| Total Entries: 10 |
prospero
Entry: 10
11/28/2002
04:52 AM
| 28th November 2002: v0.7.0 will be released today! Probably it'll be evening GMT, as I'm still waiting to get a proper domain and space set up for the Firestorm Productions website. If that doesn't work out, I'll put it up anyway - I just need to add a few more levels to give you guys something to play through, and perhaps a conversion facility to allow the level editor to load in levels from v0.5.5. |
prospero
Entry: 9
11/24/2002
04:35 PM
| 24th November 2002: v0.7.0 is imminent! It'll be accompanied by the launch of the website (hopefully hosted by remakes.org) and will feature some great retro-style chip-based music from Alexander Bulér. I want to get it released as soon as possible, but at the same time I want to make sure that the features I've added are stable =)
Unfortunately I may have to backtrack my original estimate of 20 completely new levels and go to something like 5 (perhaps 10 if I'm on a roll). Making playable and balanced levels is a little more of an involved procedure than it has been... |
prospero
Entry: 8
11/21/2002
07:40 AM
| 21st November 2002: Aargh. The highscore table is alive, and working fine. The game will recognise if you achieve a highscore and lets you enter you name quite happily. But can I get it to save the new highscores afterwards? Of course not! It knows that I only have this to do before I can get on with the far more fun part of designing levels... grrr!
More info as it comes in...
[EDIT]: Just fixed it, a few minutes after posting this. It seems BB isn't smart enough to directly overwrite existing strings in a file, so updating highscores means recreating the levelset file from scratch :P |
prospero
Entry: 7
11/19/2002
11:26 AM
| 19th November 2002: BlitzOut is dead. Sorry :( But Poing PC as it's now called is very much alive! :D
From v0.7.0 onwards, the game will be known as PoingPC to better reflect its roots. A lot of things are already added / tweaked, almost all of them related to the level editor (in preparation for asking the Blitz community to help with some level design :D ). Hopefully it's more stable and friendly to use now.
Within the next couple of days I hope to add proper support for hiscores and design the first two official levelsets (each of 10 levels). In theory v0.7.0 should be released by the end of the month! |
prospero
Entry: 6
08/19/2002
11:14 AM
| 19th August 2002: Things are going quite nicely :) I've managed to chunkify both the bat and the ball with no obvious headaches, and although the ball does look a tiny bit squarer than before, it's not at all bad. I had planned for a release over the weekend, but I got caught up with other things...
One thing I'm doing is adding an options menu to allow configuration of various game elements - versions up to this one would crash if run on a machine without some form of sound, so I'm hoping to fix that. I'm also putting in a selection for the timing method used in the game (either the old run as fast as possible method or standard timer-managed updating) as well as graphics colour depth. This is all written to a config file with some changes requiring a restart before they take effect.
Also added is Z-Ball style ball curving - hitting the ball while moving the bat upwards spins it clockwise, hitting it down, anticlockwise. This can be turned off in the options menu, but there will still be bricks that can alter the spin of the ball... ;)
There's a few more minor things but that covers the exciting stuff :) The new version will probably be out tomorrow afternoon ish. Maybe. :p |
prospero
Entry: 5
08/14/2002
06:17 AM
| 14th August 2002: After taking a short break to get some feedback on the new beta release, I'll start work on fixing / adding / improving :)
One thing that seems to need tweaking is the size of the ball and the bat. The bat can be made chunkier quite easily, but the ball could be a bit more awkward. To all effects and purposes in the current collision detection system, the ball is square - 5x5 pixels :) It'll be interesting to see if I can increase it to something like 7x7 or bigger and still maintain it's round appearance while also keeping the collision working as it should... true circular collisions are something I really don't want to have to tackle just yet (that can wait for BlitzOut II! :p )...
There's also a rather horrible bug which can trap the ball if you lose it on any level past 1. Again, this should be relatively simple to fix, most likely by adding spin to the ball so it has a far greater chance of bouncing out of wherever it's trapped.
Expect v0.5.5 sometime within the next couple of days! |
prospero
Entry: 4
07/31/2002
09:49 AM
| 31st July 2002: Progress on the level editor is going nicely. Right now it has only basic functionality and is somewhat cumbersome to use, but does allow for saving and loading and the creation of levels without mucking around with ASCII files.
I've also added in the mutli-hit brick functionality which allows bricks to have up to 8 hits before being destroyed (each hit with its own brick image of course).
Thanks to Jepp Nielsen for his BBGui library which features heavily in the editor! Expect a beta release within the next fortnight or so, if all things go well :) |
prospero
Entry: 3
07/14/2002
05:59 PM
| 14th July 2002: Made a fair bit of progress on the level editor today. Right now I have a brick palette working OK with left and right mouse button 'pen' selection active. If I can carry on making good progress I should be able to release a working version of the game in no more than a couple of weeks time! |
prospero
Entry: 2
07/13/2002
02:19 PM
| 13th July 2002: Ah HA! At long last I finally figured out what was wrong with the bounce logic. To be accurate it wasn't the logic that was causing the problem... it was when that logic was being executed!
The problem stemmed from my method of collision detection. Every time the ball is moved the collision part will see if there is a brick one pixel right, left, above or below, and if there is it updates the ball's collision status appropriately. The flaw in this was that the ball would detect a collision even when there wasn't one about to happen - imagine the ball is passing the top-left corner of a brick within one pixel, but its path doesn't actually make it hit the brick. The logic didn't realise this and was bouncing the ball incorrectly :) The picture below may sort've help to illustrate...
This was fixed easily enough by making the ball bounce only when its angle leads it to hit the face of the block - so for the example above, hitting the block would require an angle of 89 or less or 271 or higher (bearing in mind that angle 0 = heading right). Because the ball shown is initially approaching the brick at an angle of 135, it would now correctly miss the block and carry on its jolly way.
Now the game physics are (hopefully) as they should be, I can work on the level editor and actually start making this thing resemble a game! |
prospero
Entry: 1
07/12/2002
06:40 AM
| 12th July 2002: I've been experimenting with the new angular movement system and have added a couple of vaguely useful debug instructions that use the cursor keys to directly control the ball.
I'm having a hard time in pinpointing exactly what is causing these weird bounces that have plagued the game so far. Comparing my code to that of Quadamage's 360 degree demo, the only difference is the way in which I'm handling bricks. Right now I have a fixed size array for the entire level based upon Krylar's tilemap-handling code example. If a brick doesn't exist then it's image-to-display index is -1, otherwise it is 0 or above. This then links to a separate array of brick images.
For this type of game this method is a little inefficient, and I think it'd be best to overhaul the way in which I load and handle the level files so that I can do a B.BrickType = New BrickType and Delete B system. The brick data type would hold it's image index, X and Y co-ords and other attributes. This should also bring me exactly into line with QD's system so I can continue hunting down this irritating bug and release a truly 100% fixed alpha 5. :) |
