I can't reproduce this error. I've tried with test patch D25 and with official version D. Full grid of that car on that track (as in your screenshot).
If I had your computer I would be deleting any relevant files like the mod files in the mods folder and the generated path files in data\knw and any saved start grids in data\misc, to try to find out what causes the problem.
I have not heard of a similar report from anyone else, so we don't have any clues at this time, why this happens.
Would it be OK for you to install a fresh copy of LFS and see if that fixes the problem?
Thanks, I've fixed that now in my version. The mod name was restricted to 27 characters before the check for ~ in the name. So that would happen for any mod with a 27 character name (maximum length).
Here is a new update, D30.
It annoyed me too much that on exiting the editor it always warned "Any changes will be lost" even if you hadn't done anything. I knew the best solution for this would be an undo function (because it tracks changes). The trouble is the undo is quite a lot of work and it took me about 3 days. Sometimes it's simple but sometimes some rearrangement is needed. I was a bit shocked by the actual size of the vehicle editor source code file. Luckily most of the more complicated side of the undo function was already done for the track editor undo.
I found some bugs along the way and improved some things as I had to test nearly every button. I hope there are no new bugs!
Vehicle editor:
Undo function for all editor operations
Exit warning only if changes have been made
C key to clear selected point and selected tube
Delete key deletes a point or tube (if only one is selected)
CTRL+RMB (copy) SHIFT+RMB (paste) now work with frame point values
Hide the bottom right buttons now hides the whole settings block
Rim modifiers moved to top right in rim editor avoiding overlap
Special view no longer includes wheel mass cuboids by default
Error message displayed if a vehicle file cannot be loaded
Modeller:
FIX: Bottom right screen layout was wrong in 2D views
Animation editor:
FIX: Driver in wireframe mode could be obscured by the vehicle
It annoyed me too much that on exiting the editor it always warned "Any changes will be lost" even if you hadn't done anything. I knew the best solution for this would be an undo function (because it tracks changes). The trouble is the undo is quite a lot of work and it took me about 3 days. Sometimes it's simple but sometimes some rearrangement is needed. I was a bit shocked by the actual size of the vehicle editor source code file. Luckily most of the more complicated side of the undo function was already done for the track editor undo.
I found some bugs along the way and improved some things as I had to test nearly every button. I hope there are no new bugs!
Vehicle editor:
Undo function for all editor operations
Exit warning only if changes have been made
C key to clear selected point and selected tube
Delete key deletes a point or tube (if only one is selected)
CTRL+RMB (copy) SHIFT+RMB (paste) now work with frame point values
Hide the bottom right buttons now hides the whole settings block
Rim modifiers moved to top right in rim editor avoiding overlap
Special view no longer includes wheel mass cuboids by default
Error message displayed if a vehicle file cannot be loaded
Modeller:
FIX: Bottom right screen layout was wrong in 2D views
Animation editor:
FIX: Driver in wireframe mode could be obscured by the vehicle
I want to release all the recent updates (in Editor and LFS) as a new official version, but after recent time in the vehicle editor I had to do some updates first. I found it too ugly and hard to use, without tooltips, with some overlapping text and general disorganisation. This is the result of it being an old in-house development editor that was thrust upon the community without enough time to polish it. So I've spent 2 or 3 days making some improvements which should make it a better place to work and in turn can help you make nice mods.
Doing tooltips is harder than you might imagine and I haven't finished them in all tabs. My head is spinning now. I would be grateful if you would test the vehicle editor updates and let me know if it seems OK.
Changes in Editor 0.7D28:
Various improvements in layout and organisation
Tooltips are now available (allows reduced button text)
A new button to show the view from driver's eye position
Editor sliders now have both arrows on the right
FIX: Spare wheel pitch now expressed in degrees
FIX: Nudging of RPM sliders in engine editor
Doing tooltips is harder than you might imagine and I haven't finished them in all tabs. My head is spinning now. I would be grateful if you would test the vehicle editor updates and let me know if it seems OK.
