Transport Tycoon Forums

I can't wait for the next stable release! Ohhhh, how much fun it'll be!
*laugh manicly*
FUN!

I've found two bugs in beta 1, the fixes are already sent to Josef. The first bug made callback 2E ignore GRF version 7, the second made callback 2F do the same.

The first one was because of Josef's typo, the second was becasue of an incorrectly applied diff for alpha 72.

In my latest game, I get a chrash when I do certain things on a certain place on the map.
I did not attach a chraslog since it didn't get made. I can reproduce the crash every time in this save by doing the following:
1. Go to the town of Quito
2. There are several oil services running from the Latatunga oil wells to the Quito Oil refinery. Start deleting the bus services. (Remove the roads too, thats when the crash occurs.)
3. The game should crash while you are doing that.

I had this error as well in the final alpha before the beta

has somebody noticed an oddity on the max TE ? it might be that it existed from some months.


another bug; on launching of the Win version the game hanged. I tried deleting the newgrfw.cfg and the result was the same. so I tried after deleting the ttdpatch.cfg and then the game generated a crash log.

hovering teacup wrote:has somebody noticed an oddity on the max TE ? it might be that it existed from some months.

...

?

in the DOS version of the beta the max TE of trains (prop.1f of the action 0) seems ignored.

i think i've spotted a strangely high figure for the TE on the vehicle window some month ago but then i thought it would be just a bug of display without influence on the behaviour of train. (all this is if my memory is corrrect)

the fact in the first paragraph IS real tough.

Must've been Lakie, he was dealing with a new approach to TE.

Csaboka wrote:I've found two bugs in beta 1, the fixes are already sent to Josef. The first bug made callback 2E ignore GRF version 7, the second made callback 2F do the same.

The first one was because of Josef's typo, the second was becasue of an incorrectly applied diff for alpha 72.

Would this be the cause of my getting a lot of invalid sprites? It seems that now just about every second grf in my newgrf has an invalid sprite.

OOPS!

Patchman wrote:If you find that a large number of your .grf files suddenly have "invalid sprite" problems, you have probably accidentally enabled "mandatory GRF Resource Management" mode.

hovering teacup wrote:another bug; on launching of the Win version the game hanged. I tried deleting the newgrfw.cfg and the result was the same. so I tried after deleting the ttdpatch.cfg and then the game generated a crash log.

This appears to be a crash in the newsounds dll, trying to parse the .wav data, but it gets a NULL pointer. How or why, I cannot tell. Have you modified your sample.cat file?

hovering teacup wrote:in the DOS version of the beta the max TE of trains (prop.1f of the action 0) seems ignored.

I think more detail is needed. What train set, what train, what was the value, what should it have been, etc....

Thats strange, indeed.
Since TE without a callback 36 should be the default.
And indeed, please give more information, since with my test trains* worked fine.

* - one with callback 36, and the finnish set betas which doesn't use it.

~ Lakie

Patchman wrote:Have you modified your sample.cat file?

i replaced the sample.cat with an empty file of the same name because i use the win version only for testing.
now i copied the sample.cat from the DOS-TTD folder and launched the game... and it hanged :-/

Lakie wrote:Since TE without a callback 36 should be the default.
And indeed, please give more information

it just happens on my computer with any engine. actually i didn't post much info because the bug just happens universally.
tried deleting the two cfg files and used a brand-new ttdpatch.cfg with only the mountain switch edited to use the realistic aceleration --> no difference. really no clue. :-\

hovering teacup wrote:

Patchman wrote:Have you modified your sample.cat file?

i replaced the sample.cat with an empty file of the same name because i use the win version only for testing.
now i copied the sample.cat from the DOS-TTD folder and launched the game... and it hanged :-/

Where does it hang? In other words, how far does it get into the game? Does removing your newgrfw.cfg fix it?

hovering teacup wrote:it just happens on my computer with any engine. actually i didn't post much info because the bug just happens universally.

But what value does it show for the TE? Is the value in the purchase screen different from the one in-game (after enabling the display of acceleration and TE)?

Please give examples!

Patchman wrote:Where does it hang?

at the startup, with dark screen.

