Mumbles

By your display name, I presume you live in the US? Why would you buy parts or even full machines from someone over here? If you pay $100 for anything related to a computer over in the US, we pay close to about £100 for the same thing over here in the UK. But as you know, £1 is worth more than $1 (£1 is worth about $1.50)... So our price for the same hardware is more like $150 in your prices.

That was certainly true during most of LE2's time, but like a couple of months before the new engine came out, AMD got their butt in gear and sorted out a lot of the problems they had. Now a top end AMD card should work just as well as a top end Nvidia card - all just personal opinion now.

Oh, I've been using newton for too long, I'm so used to using the custom 6DOF joint that I totally forget that the normal joints need parents. But in Newton 2 (I know Newton 3 is out, but I'm still using 2), even if you do create a joint with a parent, you can call NewtonBodySetMatrixRecursive (which sounds like the SetMatrix function in Leadwerks with the recursive flag set to true) on the 'parent', and it will move all children too, and rotates all the joint pins on the parent to match. I don't know if Leadwerks allows this or not, or if you have to call EntityParent first - or whatever that function is called in the new version, so that the door is parented to the pivot. But in my mind though, I'm still sure that moving and rotating only the pivot (the parent), after they have been joined together is the way you're supposed to do it. That's certainly how you do it in Newton 2.

I would have thought you just attach the joint to the door (both totally unrotated), and then rotate the door. Would that not automatically rotate the joint -with- the door?

The hashmap approach certainly sounds like it could work, but I've never tried using function pointers in this way before. But I will warn you, all your function pointers in the same hashmap need to have the exact same parameter list and return type. A function pointer to a function that returns an int, is not the same type as a function pointer to a function that returns a float for example. And two function pointers to functions that each return an int, but have different parameter lists are also not the same as each other. However you can get around that by having a "void *" return type and a single "void *" argument. This way it's up to the individual function to cast that void pointer into a meaningful struct or object to work with, and then whatever it passes out, the calling code must then know what to convert that void pointer into. It will probably cause quite a few errors at first as you accidentally cast these pointers into the wrong types, but in time, if the hashmap console approach works, you'll quickly adapt to using void pointers as a means of passing the data to and from your dynamic functions.

I suspect the underlined bit is wrong, because it appears to be a standard void pointer, and not a function pointer. myMap.insert(pair<string, void*>("Test", pointerToFunction)); For some reason, I'm thinking it should be this instead myMap.insert(pair<string, void (* funcPtr)(void)>("Test", pointerToFunction)); The bit in red is whatever the parameter list for pointed-to function is, in this case, blank. And pointerToFunction should be declared in the same way. The bit in blue is the pointed-to function's return type, again, it must be declared identically if your function took two int parameters, then try myMap.insert(pair<string, void (* funcPtr)(int, int)>("Test", pointerToFunction)); and so on... I can't think if you need the (* funcPtr) bit or not because that's how you declare a function pointer, but I don't think you want a declaration here, so even I'm a bit confused. Remember function pointers can be confusing, even to the best of us, and even I'm a mere novice. Is there not an easier way to do what you're trying to do?

In raw Newton, you attach a custom 6DOF joint to the body you want to restrict (and specify no parent body), and set that joint's "linear limits" for the axis in question to 0. So for top down views only, with no jumping, set the min and max linear limits for the Y axis to 0. Now it can move forward/backward and left/right, but not up/down. BUT... I don't think LE 3 exposes the custom joints part of the Newton library (and I know for sure LE 2 doesn't), so sorry, but I don't think you can do it.

That's the way I've always done it. Quick mock up the rooms with CSG, because architecturally, most rooms are cuboids, or maybe a central cuboid with smaller cuboids extending from parts of a wall (like a closet or something). Proper polygonal modelling is for putting things in the rooms, like TVs and fans etc. Most real life rooms are boxy, not with rounded corners. Spiral staircases are quite a niche thing. It's not like every building has one...

