Page 4 of 5

Re: 2017-08-02 snapshot

Posted: Sat Aug 12, 2017 5:20 pm
by Big Smooth one3
jlv wrote:In its current form it won't make straight ruts. It pushes the terrain around according to the force applied without ever lowering it. I guess it might be worth having a second radius that only applies when it lowers for ruts.

So if it pushed terrain around without lowering, then is depth only referring the amount of raising around the point of force? So it's making the rut by building up around the line vs. digging into and displacing?

That being the case, I think allowing for lowering would be a good addition.

As to multiple layers, would it be possible to simply add a setting for regressive-like deformation? So on a scale from 0 to 100, 0 would 100% of erode every time force is applied, 1 would be 1% less for each successive force (each time someone takes the same line), 5 would be 5% less, etc.

Re: 2017-08-02 snapshot

Posted: Sat Aug 12, 2017 6:37 pm
by barrington314
ShackAttack12 wrote:
barrington314 wrote:The way I understand slope is that the erode setting range would be from 0 to 1.0. a slope of 0.5 wouldn't allow anything more than a 45 degree angle.
(I hope I'm correct with that :oops: )


Isn't the slope in radians?

15* = .26rad
30* = .52rad
45* = .785rad
57.3 = 1.0rad

Maybe I'm wrong... :oops:

I figured it would just be the max rise over run you want and then the radian calc would come next, right? Guess we shall see what JLV says. :shock:

Re: 2017-08-02 snapshot

Posted: Sun Aug 13, 2017 1:55 am
by jlv
I was just rereading the slope limit code and it actually only applies to the case where the dirt would be pushed uphill so it shouldn't be preventing downhill braking bumps. I'll do some testing to make sure I haven't reversed the test somehow.

checkerz wrote:Quick question on slope, is this 0-90 degrees or is it a percentage or what exactly is the max on this number? Also, can you give some examples of force ranges as well? I'm just throwing darts at a board right now basically.

Barrington has it right. A slope is rise/run. If you're going up a hill and gain 1 foot of elevation for every 10 feet it's a 10% slope or .1 in the game.

For the forces, it's in pounds not counting the vertical component. If the bike is in a turn leaning at 45 degrees the lateral force should be equal to the weight of the rider + bike. Divide that by 2 since it's distributed over both wheels.

Re: 2017-08-02 snapshot

Posted: Mon Aug 14, 2017 2:34 pm
by Joe
Just came in to write that I am so pumped for erode!
Game changer for sure!

Super stoked

Re: 2017-08-02 snapshot

Posted: Mon Aug 14, 2017 4:29 pm
by Mr. Wiggles
Joe wrote:Just came in to write that I am so pumped for erode!
Game changer for sure!

Super stoked

Joeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee we gotta ride in sim sometime

Re: 2017-08-02 snapshot

Posted: Mon Aug 14, 2017 9:27 pm
by Kawasakis
So does erode work on public servers or no? I'd love to try it out.

Re: 2017-08-02 snapshot

Posted: Wed Aug 16, 2017 2:19 am
by yzmxer608
Yep, think it's on 05 still and maybe some others I don't know about.

Re: 2017-08-02 snapshot

Posted: Wed Aug 16, 2017 5:57 pm
by barrington314
yzmxer608 wrote:Yep, think it's on 05 still and maybe some others I don't know about.

mxslobby 19801 8)

Re: 2017-08-02 snapshot

Posted: Wed Aug 16, 2017 6:10 pm
by barrington314
jlv wrote:I was just rereading the slope limit code and it actually only applies to the case where the dirt would be pushed uphill so it shouldn't be preventing downhill braking bumps. I'll do some testing to make sure I haven't reversed the test somehow.


Last night I only did 1 moto on Budds with erode on. And only a few riders. But this is what I found and am wondering if what you described is what's happening.

This is the tunnel roller after the finish. Nice braking bumps going up the hill there. They felt pretty good IMO. Formed up nicely.
Image

The tunnel single after big gulp. Again, some of the bigger bumps that formed on the track were going up this...
Image

But this is after henry hill tunnel, going down... just small chatter. No real defined bumps.
Image

This is going down hill before mechanics area... just chatter. small bumps.
Image

before big gulp. still down hill but more flat than the other examples. Still mainly just small chatter.
Image

Re: 2017-08-02 snapshot

Posted: Wed Aug 16, 2017 6:41 pm
by checkerz
This also VERY cleary demonstrates what I was explaining about ruts forming going into and to the middle of turns and then not much coming out.

Image

Re: 2017-08-02 snapshot

Posted: Wed Aug 16, 2017 6:58 pm
by J.Armstrong
I've noticed that erosion has less of an effect while the rear tire is spinning (off the start, exiting corners pinned etc.). When the tires are tracking nice and not slipping (ex. braking) then the result is quite obvious and actually really good. If you lug it coming out of corners the ruts at the exit are much more defined.

Re: 2017-08-02 snapshot

Posted: Wed Aug 16, 2017 7:03 pm
by Aaron Hall
Also i have noticed with the bumps that seem to be forming in diagonals is this due to riders being lent over and breaking/accelerating? Because I am really not liking that don't know if that the erode can change that or is it something JLV will need to change?
Image

Re: 2017-08-02 snapshot

Posted: Wed Aug 16, 2017 8:15 pm
by barrington314
Aaron Hall wrote:Also i have noticed with the bumps that seem to be forming in diagonals is this due to riders being lent over and breaking/accelerating? Because I am really not liking that don't know if that the erode can change that or is it something JLV will need to change?
[img]https://i.gyazo.com/f376fbfcc45ef27d0141df2f04e3c763.png[img]

I've noticed this a lot on unadilla.

Re: 2017-08-02 snapshot

Posted: Thu Aug 17, 2017 1:03 am
by jlv
Thanks for taking the screenshots. I'm thinking it might be because you can't brake as hard going downhill. Uphill you can brake as hard as possible but downhill you'll endo if you brake too hard. That makes more force uphill and less downhill. I guess it's obvious now that I think about it. Since the stopping distance is shorter uphill the same force is compressed into a smaller area. A smaller erode_upper setting might even it out.

If I do the more erode when slipping thing it might also help with that since the rear wheel is probably slipping more downhill.

The diagonal thing looks like more of a circular wave centered on the first bump. I don't think there's anything I can do about that. I wonder if the double slit experiment would work with dirt bikes?

Re: 2017-08-02 snapshot

Posted: Thu Aug 17, 2017 2:09 am
by Atom6246
jlv wrote:Lazy terrain reshading for better erode FPS.

JLV, was this update what you referring to when you said this?
jlv wrote:
Atom6246 wrote:Was really fun playing with erode again especially on an rF track with a decent amount of people^
Although it sucked I had to leave eventually because I started getting frame skips down to 8fps. JLV is there any hope for future optimization?

Yes. The current reshading code is very inefficient. I expect to improve its speed by an order of magnitude in the next few weeks.