The Stonehearth Graphics Test Thread

You could check the Windows event log to see if it shows anything.

As I’m too lazy to type a tutorial (in case you need it, maybe you don’t) here’s a quick link:

Event viewer in Windows 7

It might not be completely fair, bit I can reproduce the no-visible-start scenario in one of my Windows 7 virtual machines. That is to be expected, but it does show one scenario in which the game will not start.

The file debug.log in the Stonehearth folder (C:\Program Files\Stonehearth on a 32 bit machine, C:\Program Files (x86)\Stonehearth on a 64 bit machine) contains one line:

[1126/163759:ERROR:gpu_info_collector_win.cc(102)] Can’t retrieve a valid WinSAT assessment.

Could you check your folder for this file, @GeorgeCrecy

3 Likes

you are quickly earning a top spot on my Christmas card list, good sir… :christmas_tree:

2 Likes

I had already checked that folder @mderond, and unfortunately there isn’t an error log. The file just doesn’t exist at all, not even in any of the subfolders.
And @SteveAdamo, I tried on another machine, and got the GL_NO_ERROR.
The specs for the other computer are as follows:
CPU: AMD Athlon II X4 635 @ 2.90 GHz
Memory: 4 GB
OS: Windows 7 64-bit
GPU: ATI Radeon HD 4200

1 Like

I’m trying :smile:

This sort of thing is what makes my work intresting and even though SH is not my work, it still excites my spidey sense.

P.S. A chrismas card is always welcome :wink:

1 Like

ok, thanks for checking… aside from trying to re-download, and perhaps running the executable as an admin, my only other advice would be to “like” the bug report here, to lend weight to how pervasive it might be…

Another thing you can try, is to see if there is a stonehearth.ini file in the program folder. When I remove (or rename) that file, I also get the same symptoms, it starts but nothing shows and then it quickly quits again. Maybe it somehow didn’t get installed?

Just in case the file is not there, on my machine the contents are as follows:

1 Like

Fortunately, there is an .ini file. I even tried making enable_shadows as false to see if I was having some sort of memory overload, making the computer stop the program rather than continue trying. Still no luck though.
I appreciate your trying to help @mderond. You don’t have to though, no obligation. :smile:

1 Like

Sure, no problem. I know I would be annoyed if the demo wouldn’t work on my machine, and I’m fortunate enough to have quite a lot of hardware ‘laying around’, so I can try some things. I enjoy helping where I can.

I am afraid I’m running out of ideas though… If I do think of something I’ll do q quick post!

1 Like

I managed to reproduce the same behaviour on a physical machine, but unfortunately that was an old netbook with an Intel Atom processor (1 physical core, 2 hyperthreads at max 1.6Ghz) an integrated graphics. So that particular machine is not truely representative.

Anyway, it was quite slow, and I could see what happens. The program opens up, you can see the actual window being created. The contents of the window is never shown however. It stays black and for a few seconds the window title shows (not responding).

Then it closes.

I found the stonehearth.log file, and from what I can tell it appears as if the scene is being built up. As it is never shown, my guess is that something breaks (graphics driver issue) when it’s time to show the first actual rendering. Ther is also a .dmp file, and a zipped crash file. I’m not sure if these have actually been sent over as I have never had the chance to allow this (by clicking in the UI).

I also made a Process Monitor log so if someone at Team Radiant is interested I can hand that over.

I admit that this is not the most suitable machine to test on, but from what I can tell this is the same behaviour as we’ve seen above.

2 Likes

I tried running the test, but I always get an error message on startup that says:

Stonehearth.exe - Entry Point Not Found

The procedure entry point RegGetValueA could not be located in the dynamic lik library ADVAPI32.dll.

HP Pavillion A1640N
Intel ViiV
Windows XP Service Pack 2? (I’m not sure if it’s Service Pack 2 or 3)

Can someone help me?

there’s no two ways about it… i am “promoting” you to Tier 1 Bug Sleuth… enjoy the new title… :smile:

thanks for the feedback… this was captured in this previous report

1 Like

My other rig is exactly the same specs except with an AMD Phenom IIx4 and it ran smoothly. Maybe it was just a hiccup since you couldn’t reproduce?

Haha, I’m honoured, thanks!

In the end I think it is important work. Fixing a bug that you cannot reproduce is quite difficult.

2 Likes

@GeorgeCrecy Your issue appears to be identical to mine and a few others who have posted.

I loaded the crash dump file into WinDbg and it claims “C++ EH exception - code e06d7363” as the exception that caused the crash. I’m not a Win32 developer, however, so my experience on this is VERY limited. That might be a completely ambiguous error message for all I know. Both GeorgeCrecy and I have 64-bit OSes (I’ve noticed that might be a common denominator among other similar posts, as well). I have a LOT of video games installed on my system with just about every Microsoft Visual C++ redistributable under the sun (4 2005’s, 7 2008’s, 2 2010’s, 2 2012’s)…does anyone think this could be an issue with that? Does Stonehearth require a MS C++ redistributable and is so, which one?

The machine I had it crash on was a 32 bit Windows 7 machine. All other machines I tried the SH GT on, are 64 bit (Win7 and Win 8.1)

What I’m wondering the most, is if SH gets these crash dumps as you never get to approve of the upload. They might be missing out on usefull data. And while glitches are annoying, having the program not start at all is kind of a dealbreaker, I suppose :wink:

After work I could try to run a clean Win7 32 and / or 64 machine to see what happens. I already tried it on a clean Win8.1 64 bit machine, and that ran fine.

1 Like

Well I know when you first start it up it asked if you wanted to opt into sending additional data. Does that include crash reports? If not then you’re right on the money. They’re missing tons of good data.

What I’m wondering the most, is if SH gets these crash dumps as you never get to approve of the upload.

They get them. Your question piqued my curiosity…Wireshark confirms the sending of the zipped crash.dmp file to host support.radiant-entertainment.com upon running Stonehearth on my borked SH platform.

1 Like

wireshark scares me… we routinely use that at work, to check packet details, and wow… you practically need a degree to use it effectively…

but thanks for this confirmation that results are still being sent! :+1:

Heh, yeah, Wireshark is a powerful tool. I remember my jaw hitting the floor when I was able to sniff my own VOIP call, save the results and, using Wireshark, recombine the audio stream into a playable format. Scary indeed.

And you’re more than welcome. Again, anything I can do to help.

1 Like