- 1,437
- The Motherland
- Manasseh257
Terrain Testing/Exploring (Bump/Jump Maps) (OP Updated 10/3/15)
- Thread starter Quakebass
- 103 comments
- 13,582 views
- 1,437
- The Motherland
- Manasseh257
- 5,489
- mid west USA
- Otaliema
I honesty haven't looked at sector times on a custom track yet.
Do the number of sectors cap out at a particular number or is it a fixed number? Say 3, 4 & 5 sectors depending length. So much if this app is linked to distance it would make sense the sectors would be too.
- 1,437
- The Motherland
- Manasseh257
- 3,808
How long before PD think you guys are having too much fun and patch it out? they did it in GT5 which annoyed a lot
- 263
- Munich
- Hitokiri-666
The guys from PD must know, that as long as they give us this tool, we will use every millimeter/inch of it to squeeze all out what is possible. Except of this, I don't see the problem with "jump-tracks", as Seattle was already also a "jump-track" made by PD. ...well... its no longer in the game... 
Quakebass
Premium
- 1,899
- Washington
- quakebass
Large Update: Track Width + Anchor Point Cause-and-Effect Findings, and Definite Track-Laying Methods/Strategies for Rough Terrain
Ok, first off, I began with the widest track width setting, but didn't begin comparing track width effect till much later - near the end, really. The first thing I started testing was anchor point methods, and got a butt-ton of results. The definitive track laying methods are in the final summary. Additionally, everything I tested was on Eifel Flat with maximum banking settings - I will (likely) do everything I've tried here on Death Valley to see how non-static elevation affects things.
Everything here will add up into summarized findings:
Anchor Point Cause-and-Effect:
With the first test, I simply crammed in anchors using the the "Add Anchor" function:
And oddly, it resulted in absolutely no bumps, and even stripped the corner of it's banking (the first is the anchored corner, the second non-anchored, to show the difference):
Sooo, using the existing anchors, I went in and made the first straight all wiggly, which gave way to bumps:
...That wiggle was made by leaving all the anchors pointed vertically, and sliding the first anchor left to the minimum radius, and the second anchor right to the minimum radius, and so forth. I re-used the anchors, and again kept each one vertical, and slid them as close and as far right as could to the previous, as shown:
...And got these kinds of bumps:
Because it would be too time consuming to re-adjust all of these anchors, I went and drew up a new quick course, and alternated between vertical-facing and rotated anchors; interestingly, it produced sort of "elongated" bumps that were still large, but smooth due to the distance they covered:
Again, to save myself from re-adjusting anchors, I started another new course, and tried a few things at once:
I started off by leaving the anchors where they sat after drawing them (rather than moving them to their minimum distance like usual, and opposed to using the Add Anchor tool):
The straighter part resulted in near-flat road, with some minor (but still notable) bumps here and there, and the alternated straight/rotated section produced an interesting bump formation; in addition to being spaced farther apart, the actual shape of the bumps were slightly different - the front side was smoother and and more gradual, and the back end or "drop" was more steep and extreme - it was difficult to capture on my iPad, but if you look closely you can see the shading:
I then went in and compressed the anchors to their minimum distance on the first straight:
...And also aligned the anchors as straight as I could (note that it's labeled as a single distance and doesn't register a radius value) to see how it differed from the "Add Anchor" tool:
...And now the straight had mild rough terrain:
So, I tried moving all the anchors in a more jittery path (making sure each registered distance and radius values) to see how it affected the bumps, and lo-and-behold, they became more extreme. They still felt somewhat realistic near the center, but the outer parts became clear that they were results of banking functions, too smooth and bouncy in my opinion:
I then tried closing the gap with the straight/rotated placement type, and the bumps became more dense, but still kept the unique shape:
...And that largely concluded my anchor placement testing on Eifel Flat.
