Timing Gates and Position

Thanks for that. I've managed to create some smoother curves in the questionable areas. Trial and error and honestly, i couldn't tell you how i did it... :(

Now going to attempt to edit the AI boundaries.

But.... after you've edited the curves etc.. how do you actually save the file you're working on? I can't see any SAVE button.
 
Unfortunately, I have to work, but I will try to make a tutorial for this, as I feel it is a big hurdle for many people. I'm so old that when I was a kid, I had a Dos computer, so it seems pretty easy to me :D

In the meantime, here are a few pointers to help you get going.

[...]

I hope this is not too confusing. I will try to make a video on the subject to help everyone who gets to this point and give up because it is just too hard.
This is outstanding, @Speednut357 :) So well explained! :inlove: Thank you.
 
Last edited:
At work now. But from memory, type new name in the field at top and hit export. It saves as a plain txt file. Rename the txt file to trackname.AIW
Can i pick your brain again Speednut.

Here's a short clip of one of the tracks I'm having trouble with. At 47 seconds into the lap you'll see the AI all dramatically slow for a corner that is supposed to be pretty much flat out.

The original track build for this section had a blind crest apex, and a dip, which i didn't like. The AI were dabbing the brakes but were still pretty quick through the section. I reshaped the track/terrain elevation and smoothed the 'dip' out in BTB/3DsimED, exporting a new AIW file. Since when the AI are now braking hard and coming to a standstill as seen in the video. They were not doing that before.

In AIW editor I can't see any data that suggests the AI are instructed to brake hard. I've clicked on all the waypoints and in the green leftside menu for AI braking, the values (1, 1, 0,9) are the same as the rest of the track.
Is there somewhere else i can check the AI braking data?

For example. Each waypoint has this data:
wp_pos=(501.300,2.450,-276.008) // 187
wp_perp=(0.4673,0.0000,0.8840)
wp_normal=(0,0.9999,0)
wp_vect=(-4.520,0.000,2.413)
wp_width=(6.495,7.530,10.000,10.000)
wp_dwidth=(10.000,10.000,0.000,0.000)
wp_path=(4.3305,0.0000)
wp_lockedAlpha=(0)
wp_galpha=(.5)
wp_groove_lat=(0)
wp_test_speed=(-1.0)
wp_score=(0,553.382)
wp_cheat=(-1.0)
wp_pathabstractionspeed=(0.0000)
wp_pathabstraction=(0,-1)
wp_wpse=(0,0)
wp_branchID=(0)
wp_bitfields=(0)
wp_lockedLats=(1)
wp_multipathlat=(0.0000,0.0000)
wp_translat=(0.0000, 0.0000)
wp_pitlane=(1)
WP_PTRS=(186,188,-1,0)

Do you know which line of text is the braking instruction, if i could possibly change it in notepad. ?

I've smoothed the racing line in AIW editor, and opened up the boundaries to avoid any bottleneck, and i can't see any reason why the AI would register the section as a braking point or corner. They should be going through here pretty quickly.

As you can see from the rest of the lap they are on pace. So it's just this short section that's ruining the track.

Any ideas?

See the track lap here >
 
Hi, The speed adjustment for any single waypoint is the wp_cheat=
If you get that section of road loaded up in the Aiw editor and click on the individual waypoints, you should be able to spot the one which, for whatever reason, has a lower value.

One issue I have had is that if a track has had a low value put into a wp_cheat to slow the cars, Sometimes the editor will not want to overwrite the value. I don't know why but the software is free, so I cannot complain.
Just take note of the coordinates in wp_pos and use the search function in your text editor to locate the waypoint, and change the wp_cheat value in the text file. This only happened to me once, but it may be the issue you are seeing.

wp_width=road left, road right, far left, far right
wp_dwidth=collision left, collision right, cut left, cut right
wp_path=latitude for dry line and wet line (offset from wp_pos along wp_perp)
wp_lockedAlpha=(used for editor specific algorithms)
wp_galpha=alpha for groove
wp_score=sector ID, distance into lap
wp_cheat=multiplier to tweak the speeds through this point
 
A couple of other thoughts

I have had a situation where there was a very small kink in the Aiw path that I couldn't see because the detail was very small. I had to zoom right in to spot it.

