The Renderer's Rings

[WytRaven]

the ray nudge distance of 0.1mm, right?
Just thinking...

I think it's 0.01 mm (1 hundredth of a millimeter).

[PureSpider]

Nope it is 1.0e-4 m = 0.0001 m = 0.1 mm by default

ray_origin_nudge_distance
Ray origins are offset by this distance after intersection, in order to avoid false self-intersections.
type: scalar real
restrictions: must be >= 0
units: meters
default value: 1.0e-4

Blendigo default is 0.00001m = 1.0e-5m = 0.01mm
If your diamonds are just 1.6mm in diameter, 0.1mm on either side is fucking much.

Actually - with the help of CTZn I found out that scaling the diamonds down just a tad (probably below ray nudge distance) causes a huge difference in their refractions (as opposed to coplanary diamond/socket faces).

[fused]

* Changed self-intersection avoidance code.

With the changes the the self-intersection avoidance in 2.0.8 it would be interesting to know if the ray nudge distance is still relevant at all.

[WytRaven]

The diamonds are 1.2 mm in diameter and the distance between the stones and their sockets would be measured in micrometers I would guess. It's to small for me to be able to actually give you a distance. All I can say is that they are "floating" within their sockets. There is no intersection between diamond and socket.

The ray nudge is blendigo default: 0.00001

I'm not sure what you are replying to PureSpider as everything you just said agrees with what I said earlier...

[PureSpider]

I just confirmed that a gap of width < ray nudge distance actually changes the rendering result, wasn't sure about that one before!

[WytRaven]

So if you look at the floating glass test render on the last page do the diamonds look as though they are suffering this coplanar issue or not?

[CTZn]

The three stones perhaps yes, but that would take a test render to say.

[PureSpider]

See the renders for comparison.
And not, it's not cause of the SPP, the refractions converge relatively fast.

[WytRaven]

Yeah I can see what you mean...

I'm about to start remodeling the silver ring so I will try some different gaps myself and see how I go.

BTW I'm in IRC at the moment if you want to chat about this.

[WytRaven]

Ok so I have completly remodeled the silver inner ring.

I decided to do some decent research of jewelry design and in particular setting design before going ahead and discovered a couple of things worth noting.

• Faceted translucent gems of any resonable size are always mounted with open backs
• Strips of faceted translucent gems are generally set in slots or in raised claws

So taking these things into account I completely changed the design of the inner ring. I increased the size of the diamonds and therefore went for a single strip instead of three as creating three slots would have made for a ring that couldn't really be physically made; the silver would get too thin.

Despite the open back principle above the double ring design doesn't really allow for it so I have provided plenty of room in the settings for light to get in behind the stones instead where it can reflect off the inner part of the gold outer band. I also slightly increased the radius of the bevel on the inside of the gold band, but only very slightly as when looking at my wedding ring the bevel is very small (and no it doesn't hurt me )

The one last tweak of note is with the diamonds themselves. In the test render for the floating glass setup the uniformity of the diamonds sticks out like a sore thumb so each diamond in the new design has been rotated along it's axis somewhere between -1.5 and 1.5 degrees.

I said I wasn't going to use the floating glass technique again but with a little tweaking I managed to get it to perform at a reasonable speed and the double reflection has actually grown on me so I am rendering the new ring design against floating glass.

In the meantime here are some shots of the new ring design. There is a cross section of the diamond slot, a cross section of the gold band, and total shot showing the claw mounts for the three primary stones.

I tried to keep the claws small and unobtrusive as claws and raised settings are more of a feminine thing where this ring is supposed to be masculine (I'm not sure the masculinity has been preserved unfortunately)

NewInnerRing.jpg (135.26 KiB) Viewed 17277 times

[Stur]

WytRaven, the expert of technical drawings

Very nice and precise modling Wyt

[PureSpider]

• Faceted translucent gems of any resonable size are always mounted with open backs
• Strips of faceted translucent gems are generally set in slots or in raised claws

Could have told you that...

Nice design anyway, let's see how it will turn out

[WytRaven]

Latest test of new design. Just need to do the texture mapping and re-render and it's about done I think.

RenderersRingsTest2.jpg (172.57 KiB) Viewed 17433 times

[suvakas]

Is there a texture map with a noise pattern on the inner side of the ring ? That kind of noise looks a bit like Vray rendering artifacts caused by a low irradiance map settings for glossy reflections. Brings down the visual quality for me. Has a biased feel or smth.

[WytRaven]

No this is without any texture mapping. It just hasn't converged fully yet. This is at ~10K spp.

EDIT: (Well I certainly hope that's all it is and not some weird artifact....I just noticed a distinct pattern in the noise...)