I guess it’s been a fairly progressive week so far. I’m done with my optimizations for the moment. Things work exactly as I said they would (there’s a first) with one change. I mixed fixed-time step with variable-time step, so, only when your frame rate drops below 30fps will you see a sharp decline in performance. Who would want to play a game that slow anyway. In the end, things work better than planned!
Moving back to online plan today was kinda like a kick in the head. I suppose I have trouble moving on to things I left off with on a bad note. The thing that bugged me about it was the lack of smoothness. Having more time to do things right, I spent the day fixing it. In the end, the game moves very very smooth to my standards. When I slow down the connection the game still manages to move decently. It’s by no means perfect but much much better than before.
Now that I’m settled back in I have 4 tasks to do before I move on to something else.
Bundle character data from the server into 1 packet
Keep accurate tracking of things on the screen (Weapons)
Decrease the size of the packet (smaller the size the more I can do)
Support for more than 2 players
(Sigh) It’s gonna be a fun week.
Note: You may have noticed I changed the level around. I looks kinda the same huh. It’ll probably be a while before I try to make a fancy stage.
I planned to do this earlier but I got a bit tied up. I’ve added part three of The Perils of Courtship to the story page. The story doesn’t get deep like things to be, but I often enjoy changing tones which you can probably tell from previous games.
Glad that’s out the way! To accompany the finish, here’s a quick picture of Roy. I don’t think it has enough pizzazz to go into the art section so I’ll leave it at that. The design is a fair deal different from what you can remember in Dream Master. One of the questions in my current design though is the lack of padding on the fingers. Rabbits actually don’t have padding on their feet so that leaves me with a problem. I’d like to give every animal a sense of touch here so I have the option of adding pads or making up something. I doubt animals with hooves have sensory organs in their hooves (remind me to check) so I might just make up something. There have been plenty of times when I’ve bent the rules a bit. I guess having some sensitivity in the fur isn’t so far-fetched. Am getting too far into detail again aren’t I?
In other news, remember optimization? Well, I think I managed to get about 30 on screen at once at max speed. However when the number surpasses that, there is a SHARP decline in performance simply because I use a fixed-time step. While I love using a fixed time step to solve some of the physics problems that occurred last game on slow computers, it’s going to virtually KILL the game on slower computers completely now. I decided that I’m going to remove it. To combat the problem, instead of having the game try to run at a constant speed on slower computers, I’m simply going to run the game slower when things get too slow for physics to work correctly! Without the fixed-time step, I was running around 70 characters (a few collision factors out of the equation) at 30 fps! If I tried that stunt before I’d probably run it at maybe 2 fps? With this decision I’ll have the option of more on the screen and slower computers will only slowdown instead of being singled out. Good stuff!
The forum is officially gone! And gone with it are the 5 other members that never posted, so whoopty doo to that. I’ve added all of its profiles to the profiles section and now I declare myself caught up. Still missing are Walkthroughs, Character Galleries, and The Vulpine Studio tool but the fact of the matter is, I’d rather post the walkthrough when people start asking for help, the character galleries are a little outdated (especially George Jetson down there), and I don’t wanna talk about the Vulpine Studio. Not only is it using the XNA 3.0 CTP at the moment but too much has changed in the new version for the old one to even be worth testing.
In regard to my last update, I’m still working on that optimization. I had to reorganize the Level class about 3 times to get things just right, extremely time consuming but I learned a few C# techniques along the way. It seems I was right in predicting 100,000 triangles on the screen at once using the vertex buffer (although, the reorganization did a lot as well) however I learned that while my computer is fast enough to draw all of the triangles at 60fps, it isn’t fast enough to edit all of them in good speed! Ain’t life grand. Still, I wasn’t planning to use that many triangles anyhow. I’m not making the Unreal Engine! When it’s all said and done though, the levels should look much, much, much, much better!
I think 25,000 triangles is a good number. That’s about 12x the size of the largest scene on Tandem Tales and it should move at the same speed. I wonder if my little tricks would work on Characters! It would probably only take 2 days to do but is it worth it….. I guess I could put off online play for a while longer and get an estimate. There’s something enchanting to me about having a lot of characters on screen at once.
One more thing! I may start doing things in exchange for Download.com reviews… I dont’ know what or how I could go about that but I need some reviews. Reviews = Inspiration!
Not really news but I’d like to note that by Saturday I will have everything from the forum moved and the old forum will be removed. The use of html in the conversion is probably a bit unorthodox but it works.
