Forum Coordinators: RedPhantom
Poser - OFFICIAL F.A.Q (Last Updated: 2024 Nov 28 11:20 am)
I'm sure Dr. Geep knows more than I do, but I seem to remember that the angle start/end referred to the peak angle on the cones produced by the spot. The start angle refers to when the light begins to decrease, while the end angle refers to the outer edge of the cone. Thus, if you have angle start of 40 and angle end of 80, the light will be undiminished in the inner 40 (0 to 20 in every direction) degrees of the cone, and decrease linearly from 40 to 80 (20 to 40 in every direction), where it becomes 0. evr
You're right, evr. I just can't figure out the idiot values in poser and I'm fed up rendering every 20 seconds to see what effect each change has. I have a bad feeling I'm trying to get poser to do something it can't. As a photographer, I expect lights to follow certain rules, but poser lights have their own rules. igohigh, Thanks. I looked at geep's tute (after the half hour it took to load. LOL.), but it didn't tell me anything I didn't already know. I'm afraid I'm going to have to work with the light positions instead of Angle/Distance. Ah well, back to the grind. mac
The missing ingredient is the Shadow-Cameras. Each light has a corresponding shadow-cam which controls and focuses the shadow. The shadow is created as a seperate render and composited into the final pic. If you go to the camera list and choose the shadow cam for each light, adjust the percentage until the view is approximately filled by the figure which will cast the shadow. When you select the shadow-cam, the view will be thru it, as it is a camera of sorts. I received a lot of help from Nance on this subject. The damned feature is not documented anywhere I can find. The larger the End Angle of the light is, the more diffused the shadow will be. It is also good to experiment with the light's distance values.
Your pic shows the exact problem I was having (over and over) which is now a thing of the past due to proper shadow-cam adjustment. I have adopted the procedure of using a set of lights for illumination with no shadows, and a set of shadow-casting lights placed strategically to allow easy shadow-cam settings. Usually you will need to decrease the Percentage dial, which will cause the extreme foreshortening to lessen. I am learning to use P5, and to avoid the shadow-mapping by using ray-traced shadows.
The light settings DO affect the shadow-cam view, but the Percentage (in s-cam dials) ALSO affects it. Your pic above still has the "floaty-feet" shadow flaw. You still need to adjust the s-cam to be rid of that unnatural gap between the feet and the base of the shadow. It takes some experimenting to find useful correlations between all the settings.
If the shadows don't start at the feet, the min shadow bias is one parameter you could use to change that. Of course, if adjusting the view angle of the shadow cam works for you do that - if it looks right, it is right. The shadowcam is nothing weird, it really makes sense. The main difference between raytracing and shadow maps is how they determine what is in the shadow and what not: Raytracing looks at every surface point and then asks "Can I see the light from here? If not, I must be in the shadow." Shadow Maps start at camera view and ask "What can I see from here? If I can't see you, you must be in the shadow."
If the walls of the room itself are not going to cast shadows, you should disable Cast Shadows on that prop as well. The shadow 'map' has to be shared by all objects in the scene that cast shadows, so if you have some very large (or far away) objects in the scene, they can dilute the map size. The above is particularly true with omni-lights (there's a mini-tute on my artist page)... the tips Connatic gave above mostly just apply to spot lights (which is what you're using in this case, but if all else fails, you could set up an omni-light to cast shadows).
Cinema4D Plugins (Home of Riptide, Riptide Pro, Undertow, Morph Mill, KyamaSlide and I/Ogre plugins) Poser products Freelance Modelling, Poser Rigging, UV-mapping work for hire.
Spanki, Sorry. I forgot you. Unfortunately, the walls need to be set to cast shadows, at least, in the default state for the users. I have already made SET files which switch 'cast shadows' on/off for the walls and other figures, but that was something I just added as being possibly useful to some people. Nothing to do with this. I'll take a look at your tute though. mac
Further developments.
It seems that the shadowcams DO hold their settings (I think). It's just that they reset to default settings. Presumably this has to do with all the hidden Auto dials, which I think change when you change the light settings. (Anyway, on their own they do nothing).
Here's the Scale block from a shadowcam (name in poser = zoom - the getStringRes changes from omni to spotlight S cams)
scale scale
{
name GetStringRes(1028,32)
initValue 1
hidden 0
forceLimits 0
min 1e-006
max 100000
trackingScale 0.004
keys
{
static 0
k 0 0.5
}
interpStyleLocked 0
}
Note that initValue = 1 (ie. 100%), but the k value = 0.5 - or 50 % - which is what I had it set at. I'm guessing that everything reverts to default so that any changes made can be done with 100% scale/zoom. But why this should be, I haven't a clue.
