Hi all, I asked about methods to carve letters in stone a couple of weeks ago and received some good ideas. The best was from Antoine Clappier (via ChrisIrwin). The problem I have is that for some reason it seems to compound the text outlines in an arc when seen from above. As a result, you can see through the letters directly in front of you, but the further from center they are, the less visable. If I leave the camera were it is and pan to the left or right so that I am looking at the outer letters, they look fine, and the middle ones are messed up. I thought it might be something with the camera, so I tried backing that up and zooming it in, but it made no difference. I put a flat cube inside of the object to control the depth of the letters (since compounding goes all the way through.) I changed it's color to Purple just to make it easier to see the letters. I was going to attach the file to see if someone could look at it, but I can't figure out how to attach a .car file. I tried putting on my Web server, but when you go to the URL, it gives you the text for the .car file. Any ideas? Thanks, MarkV

I don't know about Windows OS, but in Mac OS we can download a .car file as text, then drag it onto an autotyper to convert it to the correct type and creator codes. To avoid that, you would zip the .car file before uploading it.

Could you post an image of what you're seeing? Thanks, -Kix

-Kix

Great Idea Kix, I was having a senior moment or I would have thought of that.

Here is how it looks in the Spline Modeler. As you can see, the numbers in the middle look fine, but if you want to see the Zero, you would have to dolly the camera to the right then pan back to the left and you would see though the Zero, part of the Nine and very little of the eight and the rest would be solid.

Rendered image to follow.

Here is the Rendered image with the Purple cube inside the stone for easier viewing.

I would be glad to email someone the file in anyone wants to look at it.

Thanks gang,
MarkV

How about designing the letters as an image and adding it in the transparency channel (see enclosed picture)

Did you build the first cross-section with the letters, then resize the front cross-section? If so then the letters were shrunk along with the rest of it at the front, making a funnel shape. It's hard to see into the small end of a funnel. Try making the basic cross-section, then re-size the front, then add the letters to the back, and then copy/paste them to the front so they are the same size on both cross-sections. Then do the compounding. I think that this would eliminate the letters getting smaller in the front, and you could bevel them(labouriously!) in the vertex modeler, something you couldn't do if you used the transarency option. Hopefully, C3 will include an EASIER method of beveling. (Do you hear us Antoine? PLEASE??!!) You should be able to attach a .car file here at R'osity. just use the "Attach file" option below, and browse to the file. After you download to your computer, (if your on a PC,) then change the file extension from .txt back to a .car and open. I've done it for this message. (I hope it works! ;>) Hope this helps! Good luck!

Bluetone, It opened just fine. I just changed the txt extension to car.

Thanks for the confirmation keithw, now lets hope that my other sugestions work for mvernon.

Lindnan1, That looks like what I want, but I am not sure how 'create the letters as an image'. Bluetone, I only had one cross-section. I am not sure you could select objects on multiple corss-sections and then do a Compound as Object. Antoine's advice was to 'Convert text to Outlines', shift click on the main object and do a 'combine as compound'. It seems to work other than the arcing. I don't really need to bevel them, so Ithink Lindnan1 has a viable solution as soon as I can figure out how to accomplish what he is suggesting. Thanks, MarkV

Beveling can help get more 'organic' images instead of a hard '3D' one. It has to do with the light catching the edges that helps define the edge in 'real' life. As far as how to do what lindnan1 says, start with your favorite 2D paint program, (Photoshop, PaintShopPro, etc.) and then create your text using black letters on a white background WHITHOUT aniti-aliasing. Save the file as a tif or jpg. In Carrara, duplicate the shader you are using for the stone block. From the mini-menu at the top of the shader, change your shader to a multi-channel mixer. Your former shader will be used for 'source 1' while 'source 2' and 'blender' will be empty. Change 'source 2' into a single-channel shader, and make the channel be transparent. Set this channel to a value, and the value to 100% For the blender, set it to a texture map, and import yout painted text file you saved earlier in Photoshop. Make sure that the settings are NOT set to tile. Set your rendering to Transparency - ON. If you have any lights passing trough your object, (shadow cones, light behind the object, important shadows,) set Light Through Transparency - ON. Your stone block should now have a hole in it for every letter in your text. With this technique, you can make many fake boolean shapes, without the hassle of extra geometry. However, you also don't get the benefits of it either! Hope this helps!

