Tuesday, 6 August 2013

Tuesday Roamer

Roam and Roll

Today was spent adding the proper player controls to the physics sand-box prototype I am working on, which means proper strafe, acceleration, smoothing, speed for walk and run, pretty much everything except jump. I threw out the terrain ground height check, and replaced it with the final Bullet Physics character controller, minus the mouse look camera rotation stuff which was kindly provided from my early 'decent demo' prototype.

The result was pretty good, and I was able to run around as before, but critically not be able to run up the steep hills as the step value was sufficient to block my player character doing that (as intended). I then dropped lots of physics objects which I have also enabled through my new DBP bullet physics command set and all was good.


Two things plagued me during these tasks. The first was that the performance was not 'great' once I added the 16 million triangles needed to accurately simulate the terrain floor for the physics system, and the general speed of moving physical objects did not seem right.

It was though I existed in slow motion and that to get any decent speed I had to be travelling at quite a high velocity to get moving. The basic problem is that Bullet expects the ratio of 1 unit = 1 meter, whereas the FPSC universe up to this point has been 1 unit = 1 centimeter give or take.  Given this, my objects are considerably larger than Bullet expects them to be, and create these slow-motion artifacts which spoil the fun. I can increase gravity, and decrease mass but it's not going to work flawlessly if I start guessing.

What I need is some real special math and lots of new Bullet settings to create a world that makes my 1 unit = 1 centimeter universe perform as though the real world has hold of my physics objects.  I suspect this means some deep Google search research come Wednesday.  Sure, quick hacks are available and I can move on to some degree, but the physics will be with us a LONG time and I want it to be mathematically sound before moving too far.

Cheeky Visual

My blog readership has dropped away somewhat since Rick's decision to post anything other than near-finished visuals, and I am a bit miffed about that. Nothing egotistical, just that you guys and gals will vote with your visits, and it tells me you tune in to more Reloaded news when there are visuals to look at.  To that end, I have posted a sneaky one below and added some disclaimers all over it so there can be NO confusion that this is work in progress art and should not be considered finished material we are going to send to the printers on Monday.

As you can see, looking past the disclaimers, I am using a sandbox test prototype with the terrain visuals stripped out so I can focus on the physics shapes and simulation.  You are looking at the aftermath of a stack of crates that fell over because they are placed on a slightly sloping terrain, and then a sphere barreled down from a high hill and smashed into them.

Signing Off

I have deliberately ignored my emails today so I could focus on getting physics to where I wanted them to be.  As good as this excuse is, the real reason is that playing with physics is SO absorbing, I spent a good deal of my hours re-running the prototype and following barrels down steep hills and watching them make subtle course corrections each time they hit a bump. 

I would not be surprised if many users will be making obstacle courses rather than games with Reloaded when we release this puppy :)  I have a few more days on physics, and then straight into A.I, so Wednesday will be about getting the 'enemy character' physics object in place and doing all the things that entity needs to do (i.e. push crates out the way, walk on the floor, e.t.c). I will also add jump so I can walk around on those screen shot crates and see what happens. Will I fall through them, will they fly away, will I jiggle up and down, tune in tomorrow!


  1. HAHAHA LOVE that screenshot!! :D I'm sure Rick wouldn't mind if you kept doing that, and I'm sure we all will be very happy to see some more disclaimer-smeared WIP pics.

  2. I burst out laughing when i read "not-real!"

    I wonder what would happen if a character was shot at the top of a hill... and died in rag-doll... lol

  3. Nice to see the Physics going in and the image shows that immediately as in Yeah we have progress today which viewers do like to see of course. It kind of helps make their day better. A bit of good news in tough times for many. It somes up quickly what the post is about and where Reloaded is at.

    Once the Physics is done to your satisfaction I am sure everyone will be very happy and will look forward to seeing AI progressing - Player and Enemy on the terrain - a milestone perhaps that should generate more of those "Viewers" needed.

    Downhill from there on in excuse the Pun. Well not quite as I am quite sure you will still have a lot of hard work to do which we appreciate, but you get the point.

    I am looking very much forward to seeing the AI and some of this functioning in the Editor at some future date.

    Thanks again for the very obvious hard work going into Reloaded.


  4. Hi Lee

    You do need to release the occasional development pic to hold peoples interest. Personally I read the blogs everyday regardless of whats on there but the screenshots really add a lot of value there.

    I agree that it would not be a good idea to fudge the physics and it would be better to take some time over this and implement it in a robust way.

    Looking good so far.

  5. Hello, That's very impressing, however, how segments interacts with terrain? I mean when fpsc:r will be ready i will make a test game and of course a lot of buildings.. but can we go "under" the terrain layer inside the segments? Imagine a 3 floor bunker where only one floor is outside the terrain and other two are under the surface.. will be possible and easy to do? And rolling barrels wil stop when hit a building? Will be i able to build a complex that is half outside and half inside an almost vertical terrain wall?
    Thank you!