I'm more familiar with WorldCraft, but it was the same thing. Very useful too for fast room building, and then you could "carve" a doorway. Carve just being their word for CSG subtraction. Very nice for LE 2 was that UU3D has an RMF importer, so those of us without LE 3 can still have fast, CSG built levels - except we use a third party program instead. (I don't believe there is a clause as part of WorldCraft's or Hammer's license that disallows such usage) I must admit, when building a house (for example) I just created a huge big brush the size of the entire house, hollowed that and then added lots of thin horizontal brushes to separate each floor, and then thin vertical brushes to define each room within the floor. This way I didn't have to worry about "connecting" the rooms together, as I would have if I'd done each room as it's own hollowed brush. I only hollowed the entire building as a shortcut - I could have done it with 6 large (but thin) brushes just as easily.

I don't know who's method is easier but I'd go with Shadmar's method... Your way of a make room button has everyone's rooms looking very similar, in the same style. When you build a prefab, all of your rooms look similar to each other, and all of my rooms look similar to each other - but your rooms are probably in a totally different style to mine. Every game developer's rooms looking more-or-less the same gets really boring really quick. Games are where you can let your imagination run wild. If you're not going to have anything unique in your game, I'd join Activision... The rooms in Metal Gear Solid didn't look anything like the rooms in Splinter Cell even though they were the same sort of game from the same sort of time.

Whereas I thought the gravel was too shiny for a real photo...

Well I know what my first thought was, but the image URL has proved that once again, I'm wrong.

I didn't think LE 2 load scene returned anything, but once done, you could then do CountChildren() on the scene (everything it loaded was a child) and then use GetChild for each of these children to access each entity (GetChild returned a TEntity type) Obviously have no idea how this translates into Leadwerks 3 but in LE 2, LoadScene returned nothing. However there was a tutorial where it guided you through the process of writing a ProcessScene() function to get the children. Edit: Actually no, I lie, You passed it a string, it returned a TScene object, which you then called CountChildren() on

Pizza, mmmmmm... The London Underground roundel is a registered trademark isn't it?

I did the same in 2008. Had a Windows XP computer, and a Windows 2000. I got a quad core AMD Phenom processor, but each of these Windows version were locked to 2 CPUs each (physical or logical), I would have to upgrade to Vista or newer. So I upgraded Windows 2000 ... to Debian Linux. The version at that time did support it, but the system clock counted twice as fast as it should have done. There was no excuse for it because Debian has been running in data centres for years, with servers having hundreds if not thousands of physical and logical CPUs, but a few months later they fixed it.

That's not official - just what I believe...

I think now, LE 2 support will be serious bug fixes only. Could be wrong, but I think it will be almost all Leadwerks 3 now

To use the older compilers, don't you need the older versions installed? Makes no difference to me, I'm an XP'er, but I know code blocks also allowed you to use VS2005 and VS2008 compilers, but you had to have them installed to do that. But that's probably down to copyright. Microsoft shouldn't have that problem because it's their product so I don't know.

He doesn't like the fact he use to use an LE 2 tool for Leadwerks 3 to be fully functional... Or so is the claim, I don't own Leadwerks 3 so I don't know.

We are superior...

In 2.2x it was available as a CPP file. But it's not guaranteed to be the same in 2.3 onwards

I'll remember you said that...

Glancing back to old TGC times, I would have thought showcase forum would be better suited to actual completed products that you want to show off. Maybe there should be an completely dedicated work in progress board here.

The command I remember is AppSuspended() returns 1 if the window is inactive (you have Alt-Tabbed away), 0 if the window is active

Source code if you really want it, but really, everything is in the one source file. It's all in the last post, but that scroll bar makes it a bit harder to read... Minor typo, but not serious. the int main has it's first value of "int argcn". It shouldn't say 'n' at the end, but there's no way I'm editing that post, otherwise I'd have to format that big block of code all over again - no thanks.