Is there a standard format size for best quality?

Basically it's anywhere between 1000x1000 to 3000x3000.

Thanks, Greatly appreciated.

It also depends upon what you are texturing. Humans need higher resolution maps than furniture, but a complicated alien spaceship interior or blasted boiler-room wall might require as much detail as the hooded and suited figures. A castle in the distance might not need every stone lovingly painted... unless your director is Peter Jackson and he's going to move the lipstick camera over its walls. ;^)

It ALSO depends on what your final output will be. For example, if you are rendering for a video, the final output is 720 x 486 or thereabouts. A texture that is really large is really wasted for the most part. On the other hand, if you're going to zoom in on an eye, the eye texture might have to be around 1000 pixels high (or double the height of the final output) for best results. On the other hand, if you are going to render a closeup for print output, you will need the highest resolution textures you can find. A 2400 x 3000 pixel image will end up being 8 x 10 inches when printed at 300 DPI, so you will have to use textures that will look flawless when rendered to pixel dimensions that large.

Some more advice yet: How close is each object going to be to your camera and are you going to be using depth of field so that distant (or extremely close) objects will be blurry? Those objects can and should have lower resolution textures to save on rendering power, since high detail on them will be wasted overkill. Next: I don't know for sure about poser's rendering engine, but almost all the others I've worked with are "happiest" in terms of efficiency with textures that are square and in the "magical" powers of two (256x256, 512x512, 1024x1024, 2048x2048, 4096x4096, etc.) You can enter other values, but I'm fairly sure the rendering engine will round off and interpolate to the nearest power of two, which can also be wasteful to you in terms of time and detail if your selected rez if far off from one of these. (Again, not positive about poser's renderer, jusr renderers in general.)

Also... is the texture detail REALLY necessary? A painted texture for skin has less information than one made from a series of photographs. Same for tree bark or beach sand. How much detail does your source have... and how important is it that it be preserved? Texture maps are huge memory hogs, and the bigger they are, the greedier they are for system resources. Let's say that you have a man-in-space wearing a typical puffy white suit and reflective faceplate. You could do that as an all-inclusive 4000X4000 tex map, plus bump map, plus reflection map. Ouch! That is more wasteful than having a full map just to do Vicky's eyelashes! If you mapped him in sections, you could have a 1000X 1000 map plus another bumpmap for the suit... and maybe even smaller, since the wrinkles behind his knees aren't very important. ;^) Then have your separate 1000X1000 map for the head/helmet, plus bump, plus reflection. By bringing the size down to something manageable, you can render the space station with figures all over it doing assembly or repair or experiments... you can have better looking figures... and your computer won't whimper and die. Carolly

The standard way is to make a small, simple texture with a few colors and no details. Then resize it to 4096x4096 pixels, and magically the fact that the texture map will now eat all your RAM and disk space somehow means that the texture is ultra realistic and great for close-ups.

LOL! PhotoShop will "add" detail when enlarging like that, but not very well. There is a fractal-based program for another $500 or so which does a great job of guessing at the additional data content. However, for that kind of money, one can buy a decent digital camera. ;)

Nutz! Coulda sworn the answer was "7".

Lots of good tips, but no specific answer really exists. As was suggested by several above, "best" size would be one that is close to the size that the area will appear in the final render.

Some have suggested a map size twice the resolution that the area will appear in the final render if you want some smoothing, or just nail it at the same size if not.

(clear as mud?)

If your figure's iris is going to cover 50 pixels in the final render, a UV map size that yields an iris area 50 pixels across would give you the most accurate application of the map with the least scaling distortions. However, because the UV template usually combines numerous materials at different scales, it's not likely you'd ever get them all to match without using a different map for each material in the UV template.

Not practical for numerous reasons, so no one really approaches it that way, ...but hey, you asked.

As Carolly suggested, I also suspect that in practice, most folks tend to lean toward "bigger is better", willing to sacrifice performance for detail, even though those details may often not be visible in the final composition.

With all that said, I usually end up making Victoria maps at 2048x2048.

Just to add to the confusion... - IIRC, P5 (and most render engines) prefer texture maps to be square. - They also like them to be in sizes which are powers of 2. (ie 2 multiplied by itself some number of times. The progression starts: 1, 2, 4, 8, 16, 32, 64 (<- these are probably irrelevant) 128, 256, 512, 1024, 2048, 4096 (just keep multiplying by 2)... That's because these are "pure" binary numbers, and computers munch away happily on "pure" binary. (Not only that, but they usually convert the actual texture size to the next value up internally, even if it just means adding a lot of blank areas around the edges. You never see it, but the resources are used up none-the-less...) The real answer is: there is no answer. Objects that are in the foreground need bigger maps, because otherwise detail gets lost. Those in the back-ground can use smaller maps, and should do so, because otherwise resources are being wasted. That's for still pictures. In animaton, things are a bit different. - Animation can get away with smaller maps, 'cos the viewer's eye/mind fills in the details and ignores imperfections. - Animation tends to need larger maps because the objects can get closer to the camera during the scene... Cheers, Diolma PS - I may have been a bit simplistic in my explanation; that's 'cos I didn't want to baffle myself with science..:-))