New games?
Miscellaneous Forums/General Discussion/New games?
Anyone?
I know Jake (greyalien - if you did not know :! ) has one on the way... Anyone else? the "Hum" seems to have gone...
the website always says my password is incorrect even if I get a new one
We're doing a new one. It'll be primarly targetted at consoles but we're prototyping on PC with Blitzmax. Hopefully a PC version will emerge from that too.
I'm doing a re-write of a classic game in BlitzMax. Will be very cool when it's done. I'm hoping before Xmas. ;)
I'm still writing my platform game. But it's the same one I've been more or less working on for years, so perhaps it never contributed to the "hum" anyway.
I'm doing a re-write of a classic game in B3D..... probably will be humming as opposed to contributin to the hum though. Who knows, i'll go for a pre christmas release... don't know which christmas... 08 or 09 ?
seriously though, i like the game play and its very early days.
I'm downloading lorry loads of porn and going blind so unlikely to see any kind of software output from my camp.
Rob to much info
Elite Multiplayer is still in development (just released alpha release 10). Once that is tested solidly I will add networking to the 3d client and then start major work on the 3d side :)
Ok, not a new game, but I'm making a hefty update to Tank Universal and hope to have a test level ready sometime in the next month.
I have a game that I've been slowly working on for a while now. It's written in BlitzMax, and is a 2d Zelda-ish game. It's a short game, intentionally, since it's basically a test of my engine, and will be freeware.
- The engine itself is complete (but hasn't been through beta testing yet).
- The main quest is 99% scripted (finishing adding monsters to final dungeon). No side quests have been completed.
- Sound is working, but I haven't assigned sounds to everything yet.
- Still using programmer art and free tiles at this point. I had two artists volunteer to help out. One made an intro sequence and title screen for me. The other made the beginnings of the main character sprite. Both seem to have lost interest or gotten busy with other things at this point, so I'll be looking for help again soon. :(
Anyway, here's a screenshot of the WIP version:
I have a flight sim in development and just started working on a second tech demo too.
I've been back working on my Vehicle Physics engine for the last month or so. Getting right into it and have a clear plan of the game I'm making. I've been mainly improving the vehicle handling and re-structuring the engine to make it easier to implement all the mini games I have in store. Added ghost car support on the race track today for TimeAttack racing.


Experimenting with headlights ..

I hope to have EngineTest7 on the way shortly and finish the game at the end of Q1 2008.
Stevie
glad your still working on your driving engine steve, i loved the demos
I'm still working on my Invaders from Space,
http://www.blitzbasic.com/Community/posts.php?topic=71599#800251 I got a little sidetracked with with my texture packer/ bitmap font packer, but I've started back up on it again.
@Stevie G: I love the look of that!
Yep, there's is one on the way...
PLS
Steve's stuff always looks cool makes me want to try harder ;O)
Sounds like there's some cool stuff on the way. I look forward to seeing it! :-)
That just looks so fricking cool...reminds me of the 90's when virtual reality was becoming the hot new thing and then abruptly just died.. :p
Neat to see flat shaded polygons put to good use. How about a shadow under the car to help with knowing whether you are actually touching the ground and where? Since you dont have textures it's possible to visually lose the association between where the car is and where it is in relation to the road.
Me, I'm still working on my funky game/graphics engine, aiming toward my first game being a 2D space shootemup. I'm currently designing an advanced next-gen holographic particle system which will handle pretty much every object on the screen all in one system - the player, the bullets, the aliens, the stats, the background, everything including all the game logic and scripting. This type of game will be a good test of many parts of the system that I want to be in place ready for a future projects which are going to be somewhat more challenging. I'm starting out with something of a redo of an existing game genre to get my feet wet and build up the engine's capabilities.
Well, I'm still working on my Rockfall remake (Repton/Boulderdash clone)...

The game side is finished now, and all I need to do is get the level editor in so I can design the tutorial levels and make up the current 71 levels to 100. Following a bit of tweaking the game should be good to go. Was hoping for a pre-Christmas release, but I suspect I'm going to miss it!
@ Imaginary Human - shadows are already there ... this shot shows them better ..