Beveling can help get more 'organic' images instead of a hard '3D' one. It has to do with the light catching the edges that helps define the edge in 'real' life. As far as how to do what lindnan1 says, start with your favorite 2D paint program, (Photoshop, PaintShopPro, etc.) and then create your text using black letters on a white background WHITHOUT aniti-aliasing. Save the file as a tif or jpg. In Carrara, duplicate the shader you are using for the stone block. From the mini-menu at the top of the shader, change your shader to a multi-channel mixer. Your former shader will be used for 'source 1' while 'source 2' and 'blender' will be empty. Change 'source 2' into a single-channel shader, and make the channel be transparent. Set this channel to a value, and the value to 100% For the blender, set it to a texture map, and import yout painted text file you saved earlier in Photoshop. Make sure that the settings are NOT set to tile. Set your rendering to Transparency - ON. If you have any lights passing trough your object, (shadow cones, light behind the object, important shadows,) set Light Through Transparency - ON. Your stone block should now have a hole in it for every letter in your text. With this technique, you can make many fake boolean shapes, without the hassle of extra geometry. However, you also don't get the benefits of it either! Hope this helps!

Bluestone, Thanks for the reply. I guess I am missing something. I tried your suggestion and ended up with the texture below. How would I apply this new texture? if I just drag it to the object, the stone is transparent everywhere except for the letters. I mst be doing something wrong? Thanks, MarkV

Hi Mvernon, I don't know if this is the type of letters in stone your looking for or not, I was just playing around? If it is = I created the rock objects in Amapi, created the 3D letters in Amapi also, did a boolean subtract = letters from rock objects. In the top rock object, I subtracted the letters completely through the rock object. In the bottom rock object, I only subtracted the letters half way through the rock object. Either way, you can still see the letters in each rock object, no matter what angle. A lot of the letter's visibility depends on the placement & angle of the lighting in the image. In this image, I placed the light at the top right side, & out in front some. Anyway, just thought I'd add my 2 1/2 cents worth, & you have a good one . . . Mark.

Hi Kruz, Those look sweet. Guess I am going to have to spend some time and learn Amapi as well as Carrara 2. Thanks, MarkV I am still hoping Bluestone can give me some kind of idea of how I fowled up his approch. Should work.

Not positive but try inverting your texture map. there's a check box for it: invert color Kevin

Hey MarkV, I bought Amapi a few months before I got Carrara 2, & I really feel comfortable using Amapi's interface for modeling, so I do almost all of my modeling in Amapi. I don't know enough yet, about all the nice things in Carrara 2, but I'm learning. Here's another image, this time with the letters cut into a somewhat round object, good luck with your project . . . Mark.

Sorry mvernon! I told you the colors backwards! (Feels sheepish and hopes for forgiveness.) MeInOhio is correct, in that you can correct MY error by selecting 'invert color.' As far as booleans are concerned, I, and others I have listened to here @ R'osity, have had lots of problems with booleans. Strange things sometimes happen at render time, while others happen when you do the boolean, then you end up spending HOURS fixing it in the vertex modeler. Not worth my time, UNLESS the effect your looking for can't be accomplished without it. I have to admit though that I haven't spent a lot of time in Amapi, and as such, maybe it treats booleans better then Carrara does. Hoping this helps better then my LAST sugestion. ;>

Hi All, Thanks so Much! Inverting the colors fixed the problem. Bluetone, thanks for the idea. I had never messed with the shade tree editor much (occasionally tweaking a few settings to see what happens.) I learned some neat tricks trying your approach. I too, have never had much luck with booleans (they worked great in ID4. Supposedly Carrara was based on ID4 and RD, but the booleans sure got messed up in the combination.) Kruzr, I really like the images. Now I have to decide whether to devote 100% of my available time to learning Carrara or 50/50 Carrara Amapi! Interesting name, is Kruzr in reference to a boat, car, Kruz'ing through life or what? Thanks to all who replied! This is a great group and I hope to get up to speed enough to be able to help also! Hahaha Regards, Mark