More progress on the front, now only missing the upper grill and headlights.

Progress on the overall mesh is slow as I spend some time modeling more detailed parts, such as the bottom part of the rear bumper and its fake grills, as well as the side mirrors.
I also improved the front bumper, and added details I was missing before; I'm not satisfied with the bumper's blade part yet though, so it's not included here.

#42 will race.

Yeah, I think I will do both normal and cup versions, the cup version only differs visually in the added spoiler and roof intake, the removed front daytime running lights and the interior being almost entirely stripped, as far as I know. The road version will certainly be based on the TCe 140bhp engine of pre-2023 though, as after that it's either 90bhp or hybrid.

Back to Blender, I ended up purchasing the blueprint I was using for reference, instead of using just the low-resolution preview, so I'm revisiting the entire thing to fix the shapes that were sometimes off quite a bit. I'm also using this as an opportunity to model cars in more detail, not just for LFS, so I'm going for a high resolution model, from which I will make the LFS geometry.

The screenshots show the overall shape of the car, it's now a single mesh with much cleaner topology, but I'm still missing some of the front bumper and door details, and still fixing the trunk and rear fender shapes around the tail lights and the C pillar. And of course, I'm now extending the mesh over the lights and will have to cut them out properly.

Once I'm done with this overall mesh, I will start making the various panels again - I've made the hood already (quick and dirty test), just because it looks so different with the guide mesh extending toward the front grill. I'll probably try to make a low-poly version of the hood pretty soon, to get an idea of how many triangles I can get away with.

Reworked the geometry around the pillars and roof, still need to tweak the volume. I turned on the subdivision modifier to start hunting bad curves and topology (which is still a big mess for the most part).

Added the top intake and "rear wing", drafted the roll cage as I came across a figure describing it in the Clio Cup technical rulebook.

Also added the rough shape of the central brake light to work on it later, as its support helps close the rear window.

I started modeling a Renault Clio Cup, as I found it interesting that the car basically drove on 3 wheels in some corners, since the inside rear wheel tends to lift. I'm happy to see that it is also possible in LFS (and while tweaking the suspensions, I realized the XFR also exhibited this behavior to some extent), although more work is needed as it likes to flip as well.

Anyway, here's the current progress in Blender, I'm satisfied with the overall shape despite the low mesh resolution and some questionable topology, but now I'll have to properly plug all the holes and refine the shapes.

The custom logo fits in the Renault rhombus, but is modified to make the letters L and T (first and last from Laurent, which is basically Renault with the letters rearranged). Hope it's okay this way.

Skin for #42

Hi! Godot is a game engine, but also has a pretty nice UI editor for non-game applications (or for games too). You don't need it to run this tool, actually the 60-ish MB executable is the engine itself (with the editor and some other stuff removed, and my own project embedded).

One important thing with Godot though, is that it supports multiple rendering APIs, like Vulkan and OpenGL, and I chose Vulkan by default, so if your GPU is old it may not support it; however, I also released an OpenGL version, you can use that one and it should work just fine.
The next release of Godot should automatically allow falling back to OpenGL if Vulkan is not supported, at that point I will be able to provide a single version, and that error message should stop showing.

Let me know if you have any other issue!

#42 reporting for duty.

BTW I think you meant September 26th.

GodotInSim v1.3 is now available on GitHub.
It introduces more format customization for GISUtils time strings, the default grey color for LFSText, as well as fixes for LFSText (some characters in the extended characters of the Latin code page were not being converted properly), and a few other minor additions or quality of life improvements.

Also, while I haven't had as much time as I would like to work on actual projects using this tool, here are a couple sneak peeks:

• Multi-class standings and live gaps, each class can be configured with arbitrary cars, and a "hidden/ignored" class can be used for safety car and other utility vehicles if necessary.
  
• Live delta, displaying the time delta to the best recorded lap of the session, and colored arrows showing the derivative (current improvement (or lack thereof) over the best lap, that's the rate of change of the delta). Also displays final deltas for each sector.
  

Sadly I did just check, and I'm getting the same value for all tyres, even non-existing ones (e.g. 0 (R1) for "all 4" tyres of the Reaper X1 bike, or 5 (road normal) for the 3-wheeler Hadraplan). At least that's what I get with the default setups.

I haven't checked the values, but the IS_NPL packet gives you the tyre compounds for "all 4" wheels - maybe 2 of those values are equal to TYRE_NUM, which would indicate a motorcycle?

#42 will be racing.

As gu3st said, the DDoS situation certainly hasn't helped with Scawen's ability to make progress; he's been quiet on the forums in the last month though, let's hope this means he's able to focus on the the new physics without too much distraction.