[EDIT]
Looks great JassieB!
Stevie G -
Wow man! Looking fabtastic as always - I'm looking forwards to seeing this come together as a final product for you - the work you've done on it so far is inspiring!
Great stuff.
IPete2.
Stevie G:
Its great!.
I have stored on my had drive all versions you posted. Please tell us more:
Are you gonna make a game with a scenario?
When it will be ready?
What you think about your engine's price?
I want to buy and play the new one NOW!
Hey that engine looks like it would be great to remake "Hunter" from the Amiga, that game was cool.
@Stevie G
I still like the overall style of it but i think that these shadows don't fit well here. Actually i would expect hard ones.
@JazzieB
Looks after a lot of testing work...
StevieG I like that you are using bright colors in your game, and that you don't care about hugely realistic polygon counts.
@Stevie G
I still like the overall style of it but i think that these shadows don't fit well here. Actually i would expect hard ones.
I challenge you to do better & faster textured soft shadows :) Bear in mind that for speed the whole level uses one 1024 x 1024 texture.
Well, if that's the best solution then i would cut it out or at least make it an option as i think the way the shadows are looking now they don't fit to the rest of the scene.
I've got a game coming out soon.
AtomHex - A 360 shooting/collecting thingy.
Currently wrapping up the web side score & registration stuff.
im about to get started on one.. though wont have anything
to show for quite some time... should be fun.
RTS kinda thingy..
Believe it or not, "Bad, Bad Bots!" is still very much alive and kicking. It's now a year since the last demo was posted and submitted to IGF. I took a step back, looked at what I had and what was lacking. I've spent the last year adding those things. And just two days ago I reached FEATURE LOCK (and I've added a ton of new features :) ). I have a crew that's going to help with the level designs while I push on with the graphics and animation.
Dustin: Take a look at the game in this thread:
http://www.blitzbasic.com/Community/posts.php?topic=73437 It made me think of Bad, Bad Bots. I think there are similarities and maybe things to be learned from that game.
Thanks to your 'free game a day' tip I managed to snag it. I haven't had a chance to try it out yet. If the "Montezuma's Revenge" comparison hold up then it would be a little more arcade like than mine (BBB has no jumping). I'd compare my game more to something like "Lost Vikings" or "Castles of Doctor Creep". Looking forward to giving Azangara a try though!
You have to install that game (and the unlock code) on the day it was downloaded - hope the zip is not languishing unzipped somewhere...
Yes your game is a bit more "cerebral" for sure. I enjoyed this game for a sort of mindless fun, but it has quite a lot of tactics too...
Stevie G: I think the shadows in your game do the job well. Can't wait to see what you do with just like your other engine. really nice stuff!
> You have to install that game (and the unlock code) on the day it was downloaded
Oops! Heh, guess I'll be playing the demo then. *smacks forehead*
Oh well, mindless fun works for me a well.
Stevie G: Fantastic work, the shadows do look fine as they are, gives you an idea of how far your object is from the ground. Cool as ever.....
I've been playing around with:
http://jv-ode.devcode.co.uk/It is great for 3D object manipulation. Very intuitive, I've been delving back into Blitz3D after finishing something in BMax (is anything ever finished?) - the above physics engine is great fun.
Stevie G - I like the look of your game (if it is a game) but I'd definitely scrap those soft shadows. They ruin the whole feel of the place. Why not try a tesselated mesh for each non-static object then line pick the ground underneath for deformation - and just use alpha - that way you can keep the hard crisp edges throughout the game world. It would look just perfect - trust me. You can make different vague shadow meshes to suit each vehicle.
Stevie G - I like the look of your game (if it is a game) but I'd definitely scrap those soft shadows. They ruin the whole feel of the place. Why not try a tesselated mesh for each non-static object then line pick the ground underneath for deformation - and just use alpha - that way you can keep the hard crisp edges throughout the game world. It would look just perfect - trust me. You can make different vague shadow meshes to suit each vehicle.
Your kidding right!! Not much overhead in doing at least 4 long linepicks for over 50 moving objects. Not to mention the fact that the shadow would not hug the landscape nicely at all. I'd also have to transform these linepicks based on the best fit for the objects pitch / yaw and roll rotations which is a pain. The whole idea would be dog slow - trust me.
Now, I can assure you that I have tried various other methods and none come close to the speed and versatility of my current method - sure they are low res but so are the static shadows.
Don't get me wrong I'm open to suggestions but not silly ones like that. If anyone doesn't like them then they can turn them off.
Are these better?
It's silly if you're not going to think about it over-much. You only need to deform the mesh if it's close enough to notice...Obviously. Stick with the soft shadows if it makes you feel un-silly.
Hujiklo,
Clearly I have thought over it, although I failed to mention that I have tried your method many iterations ago ( and yes, only on vehicles within a certain range )and it just doesn't look good. Note also that the projectiles have shadows too. Raw speed is the aim here so I'll stick with being un-silly thanks.
Stevie
I think stencil shadows would look good. Probably too slow to be useful tho...
