FPS improvement

[ALVAZ]

Hi, the question is about how to design a circuit to get the maximun FPS, because after testing my circuit I realized that i get less fps I expected, I have to say that, It´s a 5, 8 km long, with not very much number of objects, only 564, and after compariying my FPS with another circuits, i.e. Essingston, Mills and Toban, circuits that come with rfactor, They have much more FPS, and the same, more or less, number of objects that mine!!, so, the question is: How can I improve FPS number?
I follow the following rules:
1- Track: use the minimum number of panels, bigger in straights, and shorter in curves, cause I don’t like to see “square curves”
2- Terrain: use the minimum number of panels as well, and the minimum amount of terrain, I didn’t get a visible improvement of FPS, taking of all the terrain around the circuit.
3- LOD, as close as possible, but , I didn’t get a visible improvement of FPS, lowering as much as I can, 500m for all objects, and 1000 m only, special objects.
4- LOD in terrain, the same, 200 m most of the panels, and 500 in some circuit zones.
5- Objects: as simples as possible, but this is not always possible, I need to design that special object to make the circuit different, buuut I does not have to be very big.

Ok, that’s all, suggesting taking objects off, is not the solution that I want to hear (read),

anything else?

Thank you for your advises



THAnks

 

[ALVAZ]

hi, I have some findings!!, I managed to increase FPS, 15--20 units, you can say, it is not too much, ok, but when you have 20 FPS, increase to 35-40, its very good!!!, ok lets see, I got rid of all the "receive shadows" and "cast shadows" of WALLS!!!!, it is the unique event, that after modify gave me a mesurable effect!!, so the way is clear: Getting rid of "receive shadows" and "cast shadows" of all those walls placed in LFP, that is, Low FPS Point, ).obviusly let the "receive shadows" and "cast shadows", of walls located under bridges ....

I hope this help!!

Butter69, What about your experiments?

Regards

 

[Butters69]

Getting there just working on fine tuning the map that I want to run the tests on. I am going to be aiming at testing what the effects of turning off each of the following will have on terrains and objects.
- Driveable
- Collide
- Do you want this surface to receive shadows
- Does the terrain receive lights
As I am mainly focussed on dirt ovals which the infield will be the garage bays, pits etc I am interested in the potential performance gain these will have. If you have ended up over the fence your race is done so I don't need it to be drivable or contactable. This may apply to areas of your own maps especially if you are creating street circuits with walls or barriers.

Moving to a different country at the moment which is why I haven't got the track completed yet give me a month and I will have my results for yas.

 

[Rubinhofan]

For RBR roads I can place couple textures (usually 4) on one horizontal sheet and make one material from them. The supported Material Editor allows me to "paint" the texture with surface's I want (gravel, tarmac etc). Then I use offsets and scaling in BTB to choose a proper piece form the texture.

How can I scale it right and get the right offset ?
Look at the following picture:


Uploaded with ImageShack.us
I have a texture file with 16 different images. I want to set image 1 for the road, but btb repeats the whole row. So I get 1,2,3,4 and then it starts again. I only want texture 1.

 

[martinez]

My first answer is that you can't use vertical texture atlases. To solve this - you should set Material Change Per Meter (ex. 6m) or Per Panel (but then panels had to be approx.6m long) and use Material Change cross-section 100 000 times - that obviously would be insane...

I have made small BTB project, where you can study how it can be done. Only with horizontal texture plates:

http://www.sendspace.com/file/cw663b

The example is very rough and textures don't fit each other (no smooth cross-overs), please don't judge me for this . Anyway, you can see clearly how they change on the track.

I use maximum 4 textures in a row. They have to be the same width, so for example 4 x 256 pixels make 1024. Then when you apply that texture/material to a road, the scale must be 0.25 and offsets like 0.00 - 0.25 - 0.50 - 0.75 (1.00 is the same as 0.00).
For roadsides (also 4 in a row) - the same scale, but they can be flipped horizontally, then it's some kind of small headache how to set offsets again, because they are also flipped Make some notes in meantime about this to avoid constant quessing.
Another example is "road02" where is one road (coble) and two roadsides (256 + 128 +128 =512), then for road scale must be 0.50 and for roadsides scale = 0.25 - and according offsets you will find easily.

I wish we had opportunity to use more textures in a row, but using scale like 0.125 is not possible (just two diggits after comma).

In material editor you have to load a texture and paint stripes how you like.
But.... I observed strange behaviour when using offsets - the physics are messed. And some solution can be flipping textures horizonatlly and use negative numbers for offsets (like -0,25 -0,50 -0,75 also -1.00).
Or - place gravels and tarmacs on different textures (you probably need to paint texture cross-overs as well!)
Since the whole method can work also for rFactor, in this case it's necessary to do like this - rFactor does not have Material Editor and different surfaces have to be separated because physics are assigned to whole textures.

This kind of building I have used in my Rally Poland 2009 Shakedown track (RBR_RX).

The advantage of this method is that you can compress materials. And you can paint your track more interesting (especially with different roadsides and smooth cross-overs between them).
The disadvantage is that this makes more objects when you export the track because material cross-sections divide roads for parts.
In the middle I would put a word that using more Material Change cross-sections automatically solve tracks' LOD, because you get smaller objects. Remember that a BTB track has LOD like this: track length + 1000. For example - a single 2 km track (with just one=first material change crossection) will have LOD 3000 meters then.
To be exact - you can edit this in objectlist.ini (in exported _RX folder). But be aware that LOD cannot be set smaller than track length, because it will disapear in the game (you will drive on invisible surface ) So, for example when you see something like:
.................
[Object_0]
file=t_0_s0_sm0
xform=t_0_s0_sm0r0
cloneCount=1
LODIn=0
LODOut=1014,566 ///// can be not less than 14,556 !! (would be also insane to see the part 15 meters before your car
Visible=1
DrawInstanced=0
Moveable=0
ShadowCaster=0
CollisionModel=0
...............

I suggest to use in track properties - Panel Spacing: Linear and specify material change length when you apply texture to track part. For example - 256 pixels texture still look fine on 5-6 meters. Setting this in XPack has no effect, you have to input these numbers while building in BTB. More - keep in mind your choosen material repeat length and don't place material change cross-sections in too small distances. The even spaces between panels' egdes will help you in calculations. Also - beware where track's control points are - when you place shape or material crossection too close, the panels will change length. And - I avoid setting material AND shape crossection in the same place - when you move these small points, they will change also texture position.
One more thing - when you drive your track and see some thin lines between textures - it's normal... The solutions are:
1) change the offset a little (+0.01 or -0.01) Scale has to be also adjusted a little.
2) choose and place textures on the atlas that egdes of the parts (as well as edges of the atlas) have similar colors.

For more distant (standart BTB) terrain you can use single textures - I attached them in XPack too - the same textures are used on texture plate - I could maybe paint some crossovers leading to single ones - just no time.

I believe your imagination will guide you to much better results. And hope all my talking here helps a little