[Offer] 3d artist willing to make simple stuff free. ask away.

[South8028]

Wow, I hate to be right. May your recovery be swift & complete (if possible).

 

Your right about the triangle reduction being tricky as blender doesn't do a perfect job. At times it creates these weird spots with lots of overlapping tiny faces. So it always requires some manual correcting.

 

I usually do it in steps, as that helps. Using vertex groups helps too, although thats a bit tricky as its not fully implemented yet into blender. (you can't turn a group back into a selection yet for example)

 

I (re-) create my UV's after decimating the mesh & in case of the 3D scans, I bake the difuse map as well as a normal.

Works like a charm. Especially since those scans usually have a lot of tiny parts, which combine into bigger UV's.

in max you can just click on the edge loops by holding ctrl and pressing del. With practice, it is not difficult to manually optimize the model in 5 minutes, for example, from ~ 100k polygons to ~ 5k polygons. UV can be done once and the unwrapping will be saved during optimization, and vice versa, with turbo/mesh smooth. But, for optimization, a model can be considered buggy, where oblique edges are already formed. Model for optimization and vice versa for turbo/mesh smooth should have only rectangular/square polygons.Oblique edges should not be in the source, they are automatically created when converting the model to fbx / nif. Therefore, it is necessary to store the source files in clean formats. UV cuts the model into pieces and it is much more difficult to optimize it. Therefore, optimization should preferably be done before uv. Edited June 16, 2023 by South8028

[RoNin1971]

 

Wow, I hate to be right. May your recovery be swift & complete (if possible).

 

Your right about the triangle reduction being tricky as blender doesn't do a perfect job. At times it creates these weird spots with lots of overlapping tiny faces. So it always requires some manual correcting.

 

I usually do it in steps, as that helps. Using vertex groups helps too, although thats a bit tricky as its not fully implemented yet into blender. (you can't turn a group back into a selection yet for example)

 

I (re-) create my UV's after decimating the mesh & in case of the 3D scans, I bake the difuse map as well as a normal.

Works like a charm. Especially since those scans usually have a lot of tiny parts, which combine into bigger UV's.

in max you can just click on the edge loops by holding alt and pressing del. With practice, it is not difficult to manually optimize the model in 5 minutes, for example, from ~ 100k polygons to ~ 5k polygons. UV can be done once and the unwrapping will be saved during optimization, and vice versa, with turbo/mesh smooth. But, for optimization, a model can be considered buggy, where oblique edges are already formed. Model for optimization and vice versa for turbo/mesh smooth should have only rectangular/square polygons.Oblique edges should not be in the source, they are automatically created when converting the model to fbx / nif. Therefore, it is necessary to store the source files in clean formats. UV cuts the model into pieces and it is much more difficult to optimize it. Therefore, optimization should preferably be done before uv.

 

 

Not just preferably. Its pretty much the only way, as the optimization will affect the UV. Which in turn is crappy before optimization. Especially with "complex" models like stalagmites. In this case I define the seams myself, as blender doesn't know how to do it properly in these cases, ending up with a lot of small pieces.

 

In combination with that I just love the option to bake a new diffuse, with new UV, from an exisitng texture on a high poly model with a crappy hopelessly cut up UV, like the 3D scans.

 

edit:

In case of the big pillar, I created a cylinder first & made the UV fit the emtire texture.

Only then did I start increasing the poly count (to 1.5 milion) & apply a displacement map (which uses the UV too)

Then decimated it (a copy, of course) in 2 steps down to just under 6K & optimized it (removing another few hundred faces)

 

During the UV optimazation I only reduced the stretching a bit (not to much as the displacement map is based on the texture and should 'match') & straightened the edges to fit the entire texture again.

 

& finally baked me a normal to fit with that.

 

edit2:

As a footnote: The texture for the pillar is based on a picture from a fairly straight/flat wall with the 'drippers' running down. Not an actual stalagmite. I just made it seamless horizontally and wrapped it around the cylinder. Hoping it would look exactly like a huge stalagmite. (as the two are/look much the same, except for the latter being round.) & it worked perfectly, imho.

Edited March 24, 2023 by RoNin1971

[niston]

Fernicles: Could you (roughly) model a PARSONS 500MW turbogenerator set, as it was used at Fawley Power Station?

Essentially just the blue parts in the pics, and the large piping that joins them.

 

 

 

 

 

 

These are (as visible from front to back): Bearing, high pressure turbine, intermediate pressure turbine, followed by 3 low pressure turbine stages.

 

The open, blue boxes towards either side at the front contain the governor steam valves. The valves would normally be covered, but are exposed here due to the unit being dismantled.

The small, grey cabinet in the middle probably houses lubrication system components. And the 4 large, pale blue things (2 on each side) that are inset into the floor are the front covers of the steam condensers.

 

What's visible is only about the upper 2/3 of the genset; The rest is set into the floor.

 

The far end:

 

 

 

 

 

On the far end of the arrangement is the generator, the exciter and another bearing. These, and the parts from the upper picture, are connected by a shaft that runs the length of the entire set.

Looking at the steam/water feed schematic of the system, the last blue box on the axis appears to house a steam driven feedwater pump. I don't think it's connected to the turbogenerator shaft.

Maybe I can find out more about it.

 

The 3 large, gray pipes emanating from the generator are gas insulated high voltage conduits; These are basically a wire inside a gas (SF6 if I'm not mistaken) filled pipe, each.

Pretty advanced for it's time - These turbogenerators were built in the 60s.

Also, the generators really are alternators, so technically these are not turbogensets but turbo-alternator-sets. Heh.

 

 

Front end with covered steam chests:

 

 

 

 

More pics here: https://www.powerstations.uk/fawley-power-station-turbines/