Changes in Editor 0.7D28:
Various improvements in layout and organisation
Tooltips are now available (allows reduced button text)
A new button to show the view from driver's eye position
Editor sliders now have both arrows on the right
FIX: Spare wheel pitch now expressed in degrees
FIX: Nudging of RPM sliders in engine editor
Last edited by Scawen, .
About the "reduce detail" function:
There are new limits for LOD2 in Editor D26. Maximum 8192 triangles / 12288 vertices. This is because of load glitches when people leave the pits in high-polygon models, with high-polygon LOD2 (which really should not be the case).
To help with this, for those who want to work down from a high poly model, I've added a crude "reduce detail" function that acts on the selected triangles.
I suggest using CTRL+SHIFT+LMB to flood-select groups of triangles, then click "reduce detail" (or SHIFT+R) on the small groups. Inspect them each time and undo if the effect isn't good enough.
Automatic LOD reduction is a challenging problem so this crude version is about all I can do for now. I'd like to know if anyone finds it useful.
Also to help with optimisation / load glitch reduction: Groups are also compressed when exporting for test or upload. This should have no effect on the appearance of the model.
There are new limits for LOD2 in Editor D26. Maximum 8192 triangles / 12288 vertices. This is because of load glitches when people leave the pits in high-polygon models, with high-polygon LOD2 (which really should not be the case).
To help with this, for those who want to work down from a high poly model, I've added a crude "reduce detail" function that acts on the selected triangles.
I suggest using CTRL+SHIFT+LMB to flood-select groups of triangles, then click "reduce detail" (or SHIFT+R) on the small groups. Inspect them each time and undo if the effect isn't good enough.
Automatic LOD reduction is a challenging problem so this crude version is about all I can do for now. I'd like to know if anyone finds it useful.
Also to help with optimisation / load glitch reduction: Groups are also compressed when exporting for test or upload. This should have no effect on the appearance of the model.
Editor D26: https://www.lfs.net/forum/thread/102626
Mod Export:
A new limit of 8192 triangles / 12288 vertices for LOD2
Groups are compressed when exporting for test or upload
Modeller:
A new function "reduce detail" is visible with triangles selected
A point is selected if unambiguous after merge to green or average
Mod Export:
A new limit of 8192 triangles / 12288 vertices for LOD2
Groups are compressed when exporting for test or upload
Modeller:
A new function "reduce detail" is visible with triangles selected
A point is selected if unambiguous after merge to green or average
Test Patch D25: https://www.lfs.net/forum/thread/102117
Support for new pit speed limiter flashing light option
One wheel drive and no anti-roll if wheels are staggered
Click Skin ID in garage colours tab to copy ID to clipboard
FIX: Remote car using pit speed limiter did not move smoothly
Support for new pit speed limiter flashing light option
One wheel drive and no anti-roll if wheels are staggered
Click Skin ID in garage colours tab to copy ID to clipboard
FIX: Remote car using pit speed limiter did not move smoothly
Editor D25: https://www.lfs.net/forum/thread/102626
Modeller:
A new style dialog is now used for merge points (ALT+M) options
FIX: Points more likely to remain selected after scale or lathe
Vehicle editor:
For flashing pit speed limiter light - choose brake or rear fog
Width / ratio / rim width move in bigger steps with right click
Frame position adjustments use expected steps for mouse clicks
Limited sidecar support:
Lone wheel (e.g. front) can be offset left (in Suspension tab)
- offset wheel to right is currently disabled for compatibility
Pair of wheels (e.g. rear) can be longitudinally staggered
- only the rearmost wheel of a staggered pair is driven
- anti-roll is disabled when the pair is staggered
NOTE: You also need LFS test patch D25 for:
- one-wheel-drive sidecars
- pit speed limiter flashing light using fog light
Modeller:
A new style dialog is now used for merge points (ALT+M) options
FIX: Points more likely to remain selected after scale or lathe
Vehicle editor:
For flashing pit speed limiter light - choose brake or rear fog
Width / ratio / rim width move in bigger steps with right click
Frame position adjustments use expected steps for mouse clicks
Limited sidecar support:
Lone wheel (e.g. front) can be offset left (in Suspension tab)
- offset wheel to right is currently disabled for compatibility
Pair of wheels (e.g. rear) can be longitudinally staggered
- only the rearmost wheel of a staggered pair is driven
- anti-roll is disabled when the pair is staggered
NOTE: You also need LFS test patch D25 for:
- one-wheel-drive sidecars
- pit speed limiter flashing light using fog light
No, the idea of this change is that these options are individually selectable per vehicle, instead of being controlled by the race class.
