PDA

View Full Version : printing readable stencils to be placed on aircraft



dog1
March 5th, 2017, 05:53
hello
I have not been successful in printing my own stencils for a spitfire vc using photoshop which uses pixels and maybe the reason i cannot get them small enough and above all readable. Can someone pls guide me to what is required such as tools or a software which can do the job. these are stencils to be placed all over fighter aircraft mainly.
regards

UncleTgt
March 5th, 2017, 06:05
Dog1,

Use the vector tools/ fonts in Photoshop to write the stencil text as a vector graphic (separate layer). You should be able to relocate/resize these without much loss of definition.

However-

CFS2 is limited to displaying bitmaps of 1024 x 1024 pixel size, so even then you will find they lose definition when you save as a bitmap & use DXTbmp to get it to display in the sim.

An approximation is all that's possible ...

dog1
March 5th, 2017, 07:41
hello
i did try all these options with very poor results earlier , i tried writting saving to png and transfer , tried direct copy and paste as well and other tricks but the result is very poor . i have examples of skins with me with excellent visibility on wing , fuselage , tail sections , crystal clear . i do not know the authors or how to contact them . i'm tired of applying fuzzy looking stencils when i see a good job can be done .

Captain Kurt
March 5th, 2017, 11:52
dog1

Which Spit are you trying to work with?

There is probably one bitmap, or perhaps more depending on the model, in the texture folder that has the VC bits on it.

If it is an extended 16 bit 565 bitmap you can open it directly in Photoshop. If it is compressed open it with DXT1 Bitmap and save the file as the extended 16 bit 565 format. I assume you have specified Photoshop as your graphics program for DXT1 to use.

You can then open it with Photoshop. Once open in Photoshop see if it is a 1024X1024 size. If not enlarge it.

Then at the top of the image layer sequence, create a new layer to work on. This will be transparent except for whatever you add. You can then use the the Type tool to create the label text anywhere you want. You can do color label backgrounds and add the type over the top of the colored area as well.

When you have everything as you want it, save the file as a Windows bitmap (bmp).

Open the Windows bmp you saved with DXT1 and save it in extended 16 bit 565 format, named the same as the original VC bitmap file and you will then have the the best detailed labels that you can achieve in CFS2 for that particular model.

BTW I haven't seen you here for some time. I created the Japanese airfield figures you had asked me for. Have you seen them yet?

UncleTgt
March 5th, 2017, 11:57
Dog,

Another technique is to use a large size canvas whilst you're building your layered image, & combine these layers into a large bitmap that you can then reduce in size for use in the sim.

Gaucho59 has some expertise in this area. The most impotant tip is to use "bicubic" resampling when reducing the image size - it helps to retain details & sharpness, but only to a point...

dog1
March 6th, 2017, 07:04
ok i'll get down to trying all these solutions and other i'm trying out . thank you
regards

Blood_Hawk23
March 7th, 2017, 07:36
something I do that may or may not help, I increase the pixels per inch. I do this by first making a new blank image with the PPI increased. I then start working on the image. when I resize the image the res stays crisp. well as good as it can. you will loose some no matter what you do.

Note that freehand and the tools will give better results then using an image or pic that has been cropped.

I'm using paintshop pro x5. not sure if photoshop works the same way.

pstrany
March 7th, 2017, 17:54
Another way you could do this would be to float a polygon or two above the surface, and place a nice large and readable transparent texture on it. Of course, this works best for something for which you have the source code, but it is possible for non-moving parts to place these as a weapon on the model. This technique was passed on to me many years ago by Greg Law...


Paul