You made custom shadows for each object? So the helicopter in the shot above had a vertically projected helicopter-shaped shadow? And it looked worse? Worse than a blurry oval in a hard-edged world of flat shaded polygons? Because to be honest, those blurred messes erode the whole look and feel of the world you've obviously been striving to create. Clearly a matter of taste I think. I cannot imagine there being any speed difference at all especially with a simple lodded shadow system using hidden child meshes...and why 4 minimum linepicks for a 4 vertex square shadow (that couldn't deform in any useful way anyway) when only 1 is needed? Sounds very well thought out to me.
The soft shadows however, would sit infinitely better if you'd used smoothed instead of flat polys.
[Double Post]
Perhaps I don't understand your method but I can't understand why only a single line pick is necessary? Are you talking about projecting just the low poly shape onto a texture.
Please explain in more detail as I'd love to improve the shadows if possible?
Cheers
Stevie
I mean - take your models into your modeller. Select the top viewport and simply plot the outline of their shape and make a polygon mesh ( the squarer and more grid friendly the better). Then make 2 or 3 ( or how many you think will be enough) versions of the same flat mesh. Tesselate the first at what ever resolution you think you can get away with for a close up shadow, do the second less so and the last one do nothing at all to. Child the objects together in what ever order you see fit and export so that you can import them as an animmesh. Use getchild and hide/showentity for your LOD. For the basic far distance child-mesh you only have to do one linepick from its centre.
Optimize the rest of the closer resolution mesh picks by selecting one vert from the extremities of each side (pre-stored in a type of course) and only check these first - even on a higher tessellated mesh you'll only be doing 4/5 picks most of the time. If the Y picks are all roughly in the same ballpark then no deformation needed. I expect you're already making sure the shadows lie 0.05 of a metre above the ground (or some distance close to that figure to avoid flickering - infact I'd be tempted to increase that number the further away the vehicle and decrease the alpha value so that shadows effectively dissappear 'smoothly' into the distance). It isn't perfect but you wouldn't have to sacrifice any of the styling you have now.
No textures needed - just alpha and a brush colour.
That's what I thought you meant :)
Firstly, what happens if the car is rotating in an axis other than yaw? The shadow beneath would not match what is actually happening and would look terrible.
4/5 linepicks wouldn't cut it in certain scenarios - e.g. the track curb. Up close that would look awful and it IS dog slow cos I have tried it. Evey vertex in the mesh would have to be linepicked to ensure it hugs the landscape and even then it wouldn't look good.
If you extrude the shadow mesh by the picked normals then the wheels will be slightly underneath the shadow - again looking terrible.
My shadows are rendered to two big textures, one for high shadows and the other for low so that I can have shadows done correctly for under and over bridges etc... They also appear under water and do not require any kind of offset. For ensuring only nearby shadows are rendered I simply copyrect a square section around the players car, same for a second player if there is one and repeat for both low & high shadows. It automatically builds the static shadows too.
While the textures are only 1024 x 1024 I can render 50 dynamic objects in under 5ms. I'd prefer to have ok shadows which serve their purpose than have a system which brings the gameplay to it's knees - I'm aiming for low spec here. Don't get me wrong - shadows were never an afterthought and I put alot of work into what I have but to me gameplay and a consistent style are more important.
I'd love to be able to have perfect crisp shadows but Blitz just isn't up to the job it's that simple. If someone wrote a nice fast and versatile stencil system I'd be using that without a shadow of a doubt ;)
That said, I appreciate your time and input.
Stevie
I forgot about the kerbed roads. But either way You could use highly optimized invisible meshes for the linepicks and split them into zones - again using the childing system. I guess it's a bit late in the day to start experimenting anyhow. You lost me with the other axis rotations - why would this be a problem? Surely you're just going to get a normal from the picked surface and rotate the basic shadow mesh to it any way? I admit these shadows wouldn't be hyper-realistic, but therein lies the charm of the whole flat-shaded poly world you have there. If I had the time I'd look into this myself - in fact if I get the time I will for my own project!
If we could start a project backwards we'd have one heck of a slick design document!
Surely you're just going to get a normal from the picked surface and rotate the shadow mesh to it any way?
No. It's kinda like this. As you can see my proxy sphere is scaled on each axis to fit the size of the car so changes depending on the rotation.

If we could start a project backwards we'd have one heck of a slick design document!
Agreed!
Right - gotcha! Well I suppose it all looks better running than a still screenshot. I still think this can be done with basic poly shapes though. A cheap basic-shape stencil shadow using your own automatically pre-zoned terrain surface data, and then sorted into separate types for each zone for ease of use ( maybe even optimise it further depending on each type's surface coords). Then do you own linepick code for the active type-zone only. This could be quick as lightning depending on the depth of the pre-processing/zoning of the terrain data. The scaling of the 'basic' shadow could be either done by the stencil code or just scaled as you're doing now. I'm wishing I had the time to stop what I'm doing and give this a whirl now. I always fancied doing a Hunter type game. I worked with the guy who made Hunter a while back - he packed in games a few years ago, and now earns more money for less hassle - with the odd games-related foreigner thrown in!