Previously, the flashing light was enabled by F1 class, but now it is set by its own switch. You do need LFS Test Patch D24 to see the effect of the new switch, or it will still be controlled by the race class.
Previously, the flashing light was enabled by F1 class, but now it is set by its own switch. You do need LFS Test Patch D24 to see the effect of the new switch, or it will still be controlled by the race class.
Interesting thought but it won't be in this round of test patches. Everything I do in the current public version must be merged into the separate development version, which isn't always easy. Also any changes in the core of the physics system are likely to invalidate all the hotlaps, which isn't worth doing at the moment. Also, incompatible setups, online compatibility, interface updates, etc...
I look forward to releasing the development version, so we only have one version and it will be much easier to add new features then.
I just watched this 12 min video about heave springs.
https://www.youtube.com/watch?v=lNInCfCkrkE
I look forward to releasing the development version, so we only have one version and it will be much easier to add new features then.
I just watched this 12 min video about heave springs.
https://www.youtube.com/watch?v=lNInCfCkrkE
Last edited by Scawen, .
You are correct to use the brake light assignment (you can't have brake lights and pit speed limiter light).
In Light colours tab, click the TEST button below "brake" to make sure the lights work.
You need to enable:
Pit speed limiter (in Allow)
Pit speed limiter flashing light (in Light colours)
In Light colours tab, click the TEST button below "brake" to make sure the lights work.
You need to enable:
Pit speed limiter (in Allow)
Pit speed limiter flashing light (in Light colours)
Today I did a quick update to allow certain options to be set per vehicle, instead of being controlled by race class:
- Indicators / Headlights / Pit limiter light (in 'Light Colours')
- Handbrake / Horn / Starter Motor (in 'Allow')
- Oval suspension arms (in 'Suspension')
- Grooved slicks (in 'Wheel Object')
To use these settings if they have been adjusted in Editor D24, you will need LFS Test Patch D24.
- Indicators / Headlights / Pit limiter light (in 'Light Colours')
- Handbrake / Horn / Starter Motor (in 'Allow')
- Oval suspension arms (in 'Suspension')
- Grooved slicks (in 'Wheel Object')
To use these settings if they have been adjusted in Editor D24, you will need LFS Test Patch D24.
Today I did a quick update to allow certain options to be set per vehicle, instead of being controlled by race class:
- Indicators / Headlights / Pit limiter light
- Handbrake / Horn / Starter Motor
- Oval suspension arms
- Grooved slicks
You will need Test Patch D24 to use these adjusted values if they have been set in Editor D24.
https://www.lfs.net/forum/thread/102117
- Indicators / Headlights / Pit limiter light
- Handbrake / Horn / Starter Motor
- Oval suspension arms
- Grooved slicks
You will need Test Patch D24 to use these adjusted values if they have been set in Editor D24.
https://www.lfs.net/forum/thread/102117
Thanks for that. The offset seems to be inside LFS and it is reading a block of memory from a file.
From the information available, it's not possible to know which code was calling that function at the time.
As no-one else has reported this bug, I am wondering if it could be a corrupted mod file on your hard drive. Corrupted files are rare but it can happen. There is one way to test this. I don't know how much you like this idea but if you would rename your mods folder to something else (or delete it) then you will download mods fresh again.
That doesn't explain why D21 crashed and D20 didn't. According to this theory, any available version would crash if someone on the server was using the mod that is corrupted on your hard drive.
