AI Blackwood lap time - updated with version D44

Cars - no major changes - although I think they handled the corner better after the long straight in the second sector (especially on lap 3)

Motors:
D43 vs D44 accelerated very much
Vulcan - 11 seconds.
Chimera - 7.98 seconds
Reaper - 10.84 seconds

Unfortunately, during the first attempt, Vulcan overturned on lap 3 at the end of the 2nd sector. - curb ? (replay attached)

Regards

D45 update: https://www.lfs.net/forum/post/2063105#post2063105

For classic racing use (balanced grids, no motorcycles, bicycles, buses... backhoe loaders etc.) and for more fun and recreational use than competitive, the D44 patch radically alter the gaming experience.
This can be improved, but already very satisfying

I particularly appreciate the new combativeness of the AI. It is no longer possible (apparently?!) to "scare" the bots by pressing them against the edge of the track so that they slow down and move behind. It's good.

The bots are also much better distributed on the track with more varied trajectories and fight more among themselves. The races are more lively. There are still a few car crashes, lawn mowings, and stranges road trip, but it's already much better than the D43 patch. The defects I noticed have disappeared.

Thank you and well done Scawen for this major improvement. This new AI radically alter the perception of single-player gaming (for someone who unfortunately doesn't spend a lot of time playing, but who has known the game forever).

I see ai have issues when shifting gears in the bikes aswell, causes alot of instability. This seems 10x since the latest bike update aswell. Bikes shouldn't have this much gear debounce.
It's also a huge hinder for us lfs bike users aswell, regardless of lean angle the shift debounce is a destroyer.

Quote from Scawen :

Right, I did see the crash once. I need a way to reliably reproduce the crash.

EDIT:

Trying a lot, I never saw it again. In my case it was an actual crash, but not in LFS.exe and I didn't have LFS in the debugger, so no info yet. I clicked to enter game from the entry screen and got a crash, but only once.

UnknownMaster21: You called what you saw a "softlock" but what do you actually mean by that?

Alright! INGUL Sidecar. This causes AI not able to drive. It tries to generate path for AS4, nothing happens.

LFS is not responding.

EDIT: Yeah, no setting can generate path in this combo. on BL1, this vehicle AI can use this vehicle, but driving it is completely... nuts.

Quote :

--deleted post--

No, but we are not going to discuss this thing here, it goes out of topic guaranteed.

id like to share a weird happening to me just few minutes ago
i was playing on LFS 7D44 in a server when my game suddenly closed, and after some checking, for some reason windows defender had closed it and put in in quarantine.

there is a screenshot of my windows defender threats history (if anyone wonder wich language it is in, its Brazillian Portuguese)

ofc i removed it from the defender thing and reopened it normally

it happened just after i changed my game to 1280x720 75Hz resolution (i normally play in 1366x768 60Hz)

Bug Report: AI in KY3R can't find pitstop properly in certain pitboxes, they seem to attempt stop between boxes. Also some of the vehicles still manages to get pitspeed penalty due accelerating too much., such as Formula US 93


EDIT: As Confirmed, AI does exactly same on BL2 and SO6R ( Although SO6R seems to be happening a bit less often. )

EDIT2: AI on FE4, that is chaos. I suggest to check it out, lots of crashes literally every braking point. ( Also pitbox issue )

Quote from UnknownMaster21 :

Bug Report: AI in KY3R can't find pitstop properly in certain pitboxes, they seem to attempt stop between boxes. Also some of the vehicles still manages to get pitspeed penalty due accelerating too much., such as Formula US 93

Confirmed.

I saw this before when AI cars were damaged but thought it was that they couldn't brake with the damage but ...

Just tried a few 2 lap races with compulsory pit stop and (with my WIP mod) the majority of the AI overshoot the pit box and get disqualified.

Happened at WE, FE, KY and BL (0.7D44) but was OK at RO (although there was some contact).

I tried the same (WE) race with FZR and it was fine so it leaves me wondering if there is something I can do - how do they "know" the length of the car?