Track Width Cause-and-Effect:
I went with the lowest track width setting to re-try these tests. This is the layout I published:
As discovered by @Otaliema , track width affects the minimum track radius and anchor distance, so it allows denser placement. The first section used the lateral slide placement, and the other two used random placement, though the last one was on a wide corner after using the Add Anchor tool:
All the previous tests pertaining to rough terrain largely checked out. Results on track width are below:
From all of these tests, this is what I determined:
*The use of the word "bumps" pertains purely to rough terrain in this particular context:
On Eifel Flat:
- Anchored track that follows a perfectly straight or circular path, or is on a path created purely with the "Add Anchor" function will not produce bumps and will likely remove banking.
- Moving an anchor (while staying densely packed) from any of previously mentioned path types will result in some kind of bump.
- The most aggressive bumps are created with minimal radius and length in conjunction.
- There is a minute difference in bumps when using "Add Anchor" tool vs. standard track-drawing ("add anchor" seems to keep both sides largely balanced/equal, drawing seems to smoothen the front side and sharpen the backside [or "drop"] of a bump).
- Anchor points laid in a regular or repetitive fashion will result in a regular/repetitive bump formation.
- (Perhaps obviously) Increasing distance between anchors will increase distance between bumps.
- An alternating vertical (or any initial direction) and rotated (at a regular angle, usually 30-40 degrees off the initial direction) method will produce larger, but longer and smoother bumps than usual, due to banking.
- A wider track width (likely due to the banking lifting/lowering either sides of the Tarmac) will result in larger and more extreme bumps.
- In addition to doing the inverse of wide track, thin track (from the feedback my GT3 R8 has given me) also feels more realistic - whether or not this is because the track pieces/anchor distance can be made shorter (and therefore bumps more dense) is uncertain, but it is a definite function.
- This "realism vs. bump size" scales linearly with track width (Wide track = large bumps, unrealistic terrain feel, thin track = small bumps, realistic terrain feel).
- For this reason, in making rough terrain courses, you will have to decide between realistic driving feel and road space.
- All banking functions/data from previous tests apply to all data found here.
- While still minimizing distance and radius, anchor points placed in more random/non-pattern based positions will definitively produce more realistic and "random" feeling terrain.
I did a good 5-ish in total layouts this time (rather than basically one), and many of those had several edits to multiple parts - but only one test track is getting published from all these tests. Most everything I built, despite providing useful data and results, was bland and repetitive, and near-copies of one another. The last one I deemed an entertaining drive, and mainly kept it because it showcases the item at the end of the discoveries/findings that I put in bold. It's probably the most bland of my tracks so far, but I really like the feel of the road.
My next tests, as I said at the top, will be on Death Valley to try some new things with the new theme. I'll also be testing how these different track laying methods are affected by elevation, and if there are different algorithms for dealing with elevation depending on the theme you use. I really hope there isn't, because then I'll have to re-do everything I did in this post three more times... Which was tedious, to say the least.
I think these next tests will be more fun...
Ok, first off, I began with the widest track width setting, but didn't begin comparing track width effect till much later - near the end, really. The first thing I started testing was anchor point methods, and got a butt-ton of results. The definitive track laying methods are in the final summary. Additionally, everything I tested was on Eifel Flat with maximum banking settings - I will (likely) do everything I've tried here on Death Valley to see how non-static elevation affects things.