If it's OK with you, I'd be grateful if you would try this test because if the problem doesn't come up again that could mean it really was a corrupted file issue, and not something I should look into right now.
From the information available, it's not possible to know which code was calling that function at the time.
As no-one else has reported this bug, I am wondering if it could be a corrupted mod file on your hard drive. Corrupted files are rare but it can happen. There is one way to test this. I don't know how much you like this idea but if you would rename your mods folder to something else (or delete it) then you will download mods fresh again.
That doesn't explain why D21 crashed and D20 didn't. According to this theory, any available version would crash if someone on the server was using the mod that is corrupted on your hard drive.
If it's OK with you, I'd be grateful if you would try this test because if the problem doesn't come up again that could mean it really was a corrupted file issue, and not something I should look into right now.
Thanks for the report. This has happened as the mod failed an upload check, due to a new upload checker that I put in place. It's not that anything is really wrong with the upload checker, but when people don't use a recent editor test patch to export their mod, it's possible for this issue to come up due to issues in their old version of the editor.
Then, when that does happen, the website doesn't really do the right thing, as you noticed.
I've notified Victor to see what he can do on his side, and also I've reinstated the original upload checker so this doesn't keep happening right now.
I will now notify the uploaders of a few mods that have gone wrong in this way since yesterday.
Then, when that does happen, the website doesn't really do the right thing, as you noticed.
I've notified Victor to see what he can do on his side, and also I've reinstated the original upload checker so this doesn't keep happening right now.
I will now notify the uploaders of a few mods that have gone wrong in this way since yesterday.
The mod upload checker has been updated to a new version.
I recommend that mod creators should use the latest test patch, to avoid any inconsistency with the upload checker.
EDIT: I have now reverted to the original mod checker. Although now there is nothing wrong with the new one, there can be inconsistencies when mod creators don't use a recent update of the editor. This is too often the case, so it isn't yet the best time to use the new mod checker.
I recommend that mod creators should use the latest test patch, to avoid any inconsistency with the upload checker.
EDIT: I have now reverted to the original mod checker. Although now there is nothing wrong with the new one, there can be inconsistencies when mod creators don't use a recent update of the editor. This is too often the case, so it isn't yet the best time to use the new mod checker.
Last edited by Scawen, .
Sorry about that. I don't yet know how it could have gone wrong in that way.
I have reinstated the old veh checker on the server. Please try editor D23 and let me know if it works (also, D21 or others should be OK).
Only D22 should *not* be OK, because it does save the subobject names, which will fail on the reverted veh checker.
D23 *does* save subobject names if you share for development, but not if you export for submission. So that should solve the issue that kenblock30 reported, in a more specific way.
Now I'll try to figure out why the new veh checker didn't work as expected.
EDIT: (4 May 11:03 UTC) I found out why the veh checker didn't work. I have made the necessary change and tested it manually. I have put a new version there, which hopefully causes no problems (and should allow use of editor D22).
EDIT2: (4 May 12:38 UTC) I've reinstated the old veh checker again as there is some other error that I can't understand yet.
EDIT3: (4 May 13:46 UTC) I found out why the 2nd new veh checker didn't work. It was a completely different issue from the 1st new checker's problem. I have made the necessary change and tested it manually and also by doing a test submission of a private mod. I have put a new version there, which hopefully causes no problems (and should allow use of editor D22).
I have reinstated the old veh checker on the server. Please try editor D23 and let me know if it works (also, D21 or others should be OK).
Only D22 should *not* be OK, because it does save the subobject names, which will fail on the reverted veh checker.
D23 *does* save subobject names if you share for development, but not if you export for submission. So that should solve the issue that kenblock30 reported, in a more specific way.
Now I'll try to figure out why the new veh checker didn't work as expected.
EDIT: (4 May 11:03 UTC) I found out why the veh checker didn't work. I have made the necessary change and tested it manually. I have put a new version there, which hopefully causes no problems (and should allow use of editor D22).
EDIT2: (4 May 12:38 UTC) I've reinstated the old veh checker again as there is some other error that I can't understand yet.
