Page 1 of 2
Posted: Mon Jan 14, 2008 1:12 pm
by CeeBee51
Both mid-east maps crash for me but it's been a long time since I played though I do remember it's always near the beginning.
Posted: Mon Jan 14, 2008 1:37 pm
by karsten
The original RRT_Events.xml file, eh? Well, lets try to put ourselves in SMR's place as it works through the "random events" for "Chicago to the Rockies".
Passengers ... check
Mail ... check
.
.
.
and so it goes on until late in the game, when it hits
Arms ... Arms?? What is this game, splatterhouse2? No arms, what to do?!?
Now SMR is faced with a serious decision problem. Will it simply skip to the next random event? No Sir! I proudly issues a support call to the programmers (writing "random event") into the log file ... and dies.
Posted: Mon Jan 14, 2008 7:47 pm
by karsten
Well, Snoopy, you have your convictions and that's OK. I still think it is a shame that someone like Bushdoctor had to spend a very substantial amount of his time testing maps, only to finally arrive at the same conclusion, ie:
I personally have no doubts: It's a News report that makes the game crash.
And the safe event solution is very simple. As you can see in the attached file, there are no events in this file that could crash the game.
That does not mean that every map needs to use this file. Once one is certain that a new map runs stable, one can carefully add compatible news events, although why one would want to do so is beyond me, because in most cases the events don't really add usefully to gameplay.
Posted: Mon Jan 14, 2008 8:10 pm
by Lowell
For your information, New Events Must be made Global...then only referance them in your scenario's xml list. I know becsause I have made new ones and that's how they must work.
We need people to play test these maps...not ponder on a best guess on what we think should work...see it work and make it work.
Posted: Tue Jan 15, 2008 12:05 am
by Bushdoctor
*grinn*
A few days back I started a test run in order to find what I believe is a crashing news report.
There are several issues that make it a very hard task:
First of all the news reports appear random. This alone creates several problems.
During test-games you can't work up to the crash. All you can do is let a map run in double speed for hours. However, each news report makes the game pauze, so you have to be around to click it to continue.
Then when you finally reach a crashing point, you think you'd return to a save
and change the events.xml for an empty one and slowly insert events to see what's
causing the problem. No can do: A saved game seems to include the events
that it received when initializing the map.
We would need help from testplayers to find the culprit.
There are 64 events in the default RRT_Events.xml. (That's the one we were using, right?).
When cutting that list into batches of 5, we would have 13 different event xml's.
All 13 would have to be tested in a game where constant saves have to be made
untill all 5 news reports have passed by.
If a testplayer hits a crash, he/she would have to load from the last save
and confirm that instead of a crash now a news report appears.
There is a 1 in 5 chance you hit that same crashing report again,
so you might have to repeat that last part a few times.
It's worth mentioning that there only should be 1 of 13 batches that actually crashes.
Once we find that specific batch, we split it up into a single event per xml
and do the final test run with 5 files.
Anyone up for joining a test team?
Posted: Tue Jan 15, 2008 12:18 am
by snoopy55
The first thing your team might try is to list the Goods used in the map and create a Events XML (map specific) with just those in it. This may or may not be part of the problem, but it wouldn't hurt to do it that way. This would also get rid of any events that have nothing to do with the map.
Posted: Tue Jan 15, 2008 12:28 am
by Bushdoctor
I can see your point, but my idea was to work with the version of the Rift scenario
of which I'm sure has perfect working goods. I've played that version with an
adjusted events file and it worked like a charm. For hours on end.
If we take that map and use the default RRT_events.xml I think we'll reach our goal.
I'm pretty sure I'm right about that news report but if not, we can always
do another test run concentrating on the goods.
Posted: Tue Jan 15, 2008 12:31 am
by Lowell
I have tested all the events for the US Events...there are a few that are set for EURO Events.
What map is crashing default events? Rift? Default or New Event?
Events must be entered into the global rrt events.xml list.
I can test it...place an events file inside the map...list only the new or old event you want to test...Wham...the first event you get is the single one you listed.
That's how I test mine.
Let me know, I give it a crash test for sure and make sure it works.
EDIT///Yeah...why even test that version, delete it. All of them should have been on one thread anyway, until the map worked. If the old ones are broke...get rid of them.
Snoopy...how can you have tested all the maps here when just yesterday you said that your computer couldn't play Basin and Range because it was a size 15 and too big. Chicago to the rockies is a size 15 map. Hmmmmm I edited my last statement, I said what needed to be said.
Posted: Tue Jan 15, 2008 12:31 am
by karsten
Anyone up for joining a test team?
Well you were testing Rift V2, and since Rift V3 is event-crashproof, I am pleased herewith to make a prediction if you really want to test Rift V2:
Any event relating to Pass., Food, Goods, Grain, Oil, Textiles, Gold, Livestock, Milk, Nitrates, Cornmeal or Cotton will be fine in Rift V2.
Any event relating to
any other goods:
CTD!
So it should be easy to compile an event file that is guaranteed to crash Rift V2 first time, every time!
Posted: Tue Jan 15, 2008 2:07 am
by Bushdoctor
Well, yea heck. Replacing the events for a fresh tested batch
would solve all problems in one big bang. Let's go for that
and instead of putting energy in a test team I can concentrate
on learning more about this interesting world called XML.
Posted: Wed Jan 16, 2008 12:38 am
by CalmDragon
On a map I created and turned on logs for that was using RRT_Events.xml it was going CTD on several Random Events. It got to the point were I was saving my games as pre-crash-01 etc.
Posted: Wed Jan 16, 2008 1:22 am
by Lowell
Were they new events you are trying to add? If so add the full listing of the event in the global events file, then referance the couple lines for your event in your scenario's event xml list. I placed the full listing in both places once and it will cause a crash. The full listing needs to be global, then only the name of the event gets listed in the map.
EDIT// Oh...you mean somehow the SMRI update has an event that is msipelled or something...the thing to do is open up the Assets/Events.xml file in the game proper file in ProgramFiles/2KGames...etc. Compair every Event with the default list in there, then you will find your error. Most likely it is a typo...and by the way I think I corrected one in my recent maps and had the script and event photo placed in all my assets...just now remembered...sorry. I can't remember which one, but I do remember it was one of the ones that was in a fraxis update fpk I think.
Posted: Wed Jan 16, 2008 6:55 am
by snoopy55
Lowell wrote: The full listing needs to be global, then only the name of the event gets listed in the map.
A Tycoon's Tale (Santa) <EventsXMLFile>RRT_Events_Santa.xml</EventsXMLFile> No GLOBAL RRT_Events.xml
Posted: Thu Jan 17, 2008 8:55 am
by Rodea2007
snoopy55 wrote:Lowell wrote: The full listing needs to be global, then only the name of the event gets listed in the map.
A Tycoon's Tale (Santa) <EventsXMLFile>RRT_Events_Santa.xml</EventsXMLFile> No GLOBAL RRT_Events.xml
Yes...you correct Snoopy. New event no needs to be global, but you must to be full listing in Scenario map.
Posted: Thu Jan 17, 2008 11:15 am
by snoopy55
Rodea2007 wrote:Yes...you correct Snoopy. New event no needs to be global, but you must to be full listing in Scenario map.
"Thank-You.....Thank-you Very Much" (my Elvis impersonation)