2.6 nightly/alpha discussion

DaleStan · **Joined:** Wed Feb 18, 2004 3:06 am **Posts:** 10285

athanasios wrote:

If it doesn't create a CRASH.TXT how can he post one?

He failed to specify such file failed to appear. Proper assumption in that case is that the file appeared and he just doesn't realize that it contains UI.

athanasios wrote:

Now some hints, cause as you can see mine crashed too:

That also fails to be an OlldDbg screenshot. gnome's image, however, is.

The first thing you should do is check that you have the correct mkpttxt.exe and/or ttdpttxt.dat for your version of TTDPatch. If so, post ttdpttxt.dat. If it's not present, report that too.

Lakie: It's the ud2 at <initourtextptr.ok-2>. Do you have an x86_64 processor? I haven't seen this crash, so it'll probably be hard for me to debug if it's a 64-bit issue.

eis_os · **Posted:** Sun Mar 09, 2008 9:54 am

Hmm, how much memory do you have? We already had once issues with pointers and clever using thought to be fair game bits. Anyway, that ud2 looks like some game text problems, as DaleStan mentioned...

Lakie · **Posted:** Sun Mar 09, 2008 2:54 pm

DaleStan wrote:

Lakie: It's the ud2 at <initourtextptr.ok-2>. Do you have an x86_64 processor? I haven't seen this crash, so it'll probably be hard for me to debug if it's a 64-bit issue.

I'm finding TTDpatch r1833 works fine here, however it should be noted I'm using an Intel x64 Processor which doesn't use AMD's x86_64, it uses Intel's EM64T...
And I no longer have my AMD system, as I gave it to my brother.

~ Lakie

eis_os · **Posted:** Sun Mar 09, 2008 3:11 pm

Windows Server 2008 64bit in VMWare Player with an Athlon64 seems to start and run without any problem (No compatibility settings used either), except as usually the mouse works a bit strange in emulation. I don't have the focus problem that Lakie described earlier...

If you still have the problem with a fresh installation and ttdpatch nightly without any grfs please try to find the switch/grf causing this problem.

gnome · **Joined:** Tue Nov 11, 2003 8:16 pm **Posts:** 7

the latest nightly r1833 still crashes here with my old configuration.

I've made some tests and find out, two configuration-switches makes the game crashed on start - with value 'off', the game starts without fault and is playable.

Here are the scwiches
startyear=1942
or
enhanceddifficultysettings=on

Could it someone explain me?

Regards
gnome

athanasios · **Joined:** Thu Jun 23, 2005 12:09 am **Posts:** 3138

DaleStan wrote:

#@#@$#@$

Please read my post again.
* I said I didn't have time for further tests.
* I mentioned that my problem may have to do with a parameter in config.

Finally my issue is irrelevant:
Has to do with resolution. Unless set to "off" it crashes. Why should I get such a message with standard values

:

regards
athanasios

Rob · **Posted:** Mon Mar 10, 2008 6:01 am

Have you read the disclaimer on the Nightlies page ?

Disclaimer wrote:

Feel free to try it, but be aware that this code may be even more unstable than regular alpha or beta versions. Furthermore, it is compiled without version data, so TTDPatch will warn you about this when you run it.

Because it is compiled without version data, the patching may fail with a message like

Failed to find string #000D at 0063EA10, found 00/01 at 00600000 for oldaitestcargotypes in patchnewcargos

If this happens, try the following:

* Make sure you're using an unmodified version of the TTD executable (gamegfx.exe or ttdx.exe). In particular, it must not be a non-standard translation or modified by TTD Alter. Try making a backup of gamegfx.exe/ttdx.exe and replace it with the original version.
* If that doesn't help, make a backup of your ttdpatch.cfg and delete it, to force TTDPatch to create a default configuration. If this helps, submit a copy of your original ttdpatch.cfg to the patch developers (or the bug tracker) along with the exact error message you got.
* If that didn't help either, something is very wrong, so please submit a detailed bug report.

eis_os · **Posted:** Mon Mar 10, 2008 7:03 am

Specially it's even documented behavior:

http://wiki.ttdpatch.net/tiki-index.php?page=Resolution

DaleStan · **Joined:** Wed Feb 18, 2004 3:06 am **Posts:** 10285

gnome wrote:

the latest nightly r1833 still crashes here with my old configuration.

I've made some tests and find out, two configuration-switches makes the game crashed on start - with value 'off', the game starts without fault and is playable.

I still can't reproduce this. Your full config files, please. Both newgrf[w].cfg and ttdpatch.cfg.

athanasios wrote:

* I mentioned that my problem may have to do with a parameter in config.

One of how many switches? And I'm supposed to test every single possible configuration (of somewhere around 10^120 possible) to determine which particular one is causing your crash? I think not.

gnome · **Joined:** Tue Nov 11, 2003 8:16 pm **Posts:** 7

DaleStan wrote:

gnome wrote:

the latest nightly r1833 still crashes here with my old configuration.
I've made some tests and find out, two configuration-switches makes the game crashed on start - with value 'off', the game starts without fault and is playable.

I still can't reproduce this. Your full config files, please. Both newgrf[w].cfg and ttdpatch.cfg.

here you are

I'll hope, it can help

Regards
gnome

DaleStan · **Joined:** Wed Feb 18, 2004 3:06 am **Posts:** 10285

Even with that, it still works for me. I still think this looks like massively broken text data, which could be either internal or in ttdpttxt.dat. I can't explain how either could happen, though.