EDIT3: (4 May 13:46 UTC) I found out why the 2nd new veh checker didn't work. It was a completely different issue from the 1st new checker's problem. I have made the necessary change and tested it manually and also by doing a test submission of a private mod. I have put a new version there, which hopefully causes no problems (and should allow use of editor D22).
Last edited by Scawen, .
The LFS developers should host instructions on how to reverse engineer someone's software without their permission?
Honestly, I'm not even sure programs should be allowed to work the way LFS Lazy does, hacking into LFS. But it was allowed in this case, so that is beside the point.
I've added a couple of the most requested features from LFS Lazy into a recent test patch.
I know there are many other features of Lazy that are still wanted, though I assume you would prefer me to to continue working on the new tyre physics, instead of trying to implement the features of LFS Lazy?
It's possible to recreate many of the features of LFS Lazy without resorting to hacking obsolete software which is itself a hack, even if a very good one. I mentioned above, the PACT Driving Assistant which is currently in development has many features that may be of interest (although I don't know which features of LFS Lazy can or cannot be done by PACT).
Honestly, I'm not even sure programs should be allowed to work the way LFS Lazy does, hacking into LFS. But it was allowed in this case, so that is beside the point.
I've added a couple of the most requested features from LFS Lazy into a recent test patch.
I know there are many other features of Lazy that are still wanted, though I assume you would prefer me to to continue working on the new tyre physics, instead of trying to implement the features of LFS Lazy?
It's possible to recreate many of the features of LFS Lazy without resorting to hacking obsolete software which is itself a hack, even if a very good one. I mentioned above, the PACT Driving Assistant which is currently in development has many features that may be of interest (although I don't know which features of LFS Lazy can or cannot be done by PACT).
Last edited by Scawen, .
Yes, that would be good to know.
The editor test patch (D19) has a lot of good updates: https://www.lfs.net/forum/thread/102626
I've received another South City update from Eric so did a few track editor updates that were relevant at this point. In the process I updated a few features from the public editor, so here they are.
Editor Test Patch D19:
Hotkey for "hide/unhide (un)selected" changed from X to H (like blender)
- small H button at top left to hide/show message history (old H function)
Left click increments should now always be smaller than right click
- previously this was not consistent for all types of buttons
- as before, CTRL may do smaller steps, SHIFT may do larger steps
- this change should apply to distance, colour, angle and scale
Find button for error triangles or points now selects all as expected
- previously only selected the first point or triangle in the list
Editor Test Patch D19:
Hotkey for "hide/unhide (un)selected" changed from X to H (like blender)
- small H button at top left to hide/show message history (old H function)
Left click increments should now always be smaller than right click
- previously this was not consistent for all types of buttons
- as before, CTRL may do smaller steps, SHIFT may do larger steps
- this change should apply to distance, colour, angle and scale
Find button for error triangles or points now selects all as expected
- previously only selected the first point or triangle in the list
I'm working on it. Progress is not stalled.
I did spend a lot of time last month on test patch and editor updates, but that's not my focus now. I won't go into all the reasons why that happened, but there was an important update, then I saw the E-Challenge, saw something that could be improved there, thought of a few other useful improvements, one thing lead to another...
Not a problem, that's all good. The updates were significant and important.
The idea of the progress reports wasn't really "this is what I will do forever". It was an insight into the development process, and I went into extreme detail, maybe even too personal, so people could really understand what being a programmer is about.
Tyre physics doesn't go the way multithreading goes. For me, it's not really suitable for rapid progress reports. There's a lot of experimentation and testing, research and hard thinking.
However much you want the new version released, multiply that by 100 (rough estimate) and that's how much I want it released. The new version is what I work on and think about for many hours every day. All I can really say at this point is don't worry, I'm on it.
I did spend a lot of time last month on test patch and editor updates, but that's not my focus now. I won't go into all the reasons why that happened, but there was an important update, then I saw the E-Challenge, saw something that could be improved there, thought of a few other useful improvements, one thing lead to another...
