The new Beaufighter I 3dz is on it's way. Worked all day long on it. One half might be ready tomorrow or on Thursday, depending how things go on, but the worst part -the nose!!!- is done. RS problems seem to be solvable so far (even for me ).

Now the big question: When I have one side ready, can the other side simply created by changing the signs of the y-values? Just a stupid idea, but maybe it works.

In a word - no.

I wish it was that easy

Andy,

my guess would be that all the elements have to be read the other way around so they appear from the other side.... that is, all the elements that are read clockwise on one side of the plane need to be read anticlockwise on the other....and vice versa.......


cheers,


Pobs

S**t. That's a real pitty. Now what's the reason that it doesn't work?

I'd rather change the orientation of 200 elements than going through this 3dz hell again...

Um, don't hold me to this, but:
Because EAW uses a clockwise(???) coordinate system, it means that the points of an element are ordered in the 3DZ file moving in this direction. This in turn influences the direction that the normal points and also which side the element is visible from.

So IN THEORY (and this is only working in my head ) to make it visible, you need to:

1. change all the y-coordinates
2. reorder all the points of all the elements
3. multiply all the normal values by -1
4. test it and hope that it works

Cheers,
Gurney

PS. Just in case you want to know, I haven't done anything on the RS calculator. As I've said in other posts I'm really busy till about mid-December. But if I get a chance on the weekend, I'll to to make a really small, program to change those values for you, just so you dont have to do it all by hand. But I can't promise it though.

1) Reversing the normal value from positive to negative tells the game to viewing angle as seen from either the left side or the right side. But it does not address the RS sequence at all.

2) The rotation that Gurney talks about also determines whether an element is seen from one side or the other. In other words, the order that the points and PCX mapping coordinates are listed in, in the code line tells the game which side of an element to place the skin onto. But it does not address the RS sequence at all.


For example using a simple 3 point element:

5 3 100 99 22 101 109 26 102 119 32 0 216 255 could be seen from the right.

Reversing the order:

5 3 102 119 32 101 109 26 100 99 22 0 216 255 would then cause the element to be seen from the left.

But it does not address the RS sequence at all.


3) The Y coordinate determines whether a point lies on the left or right side of the center point of the plane. A negative point value means it is on the left side, a positive value means the it is on the right side. This would also need to be changed. (Similarly, a the point location for above/below and fore/aft of the same center point are determined by positive or negative point values.) But it does not address the RS sequence at all.

4) Did I mention none of this addresses the RS sequence at all?

Number 4 is the answer - the rendering sequence for one side does not work for the other side - or at least I have never been able to get it to. I know Col. Gibbon gave this idea a shot at one time too, when we were working on the P-38 models I think, with a similar level of success as me.

Further complicating the issue is the fact that one side is usually not the same as the other side on the model. Just one additional element like a pitot tube, or a nose gear door, etc., and the entire rendering sequence is changed. And we're screwed.

Gurney, I can only wish you luck and perhaps some divine inspiration - or maybe even intervention. But we really, REALLY need an RS calculator.

Cheers guys,

Hoi There!

well ok the simpliest answer is no. but as Kurt sais, it can be done.

about the 4th question of Gurney:

Kurt is right about that the RS aint the same on both sides, but the tick is:

i have once tried the way of doing what Andy mentions: to change the points y value and the normals. it did worked, but as the guys mentioned the points order (and yes, Clokwise it is) had to be changed, or elseway the 3dz will be "upside down and downside up" seenable...

i have created an Excel table for that where i have converted in the 3dz and thrown that ionto the tabel and thad did the work instead of me and the two sides were OK...

sorryfully i don't have that stuff anymore...

but to answer the main question:

YES IT CAN BE DONE !!!

Mr. Johnson.

Gurney:

Let's have a try. If it works it's good and two engined planes can be modeled in half the time. I don't care about other elements on the other side. In most cases these can be added with subelements after the conversion process. Also unmirroring would still have to be done, but ok, these things are done quickly in comparision to the ugly 3dz modelling.

Hi guys,

anyone know the answer to this?
I am trying to get 12 or more simple squares in a single line on the Y axis, one behind the other viewable from both sides (2 elements per square)and pasted all points and elements from scratch. Used vector to calculate all normals and used this R/S method:
E000: 0 0 1 255
E001: 2 3 255
etc.
but only the first element shows (front and back)
To make sure the R/S worked I have placed every square in a regular order, they all have the same size and line up on the Y axis, 1st square in front last square all the way at the back. I have tried reversing their order , their place in the line up, reversed the whole Y sequence, read all points backwards forwards and side ways, up and down still only one element is shown. My limited knowledge on the R/S says this should be possible, I managed to paste two single trees behind eachother too this way (allthough not working perfect they can both be seen).
Any ideas?

VonBeerhofen

Gurney. Ehm, may I ask you another favour?

I could use a normal calculator that automatically calculates all normals, or at least only normals of elements with three points only (because the normal must be right for such elements). I'd do the calculation by myself for all other elements.

BTW: To change the normals for this side switching only the y-coordinate needs to be changed, at least that's what vector algebra tells me...


Von Beerhofen. Better you ask your question in a separate thread. I'm sue you get an answer then.