Executables updated, crash on close (maybe) resolved, OSX build statically links to freetype (X11 support shouldn't be needed any more) , maybe it helps. (if you get 404 try again in 5 minutes)

I'm just seeing this now. I'll check everything when I get home (especially Console).

Quote: Original post by Oluseyi

Quote: Original post by Dmytry
BTW, I used to develop for mac 'professionally' for some job, and got very intense dislike for apple's user interface, the worst thing being unusual mouse curve hard coded (oh horror) it seems right into the driver [history: OSX never had UI for adjusting mouse parameters, but befori 10.4 there was API and you could set parameters with small C program. But, it seems, no more.]. I still did not find any solution that works. The annoyance seem to multiply tenfold when you have two computers on one desk, with 3 operating systems, and hop between them all the time. The sheer contrast. (Neither Linux nor Windows nor OSX come with my preferred (linear)mouse curve; but former 2 can be configured, and latter cannot and deviates from my preference a lot.

...

You too wrote about that on your journal, IIRC... maybe you got any solution?

Yes, I find this extremely aggravating, too. In fact, I've found myself subconsciously avoiding mouse-based design work on the Mac (at a time when I'm supposed to be redesigning GDNet's UI for V5!) because it feels like moving your hand through sludge. Ugh. The only saving grace is that I have a tablet, so if I can use that instead then I use it as a workaround.

Some have reported success using Microsoft mouse drivers, among others, but I haven't found a comprehensive solution yet. Very annoying.

I plugged an old Microsoft Comfort Optical Mouse 3000 into my Macbook once and never looked back. As far as I can tell, it uses the same acceleration curve as on Windows.

Mouse:
I found apple driver's settings, available online here All the important settings are base-64. IOHIPointing.cpp and IOHIPointing.h seem to be the implementation. I'll refrain from commenting on the coding style. I only can say that I'm kind of curious what exactly happens if mouse moves really fast and one of loops here fails to terminate as expected. Ohh the undecidability of halting problem.

So, did it work on osx for anyone here or not?

Success!



It would occasionally stutter, apparently when too many enemies were too close? I'm not entirely certain. Other than that, it was excellent.

In hundred percent unrelated: I just noticed that firefox zooms in images along with page, and that I somehow got zoom ratio very close to 1 but not exactly 1 so screenshots here looked as if there was some weird tiny font rendering glitch that I had for real long time ago but fixed. (I thought I still got that glitch on ATI cards and even started trying to figure out why.)

Updated, now with configuration GUI (press Esc to bring it up) and full-screen support. Saves configuration, as well.
What may fail: restarting itself when switching mode.
For full-screen modes different than your current desktop mode, it shows a pop-up asking if you see it (for cases when mode could not be displayed).

I have some question... what is the common method of handling situation when settings make game (including in-game UI) entirely unusable? Like a video mode that does not show up on display.
I'm popping up "do you see it?" dialog with timeout.
I thought of making separate settings program instead that uses regular UI controls, but it would be difficult to get it to look good on all supported platforms, and it would not permit tweaking of other visual settings with immediate feedback.

Also, list of available video modes. Anyone knows why it would be bad idea (on some displays) to only list those of available video modes which are smaller than desktop mode, and have approximately same aspect ratio?

Quote: Original post by Dmytry
I have some question... what is the common method of handling situation when settings make game (including in-game UI) entirely unusable? Like a video mode that does not show up on display.
I'm popping up "do you see it?" dialog with timeout.

I think that's the best way of handling it. I've seen many games that do not handle it at all.

Quote: Also, list of available video modes. Anyone knows why it would be bad idea (on some displays) to only list those of available video modes which are smaller than desktop mode, and have approximately same aspect ratio?

I don't know. Doing that makes no sense to me. I have often used monitors in e.g. 1280x1024 mode while the hardware was capable of much more (like 1600x1200) simply because the lower resolution was much easier on the eyes when doing typical desktop tasks like reading e-mail and browsing websites. In such a scenario I'd definitely would want to be able to run the game at the higher resolution.

One thing you may want to do is dual monitor support. For fullscreen mode you should give the user the option of either using one monitor (and which one) or span all available monitors. Some people prefer to use all their monitors while other people don't. E.g. with two monitors the crosshair would be on the edge between the two monitors, making playing difficult. That problem doesn't exist when using three monitors where the crosshair would be in the center of the middle monitor.

On Linux there's also the problem that under certain configurations only one of the monitors in a multi-head setup has hardware accelerated 3D. But you could simply ignore this. Such configurations are uncommon and primarily limited to the closed-source ATI drivers using an older Xinerama-style multi-head setup. Merged framebuffer setups or newer XRandR-based setups should always be accelerated on all heads.

MBP 17" Core Duo ATI x1600.

I get a window with some hud instruments but nothing else other than garbage onscreen.

Not sure how your GL works but it looks like some FBO or something else is not getting allocated properly, because it's running but some component of the view is not initializing..

What logs do you want to investigate with?