Tried the new bike physics, and just like some of my friends said, it feels a lot more twitchy and unstable now in terms of steering inputs. We all reverted to the previous patch. Our suggestion is: please keep the old bike physics but make bikes able to go offroad instead, that's something that everyone has been wanting. Also, if possible - make light mappings work on rotating subobjects (light is mounted to fork of many motorbikes in real life) Thank you

I did my best laps that i could do on D43 and D44 to compare the physics. I've attached replays and setup if anyone wants to check it out. I ran it with 30% of fuel, with wheel and pedals set to 120 degrees rotation in-game and 1.0 wheel turn compensation and 25% ffb on my TMX wheel.

Surprisingly managed to do a faster lap on a D44 even with wobbly nature of braking but i tried to be as smooth as possible. Turning and aiming for apexes is much easier so that way you can get the best line without much guessing. I might have done some mistakes but it's hard to push those tires without them losing traction in a split second.

For nearly 2 years now i've been driving mostly bike mods and i pretty much got used to the old bike physics. But something was odd and i guess it's that sideways movement while cornering. You have to lean into the corner way earlier, sort of guess and then hope that you hit the apex. Of course i know the WE2 track very well so i didn't need much guessing how to drive on it. But yea you have to lean a lot sooner and then when you lean a lot, you have to lean a little bit more and the bike seems to corner way better, but it's only at the near fall limit, very weird. And of course i abused that on this lap, or at least i tried. Also it seems to take sharper turns slower as well i suppose because of it running wide.

All in all, i think D44 feels way better, more responsive and needs less guess work to take turns, especially on these big wide circuits.

Quote from Drifteris :

....

I get a 'replay OOS error' when I try to play both of your files.

I'm using 0.7D44.

Quote from sinanju :

I get a 'replay OOS error' when I try to play both of your files.

I'm using 0.7D44.

It seems like SPR files don't download mods, so if you don't have it then you get the error. Try downloading it through mod browser in-game first then launch replay

Thank you all for the feedback.

About bikes, I can clearly see the issues you report, not only wobbling when braking from high speed, but a lot of movement when coming off or applying the throttle, or tapping the brake a little when leaned over.

I've experimented quite a bit but haven't got any conclusive improvements yet. It seems like I can dial out any issue but doing so always causes other issues at other speeds or conditions. For example I've been able to get really nice accelerating and braking while leaned over at medium to quite high speeds, but then it's a lot worse when the bike hits a relatively small bump and seriously overreacts. The trouble is that what I call the "damping" part of steering (that tries to reduce wobbling by bringing the lean rate near zero) needs to be quite high to react quickly to certain things, but it needs to be gentle in other situations.

I also tried the early stages of a more physical model which tries to simulate gyroscopic forces and really apply steering forces to that. To separate a bike's natural handling, from forces applied by the arms. But it seems a little distant for me to produce such a genuine physical model at this time and I'd rather not do that with the old tyre physics. And a complete model would need body movement too, which isn't for now.

So in that case I'm left with the original plan to improve our semi-physical model (the bike is in physics but the steering is just doing what it needs to do to balance that bike at the required lean). I still hope to improve it, given my increased understanding learned last week, now that we have better bikes and more of a range to test it on, it seems like I should be able to improve it in some way.

Quote from Scawen :

To separate a bike's natural handling, from forces applied by the arms.

Why can't I disable the "assistant" in management? Will it be impossible to drive a bike without it? Is there a gyroscopic effect on wheels at the moment?

The bikes in the LFS look very cool, it seems that the LFS engine is able to bring them to ideal conditions, as in reality, you just need to refine some points. We all really liked this experiment with bikes. I think you yourself understand that the bikes have gained high popularity. We all really hope that you will bring them to a perfect state.

Definitely at the moment, bike physics has a huge potential! It would be cool in the future to see additional interacting forces so that we can drive the bike even better, for example, add a pilot's weight shift to do "Wheelie" or get on the front wheel by braking. It is possible to do this even artificially through additional buttons. I hope you will not let the bikes die, as a separate full-fledged class of transport in the lfs.

P.s. Sorry for the bad English translator.

Im wondering will we have track editor too just like car editor? If u already answered to that question Scawen sorry i havent seen

My guess is that both car and track editor already existed in some form and were used by Scawen and Eric to make cars and tracks.

Your question has already been asked before and we had quite a discussion about it. The real question is if Scawen plans to release the track editor at some point and what about the track licensing issues? But we should not go offtopic, this thread is for latest test patch issues only.

