If you only have capx, you can unzip them with 7zip. The resultant folder contains your caproj and related files. Edit the save version tag in the .caproj file (use a plain text editor, notepad++ is good) and resave it. The versions are padded with a couple 0's (18800, 18400, etc.).

Had to google it...third result down:

Two-stage Unscented Kalman Filter for Nonlinear Systems in the

Presence of Unknown Random Bias

Please tell me that this is a real paper. Please tell me that there are such things as Unscented Kalman filters.

I don't need proof.

I want to believe.

Colludium , heh, it's like 16,7 ms is repelling the data! Perhaps this is now on the realm of wave functions, probability curves and quantum fluctuations! Maybe the laws of physics doesn't allow for perfect syncing without jerkiness. Who are we after all, that we wish to tame all those electrons for our entertainment???

All data points indicate that the render thread is undergoing a massive Laplacian deconvergence. This is almost unequivocally the result of a inverse paradox transform, causing non-linear chaotic gain and timey-wimey-wibbley-wobbley...stuff.

Alright, thanks. I thought there might be a shorthand for string comparisons where I wouldn't have to repeat a static comparator for each case, but it's really not a big problem...if I was dealing with more than a few values I'd just use an array instead.

This has been discussed before...there are probably better answers, but here is an example of mine where this is happening (tile type filtered cols):

C2 is complaining about this expression ("= is a type mismatch, etc..."):

Compare Two Values:

left(Controls.Get(ControlMenuObject.Type),3) = (MouseString_LMB | MouseString_MMB | MouseString_RMB)

equal to

TRUE(1)
[/code:19gpt8wm]

Basically, I'm trying to see if a given string matches 1 of 3 values(global string vars). So, put more simply:

[code:19gpt8wm]string foo = "foobar"

Compare Two Values: 

foo = ("foo" | "foobar" | "foobert")

equal to

TRUE
[/code:19gpt8wm]

Hmmm...that couldn't be right...could it?

Colludium

How do your graphs come out in the new node vs. the old version?

There is no link to the old version of node? I thought the idea of making node modular was partly to deal with this very contingency?

...ah, it looks like only the new version can export 64 bit binaries, so older versions are blocked.

We need 64 bit...why? Whose game is using more than 4GB of memory?

Ashley

Please, please revert this for the next stable...chrominium 38 (and thus, node-webkit v0.11.0) has totally nerfed performance.

I think someone is working on a doom clone project with the idea of it being a good start for beginners. That being said, 3D game logic is more complex than 2D, and q3d, by necessity, is a complex plugin.

Maybe this would be of interest to you:

Yeah I don't think you're ever going to get a reliable way to auto-detect and change settings during gameplay - provide an option and maybe use the framerate of the first level to set a default and then leave it.

Yup, in my current project I default to 'auto' settings which basically do just that (except, I'm running my tests prior to the actual gameplay). However, I also have "High, Med, Low" GFX settings.

While I've never had a lot of luck doing 'realtime' tweaks, I certainly think it's possible...it just requires more forethought about how such changes will affect the game, and how to avoid 'ping-pong' effects.