It would be great to get a few new sneak-peek screenshots from Eric's side as well, I'm not sure whether he is working on the new tracks we saw in the 20th anniversary news post, or improving existing ones (South City being the main one I believe, maybe Fern Bay?).

Entry number : 42
Entry team Name : Yes We Kan
Car: Protech 92
Entered driver (Ingame name / LFSW name / Country) : (Cyk / Bokujishin / France)

Maybe not a bug but rather a design decision, but in hotlap mode, crossing the start/finish line for the first time (starting the first hotlap) does not send an IS_LAP packet, which means we have no way of knowing when this lap starts. This can be an issue for many possible reasons: triggering a new telemetry dataset, starting custom timing or delta time, etc.
Following IS_SPX and IS_LAP packets are properly sent, starting with split 1.

Given Flame's screenshots, it looks like you should only need to change the mapping so 10-160 maps to 16-255.
Checking a couple values by taking similar screenshots with diffuse lighting only would ensure it is correct.

GodotInSim v1.2 is now available on GitHub.
I forgot to announce v1.1 last weekend, which added the new IPB packet in LFS 0.7F.

v1.2 goes back to using the "standard" plid syntax instead of player_id, which is a breaking change. It also fixes MST packets not interpreting messages starting with a slash a commands.

I'm also working on a more advanced tool as a demo, which will display race standings and driver statistics as well as a minimap showing blue/yellow flags and coloring cars by car/category/lapped/etc.
The goal is to allow preparing a race by specifying categories (e.g. P2/GT3), separating safety car/rescue vehicles in a dedicated category, and showing standings by category.

FWIW, I had a look at the colors corresponding to 10-10-10 and 160-160-160, they are respectively 12-12-12 (0c0c0c) and 200-200-200 (c8c8c8).
However, I believe mapping 10-160 to 0-255 is the way to go, as limiting to 200 would definitely wash colors away.

Hi Scawen, thanks for the update! I have a question related to the new ISP_IPB packet: its size is 8 + numB * 4, and there's BanIPs [IPB_MAX_BANS], of type in_addr which is not documented. It seems quite logical that in_addr is size 4, and this makes me think that each of those bytes represents one part of an IPv4 address; does this mean that IPv6 addresses are not supported here?

Short answer: yes to 2) and no to 1) and 3). The editor allows you to create complete vehicles, as well as edit any existing one to your liking. There is also a "Test in LFS" button for local testing, and there is no obligation to publish mods.

As far as engine tweaks are concerned, I'm not sure 10k hp is possible, as values are based on physical parameters; you can definitely go over 1,000hp though. Also keep in mind that engine mass is calculated automatically, a more powerful engine will be heavier.

I would suggest going through some of the editor tutorials to see what and how you can tweak things, I believe you will find them in the vehicle mods forum.

While catchfencing may require some additional work, I definitely agree kerbs should be a rather easy addition that shouldn't require the new graphics and physics, plus they should not affect object count too much (at least in places where other objects were used instead). I hope they can be included in the next InSim incompatible update (at the very least 0.7F will have some InSim changes, so maybe this could be an opportunity to update object index data?)
Those additional sausage kerbs from the new year screenshots would be a welcome addition at the same time.

If the preferred way to tackle this would be to improve the layout editor first, it will also have an impact on InSim, as it will need to take the additional features into account. The most obvious change would be in the ObjectInfo struct, I think; it would need to increase in size from 8 bytes to whatever is needed for higher precision placement of objects (especially along Z axis and heading), and this could also affect other packets that have parameters using the same format as ObjectInfo (JRR, UCO and AXM use ObjectInfo directly, OBH uses same format for XYZ and index).

With that said, and following what thegamer23 said, I think it could be nice to have grass/gravel/asphalt (and any other type of surface, really) as a flag in what are currently concrete blocks only. That way, combined with higher precision placement, you could make proper terrain within the layout editor. It would most certainly require 2 things, though: a higher object count, which is already an issue for more advanced layout, and as a nice-to-have bonus, a layer system, so the entire layout wouldn't become a cluttered mess from "terrain blocks".

As far as InSim is concerned, the increased ObjectInfo size, plus the increased object count limit, would require more bandwidth, but some clever grouping of values could help alleviate this (this is already the case for some data across InSim).

Hi, glad you got it working! Did you simply delete all files related to the app? I believe that as long as you download version 0.2.1, and follow along the above video, everything should be working (I just checked again on an old Windows PC and it works fine). Feel free to tell me if you have any other issue.

Did you download the latest version? (hotfix for file names) Recording would fail on Windows before that.
I just checked on a Windows 10 install I have lying around, and it works as intended there. Check the video I posted for @superlame a few posts back if needed.
If you still cannot get recordings to work, could you please try and launch the tool from cmd or PowerShell, and post the log here after recording and trying to load data?