Fair enough. Right now I was really just trying to explore whether or not there was an obvious explanation for it, like several large sprite sheets overloading fillrate where loads of separate sprite frames wouldn't or something like that.

I just tried nw.js RC4 and long story short, I'm seeing a significant performance difference between preview and final builds on my test laptop. Contrary to what I expected, the game runs faster (as in perfectly smooth) in preview but get the same kind of jerky lagspikes we've seen in earlier nw.js versions for win32/win64 export builds.

Laptop in question:

i5-5257U — ghz

intel iris graphics 6100 (newer integrated chip)

8 gig ram

Windows 8.1 64-bit

So, nw.js export builds seem to be bad news for integrated chips somehow while preview builds are doing fine. Ashley could this be a texture atlas/fillrate thing and if so, is there anything that can be done to optimize it?

Right, changed it.

Easy fix. Just include dbghelp.dll from the official nwjs package into the C2 bundle.

Yeah try and backup/delete the "User Data" folder in Users/<username>/Appdata/Local/<yourgamefolder>/. I had the same problem and this fixed it for me. Some kind of bug in nw13 beta 2 if I remember right.

> I think all this slagging off of C2 and in the process promoting a competing product on Scirra's own homepage is pretty tasteless frankly.

>

Not at all, these are serious concerns.

People should better understand the limitations of C2, despite it being a great engine for intuitive coding.

Critique is fine, it's just when it spins into hyperbole it projects a skewed image of the whole situation, which I feel really isn't helping.

But yeah I was being an ass about it. Sorry 'bout that neverk and others.

I think all this slagging off of C2 and in the process promoting a competing product on Scirra's own homepage is pretty tasteless frankly.

Oh hey, thanks

Sure I'll add it in a bit.

In the meantime tho you can turn it on and off in the shader code at this line here:

[quote:2kly1qbs]vec2 warp=vec2(1.0/64.0,1.0/24.0);

If I remember right setting warp = 0 should do it. You can also play around with 1.0/64.0 and 1.0/24.0 to adjust the amount of distortion. Dividing by higher values equal less distortion.

Yeah, I do that sometimes myself. But in this case I have a whole laundry-list of stuff I want to happen if either A, B, C is true *or* D and E are both true. I don't see a way around that unless I do the event twice, which means heaps of excess code. Maybe I'm just too tired right now...

As everyone who's tried it knows, events can't take and/or conditions within the same block. I thought a good way to circumvent this might be to create a local variable that I set to '1' whenever the 'A and B' event comes true and then further down set the main event block up like this...

| C = 1 | Do stuff...

- or -

D = 1

- or -

LocalVar = 1

... but it doesn't work. LocalVar gets set to 1 when I want it to be, but the next event block doesn't seem to catch it. How do you all handle this kind of situation?

This is what I get from x86 and x64 builds both before and after killing the nw.exe processes in task manager:

[0210/030956:INFO:nw_package.cc(175)] "D:\Users\Erik\Desktop\Prototype\win32\nw.exe"

[0210/031229:INFO:nw_package.cc(175)] "D:\Users\Erik\Desktop\Prototype\win64\nw.exe"

Maybe that's an error code of some kind? Doesn't look like it though.

Rambletime: The special coddling that NWjs requires to get going man... Lag spikes that come and go, demands fresh gfx drivers at all times, doesn't like integrated gfx chips, doesn't like Optimus, doesn't like Macs or Linux OS'es, doesn't like recording software, etc etc...

I did. Tested a little and it seems to work fine on other machines. I guess the Win install got borked somehow. Re-installing...