  6. I think many of your followers come from the game development field and are used to seeing progression of a project via a series of incrementally better screenshots. I am less interested in the final product, and more interested in the steps you took to get there, as that helps me more in overcoming similar obstacles in my own projects. If you create an under-ground blog, I would prefer that over a blog that just shows "pretty artwork".

    Although from a marketing standpoint, I know where Rick is coming from. The way I see it, you have two types of followers: Those that will use FPSC:R out of the box and more concerned with features and appearances, and those that will load it up in DBPro and start tweaking the source code to make their project stand out more. They will need to know how you built it so that they can modify it. You really do need two blogs, each tailored to one of those groupings.

    Also, my RSS feeder stopped showing your blog posts, and then gave me 8+ all in one day. If that problem is more widespread, that might account for the loss of viewers.

    1. I don't think it's quite as black and white as that. You've got people who are generally interested, perhaps wanting to use the knowledge to help them in their understanding of the engine, people who enjoy seeing things grow, people who want to see what the engine is capable of and try to influence it's design early on to be helpful.

      The list goes on.

  7. In response to dustin. I don't think it will be open source as it is new and will be wanting quality control. I for one am glad it will be none open source as it gets messy and some poor mods and things. But if you buy reloaded why would you want to put it in dbpro. I guess you have your reasons. I believe he will update the pro with all the commands etc with bullet and things. Although I may be wrong as I don't run tgc or own it. But I do get like you interested in how it all works

  8. Reloaded can't be all things to all people though from what I understand it is to be together with AGK the leading TGC product range at least for quite some time.

    I have my doubts that the majority of Reloaded purchasers/users for whom the product is largely aimed will want to touch DB pro and go there once Reloaded is under way and to be further developed. I very much doubt also that at this stage in DB pro development as it stands at the moment anyway you will be able to make games with it as good as those Reloaded users will be making and deploying in a RAD environment.

    There would be few wanting to tinker with it in DB pro me thinks. For the majority of indies there will just be no need. For a few I guess they will just enjoy that but if the aim to get a good game completed and released either for free or commercially Reloaded will be much more likely to get you there.

    Due to the way Reloaded is being developed from the ground up as you can see I doubt DB pro will catch up with it and there's by and large no need for it if FPS games is your bag.

    Reloaded is designed for indies and promises to give them largely what they have needed for a long time. Hopefully if successful and time will tell, reloaded will get better over time being concentrated on by TGC to become the mainstay FPS product as they have indicated as being together with AGK as said former the main product range to which endeavour will be given.

    Sorry I know little of DB pro and may quite well be wrong. I don't know how easily Reloaded can be integrated to DB pro but hopefully no source will be given out to that or any other end unless of course there is a need to from a commercial standpoint for TGC and product survival.

    If Reloaded becomes successful due to technical development and quality control aspects of the work now going into it and it becomes a commercial success for TGC then changing that by releasing the source is playing with fire and better to continue internally with developing Reloaded and ensure any success is enhanced and built upon with the same continued level of commitment and control to that end.

    Hopefully if successful then no risks will be taken but steady as you go advancement for the engine for some time into the future.

    If you want more then perhaps when TGC decide to make another version of Reloaded and for that it has to be successful for some time to come I guess.

    So far it looks to me that TGC have it right and hope that continues as is.

    You certainly don't need to be looking at concentrating on two Blogs, two products and so on. Thats a hopeless approach to Reloaded development in my opinion for what its worth which is not a lot. Lee certainly has more enough on his plate at the moment getting Reloaded right and in anything like a timely manner.

    Sooner or later the engine will have to find a place in the market and I do hope it will stand the test of time and encourage a great deal of users to take it up and ensure its success for the benefit of TGC and all concerned. Not forgetting of course Indie developers in general and current users who I am sure look forward to Reloaded as a way forward for their needs which has not really ever been available to them ever before.

    Hopefully that will ensure its success but its a funny old world and who knows.

  9. As to the number of views, if you look at the recent blogs posts below you can see clearly :

    That those with just text have few comments and or feedback.

    Those that have an image by and large have more.

    Those have that show video have the more.

    Those that have the most images and videos have the most comments and generate more interest.

    Particularly when a major milestone is reached such as Physics is in, Boxes and barrels are in and so on which is to be expected.

    The most development progress shown and interest generated by images and videos the more feedback and comments as users have more to comment about. That's to be expected.

    When you get AI in then you should get more interest and feedback/comments again especially if its visually shown. Obviously Imaging is more interesting to larger wider audience than text alone no matter how interesting as text alone cant quite match the dynamic interaction with the human senses in the same way as imaging can.

    Major engine feature milestones being reached clearly will bring the most attention from viewers.

    Once the editor become predominant in the Blog things should pick up in the commenting dramatically as Reloaded reality gets nearer.

    Thats just normal as people have most interest in something that is going to happen getting nearer and the excitement and tension mounts.


  10. Actually, Reloaded is NOT being developed from the ground up and it IS being developed in DBPro. Lee has taken the source code from FPSCx9 is re-writing most parts, one at a time.

    @Peter Coleman:

    I agree completely with everything you've said so far, as usual :)