---

Greetings

I spent a bunch of time in the past explaining to people why FPSC is the best indie game development tool out there but I still see comments coming up on the forums questioning this.

Therefore I've gone ahead and developed a full level demo which demonstrates FPSC is capable of maps with counterstrike, unreal tournament quality.

Features of this level include:

- Full scale level using entire map editor surface area
- Seamless Indoor / Outdoor environment
- Spawning of dynamic entities (for improved performance)
- Level ambience controlled with scripts
- Custom huds, AI and terrain
- Action orientated gameplay
- Developed using latest v106 RC4

Video Trailer:

http://www.youtube.com/watch?v=--oc9fZNonU

Download link for demo will follow screenshots

Comments / Feedback welcome

---

Indoor screenshot 1/5

All pictures in jpeg format

---

Indoor screenshot 2/5

---

Indoor screenshot 3/5

---

Indoor screenshot 4/5

---

Indoor screenshot 5/5

---

Very nice. Can we see some outdoor screenshots?

---

It looks cool! Is it going to have linear gameplay? Or are you able to take wich way in the map that suits the player well?

---

Looks great.

---

Coming up now!


The gameplay is fairly linear at the moment but I'm working on a multiplayer version of this level which will be completely open ended.


Thanks

---

Outdoor screenshot 1/5

---

Outdoor screenshot 2/5

---

Outdoor screenshot 3/5

---

Outdoor screenshot 4/5

---

Outdoor screenshot 5/5

---

I love it! those hills make the outdoor screens look like a seamless world.

---

I think it is a really stupid idea to have the level take up the entire grid.

---

That would make a really awesome multiplayer level.

---

It seems these days in FPSC I always see people placing toilets in the weirdest places

As for multiplayer that would probably follow this simple equation:

Lag = (Full Grid + Lots of Detail + Multiplayer Function Routines + Player Connection Speed) = ((√5)/2)+2*∞

For those mathematically challenged, this basically means you probably won't be able to play the game

---

Hahaha, that's SO true. lol

---

Thanks. This was something that I wanted to address early on in development and found this to be the best solution.


It's a reasonable comment to make for the time being, if a little short sighted.


Cheers. I'm working on this at the moment. This was always my intention as I wanted to develop a level that would cross over from single player to multiplayer really well.


Well 'probably' sounds like a good enough reason to try.

---

Ok

I've now packaged up the demo and uploaded it.

File size: 84meg (compressed from 153meg)

Download link:

http://download.yousendit.com/BE10676005C91219

Have fun!

---

How is that short-sighted?

I downloaded and attempted to play this demo. The highest my framerate reached was 12 FPS. Sometimes I got as low as 1 FPS. It was unplayable.

I know what I'm talking about. So don't dismiss my posts as short sighted.

---

Sorry Nomad Soul, but those kind of frame rates are unbearable, it is possible to use the whole grid and not have lag, but you have to be really good with FPSC and know how to streamline lag spots. Simply making a level jam-packed with detail won't cut it.

---

I would've taken a different approach by making seamless progression between levels instead of in one level. That'd make it not so easy to convert to multiplayer, though.

---

I'm not questioning your knowledge of the engine or credibility, merely suggesting that this might not be a development limitation going forward.

All of the current v106 FPSC updates (RC1-RC4) are stability / bug fixes, therefore a future performance focused update could help make this type of level design a more legitimate approach. I certainly don't think that it's foolish to try.

I haven't seen detailed specs for X10 but I'm sure we'll be able to import X9 maps and I doubt an X10 configuration is going to struggle with large maps.


I'd like to start learning how to create texture baked segments like Rolfy uses as he states he's getting a constant 30+ frames per second on his masterpiece which is insane. This would also overcome the issues with using normal maps with segments that haunt me so much. Lee's having a look into that after he's completed X10 so I'm still hopeful.


Noted. At least this has demonstrated that maps can be made to be seamless. I've never tried taking on a multiplayer map before so this is new territory for me. I'm looking forward to it and hoping it's going to come along much quicker without the requirement for arduous AI testing etc. Although I'm sure there'll be plenty of other interesting challenges involved to keep me up half the night.

---

I hope that TGC will keep on updating FPSC long after the release of x10.

---

Actually, my post is somewhat misinformative. As my equation makes use of infinity, your screen should never update. Ever

---

'

Then sorry about the misunderstanding.