Not a problem, that's all good. The updates were significant and important.
The idea of the progress reports wasn't really "this is what I will do forever". It was an insight into the development process, and I went into extreme detail, maybe even too personal, so people could really understand what being a programmer is about.
Tyre physics doesn't go the way multithreading goes. For me, it's not really suitable for rapid progress reports. There's a lot of experimentation and testing, research and hard thinking.
However much you want the new version released, multiply that by 100 (rough estimate) and that's how much I want it released. The new version is what I work on and think about for many hours every day. All I can really say at this point is don't worry, I'm on it.
Here's another update. Hopefully you should find it is more tolerant with bad normals (finds less normals to be bad) and more informative about small or thin triangles that can cause bad normals.
Merge by distance can now be used over a whole mesh and instead of aborting after 12 seconds, it shows progress every second so you can stop it if you want.
Editor Test Patch D18:
New function "invert" to invert selection (selected - unselected)
Point merge options now on one line and includes distance merge
Improved merge by distance to update every second if merge is long
- you can watch the progress and stop at any time by pressing ESC
Improved handling of bad normals and some of the reasons for them
- triangles that are too small or too thin are separately listed
- triangles as small as 1/3 of a millimetre can contribute normals
Merge by distance can now be used over a whole mesh and instead of aborting after 12 seconds, it shows progress every second so you can stop it if you want.
Editor Test Patch D18:
New function "invert" to invert selection (selected - unselected)
Point merge options now on one line and includes distance merge
Improved merge by distance to update every second if merge is long
- you can watch the progress and stop at any time by pressing ESC
Improved handling of bad normals and some of the reasons for them
- triangles that are too small or too thin are separately listed
- triangles as small as 1/3 of a millimetre can contribute normals
Last edited by Scawen, .
Reason : typo
Editor Test Patch D17:
I'm short of time today but added a couple of features that make yesterday's model error reporting more useful.
New 'find' buttons to select error triangles or bad normals
New button to select a point by entering the point index
Index of a single selected point or triangle is displayed
Rearranged buttons in tri mode to line up with point mode
I'm short of time today but added a couple of features that make yesterday's model error reporting more useful.
New 'find' buttons to select error triangles or bad normals
New button to select a point by entering the point index
Index of a single selected point or triangle is displayed
Rearranged buttons in tri mode to line up with point mode
Editor Test Patch D16:
Vehicle editor:
Spare wheel can now be offset laterally (set RIGHT value)
FIX: Save as SIT / STR name now limited to 7 characters
FIX: Dashboard texture can now be updated in a 2D view
Modeller:
Removed message spam about bad normals - instead a report is displayed at the top right in the modeller. This feature is not yet finished as I would like to add a button to help by selecting one of the offending points or triangles.
Vehicle editor:
Spare wheel can now be offset laterally (set RIGHT value)
FIX: Save as SIT / STR name now limited to 7 characters
FIX: Dashboard texture can now be updated in a 2D view
Modeller:
Removed message spam about bad normals - instead a report is displayed at the top right in the modeller. This feature is not yet finished as I would like to add a button to help by selecting one of the offending points or triangles.
I just found out that I can allow the spare wheel to be offset, with no compatibility issues. I hope to release an editor test patch today or tomorrow. I'll check your bug reports thread as well.
OK, this is now fixed in Editor Test Patch D15.
The solutions:
1) Vehicle editor will not crash (in Textures tab or when creating car from model)
2) There is a message "Too many materials"
3) It will not export for testing or upload if there are too many materials
This must be the best way, as LFS itself will crash if there are too many materials. I believe there should be no need for so many materials, if you can share texture pages, etc.
Thanks for the crash report!
The solutions:
1) Vehicle editor will not crash (in Textures tab or when creating car from model)
2) There is a message "Too many materials"
3) It will not export for testing or upload if there are too many materials
This must be the best way, as LFS itself will crash if there are too many materials. I believe there should be no need for so many materials, if you can share texture pages, etc.
Thanks for the crash report!
FGED GREDG RDFGDR GSFDG