Everything here will add up into summarized findings:
Anchor Point Cause-and-Effect:
With the first test, I simply crammed in anchors using the the "Add Anchor" function:
And oddly, it resulted in absolutely no bumps, and even stripped the corner of it's banking (the first is the anchored corner, the second non-anchored, to show the difference):
Sooo, using the existing anchors, I went in and made the first straight all wiggly, which gave way to bumps:
...That wiggle was made by leaving all the anchors pointed vertically, and sliding the first anchor left to the minimum radius, and the second anchor right to the minimum radius, and so forth. I re-used the anchors, and again kept each one vertical, and slid them as close and as far right as could to the previous, as shown:
...And got these kinds of bumps:
Because it would be too time consuming to re-adjust all of these anchors, I went and drew up a new quick course, and alternated between vertical-facing and rotated anchors; interestingly, it produced sort of "elongated" bumps that were still large, but smooth due to the distance they covered:
Again, to save myself from re-adjusting anchors, I started another new course, and tried a few things at once:
I started off by leaving the anchors where they sat after drawing them (rather than moving them to their minimum distance like usual, and opposed to using the Add Anchor tool):
The straighter part resulted in near-flat road, with some minor (but still notable) bumps here and there, and the alternated straight/rotated section produced an interesting bump formation; in addition to being spaced farther apart, the actual shape of the bumps were slightly different - the front side was smoother and and more gradual, and the back end or "drop" was more steep and extreme - it was difficult to capture on my iPad, but if you look closely you can see the shading:
I then went in and compressed the anchors to their minimum distance on the first straight:
...And also aligned the anchors as straight as I could (note that it's labeled as a single distance and doesn't register a radius value) to see how it differed from the "Add Anchor" tool:
...And now the straight had mild rough terrain:
So, I tried moving all the anchors in a more jittery path (making sure each registered distance and radius values) to see how it affected the bumps, and lo-and-behold, they became more extreme. They still felt somewhat realistic near the center, but the outer parts became clear that they were results of banking functions, too smooth and bouncy in my opinion:
I then tried closing the gap with the straight/rotated placement type, and the bumps became more dense, but still kept the unique shape:
...And that largely concluded my anchor placement testing on Eifel Flat.
Track Width Cause-and-Effect:
I went with the lowest track width setting to re-try these tests. This is the layout I published:
As discovered by @Otaliema , track width affects the minimum track radius and anchor distance, so it allows denser placement. The first section used the lateral slide placement, and the other two used random placement, though the last one was on a wide corner after using the Add Anchor tool:
All the previous tests pertaining to rough terrain largely checked out. Results on track width are below:
From all of these tests, this is what I determined:
*The use of the word "bumps" pertains purely to rough terrain in this particular context:
On Eifel Flat:
- Anchored track that follows a perfectly straight or circular path, or is on a path created purely with the "Add Anchor" function will not produce bumps and will likely remove banking.
- Moving an anchor (while staying densely packed) from any of previously mentioned path types will result in some kind of bump.
- The most aggressive bumps are created with minimal radius and length in conjunction.
- There is a minute difference in bumps when using "Add Anchor" tool vs. standard track-drawing ("add anchor" seems to keep both sides largely balanced/equal, drawing seems to smoothen the front side and sharpen the backside [or "drop"] of a bump).
- Anchor points laid in a regular or repetitive fashion will result in a regular/repetitive bump formation.
- (Perhaps obviously) Increasing distance between anchors will increase distance between bumps.
- An alternating vertical (or any initial direction) and rotated (at a regular angle, usually 30-40 degrees off the initial direction) method will produce larger, but longer and smoother bumps than usual, due to banking.
- A wider track width (likely due to the banking lifting/lowering either sides of the Tarmac) will result in larger and more extreme bumps.
- In addition to doing the inverse of wide track, thin track (from the feedback my GT3 R8 has given me) also feels more realistic - whether or not this is because the track pieces/anchor distance can be made shorter (and therefore bumps more dense) is uncertain, but it is a definite function.
- This "realism vs. bump size" scales linearly with track width (Wide track = large bumps, unrealistic terrain feel, thin track = small bumps, realistic terrain feel).
- For this reason, in making rough terrain courses, you will have to decide between realistic driving feel and road space.
- All banking functions/data from previous tests apply to all data found here.
- While still minimizing distance and radius, anchor points placed in more random/non-pattern based positions will definitively produce more realistic and "random" feeling terrain.
