FAR pathing issues?

delta9 · February 07, 2009, 11:57:57 AM

Take a look at these.

See those cars shooting off? At least they do graphically, but on either side and in between the numbers stay the same. I see a gap in the paths on the right lane but is that just graphical because of the ground? But when you query the two squares on the transition piece that have the least actual road on them, they only show the commutes of whichever lane is in that square. Is that supposed to happen? Not all houses follow that glitch path, but I have found some of the ones that do and their commute is still listed as short. But the odd thing is that there is an option for it to be taken or not, which makes me think there's some odd glitch in the path. At the same time though, it seems to be of no consequence because it doesn't increase the commute time. A wormhole

Tarkus · February 07, 2009, 02:53:38 PM

delta9, that occasionally happens just after you build a new stretch of road out of puzzle pieces. If you run the game a little longer, it should correct itself.

-Alex (Tarkus)

delta9 · February 07, 2009, 03:26:21 PM

They've been there for over 14 game years.

delta9 · February 07, 2009, 07:11:49 PM

And they are certainly persistent. I've reconstructed and made new FAR sections and all of the 18.4 to ortho transition pieces experience the same defect on the inner curve lanes.

I've also found to the contrary of what I said before, all cars on that route ARE forced through the "wormholes". They leave and return upon the same line, and as I said, all cars on the inner curve lane follow that broken path.

CaptCity · February 07, 2009, 07:47:08 PM

I, too, have seen this in the game and it doesn't seem to go away. I think this came up once on the 3RR thread during testing, but I don't know what came of it.

I did take a close look at the piece where the 'extra' arrow seems to take off...

Looking closer at just the paths...

Looks like an extra arrow there...

Don't know what it all means. It doesn't seem to affect commuters that I've noticed.

CaptCity

dedgren · February 07, 2009, 08:26:40 PM

I think I know why this is happening- a zero length path on one of the gridsquares. I made a note when I first saw it- must've lost it.

If I'm right, it can be fixed ASAP. There's some other stuff due for a NAM update soon, so an official fix won't be far away, but I'll see what I can do in the meantime.

Sorry, folks. Don't mean you should suffer from my mile-long to-do list.

David

sithlrd98 · February 07, 2009, 08:35:28 PM

Although I was a beta tester for the FAR , I had never had those problems. I decided to try and replicate delta9 and CaptCity s problems. What I noticed was kinda both issues when the pieces were used in different configurations. I had the problem that delta9s second picture when I used a 45, 45toFAr , Far straight piece and the transition. If I left out the transition piece , I had the issue that CaptCity had except it didn't "shoot off" It only showed one route and it just followed the grid , kinda what David was saying...I think. Is it possible that the paths throughout from the begining of piece 1 piece to the end piece are causing these problems?

Jayson

CaptCity · February 07, 2009, 09:37:13 PM

Just to throw more wood on the fire... (so to speak)...

Configuration did matter in my case but only whether I was viewing the morning commute or evening commute (which makes sense as the outer curve path seems normal), so I tried placing two pieces (it is the 45-to-FAR piece) back to back and noticed the following paths...

In both cases, a 'shoot off' resulted toward the northeast, one in the morning and the other in the evening. Becoming more curious, I laid out all the 45-to-FAR pieces and noticed that the extra path 'arrowhead' always appeared to be facing the same direction...

Anyway, I don't know if this helps now or not, but I thought I would offer it up...

dedgren · February 08, 2009, 12:24:19 AM

CC- Your top pic says it all.

I'm almost sure thast, when I check, the SC4PATH files for the two FAR puzzle piece gridsquares marked with the "X" will have a "zero length" path in it. That's because of where the path ended (precisely crossing the corner) on the adjacent gridsquares. Remember, I was learning to path as I did these- November, '08: seems like porever ago.

I broke it in the first place- I'll fix it.

David

delta9 · February 08, 2009, 09:11:15 AM

Yay, I helped.

I've decided to be more active in this community because I've gained a strong appreciation for all the hard and creative work everyone does here (especially after the last NAM), on their own time (and money), for the enjoyment of everyone else. It's really great stuff.

smoncrie · February 08, 2009, 09:41:07 AM

David, I had a similar problem when I was pathing the elevated rail over road + intersection. I found that SimCity does not like paths that go through the corner of a tile. So I used a path that goes VERY close to the corner:

-7.999,8.0,15.5
-8.0,7.999,15.5

Corners should be avoided (by a small amount) even when the paths are longer.

CaptCity · February 08, 2009, 11:05:10 AM

Ok... curiousity is getting the better of me here (and what must be a great desire to show my ignorance)...

When I placed the two pieces back-to-back (as in the image above), I expected the extra 'shoot-off thingies' to go in opposite directions. But not so, which is why I looked at all the pieces and saw that the path arrow always seemed to be from the same direction. Is it because it's so close to the corner (as smoncrie mentioned), and the game always reads it in that direction?

If the answer is overly complicated, just tell me "that's the way it is" and I'll understand. I was just a little curious...

CaptCity

Jonathan · February 08, 2009, 11:10:26 AM

Probably as the corner is on both sides the game gets confused and as you said it always chooses one side.

Jonathan

dedgren · March 03, 2009, 10:59:34 PM

This bug is, as of today, officially squashed.

I just didn't do very well with this one the first time around.

David

SimFox · March 05, 2009, 04:39:00 AM

Dedgren, I wonder if it would be possible to learn pathing any other way than just "figuring it out oneself"?
I sort of get the whole coordinates system, but the other numbers do seem a bit confusing. In your previous example I sort of assume the number 4 on the 3rd line signifies WEST, right? what are then 1.2, and zeros? Is there any tutorial or explanation?

daeley · March 05, 2009, 06:00:00 AM

All of those can be found in the SC4PATH spec: modthesims [link] and SC4D wiki [link]

Alternatively, if you want to play around with paths, you should try SC4 Path Creator (discussion link, LEX Download link)

homefryes · March 12, 2009, 08:22:43 PM

Quote from: dedgren on March 03, 2009, 10:59:34 PMThis bug is, as of today, officially squashed.

Hi ... forgive the bump, but I am fairly certain I downloaded all the latest stuff on 8 March, and I am still experiencing this pathing issue. Did this corrected file make it into the update?

— Don

CaptCity · March 12, 2009, 10:57:26 PM

homefryes...

Are you sure you got the most recent set? And got rid of all the older stuff?

After installing the newest (9 March 09), it seemed to have been fixed...

...and traffic...

... appeared normal.

Just a thought.

-Charlie

arl85 · March 13, 2009, 02:04:34 PM

I just want to put in evidence that a similar problem occurs on S-shaped turn for STR RAM.

homefryes · March 13, 2009, 07:46:04 PM

Quote from: CaptCity on March 12, 2009, 10:57:26 PMhomefryes...Are you sure you got the most recent set? And got rid of all the older stuff? After installing the newest (9 March 09), it seemed to have been fixed and traffic appeared normal. Just a thought.

I'm sorry, but I'm just not getting it at all. I have just removed all NAM stuff and re-downloaded, and I'm STILL having the issue on the diag-to-FAR curve. I've checked and checked, and I don't have any rogue files that I can find. I ran the Cleanitol file prior to my most recent installation (of a few minutes ago). What am I doing wrong?

— Don