Milestone 5

[RonHiler]

Alright, time for a brand-spanking-new milestone thread!

To me, anyway, this is where things are going to start getting a bit more interesting. We're going to be doing the lion's-share of creating a world in this build, so it should be some good stuff.

This is the longest milestone to-date, with 24 entries in it. However, a whole lot of them are pretty trivial (there are a bunch that are nothing more than UI setup). So we'll see how it goes. I'm hoping this one isn't nearly as long as Milestone 4 was, in terms of coding time. But you know how that goes.

I spent the last couple of days going over the design doc, and I think it's ready to be posted, so I'll be doing that shortly. It's grown by another 3 pages, hehe. Not a lot of new text in this one, I just finished the list for M6 and added the list for M7, and did a bit of minor rewriting here and there. It didn't seem to me that we needed a lot to go ahead with M5 for this one, I did a lot of work on it between M3 and M4.

It also occurs to me I didn't post any screenshots for M4. So I'll go ahead and do that as well before I get started with the coding.

So here we go with another one!

[RonHiler]

After finishing Milestone 4, I took a couple of days off before plunging back into the code (need to recharge my batteries with a little bit of downtime, after all, hehe).

I started M5 this morning. Just now I finished the 6th item on the list, out of 24. Now that's the kind of progress I like to see, fully one quarter of the milestone list finished in about two hours

Of course, it won't all go that fast, unfortunatley. There's some serious coding in this one to be done, but I think there's a solid plan in place to tackle even the more complex stuff. This may be the longest milestone (in terms of number of entries at 24) through the first 7 of them, but so far so good.

[RonHiler]

Alright guys, I'm now a good chunk of the way through item 11 of M5, which at least numerically, puts us almost halfway through the milestone. So far so good. There hasn't been anything too difficult so far. I'm going to make a few internal changes to how race files are handled (as a part of item 11) which will also resolve one of our known issues from M4 (non-persistant race data in the race builder screen).

The devs over at DDO have just put Mod 7 on the public test server, and Mod 7 includes a new class (Monks). What that means is that I'm going to have to stop working on MD for the weekend and update the DDO character planner. Shouldn't take too terribly long, I'm thinking I should finish it over the weekend with a bit of work.

[RonHiler]

Lest you guys think I've been slacking on MD while building my new rig and working on the DDO character planner update (which is done and in testing phase), I have in fact made progress. We've slowed down a bit on the milestone list, and I'm still working on Item 11, but that's because I detoured a little and decided to take care of one of our known issues, which was the non-persistant race data in the race design screen (I guess I mentioned I was going to deal with that in the last update).

Well, we now have persistant races. Yay! This was actually related to item 11 on the list, which is why I decided to tackle it now rather than later (I know adding a "Browse" button doesn't seem like it has anything to do with race data, but it does, in a somewhat roundabout way).

Anyway, all of this required fairly significant changes to the race desgin screen class, a whole new class for player data, and naturally changes to the new game design screen class (whcih, being brand new, is in a constant state of change anyway).

Hopefully by this weekend I will be swappinig all my development projects over to my new machine, which includes changing over to Vista x64. I hope this does not cause any new delays or problems, but it very well might. So keep your fingers crossed there.

[RonHiler]

Hopefully by this weekend I will be swappinig all my development projects over to my new machine, which includes changing over to Vista x64. I hope this does not cause any new delays or problems, but it very well might. So keep your fingers crossed there.

Just to let y'all know, MD was transferred to the new OS with no ill effects. Well, there was ONE ill effect, but I de-illed it

[Shane Christopher]

So when can we see the DX10 build?

Because you know.. everyone else is doing it...

[RonHiler]

So when can we see the DX10 build?

Let's see, what's the word I'm looking for?.....

Oh, yeah, "NEVER" lol. This game SO doesn't need DX10 level graphics

So I'm about done with item 11 on the milestone list. I had a devil of a time seperating the player data out of the race creation interface screen. It took me a couple of days to get save/load files working right again (which reminds me, saved race files from the M4 version are not going to be any good anymore, I changed a format on one of the entries to compress it). But all seems good now, I should finish 11 tomorrow (all that's left is to add a directory crawler and load the file up and display the race name). Should be a cakewalk at this point.

So, anyway, on to item 12.

[RonHiler]

Finished through item 16 now, working on item 17, which is just the interface to allow you input your own random number seed (rather than use a true random one) if you desire.

Then it's on to world generation. It will be good to be doing something other than UI work for a little while

[Shane Christopher]

World generation is always fun. As soon as I get my camera view controls finally worked out I'm going to go ahead and try and port some of my old generation code (or completely rewrite it probably) into my new project. Right now I'm just using terrain I made up in terragen to test with.

I'm having a hell of a lot of trouble with what should be a simple cosine operation getting the angle from a given cosine value. It refuses to give out the correct angle no matter what I try. The 3rd person camera can swing around a sphere around the player, you can change the radius of the view and swing it around and up and down with the mouse, everythings good until it gets to making sure it doesn't go through the ground.

I'm trying to get it to stop when it hits 10pts above the ground (roughly 10cm). I can make it stop but then when you keep moving the mouse down the pitch it is trying to get to keeps getting lower. The camera stays still but as soon as you try pitching up again it takes ages to get the pitch back to the value it was when it hit the ground. I've tried cacheing the pitch value when it hits the ground but then as soon as you start moving over different altitude terrain that can kick up a similiar problem.

What I really want is to be able to adjust the pitch to where the camera is actually at after every update. For some reason I just can't work out how to get the pitch. I have the cos value of the angle I'm looking for it's just when I put it through the Maths.Acos() function it comes out with something completely different then what I'm looking for.

[RonHiler]

So I finished the initialization work today on the world builder. More or less, it's the same functions that we had in the old version of the program. One of the big reasons I was not happy with the old version was the amount of time it took to initialize the world.

Well, after doing quite a bit of work (both in terms of code cleanup and optimization), we now initialize everything in about 45 seconds. And that's on my laptop (which is a PoS) with a debug build and with quite a bit of debug output, which slows things down. So I'm pretty happy with it so far.

The world for the builder consists of a bit over 400,000 points (translated to a mercator projection, it comes in at about 500,000 or so). As I've said before, eventually each of those points will turn into an 8x8 grid of points for the game portion (which all together comes up with about 26.4 million points), so y'all should have plenty to explore

[edit] I want to stress for those that may not be clear, you aren't going to be running the world builder all that often. Only once at the beginning of a game (and then only by the host, not by the players). Since a typical game can last months, you won't be building worlds very much. 45 seconds is way too long a time to spend if we were having to do this every single time someone opened the game, but to me it's probably fine for the occassional new game build. Of course, we haven't yet started the iterations, which will take some amount of time as well. But to offset that, I expect that initial 45 seconds to be far reduced in a release build without all the OutputDebugString messages (of which there are about 520) and on a half decent computer. If we can keep the world building under 5 minutes, I will be happy with it (I'll try to get it less than that, we'll just have to see what happens).