ,nice glass material btw.

[s0real]

thanks for the comments.

i think the label looks ok IMO...

my aim was to give it that 'studio photoshoot' look, which i'm quite happy with.

i used a direct spot for the caustics + HDRI overide for GI + reflection.

i agree that the caustics look a bit on the strong side, but i don't think they detract too much from the image, and clients always love shiney reflective things and nice caustic effects!

thanks again

[J_Bug]

I really like the way your background fades from the champagne color to light blue. It's very elegant and natural. I also like the noise which looks intentional. Along with the subtle color variation in the spec it really gives the feeling of a well thought out studio setup. Good job all around. Is that a custom HDRI?

--Jon

[s0real]

no it's not custom, i found it somewhere after looking on google for HDRI i think...
it's the first timne i've used HDRI, and i like it! although i don't really understand how it works propery yet...

how do you create custom HDRI maps?
is it easy?

[J_Bug]

No I wouldn't say easy but not so hard. A range of photographs can be taken at different exposures and edited together in HDRSHop or an HDRI can be created with a 3D rendering engine by calculating unclamped colors.

--Jon

[s0real]

so could a HDRI map be substituted with a panoramic jpeg set to spherical environment with an output over the jpeg to control the intensity?

what's the difference between doing that and using HDRI?

::edit::

ok, i did a quick test (without caustics)



image1 = HDRI 17m41s ------------------------------ image2 = JPEG 7m39s

[mc-escher]

in a JPG, each color channel has values from 0-255 meaning you can
never have a pixel darker than 0 or brighter than 255. However, we can all tell the white in the Sun is Much much brighter than the white of a sheet of paper, and this is not captured in jpg images.

HDRI is a method for capturing at least a little more information than exists in
these types of formats, by providing a floating point format for each color channel. thus I can represent color intensities of .001 .1 1000 10000, etc.

Thus with this in the environment slot, you can see how a brightly encoded sample can have a more realistic effect on your vinegar bottle as in the image on the left.

for more information on this topic, a great place to read about it is:
http://debevec.org

Hope it helps! (By the way, your image has a great feeling to it)
Escher

[s0real]

but if you put an output channel over the jpeg, it's possible to tweak the intensity of the rgb curve/values and the output intensity, isn't it?

wouldn't that offer more control than using HDRI?

::edit::

just had a look at the debevec page... looks pretty technical!

i'll have a proper read through it when i find the time.
at the moment i'm just trying to understand the basic principles.

thanks.

[mc-escher]

It is possible to get a slightly more realistic result, but what you're really doing is driving your image into saturation by doing that. what you will truly be missing is the high contrast in the light to dark areas in your reflections (Seen by the flat reflections on the second bottle)

Another test you can play with is if you have a translucent material with an environment shining through from behind it. even if you up the output parameters and adjust curves, the light will get uniformly darker because of your material, but what you'll see with HDRI is that the brightest area's are dispersed more like real light would be (instead of just darkened)

Also, if you place some defocus on the camera with HDRI, the light will be dispersed to get that halo effect our eyes associate with an image that was
taken with a very bright light source.

If you want to see some pictures of what I mean, check this PDF and scroll down to page 49:

http://www.debevec.org/IBL2001/Image...2001-part1.pdf

I'm not saying you can't get something that looks cool with your method, only that with a jpg image all you have is the information that is the small
dynamic range that was captured when the image was taken, but not what fully exists in the real environment.

For instance pretend you have a camera and have taken 3 pictures: 1 with aperature wide open (mostly saturated), one with medium aperature (details in image well balanced) one with aperature mostly closed (very dark image). With the aperature on a camera wide open, only the darkest portions of the image will be seen, but detail in darker portions [read: contributions to lighting solution from those areas] will become apparent that were not visible at a normal aperature. Likewise, as you drop the aperature size to the smallest setting, you'll see most darker areas fall off to black, but the brightest areas will now contain detail that could be used in your scene lighting solution. (For instance something that looks purely saturated in your jpg might actually have contained leaves or something contributing green in the real environment. which your eyes would pick up,
but the camera would not.)

So it seems, to me at least, with a jpg you will never have very much more than a flat lighting solution. (adding your own lights of course could change this)

Of course, there are several proponents on this board who would advocate
making a realistic setup without using HDRI at all, it's just a convienent way to capture the lighting conditions in the real world and translate those to the virtual.

I guess I'm rambling
Escher




That dynamic range will never be captured.