How about a "Digg" style page somewhere on Scirra.com that we can use to post these feature suggestions and "sales pitches" and then let others vote them up/down to build a C3 feature list? How useful and fun would that be?!

It was just an idea. No need to switch, and using a guid would be better if you run multiple games.

I wasn't sure if maybe Pick by UID is faster code than Pick by InstVar = ???

Having to store references to other objects by storing their IDs as a number, then having a sub-event pick them out is garbage. Same goes for polluting your function namespace by creating pseudo "methods" that take an object ID as a parameter.

I'm gonna use the word "easy" very loosely here but couldn't Object types be defined as one (or multiple) of the Function params? Then they're passing UIDs (or whatever, underneath) so that in the Function, we immediately have the "picked" reference to our Objects. Seems like this would be an "easy" extension and improvement on what we already have in C2.

I ask the following having done zero 3D game dev: Wouldn't the 3D stuff need some sort of native compile/export? I can't imagine JavaScript running in the browser having anywhere near the horsepower to handle 3D games.

R0J0hound All of the Card sprites are created runtime from Xml data. The configuration Xml is on a server. So that GUID is in the Xml which means I know, with 100% certainty, all Cards on every client have the same GUIDs. They will always be picked correctly and sync'd correctly.

If there is a compelling reason, say a performance reason or similar, I could switch to UID, but I'd rather stay with what I have unless it's somehow "bad".

Right now, all Cards are in ONE sprite with many frames. Example: a standard playing card would be a sprite with 53 frames, the card back at frame 0 then all the Card values in subsequent 52 frames. I create 52 instances of every different card and set the visible frame accordingly. So one sprite handles every card on the screen.

Perhaps this is not a good way to do this. I could put each Card in a separate Sprite with just 2 frames, back and front. Maybe this will change something.

Ok that's interesting. I changed the System Compare Frame to FamCard Compare Frame, as you suggested. Now the very first flip does work on the card hovered. But then subsequent flips work on another (not the one hovered) card. So I'll say that's a good step.

Also, I think we're disconnected on the UID question. Will all Card.UID values be the same on each player's running instance? We could assume that every player has the exact same version of the exported app, but it's possible they may not. How can UIDs be guaranteed to match for all players?

Also, if you're sure it's the GUID compare, I'm in trouble. I have this compare in every single function and event. A hundred places probably. So System compare cannot accurately pick object instances by a var?

GUID comes from an Xml data file. Are we 100% certain that UIDs will always exactly match on multiplayer games? My assumption is no, so I went with the GUIDs that I am 100% sure will match.

I did use the Mouse overlapping initially, switched to this to see if it would fix. No change.

Frame = 0 is the card back, 1 - X are the card faces. Not sure how else to toggle the visible frame? How else?

LittleStain I added two more screen clips. The function called and my console.log. The global var CARD_INPLAY = 5. In the console, the first card, GUID=20 is the card that my mouse is over, the second card, GUID=180 is the one that flips.

Yeah, I know

I move the mouse to hover over a card sprite, press "F" to flip it over. How can this code flip the wrong card? It flips some other card that is no where near the cursor!? The card that it flips does not even have Location = CARD_INPLAY! There is only one event in all of my Event Sheets that fires on F released. WTH is happening?!...

> Not sure if you guys have all used Sublime Text 3 for code editing. If not, you should. If you have, you will whole-heartedly agree with my #2 C3 feature wish

>

I've been using Atom/WebStorm to do things like you mentioned in #2, but I agree that it would be nice to have built-in to Construct 3.

Never tried Atom, that was always a Mac IDE. Did they ever make a Win version? I'll have to check. I also have a licenced WebStorm IDE. I like WS, but sometimes it feels like too much "stuff". Know what I mean?

I would kill to have that fast fuzzy search find-jump-to-anything in C2 and hopefully C3. It has to be in C3.