I found it weird that though that my first news post on the forum was Friday, June 13th, and that my last one was Friday July 13th. I’d swear I was doing things like that on purpose if I didn’t know that I wasn’t. I hope that made sense.
In short, I will have no more old content to tack on. Slowly but surely, the universe is going to start expanding a bit!
Lately, I’ve been devoting a lot of my time towards improving errors and building on to the Level Studio, however I believe I’ll very soon be optimizing things again.
In earlier chapters, I had given up on optimization, thinking I had reached my limit without sacrificing too much. Boy was I wrong! Today in the level editor, I was drawing a level with 4000 triangles. That might seem like a lot to some people but believe me when I say it’s chump change. I noticed the editor slowed down at such a low count. I should’ve been getting at least 4 times that amount before any slowdown.
I did some research and learned a few things. It seems a vertex buffer could improve the speed of things a bit, although the number of texture calls in the level is outlandish. I did a few tests and learned that, textures aside, my speed could double if I use the vertex buffers. When it was all said and done, I could draw 200,000 triangles and run the editor at 30fps. I’m guessing my computer could handle 100,000 at 60fps. Since I’m only using two textures at the moment, that number will surely stick after the change. Considering other computers of mine would slow down on “Nightmare Castle”, which totalled around 1,400 triangles that were never all drawn at once. I’ll take this as a worthwhile discovery.
My reasons for speeding things up are
My target audience will mostly not have very expensive computers
I’m getting much more detailed with the levels than in previous games so I’ll need the boost.
After speeding things up, I plan to add a few minor touches, then back to online play. I’ve been having a lot of good distractions! \: ;j (clever mustache face if you’re wondering)
This Monday may be considered a bleak Monday, but it probably won’t be for most : ) . I’m removing my Monday updates for the time being. I believe it makes much more sense to do things such as this, AFTER the game has been made rather than being a hindrance to the the production of it. I will continue to add content periodically as I’ve done before but I”m not going to put complete focus on Monday anymore.
This Monday features, the continuation from the story of last week. There are 3 parts total, so much for a short story, but you can look for the third part sometime this week. The story is actually done but I wouldn’t want to lay it on all at once.
One more piece of news, I’m going to be transferring a lot of information related files to htm. They’re just so much easier to work and keep up with. You can find the story section using an htm file over on the side bar.
I’m happy to report that Download.com is now up to date on Felix games. Previously Samba of the Wolves did not work on Vista and Tandem Tales had a few sound issues.
Slow times for new content but I’ve been busy with the actual game lately. I’ve ported the Evergreen background page from the forum. New web oriented content is pretty much why special Mondays are around.
Genre: Beat-em up’/Adventure/Platformer (1 Player) Production Range: February 9th - November 16th 2006
(Worked 4 months on seperate project, blame the rest of this long period on the level editor and 3D collision detection. (acursed hills))
Summary:
A bit of a hybrid between SotW and DreamMaster and an older game I’ve yet to release. This game was supposed to consist of several tales depicting stories around the Felix universe. Online Co-op was planned and I was really going to take advantage of it but as you know plans change. I introduced too many things in the game that must be addressed before I attempt such a feat, and my tools are not capable enough to handle something like that. I’ll probably attempt an online battle mode to test online but I digress. What is available is my first attempt at that. It’s Sort of a demo tale.
Important Notes:
Hold “B” to do a strong attack and Press “L” to Roll. They both become important at certain parts. The rest should be easy to figure out..
We’re back on track for special Monday updates and this Monday I tried writing a short story. I suppose I’m trying a bunch of different things to see which one clicks. It’s probably noticeable that I haven’t done the same thing twice yet. Look below to find a short written story, I know it says to be continued but I cannot say for sure.
Aside from that, I now only have one more game to add and I’ve added the general story to the Felix series to the sidebar. You may also notice a art page, however that was in fact added yesterday. I thought I’d bring it up. That’s all for now.
Genre: Beat-em up’/Platformer (1-Player) Production Range: September 21st - December 5th 2005
Summary:
My first attempt at a complex storyline. Felix finds himself in perils crafted by non other than his imagination. He must progress through his dreams in order to escape them.This is game is considered to be a test for the 3D realm. It was made in a bit over 2 months as opposed to to the 5 months invested in Samba of the wolves, which technically is still incomplete. I think it turned out fairly well considering.
I guess it’s been a fairly progressive week so far. I’m done with my optimizations for the moment. Things work exactly as I said they would (there’s a first) with one change. I mixed fixed-time step with variable-time step, so, only when your frame rate drops below 30fps will you see a sharp decline in performance. Who would want to play a game that slow anyway. In the end, things work better than planned!