I dunno. It's all a bit strange, to say the least, but I'm hoping to crack it sooner or later.
And all I wanted was a simple spotlight setup to throw into my pack. LOL
mac
Maclean, I didn't mean you had to use the default GROUND plain, I just meant put a square under what ever you are using for the "ground". Actually what are you using for the ground/floor? If you are using a square try replacing it with as solid figure, a flattened cube, and you will probably find that the more depth the cube has the better the shadows performe. There is no "min shadow bias" in Poser 4, don't know about PP or P5.
I forgot to mention the little detail about the s-cams' inconsistent naming behavior when changing light names and adding new lights. It gets to be a confusing mess real fast, unless you keep up with the renaming. The various companies which have published Poser have failed miserably at informing the users about this undocumented feature. I wish somebody knowledgable about the entire spectrum of details on the shadow-cams would divulge the secrets of poser lights and shadows. I have only scratched the surface, but what I have learned has gone a long way in speeding up the time it takes to get the render looking how I want.
I think the "Zoom" in a SHADOW_CAM_LITE parented to a LIGHT is actually a scale channel, but for some reason whan the SHADOW_CAM_LITE has a LIGHT as its parent it gets named "zoom", and when the parent is a spotLight it's just named "scale". Any way I think you would be better using all LIGHTS or all spotLights, rather than a mixture. Also unhiding the translateZ channel can be usefull.
I mix them sometimes and have no problem. The Infinite lights are good for sunlight and ambient light. Generally I use the spots. Sorry I made an error in the term for "zoom" and "scale" in the shadow-cams. Les, I love your Omni-Light! The dials are a bit confusing, but wow does that OL enhance bump-maps! Maclean might find your light useful.
Re posr #25. "Please explain the best way to unhide the TranslateZ channel". Save a pz3, in a text editor find the approprate SHADOW_CAM_LITE. find its translateZ channel, change the line "hidden 1" to "hidden 0". You can also construct a pose file to do this for files "in play", you have to reference each SHADOW_CAM_LITE by its internal name/number. Note that a SHADOW_CAM_LITE parented to a spotLight will have a space before the number, one parented to a LIGHT will not.
les,
The floor is a flattened cube. It's a body part of the cr2 room figure (with both inner and outer materials).
I am using all spotlights. 8 of them. But poser, by default, calls the first 3 lights LIGHT1, 2, 3, even if you change them from omnis to spots. After that, it starts calling them spotlight1, etc. AFAIK there's no way round this (other than hacking). And that's why the first 3 have Zoom instead of Scale.
connatic,
This is the Ztrans channel from a light (.lt2 file). Just change 'hidden 1' to 'hidden 0'
translateZ ztran
{
name GetStringRes(1028,14)
initValue 4.15881
hidden 0
forceLimits 0
min -100000
max 100000
trackingScale 0.001
keys
{
static 0
k 0 5
}
interpStyleLocked 0
}
mac
I got them working. Thanks, Les! I am trying the enhanced shadow-cams. The extra dials give them much more mobility. I used a stupid work-around, as I have no idea how to save a .pp2 file from notepad. I copied an existing .pp2 file, opened it by double-clicking, pasted your text file into it, saved it, and renamed it.
Mac, have you tried y scaling the floor, does that help? Connatic, glad you like the new SHADOW_CAM_LITE's, I was thinking of posting them in the free Stuff, but then I thought most people would not know what they were and just get confused. Speaking of confusion, please note that the numbering of the SHADOW_CAM_LITE's in the menu can get confused in some circumstances, and not match the number of the associated light. Fore instance, load a doc that contains only the 3 default lights, delete the 3 lights, now apply the ShaDCamsPlus2.pp2. The lights will be numbered 1 to 3, but the SHADOW_CAM_LITE's will still be numbered 4 to 6, they will still work OK, It just gets confusing trying to figure out which cam goes with which light. One of the major dissapoinments in my life has been that I haven't yet found a way to change the focal length of a SHADOW_CAM_LITE :( P.S. The focal length of a SHADOW_CAM_LITE attached to a LIGHT seems to be longer than one attachet to a spotLight!!!
'Mac, to use all spotLights, delete the 3 LIGHTS that Poser usually adds to a new doc by default, then just new spotLights via the Light Controles pallet' les, I tried this and it doesn't work. No matter what you do, poser names the first 3 lights LIGHT 1, 2 and 3. I even set the default state with no lights, then added them, but they're still called LIGHT1, 2, 3. Thanks for the files les. I'll try these out. The numbering of shadowcams is a bit of a nightmare. I noticed that it always starts from 4 when adding spotlights. I couldn't figure out poser's logic at all. That's why I renamed all the shadowcams to match the lights. It's a lot easier.