I did a good 5-ish in total layouts this time (rather than basically one), and many of those had several edits to multiple parts - but only one test track is getting published from all these tests. Most everything I built, despite providing useful data and results, was bland and repetitive, and near-copies of one another. The last one I deemed an entertaining drive, and mainly kept it because it showcases the item at the end of the discoveries/findings that I put in bold. It's probably the most bland of my tracks so far, but I really like the feel of the road.
My next tests, as I said at the top, will be on Death Valley to try some new things with the new theme. I'll also be testing how these different track laying methods are affected by elevation, and if there are different algorithms for dealing with elevation depending on the theme you use. I really hope there isn't, because then I'll have to re-do everything I did in this post three more times... Which was tedious, to say the least.
I think these next tests will be more fun...
LeGeNd-1
Premium
- 7,264
- Australia
- GTP_LeGeNd-1
Simply stellar detective work you're doing here, @Quakebass 
I'm still wondering if we can make a "Wings of the Red Dragon" style jump though. I've tried your jump tracks, and they are a lot of fun, but it's still just a series of sleeping policemen style of bumps
I'm thinking we should scour the terrain for a steep drop off point, then make a straight leading up to it, then use the close packed anchor technique to add a steep ramp just on the edge
I'm still wondering if we can make a "Wings of the Red Dragon" style jump though. I've tried your jump tracks, and they are a lot of fun, but it's still just a series of sleeping policemen style of bumps
I'm thinking we should scour the terrain for a steep drop off point, then make a straight leading up to it, then use the close packed anchor technique to add a steep ramp just on the edge
Quakebass
Premium
- 1,899
- Washington
- quakebass
Sorry for the delay, some real life happened; I actually made some really fast findings, just didn't have time to make a post about it.
I'm pressed for time so I'm not gonna go ham on the formatting of this one, sorry if that's an issue. Additionally, I didn't see a lot of need to photograph the findings I found - you'll understand why.
1).
After starting a course on Death Valley, I immediately found that the course laying methods and algorithms are the same as Eifel Flat - so I have reason to believe that I can confirm it to be universal across all themes.
2).
I have confirmed that any track laid on level terrain can be made into a rough terrain-type course. Keeping a realistic road surface becomes increasingly difficult when track is laid on jagged terrain or an increasing hill gradient - and this is in addition to the functions of realism I found in my last few findings (banking, track width, anchor placement, etc.).
3).
I can re-confirm that laying dense anchors on steeper hill gradients create larger bumps/jumps. Darker/lighter terrain coloring hints at the gradient severity:
However, there's more to it:
4).
I started a layout, and I was planning on publishing it immediately like my first few tests, but I wasn't 100% satisfied with the way the jump came out - there was a slight kink that usually made you bump the right-hand wall unless you went waaaay off the recommended driving line. So, I decided to straiten the section out, slightly adjusting the rotation and position of a few anchor points, but when I transferred the new version, the thing got completely changed. The initial jump was flattened out, and the second bump was far more severe. The original sort of "launch and catch" set up was entirely different, and not really fun. Unfortunately my iPad died and I wasn't able to undo. So, I tried moving the anchors back to their original position - and again, got entirely different results. I still feel the original variant was my best one, but this is the version I settled on (which was the upward-of-like-15 adjustments):
I still haven't published it 'cuz I haven't settled on my favorite banking severity yet, and I still want to test a low-downforce car. But what I've found is that even though it's confirmed that extreme elevation changes cause more extreme jumps/bumps, even just the slightest adjustment can create drastically different results for jumpy/bumpy terrain - so if you aren't satisfied with a layout (be it an attempt for realism, rough terrain, or chaotic jump maps), just play around with anchor placement, and you may find something you like/prefer. It's hard to say if there's really any sort of method to follow because of the level of intricacies you can make with the adjustment tools.
I'll try to get this one published tonight. My tests are largely gonna go back to elevation/map exploring to see which areas and track placements allow the most extreme track conditions, but there's a few smaller things for me to examine as well. I wouldn't be surprised if other unexpected things pop in the process, as they have for most every test I've made so far.