Patchman wrote:does removing your newgrfw.cfg fix it?

no

Patchman wrote:what value does it show for the TE? Is the value in the purchase screen different from the one in-game (after enabling the display of acceleration and TE)?

purchase screen gives correct figures.

i'm not sure if this helps, but there is one engine whose TE increases with the number of wagons it pulls until it hits 21000 kN something.
more precisely, 4707kN with no wagon, 9745kN with one, 14783kN with two, 19821kN with three, 21402kN for further.
(i attach the grf reproducing it on me. the phenomenon happens only when the train's speed is above 101 km/h. --this uses the callback 36--i've not named the engine but its vehicle slot is 1C. it has odd apparence to ease distincion)



just in case, below are randomly picked examples of correct TE (left row) and the actual values engines have (right row) for other than the slot 1C.

104 1600
181 5685
189 3334
124 1836
227 4745
203 10671
330 5368
326 21770
326 991
326 20313
201 4807
307 4308
252 5848


tell me what i should do else.

It seems that the pathfinding has been altered with this beta : now trains that can't find a free exit out of a PBS block are cabable again of choosing a path against a single PBS entry signal.(see picture)
Another thing that I noiced that the signalwaitingtime at a PBS entry signal has been set to 11 days, dispite the setting of 250 days in signal1waittime that I have in my ttdpatch.cfg.
If you need further stuff I'll post that asap.

Could someone please look into the problem I posted earlier? If someone already has, I thank you for that, but could you also post about it

Rob wrote:It seems that the pathfinding has been altered with this beta

The only thing I changed in PBS was in regard to enhancetunnel bridges. However there aren't any in that screenshot... so I guess I'll require the usual to try and debug it (configs+savegame).

squishycube wrote:Could someone please look into the problem I posted earlier? If someone already has, I thank you for that, but could you also post about it

When someone has time, it will get looked into.

Patchman wrote:

Rob wrote:It seems that the pathfinding has been altered with this beta

The only thing I changed in PBS was in regard to enhancetunnel bridges. However there aren't any in that screenshot... so I guess I'll require the usual to try and debug it (configs+savegame).

The attached file contains my newgrf folder my game my ttdpatch.cfg and my newgrf.cfg.
The game is saved about 1.5 days before the train selects the "wrong" path, but it is reproduceable all the time if enough trains are in the station.

squishycube wrote:Could someone please look into the problem I posted earlier? If someone already has, I thank you for that, but could you also post about it

squishycube, the game locks up yeah? it doesn't actually crash!?
What's your method of deleting? Do you drag big squares with the dynamite tool?

UPDATE:
Damn, I'm relating this back to MegaRail being busted.
There seems to be a lot of "bad data" left in the tiles and it pops up because I use that data-area for my trams.

It seems to be the road tile to the NW of the NW->SE road-bridge down near Quito that's doing the damage. When you delete the road tile everything locks up (see attachment).
I can't diagnose the exact reason though. This works in every other situation (save game/new game/etc...) but MegaRail!
If you could just delete all the road (tile-by-tile, don't drag!) and then rebuild it all will be fine.

Update: Thanks squishycube!!!!
You just found a REALLY stupid bug in my code... It'll be repaired in the next beta revision!

I have a couple of pathfinding error reports, and something that might be part of UKRS.

1 shows a train trying to go into the station but decides a viable route goes through the one way out signal, even though the pre-signal is red

2. Shows an irregular station in a terminous format, 2 platforms are meant for coal collection, 2 for oil (with appropriate waypoints). But the caol trains refuse to go to the coal platforms, and isntead find a rout through to platforms 1 and 2. (out tracks go nowhere as I was in the processes of converting to RO-RO, but this happened before I added the outbound tracks)

3rd. (no pic) After creating a steal mill, for the first few months after receiving coal and iron ore, it put steal on the nearby station, even though no train to carry steel had ever been there (I have the good wait switch enabled). There is one train carrying lumber that was refitted from steel flatbeds.

Pathfinding problems can't be debugged without a savegame from before the train passes the offending signal and/or enters the offending junction.

And the canon name for the config file is ttdpatch.cfg. In both versions. Are you sure you posted the config you're actually using?