Ep4 crash (formerly "It's getting worse")

[dr_debil]

seems like the programmers are making progress... in episode 4 i can't even see the intro. it crashes right after pushing the launch button...

episode 1 works without any problems... 2,3,4 don't. why the hell did you modify the engine???

THIS SUCKS!!!

[Emily]

Hi,

I split this into a new thread because it's not clear to me if this is the same problem as in the thread you originally posted it in or not. It can be really hard for us to help when people post about different problems in the same thread.

A few questions for you:

1) What EXACTLY happens when you press the Launch button? Does it just crash to desktop? Do you see an error message? (If so, what does it say?) When you press ctrl-alt-del and look at the processes running, is Sam & Max one of them?

2) What are your system specs?

3) What problems did you have with episodes 2 and 3? Memory violation errors? Random crashes?

Thanks.

[dr_debil]

wow extremely fast answer. i'm really impressed.

the error messages i get are the access violation errors. Error:Access violation at 0x00650505 (tried to read from 0x1A33D786),program terminated... in case of episode 4.

In episode 2 i got this error after playing some time. after getting it, i couldn't start the game anymore without getting the error except when i was deleting the savegames.

In episode 3 the error occurs right after the intro.

In episode 4 right after pressing the launch button.

Hardware: Athlon Venice 3000+, NVidia Geforce 6800 (AGP), 1024MB DDR, ASUS A8V Mainboard, Soundblaster Audigy.

I tried to run the games under Windows XP SP2 and Windows 2000 SP4
No success! Only episode 1 is running...

[Welshy]

Hey guess what...im back

...yeh its occured again, doh!! Im beginning to think its my machine so im just gonna go and invade my brothers house this time round and enjoy it, crash free, instead of having to restart every few minutes.

[Emily]

dr_debil - when exactly did you get the error? Had you pressed the LAUNCH GAME button, or did it crash on its own while the "retrieving key" message was visible?

Sorry if it seems nitpicky, but a few of us here just experienced an issue with a different game that may or may not be related.

EDIT: Just noticed you said "launch button" above, so I think you answered my question already.

Im beginning to think its my machine so im just gonna go and invade my brothers house this time round and enjoy it, crash free, instead of having to restart every few minutes.

Well it's definitely not JUST your machine, but if you can play it on your brother's without the crashes, that's better than nothing.

This time, when did you get the error message? It's always been a recurring, throughout the game thing for you, right?

We are still doing our best to get to the bottom of this, guys!

[Welshy]

Yeh its still recurring throughout the entire game in this episode as well by the looks of things. Thanks for the quick response as usual...you probably seen me coming right?

[Emily]

Hey dr_debil - can you please run a DXDiag report and send it to support@telltalegames.com? Mention this thread so we'll know who you are. Thanks.

[Sgt. Chimp]

I'm getting this problem as well with episode 4. As soon as I click Launch Game, I get "Error: Access violation at 0x007500AC (tried to write to 0x1DF9EDC5), program terminated." Compatibility modes do not help.

Episodes 1 and 3 have no problems. (I haven't reinstalled 2 on this machine yet.)

[Emily]

Thanks, as always we're looking into it.

Sgt. Chimp, if you could send over your DxDiag too, that'd be great. Thank you.

[Cone of Tragety]

The laucher starts normally, but after I click "launch game" there's an error dialog about an access violation.

For Ep2, at 0x74FFFF (read 0xCF725A35 (not even in user mode)). There are two instances of "Sam and Max...exe"; the child process is the one that's crashing. A stack trace of the main thread (the one that owns the error window) contains:
Sam and Max - Situation Comedy.exe+0x34ffff
Sam and Max - Situation Comedy.exe+0x35160a
Sam and Max - Situation Comedy.exe+0x35689e
Sam and Max - Situation Comedy.exe+0x3470b9
d3d8.dll!CBaseObject::AddRefImpl+0x1b
d3d8.dll!CCubeSurface::AddRef+0x13
Sam and Max - Situation Comedy.exe+0x347723
Sam and Max - Situation Comedy.exe+0x107db9
Sam and Max - Situation Comedy.exe+0x10b7b2
A small window with a white background is visible beneath the error box.

For Ep4, at 0x7500B4 (read 0x6A817717). As before, there are two S&M processes, with the child crashing. A stack trace of the main thread:
Sam and Max - Abe Lincoln Must Die.exe+0x3500b4
Sam and Max - Abe Lincoln Must Die.exe+0x119399
In this case, it created no windows besides the error dialog.

Episodes 1 & 3 run fine (well, except for excessive memory usage).

This is on an Athlon 2800+, Asus A7N8X-X (nForce 2 based), 1GB of memory, WS2003 SP1, a GeForce 3 Ti 200 with ForceWare 93.71 (6.14.10.9371 internally). I've tried deleting prefs.prop to no avail. I can send dxdiag info if you want, and can use a debugger FWIW.

[Sgt. Chimp]

Heeeeyyyyy!

I just reinstalled S&M episode 2 and got a similar problem to Cone of Tragety [sic].

Error: Access violation at 0x00750002 (tried to read from 0xFFFFFFBB), program terminated. If I attach a debugger, the latter address is 0x0000003F.

Not only that, but I've got several system specs in common. AMD CPU, nForce4-derived motherboard (Shuttle, not Asus), ForceWare 93.71, and Server 2003 SP1.

[Drezic]

Same problem I can't play any episode but the first...

[Cone of Tragety]

Ack, I've misspelled my user name. How embarrassing...