It also occurred to me that if you changed the height data of the track, the waypoint may be wrong and you may need to adjust the z coordinate? or possibly start again with btb.
 
Fascinating. I've been playing with this editor sporadically with little luck; I'm reminded of much of the GPL software - free, you fire it up and stare blankly "now what", documentation consists of endless forum posts "how do I....?"

Most online data I've found is more concerned with creating an AIW file than with editing one. I have several tracks (rfactor and GTR2) with irritating quirks which prevent decent racing. Such as Cadwell in rfactor, where the AI cannot handle the quick right-left flick at the top of the hill, invariably running wide on the exit, spinning in the grass, flying back across the track to the left, causing further mayhem by trying to get back on track in front of oncoming cars (usually me); even in testing, with a mere 8-10 AI on the track, there is a near constant yellow flag there.

Now I have more ammo to tackle this idiocy (though "a little knowledge is a dangerous thing"; at least i make backups first). Thanks.
 
Speednut, you're a legend.

I did actually go back into BTB and generate a new AIW file with the correct elevations but that didn't solve it.

I had actually opened the two AIW files (older file without the slowing AI, and the newer with the slowing error) located the individual waypoints for that section of track, and in notepad, copied/paste the 'good' waypoints in to replace the 'bad' ones. When i opened the AIW file in editor all the boundaries and edges were missing. Go figure. and when i tried to load the track in the game it loaded the track but i couldn't exit the garage.

Clearly that method doesn't work :(

I suspect your theory of a minute kink somewhere in that section, or an incorrect wp_cheat value is the issue. Now i know wp-cheat is the text to look for that, and zooming in on the racing line will be my next focus.

Thanks for your help.
 
Zoomed in to the line and saw the section by waypoint 398 is long and flat. Isn't broken into multiple minute segments as with the other waypoints. Have tried smoothing and 'creating curve' but it's not altering the segment. The Z value was also 0.1 lower than adjacent waypoints 397 and 399, which would effectively make a trough across the track. so i edited the Z value to make its value fall between the two.

Was still getting the slowing AI problem so went back into BTB, cleared the entire Lines, appended / regenerate the AIW racing lines, and now it's running ok.
 

Attachments

  • offending waypoint.jpg
    offending waypoint.jpg
    418.6 KB · Views: 49
Last edited:
Glad to hear @Caribstu that the problem has been sorted! :) And thanks for sharing your solution.

For future reference, I've just remembered something else that could be helpful. If you press "h" for help in the editor, there is a list of useful keystrokes. One being "Alt-Ctrl-i", which inserts a track or pitlane waypoint after whichever waypoint you have highlighted.

This maybe could have helped with your problem... since an extra waypoint could possibly have helped any curve you created to be more gradual. Only challenge is that when I've used this a couple times, I've noticed the auto-generated AI corridors and track limits at the new waypoint can be messy – so you might need to clean them up manually.
 
Last edited:
... I'm so old that when I was a kid, I had a Dos computer, so it seems pretty easy to me...
My first computer experience wasn't until college - HP2000e, time shared basic, TTY terminals, acoustic couplers (aka cradle modems), and punch tape recording.
 

Latest News

What would be the ideal raceday for you to join our Club Races?

  • Monday

    Votes: 13 12.1%
  • Tuesday

    Votes: 10 9.3%
  • Wednesday

    Votes: 10 9.3%
  • Thursday

    Votes: 12 11.2%
  • Friday

    Votes: 40 37.4%
  • Saturday

    Votes: 61 57.0%
  • Sunday

    Votes: 42 39.3%
Back
Top