As for the OP, I know for certain I won't have any time to meddle with it until Thursday or Saturday this week. I will be trying to make that one of my priorities, though. There's a lot to go through.
*Edit: Apologies - I made all the adjustments, but forgot to save and publish the track. It's up now, titled "Death Valley Terrain Test 1". Sorry.
I'm pressed for time so I'm not gonna go ham on the formatting of this one, sorry if that's an issue. Additionally, I didn't see a lot of need to photograph the findings I found - you'll understand why.
1).
After starting a course on Death Valley, I immediately found that the course laying methods and algorithms are the same as Eifel Flat - so I have reason to believe that I can confirm it to be universal across all themes.
2).
I have confirmed that any track laid on level terrain can be made into a rough terrain-type course. Keeping a realistic road surface becomes increasingly difficult when track is laid on jagged terrain or an increasing hill gradient - and this is in addition to the functions of realism I found in my last few findings (banking, track width, anchor placement, etc.).
3).
I can re-confirm that laying dense anchors on steeper hill gradients create larger bumps/jumps. Darker/lighter terrain coloring hints at the gradient severity:
However, there's more to it:
4).
I started a layout, and I was planning on publishing it immediately like my first few tests, but I wasn't 100% satisfied with the way the jump came out - there was a slight kink that usually made you bump the right-hand wall unless you went waaaay off the recommended driving line. So, I decided to straiten the section out, slightly adjusting the rotation and position of a few anchor points, but when I transferred the new version, the thing got completely changed. The initial jump was flattened out, and the second bump was far more severe. The original sort of "launch and catch" set up was entirely different, and not really fun. Unfortunately my iPad died and I wasn't able to undo. So, I tried moving the anchors back to their original position - and again, got entirely different results. I still feel the original variant was my best one, but this is the version I settled on (which was the upward-of-like-15 adjustments):
I still haven't published it 'cuz I haven't settled on my favorite banking severity yet, and I still want to test a low-downforce car. But what I've found is that even though it's confirmed that extreme elevation changes cause more extreme jumps/bumps, even just the slightest adjustment can create drastically different results for jumpy/bumpy terrain - so if you aren't satisfied with a layout (be it an attempt for realism, rough terrain, or chaotic jump maps), just play around with anchor placement, and you may find something you like/prefer. It's hard to say if there's really any sort of method to follow because of the level of intricacies you can make with the adjustment tools.
I'll try to get this one published tonight. My tests are largely gonna go back to elevation/map exploring to see which areas and track placements allow the most extreme track conditions, but there's a few smaller things for me to examine as well. I wouldn't be surprised if other unexpected things pop in the process, as they have for most every test I've made so far.
As for the OP, I know for certain I won't have any time to meddle with it until Thursday or Saturday this week. I will be trying to make that one of my priorities, though. There's a lot to go through.
*Edit: Apologies - I made all the adjustments, but forgot to save and publish the track. It's up now, titled "Death Valley Terrain Test 1". Sorry.
Last edited:
- 1,629
- Texas
- RC45
- RC45
- 1,629
- Texas
- RC45
- RC45
Managing terrain surface is simple.
Childs play actually, if you have seen how civil engineers layout a railroad road base - simply anchor the terrain in 2 places of similar altitude and the course maker will join them making a cut through high ground or filling in low ground.
Use this logic to create what ever level of severity in road surface you desire.
You can have mild dips, bumps, crests or even gradual slopes ramps or none at all.
- 124
- Adelaide
- Belbanj12
- Are You Kidding Me? N/A
Hey guys, I made a track exploring the Andalusia Map and I found a large jump. I don't remember where it is however it is on a corner along the border of the map. Course Editor screenshots are below:
https://www.gran-turismo.com/au/gt6/user/#!/friend/Belbanj12/course/1262859/
I suspect that it was here, however it may be elsewhere along the border.
I suspect that it was here, however it may be elsewhere along the border.