And you did post those already, didn't you? Sorry, I seem to be half-asleep right now.

gnome · **Joined:** Tue Nov 11, 2003 8:16 pm **Posts:** 7

DaleStan wrote:

Even with that, it still works for me. I still think this looks like massively broken text data, which could be either internal or in ttdpttxt.dat. I can't explain how either could happen, though.

And you did post those already, didn't you? Sorry, I seem to be half-asleep right now.

No, I didn't sent my ttdpttxt.dat before.

Maybe you've right, my Installation isn't clean anymore and I've to reinstall it.

Regards
gnome

athanasios · **Joined:** Thu Jun 23, 2005 12:09 am **Posts:** 3138

eis_os wrote:

Specially it's even documented behavior:

http://wiki.ttdpatch.net/tiki-index.php?page=Resolution

Opps! Never heard about it. Rather odd though.

Rob wrote:

Have you read the disclaimer on the Nightlies page ?

Disclaimer wrote:

Feel free to try it, but be aware that this code may be even more unstable than regular alpha or beta versions. Furthermore, it is compiled without version data, so TTDPatch will warn you about this when you run it.

Because it is compiled without version data, the patching may fail with a message like

Failed to find string #000D at 0063EA10, found 00/01 at 00600000 for oldaitestcargotypes in patchnewcargos

If this happens, try the following:

* Make sure you're using an unmodified version of the TTD executable (gamegfx.exe or ttdx.exe). In particular, it must not be a non-standard translation or modified by TTD Alter. Try making a backup of gamegfx.exe/ttdx.exe and replace it with the original version.
* If that doesn't help, make a backup of your ttdpatch.cfg and delete it, to force TTDPatch to create a default configuration. If this helps, submit a copy of your original ttdpatch.cfg to the patch developers (or the bug tracker) along with the exact error message you got.
* If that didn't help either, something is very wrong, so please submit a detailed bug report.

EITHER:
The ISO available for download in the forums, Owen's pack, and rest, are modified. The same for my DOS version which I 've got from before TTDAlter or TTDPatch were born.
OR:
ttdpatch is LOST IN SPACE. :lol:

orudge wrote:

... (Multilanguage, Original CD-ROM ISO, Win95 Version)

So I assume the latter.

regards
athanasios

DaleStan · **Joined:** Wed Feb 18, 2004 3:06 am **Posts:** 10285

Anyone for whom TTDPatch quit working at r1829, please see this post.

OzTrans · **Joined:** Fri Mar 04, 2005 1:07 am **Posts:** 1439

Quote:

Game variable A2/22 - Difficulty level a pointless exercise ...

This variable gives us coders nothing to go by. Hardly any player plays a game with a pure 'easy', 'medium' or 'hard' level; almost all players use a difficulty level of 'custom' including myself.

Example : Select difficulty level of 'hard', then just set max AI players to 0 and the difficulty level of 'hard' becomes 'custom'. How would we know now what the real difficulty level is ?

We should have the following values :

00 – easy
01 – medium
02 – hard
04 – custom, resembling easy level
05 – custom, resembling medium level
06 – custom, resembling hard level

values 4 ... 6 should be determined by evaluating the closest difficulty level by looking at all individual variables. Or just forget 'custom', just determine what difficulty level the game resembles.

DaleStan · **Joined:** Wed Feb 18, 2004 3:06 am **Posts:** 10285

Talk to George. George wanted 00..03, apparently. I just copied what Belugas said George wanted.

George · **Posted:** Sat Mar 29, 2008 4:36 pm

DaleStan wrote:

Talk to George. George wanted 00..03, apparently. I just copied what Belugas said George wanted.

That's correct. I did not want to allow endless mines on hard settings.
Yes I understand, that it is possible to have custom level, that is harder, than custom. I can only suggest to provide new default difficulty level "Very hard".
For me difficulty settings other than "custom" are interesting from the point that the level of difficulty is saved inside the game and any change is indicated by changing the level to custom. The same as cheats menu in OTTD and sign cheats in TTDP - the information about feature usage is stored.
As for me, usually I do not play, but when I do it, I use HARD settings only.

DaleStan · **Joined:** Wed Feb 18, 2004 3:06 am **Posts:** 10285

OzTransLtd, if you can come up with pseudocode that defines "resembling" in a meaningful way, I may be convinced to put it in.

Of course, the functionally harder settings are sometimes are the "Easy" ones. Especially inflation; with 4% inflation, costs inflate only one third faster than income, while with 2% inflation, costs inflate twice as fast as income.

OzTrans · **Joined:** Fri Mar 04, 2005 1:07 am **Posts:** 1439

DaleStan wrote:

... if you can come up with pseudo code that defines "resembling" in a meaningful way, I may be convinced to put it in. ...

I'll see what I can do ...

Quote:

... inflation ...

Are you mixing up interest rates with inflation ? Isn't there just one issue with inflation ... a loan becomes cheaper to repay, as it is repaid with income, boosted by inflation. Inflation increases everything equally over time; i.e. costs and revenue, but not the loan amount.

belugas · **Posted:** Fri Apr 04, 2008 3:18 pm

OzTransLtd wrote:

values 4 ... 6 should be determined by evaluating the closest difficulty level by looking at all individual variables. Or just forget 'custom', just determine what difficulty level the game resembles.

I'd rather go by exposing the value you want to get, rather than make evaluation of the closest values to determine the closest level of difficulties. and IIRC, there are some settings that are the same between 2 difficulty levels.
So, it may lead to confusion pretty fast.
and I do not have custom settings, but rather easy or hard, depending on what i work on.
I know there are quite a few people who prefer to go by standard difficulty levels too

Transport Tycoon Forums

2.6 nightly/alpha discussion

Who is online