Quote from rane_nbg :

My guess is that both car and track editor already existed in some form and were used by Scawen and Eric to make cars and tracks.

No way :o

Quote from Scawen :

Thank you all for the feedback.

About bikes, I can clearly see the issues you report, not only wobbling when braking from high speed, but a lot of movement when coming off or applying the throttle, or tapping the brake a little when leaned over.

I've experimented quite a bit but haven't got any conclusive improvements yet. It seems like I can dial out any issue but doing so always causes other issues at other speeds or conditions. For example I've been able to get really nice accelerating and braking while leaned over at medium to quite high speeds, but then it's a lot worse when the bike hits a relatively small bump and seriously overreacts. The trouble is that what I call the "damping" part of steering (that tries to reduce wobbling by bringing the lean rate near zero) needs to be quite high to react quickly to certain things, but it needs to be gentle in other situations.

I also tried the early stages of a more physical model which tries to simulate gyroscopic forces and really apply steering forces to that. To separate a bike's natural handling, from forces applied by the arms. But it seems a little distant for me to produce such a genuine physical model at this time and I'd rather not do that with the old tyre physics. And a complete model would need body movement too, which isn't for now.

So in that case I'm left with the original plan to improve our semi-physical model (the bike is in physics but the steering is just doing what it needs to do to balance that bike at the required lean). I still hope to improve it, given my increased understanding learned last week, now that we have better bikes and more of a range to test it on, it seems like I should be able to improve it in some way.

What causes bikes to fall going offroad? Will we see bikes able to go offroad? (Grass surfaces) Biggest problem about this is just clipping a tiny bit of grass going through a corner and immediately losing control/falling. It would be heaps better if this was countered somehow, but i can somehow see how making this with the current tire model is tricky. Curious to see your opinion, thanks.

Would it be possible to have an option to turn off the “backlight” effect for the clock texture? It would be useful when using a custom backing texture for old cars which don’t have backlit gauges.

As an example, I've attached a gauge for my HADRAPLAN mod I'm trying to replicate. The bottom cutout is a regular texture and the inner part with the numbers is a backing texture.

Quote from Scawen :

Thank you all for the feedback.

About bikes, I can clearly see the issues you report...

Thank you for the update.
My issue remains with the gear debounce when shifting. It's quite exagerated and causes alot of issues. Anyway to correct this?

Quote from JayDeM :

Thank you for the update.
My issue remains with the gear debounce when shifting. It's quite exagerated and causes alot of issues. Anyway to correct this?

I'm not sure what you mean. At first I thought you meant a big wobble when the bikes shift (up or down, for different reasons). But I don't really know what you mean by gear debounce.

Gear Debounce is like having a "clutch bomb". It's like when releasing clutch on bike at full throttle from absolute on to absolute off in fraction of a second, bike will lift up it's nose like a stallion. ( If there is enough power and being enough lightweight, that is. )

It seems to be more effective the more rpm you have. Very deadly on cornering. It is deadly on real life too, but it is quite exaggerated in LFS. ( Depending on the bike of course )

The effect is not exaggerated, it's just that engeneers in lfs allways install the best quality clutch packs in all wehicles, it's made from unobtainium so it also wears less.

What about new layout objects?
As I understand it, they don’t need a dev patch with graphics and physics in order to fully work, can they be added to this test patch?

Quote from Aleksandr_124rus :

What about new layout objects?
As I understand it, they don’t need a dev patch with graphics and physics in order to fully work, can they be added to this test patch?

It will make the older versions incompatible, and the goal of "test patch" is to remain compatible with all of the 0.7D versions ( i think )

Quote from UnknownMaster21 :

Gear Debounce is like having a "clutch bomb". It's like when releasing clutch on bike at full throttle from absolute on to absolute off in fraction of a second, bike will lift up it's nose like a stallion. ( If there is enough power and being enough lightweight, that is. )

But I still don't know what this "gear debounce" is, where it is set, when it appears. Is it something intentional or a bug?

Quote from Aleksandr_124rus :

What about new layout objects?
As I understand it, they don’t need a dev patch with graphics and physics in order to fully work, can they be added to this test patch?

A lot of complicated code was added to support the new layout objects. It is not something that can be used in the old version.