---

The use of texture baking in scenes is primarily an aesthetic one I just like the lights and shadows achievable in a top end modeler like Max.
The real frame rate saver is in the modeling,by creating most of the scenery using planes instead of boxes you can cut your poly count immensely.

---

Looks good mate but the link has expired.

---

"The real frame rate saver is in the modeling,by creating most of the scenery using planes instead of boxes you can cut your poly count immensely."
not to mention a far cleaner model (in both shape and polygon usage)

---

Thanks for the comment and I've now registered with Filefront as the free registration with Yousendit doesn't seem to be up to all that much.

Thanks also to Rolfy and Xsnip3rX for the useful information regarding texture baking and modelling.

Here is new download link for the demo:

http://hosted.filefront.com/Nomad5oul

Best

---

Looks good!

---

i think it looks good a little slow but my computer a peice anyway
other then that really nice.

---

The link is not there..

---

On the Lagg and framerate issue. It is quite possible for FPSC to accommodate highly detailed complex maps of the full world size containing both outdoor and indoor areas if you know how to optimise levels effectively. That does not mean you have to restrict unduely your level design but you may have to work hard to watch what FPSC sees (not what you see) at any one time at every location in your map and make adjustments as necessary using whatever optimisation techniques you can call upon. You can see what the engine sees in wireframe mode in test level.

The restriction FPSC has is limited to the number of enemies or characters you can successfully accommodate in your map and in view at any one time. The issue is AI related and is not an issue users can overcome. If your game would call for only one or two active characters in any location at any one time you should not have a problem, though the AI issue can be erratic in its effect. Some other Dynamic AI objects also have some issues but in general they can be worked around. e.g. Lifts can be a problem in large, complex, highly detailed maps with many AI calls to the engine.

Whatever you game level design optimisation is always going to be necessary if you want to push the boundaries of the product. To do that you have to be prepared to put in a lot of effort.

Character AI has a much greater impact on the engine than anything else. If you get an improvement or fix to the issue then that should free you to do what you will given that you may still have to optimise and give regard to the way FPSC compiles and renders any scene and the polygon counts it returns at any given location and view. That means that at any location in your map around a 360 degree radius one has to look to achieve a resonable poly count and fps return. Its unlikely that this engine will ever be efficient enough in its compile and scene rendering that you would be able to ignore that need.

Currently irrespective of your map size or complexity, AI activity can very quickly reduce your gameplay speed to unacceptable levels. One other note on AI and its impact. Erronous Character or other object AI scripts (and even models too) can cause havoc and have a serious knock on effect. FPSC as with most other engines does not react well to bad code (or models) in a level.

In the past in differing versions, the AI issue and its effects have varied on the impact on the result in game with varying degrees of severity. As far as the future goes and any improvement to that situation in any version of the engine we will all have to wait and see. There are no guarantees until something becomes a reality.

---

Agree with the AI issue. I am designing a game where only 1-3 characters can be active and interacting with you at the same time. I've designed my levels in that you have to shoot them before the game will allow you to progess (shoot the characters to open a door or make a key appear for example) I am having some success so will let you know when the first level is ready.

---

Here is a problem with the game that drops the framerate and creates serious lag,,,after you use the crate to get out of your cell, and you go to the next room, there is a window,,I took the crate with me the second try and tried to break the window,,it floated thru the window. Everytime I had that window in my view,,the lag went crazy. I dont know how you have the properties set for that window,but try and change it and see what happens to the lag and framerate. Also,,on a side note,,WHERE THE *(&)( IS THE KEY FOR THE NEXT DOOR???...lolol

---

Thanks.


Cheers. I was pretty determined to get a seamless indoor / outdoor environment in there which was always the focus point of this demo.


I always forget those damn links expire. It's up again now, I'll keep an eye on it.

Uman / Tzap - Thanks for the info. This was only really supposed to be a tech demo to show the environment potential of FPSC maps so in order to get the thing done, I made some performance trade offs during development which I would tackle differently in retrospect.

I'm working on a new demo which will be performance focused whilst still delivering an interesting environment to explore.


There's no glass in the 2nd window after you escape the cell. This was due to the lag issue. I took the glass out and messed around with the room a bunch but that lag spot remained.


The key for the first locked door is on top of the barrel in the room with 2 ladders to climb with a guard at the top (stands next to the barrel).

Thanks to all for the positive comments regarding the general look and design of the map. I'm aware of the performance issues on the outdoor areas but I'd rather address these in a new game demo and just let people check this one out.

It'll be an interesting challenge to go for a near constant 33 frames per second for my current project but thats the plan.