I find it quite interesting that Sgt. Chimp and I are both using exactly the same video driver version, plus that Ep2's stack trace includes the Direct3d 8 library, which seem to imply a video driver incompatibility. I'm using 93.71 in particular because it's a modern version that works properly with my old video card and contains the workarounds for Aliens vs Predator.

When I get a minute, I'll try a more recent video driver version.

[Kevin]

How are you getting a stack trace? Attempting to run the game with a debugger active will cause the game to crash....

[Cone of Tragety]

A process can be fully and officially connected to a debugger by calling kernel32!DebugActiveProcess. However, many debuggers can still attach to a process in a reduced functionality "Noninvasive" mode (that's what windbg calls it) which does not officially (according to the OS) involve attaching a debugger. It is still possible to create a stack trace and read memory in this mode.

Specifically, the Sysinterals Process Explorer can extract a stack trace from a thread without attaching as a debugger. That's what I used to get the traces I posted.

I realize that the game prevents full debugging as a copy protection measure (as many other titles do). I'm not trying to bypass that protection. I was hoping that a stack trace would be helpful in diagnosing these mysterious crashes. I know they're useful when I'm debugging my own programs.

[Drezic]

I am getting tired of quotations of system specs and questions as to what excactly happens. I have seen these questions answered like a million times and I even send my system specs, on demand, to one of the telltale employers persoanlly; however, I never got a reply. I send an email; I never got a reply.

Is there going to be a new version or is there any sign of a solution? I cannot play episode two three and four and I would really like to.

[Emily]

I send an email; I never got a reply.

We try very hard to reply to all emails that are sent to us. Sometimes I'll send an email about a technical problem over to Kevin or one of the other developers and I'll forget to respond to the person who wrote in to say that I've done this, so if that's what happened in this case, I'm very sorry, it's my fault. Otherwise, it's possible we didn't get your email, or that we did respond and you didn't get our response.

For people having these problems, I really do understand your frustration. You have to understand that we are doing the best we can. The fact that we have not been able to isolate the cause of these problems is not due to lack of effort or caring on our part. There are a lot of moving parts involved and not every error is caused by the same problem, which is why it's important for us to ask for more detail. We really thought that episode 4 would have many of these issues fixed, so the fact that it doesn't means we're back to square one in some respects.

As I've said several times before, anyone who doesn't want to wait for these problems to get figured out (which would be completely understandable) can request their money back and we'll take care of it immediately. I know it's not as good a solution as getting the games to run for you, but frankly it's a better solution than you'd get from a lot of other companies.

[Drezic]

Originally Posted by Emily

We try very hard to reply to all emails that are sent to us. Sometimes I'll send an email about a technical problem over to Kevin or one of the other developers and I'll forget to respond to the person who wrote in to say that I've done this, so if that's what happened in this case, I'm very sorry, it's my fault. Otherwise, it's possible we didn't get your email, or that we did respond and you didn't get our response.

As I remeber, I received an automated e-mail that you guys received my e-mail. I also send my system specs (private message) to Kevin because he asked me to and I never got a reply. It's not that I don't believe you guys are trying but you keep asking the same questions about system specs and all which comes accross as if you aren't making any progress.

Originally Posted by Emily

For people having these problems, I really do understand your frustration. You have to understand that we are doing the best we can. The fact that we have not been able to isolate the cause of these problems is not due to lack of effort or caring on our part. There are a lot of moving parts involved and not every error is caused by the same problem, which is why it's important for us to ask for more detail. We really thought that episode 4 would have many of these issues fixed, so the fact that it doesn't means we're back to square one in some respects.

I find it hard to believe that several people with the same problems over several episodes have lots more than one problem. How are you so sure that there are lots more problems? On the other hand; I am not a computer technician, so that might be my mistake.

Originally Posted by Emily

As I've said several times before, anyone who doesn't want to wait for these problems to get figured out (which would be completely understandable) can request their money back and we'll take care of it immediately. I know it's not as good a solution as getting the games to run for you, but frankly it's a better solution than you'd get from a lot of other companies.

As I said times before, I don't want my money back. And by the way; if software doesn't work I always get my money back (here in Holland at least) and the games that I bought that didn't run properly or had problems running smoothly had downloadable patches available in no time.

I appriciate you people running a forum and an informal style of communicating; that is something other companies do not have. You should not forget that you are a company and if something does not work people will start complaining. Hence, you can keep saying that you're trying but that will not do as long as there are no visible results.

You guys have my blessing and I hope things will work out for the best. After all, it's just a game. Well, maybe not just a game, it's Sam and Max, but still. As I said before, I never expected for Sam and Max 2 to ever be brought to life anyway, so I am very thankful for your work.

[Cone of Tragety]

Apparently, 93.71 IS the latest nVidia video driver. I tried 66.93 but that seemed to have no effect.

I've noticed that both the faulting instruction address and the data address fluctuate, although the instruction address is always near 0x00750000, which is indeed within the S&M binary but does not seem to be a valid code section of the image. I would guess that either an invalid jump into a data section is occuring or the image needs to be decrypted as part of a copyprotection measure and which isn't happening correctly.

[Emily]

You should not forget that you are a company and if something does not work people will start complaining.

Believe me, I know that all too well!

Some of the people who were getting these memory access errors with episodes 2 and 3 aren't getting them anymore with episode 4. That's how I know we're making progress.

Also I'm sorry if my last post came across as me losing it a little. I know it's frustrating for those of you getting the error not to have a solution from us, but it can also be very frustrating for us to have people suggesting that we're not doing enough to track down problems (which in most cases aren't even due to our software!) when you've just put in a twelve hour day doing just that... or in this case, when you've just woken up early on a Saturday to answer support emails.

We really are trying, guys. This stuff keeps me up at night.