Maxwell Render | Official public forum

Say you have flamed granite and you know that the average distance between the valleys and ridges is .75mm, it would be nice to tell MR that this is the total depth between the extreems of the bump map.
Just a thought, because the -999 to 999 does not really mean much in real world terms...

Hi stonelli,

This is a relative coefficient and it depends on bitmap size.
I mean it's not same setting bump value 20 for 100 x 100 and for 1000 x 1000.
So, you have a large range to be able to finetune better.

Best regards,
Tom

Hmm, sounds difficult to use.

So, if I find a good value for ground plates in a scene, I must search a new value in the next scene with an other map. I would wish, it is independent from the map size and the values are not limited by 999.

tom wrote:Hi stonelli,

This is a relative coefficient and it depends on bitmap size.
I mean it's not same setting bump value 20 for 100 x 100 and for 1000 x 1000.
So, you have a large range to be able to finetune better.

Best regards,
Tom

Yeah, the problem is that it is a "shot in the dark" approach whereas one generally has an idea on low to high depth he/she is trying to achieve.
Since you guys are going for "real" I hope you will give this a bit more thought, since it is, I feel, a more realistic approch than the relative coefficient one

Along the same lines -

For roughness we have U and V - is there a 'real world' way of expressing this.

For example an injection moulding usually has a 'spark' texture on it - caused by the way the tool cavity is made. Little sparks of electricity come frome from a copper electrode and chip away at the tool metal. The copper electrode is the shape of the thing you want to mould.

I have my Sparcatron Spark Texture guide in front of me now -

VDI 12 is very fine almost satin. - Ra - 0.4 micrometers
VDI 45 is a very coarse texture - Ra - 18.0micrometers

the most common on consumer products is about 30 -33 ( Ra -3.15micrometers)

It would be nice to just select a VDI number as a form of roughness and then it is independant of the size of the component.

God know what difficulties there are in implementing this

TTFN


Jonathan

jjs wrote:Along the same lines -

For roughness we have U and V - is there a 'real world' way of expressing this.

For example an injection moulding usually has a 'spark' texture on it - caused by the way the tool cavity is made. Little sparks of electricity come frome from a copper electrode and chip away at the tool metal. The copper electrode is the shape of the thing you want to mould.

I have my Sparcatron Spark Texture guide in front of me now -

VDI 12 is very fine almost satin. - Ra - 0.4 micrometers
VDI 45 is a very coarse texture - Ra - 18.0micrometers

the most common on consumer products is about 30 -33 ( Ra -3.15micrometers)

It would be nice to just select a VDI number as a form of roughness and then it is independant of the size of the component.

God know what difficulties there are in implementing this

TTFN


Jonathan

Good point.

Maybe, the roughness means 1 = 100% = sample angle 180°.

I would like to know, why NL decided for a map resolution dependet bump intensity? It's so difficult to find the right values. In one scene you can try to get the same bump effect with different maps, but you need different values.

And at a low sampling level or at a small test render with high SL is it very difficult to prove the right value of the bump. Sometimes I have seen after hours of rendering and after the noise is gone, that a small bump was to strong.

DELETED

tom wrote: This is a relative coefficient and it depends on bitmap size.
I mean it's not same setting bump value 20 for 100 x 100 and for 1000 x 1000.

Why would the resulting bump depth be dependant on bitmap size though?
If I have a 300x300 bitmap with a 20 greyscale value, that should push the surface say 20mm in. If I then change that bitmap for a 3000x3000 map but keep a 20 greyscale value, the resulting push should still be 20mm.

stonelli wrote:Say you have flamed granite and you know that the average distance between the valleys and ridges is .75mm, it would be nice to tell MR that this is the total depth between the extreems of the bump map.
Just a thought, because the -999 to 999 does not really mean much in real world terms...

The same approach (and units) should be used for displacement also.

jjs wrote:Along the same lines -

For roughness we have U and V - is there a 'real world' way of expressing this.

For example an injection moulding usually has a 'spark' texture on it - caused by the way the tool cavity is made. Little sparks of electricity come frome from a copper electrode and chip away at the tool metal. The copper electrode is the shape of the thing you want to mould.

I have my Sparcatron Spark Texture guide in front of me now -

VDI 12 is very fine almost satin. - Ra - 0.4 micrometers
VDI 45 is a very coarse texture - Ra - 18.0micrometers

the most common on consumer products is about 30 -33 ( Ra -3.15micrometers)

It would be nice to just select a VDI number as a form of roughness and then it is independant of the size of the component.

God know what difficulties there are in implementing this

TTFN


Jonathan

I not so sure this would be good for everyone. Industry specific units or conventions should be avoided where a more generic approach can be used. I for one wouldn't know a VDI from a VDU.

I think as Maxwell is based on the physical, whenever possible it should use S.I units. One the unit systems described here (link) would be more appropriate.

http://en.wikipedia.org/wiki/Roughness

Hugh - I agree with you - the base way should be to stick to Si units, if at all possible. Then people can add 'macros' etc in the future for industry specific conversions, such as VDI.

VDI numbers are just an expression of the Ra equivalent roughness which I mentioned. It is a 'jargon' measurement that most product designers who create injection moulds should know and which most toolmakers can use.

TTFN


Jonathan


Culled from the web -

http://www.rubert.co.uk - if you want to buy a gauge - mine was a cheap one at £70 - 10 years ago!!

Just as a thought - maxwell ought to approach Ruberts and see if there can be some sort of Ruberts - plugin for material roughnesses. I think this is the way things will go in the future.

"VDI = Verein Deutscher Ingenieure, the Society of German Engineers. There is a VDI scale for surface finishes, which is widely but not exclusively used in the EDM field" EDM - Electro Discharge Machining -