Attached Link: shadow bias
Sorry for the delayed answer...I don't enable email notification for every thread, I hate overflooding mailboxes.. Anyhow, the explanation of the shadow bias: Quick explanation: It is the value of how close an object can come to a surface and still cast a shadow. When you set the value too small, self-shadowing will occur. When you set it too high, shadows will not start where the two objects meet. Long explanation: See "Advanced Animation and Rendering techniques", pages 166 through 171. And no, it's nothing Poser specific. as far as I know, every renderer using shadow maps has such a value (not everyone may expose it to the user, though). I attached a link showing examples of shadow map bias in an early version of 3ds max.'When you set it too high, shadows will not start where the two objects meet' Thanks stewer, That was a big part of my problem. The Angle End seems to be the only way to cure it, but it increases the falloff too, unfortunately. But I still don't know what this 'value' you mention is or how I can set it. Nor can I find the long explanation. I looked through both poser 4 and 5 manuals and those pages don't deal with animation or rendering. mac
The shadow min bias value is in the properties tab of the Poser 5 parameter palette. I don't think it's available in Poser 4. The long explanation I was referring to is not in the Poser manual but in a computer graphics book for programmers called "Advanced Animation and Rendering techniques" by Alan Watt an Mark Watt.
Re post #39 (). "No matter what you do, poser names the first 3 lights LIGHT 1, 2 and 3". Mac, are you sure? What version are you using? This definatly works fine for me in Poser 4.0.3.126! I am talking about the internal names here, not the menu names. If you delete the 3 default LIGHT's and add some new spot's via the Light Controles, the menu names will still be Light 1, Light 2, etc, but the internaln names should be "light spotLight 1" ect. I keep a pz3 with no lights in my Poser directory so if I only want spotlights in a doc I just load that, it definatly works for me. I am using Launch to Prefered and Use Previous State, don't know if that has anything to do with it.
Ah right, les. Now I see where I went wrong. I didn't bother checking the lt2 file. I deleted all the lights and added fresh spots, but still got LIGHT 1, 2, 3. It didn't occur to me to check the file. I was just assuming they were the 3 default lights. That's a good idea having a pz2 with no lights. Saves time. Although, to be quite honest, I'm hoping to avoid lights after I finish these ones. LOL. I haven't much patience for it. Your shadowcams with extra param dials are really good. I think my problem is slightly different though. I really have a grudge against poser lights because they don't behave the way I expect them to. If I use spots when I'm shooting, I have all sorts of tricks to make them do things, and none of these tricks work in poser. Drat! Oh well, I'm almost finished this now. Thanks for the continued interest and help. This thread should be useful to someone in the future. cheers mac
One thing I've figured out about the shadows (with spotlights), is that the further away the light is from the figure, the LOWER the Angle End setting has to be to get a decent shadow. Of course, this also decreases the field of usable light - exactly the opposite of the way lights behave in reality. In real life, the further away a light is, the more it spreads. Example - For a light that is 1.00 trans (or 1 poser unit) away from a figure, the End value can go up to 120.00 (but no more), and still give a reasonable shadow. Increase the trans to 2.00, and End has to go down to 60.00 or even 40.00 to get the same shadow quality. This is without touching the shadow cams, and obviously, you can improve the shadow by using them, but the above values can't be changed by much. I've finished my light sets and left the shadow cams at very general settings. But I've added a short explanation of how to use them to the Help File, so users can set them up for to their own scene. More than that.... I can't do. mac PS les - An interesting demonstration.
Hmmm... A question has just occurred to me, which I don't have time to follow up. Does anyone know what the difference is between an omni and a spot? OK. What I know. - The spotlight takes it's geometry from lights/spotlite.obj. Fine. But the spotlite.obj is nothing but a cylinder and 2 box props, and in a light file it's set to display EDGES ONLY. So the spotlite.obj is only a cartoon, basically. And the omni's resource is listed as geometry 7, (presumably, that idiot circle thingy), so it seems that the only difference is the light type - 1 or 0. The channels appear to be identical, except that an omni has the Angle/Dist channels hidden. I don't know if I have time, but I'm going to try unhiding these shannels and see what happens. Les. You mentioned a hybrid light. Where and what is it? If we could mix omni/spot functions, we might be able to get somewhere. mac
Content Advisory! This message contains nudity
This site uses cookies to deliver the best experience. Our own cookies make user accounts and other features possible. Third-party cookies are used to display relevant ads and to analyze how Renderosity is used. By using our site, you acknowledge that you have read and understood our Terms of Service, including our Cookie Policy and our Privacy Policy.