Thursday, September 4, 2008

A quick update, just to break the silence

I've posted a new Windows build, and am currently trying to get the client to work fully with the python-ogre deb on Ubuntu. As expected, getting python-ogre to work on Mac is quite a chore, I'm almost there, got everything compiled but running the demos gives me a "wrapper" error, and I'll probably have to understand how the code generation in python-ogre works to figure it out.

Packaging issues aside, I've made a few updates to the client. The lines in the screen represent "Wormholes", which are basically routes between star systems. Currently, only the Risk ruleset uses this feature, so lines won't display anywhere else, such as on the server which usually runs Minisec.

I've also tweaked the "Information" window to show more useful information such as the planetary resources (I tested it with the "Armies" resource in Risk). In general, I'm trying to locate more crashes by testing the client with different systems and python-ogre builds. OgreAL has been a little problematic, and it's one of the things which I am trying to get working with the linux python-ogre deb file.

My main concern about the client is the performance on older hardware. Mithro informed me that even the login screen runs at a crawl on his laptop, which is probably due to the large textures used in the skybox (1024x1024). I guess a solution to this would be to offer a lower-resolution skybox thats selectable from the options menu.

I've started my industrial attachment, so I'm pretty much limited to weekends. Nevertheless, I'll be continuing to update the client, and when tp04 support has been added content management will be a lot more elegant, allowing "media" tags which mean no more hardcoded graphics. Another thing I'm looking forward to is more battlexml support in the servers. On my end, I also hope to include some shaders to give game objects a more realistic look, but there are still many things to work out before I have time for that.

P.S. If you zoom in on the pictures it kinda looks like those old VGA games.

From Dev Hobby

Sunday, August 17, 2008

Configuration options

This is the last week of GSOC, the date in which our evaluations begin is tomorrow. It's been a pretty short and exciting trip, sometimes I'm still amazed that I have the chance to work on this at all. On to the additions!

A checkbox for "saving details" was added to the login screen. When checked, the username, password, last server connected to and some other preferences are saved in the user's home directory, so that reinstalling the client will not affect the preferences. The password text is masked now, and saved to disk in hexadecimal format. Pretty primitive, but the idea is just to prevent casual lookers from getting the password.

The configuration dialog was expanded with a couple more options as well as two additional dialog boxes for graphics and sound configuration. The main dialog now has the option to tweak the zooming speed, as well as to adjust the distance scaling. This is meant to help deal with the variety of rulesets which use different units of distance for their maps. The distance unit ratio makes stars further apart the lower it goes, as all distances will be divided by this value when starting a new map.

Graphics options allows the player to change the renderer (although I recommend Direct3D for Windows), resolution and anti-aliasing options. Unfortunately, only the full-screen option works instantly, the rest take place after a restart as only full-screen had a convenient setFullScreen() method which I could use :). Also, each renderer (opengl or direct3d) takes in different values as their configuration. For instance, direct3d takes in resolution and colour depth as a single string, while opengl takes in resolution and colour as different parameters, which means that the user should change the either the renderer alone or the other options without changing the renderer.

For sound, the options are pretty basic (I know it is missing a volume control :)). Setting the driver to other values provided by openal (other than the default 'Generic Software' option) doesn't work for my system, perhaps because I am using on-board sound and not an actual sound card.

Some work was done to allow the client to adapt better to different rulesets. The map centers properly now (I had a dumb bug where I should be reading x and y values, but instead read both values from x). Zoom distance, which is like the 'step size' for zooming, takes into account the distance scale so that larger maps will not take forever to zoom in. Panning takes into account the current zoom level, so it zooms more when zoomed out. Also, a map boundary was added so that the user will not move out of bounds and get lost, although I'm wondering whether this is a good idea as it sometimes results in apparent 'locking' when the camera runs up against the invisible walls. Perhaps the camera could slide along the walls instead, so that the user can still move about smoothly.

One issue with having an adjustable distance scale is that for large distances, the background of stars (which was a particle system) would have to be expanded to encompass everything, thus resulting in a sparser starfield. Although I like the idea of generating the starfield at runtime, increasing the number of stars would have made performance even slower than it already is, so I made a skybox in blender instead. This skybox also replaced the one in the login screen, as it is from the default ogre media - something I am trying to move away from.

The following shot shows a camera placed inside a sphere, with it's field of view set to 90 degrees. The starry texture was done by blender's built-in texture generators: some noise, clouds and playing with layer settings. After that, I rendered a screenshot in every direction and set it as a skybox. Easy to do, but getting good settings for the texture is hard, I haven't really figured it out yet.

The next shot shows the starmap with the skybox in use. Sadly, my computer chugs pretty slowly in this server with 34 players. I believe it's the overlay placement code that's the problem as the polygon-heavy models should have been hidden already.

To end on a high note, I received a nice package this weekend, courtesy of the lead dev mithro. Thank you very much! This made my weekend (well that and Singapore winning it's first Olympic medal in decades, woohoo!)

Sunday, August 10, 2008

Interface additions

I adjusted my priorities a little this week, due to the sudden realisation that GSOC is drawing to a close.

Firstly, I added a logo to the main menu.
The logo was taken from the Thousand Parsec website and put into a CEGUI ImageSet, which is the combination of an image file and an xml file which defines the pixel locations of all images within the set. Right now, it is rather empty:

I hope to fill it up with a lot more icons and image elements to add more spice to the interface.

Next, I added a map navigation panel to the lower right of the screen.

The functions of the buttons from top to bottom are zoom in, zoom out, deselect, focus on object and fit map to screen (although the last two buttons are indistinguishable at this size, one points inwards and the other outwards). The icons may be familiar already, as they come from the famous famfamfam silk icon set. I tried creating my own icons at first, but some creations should never see the light of day. With this, there will be a lot less reliance on keyboard shortcuts, hopefully making the client easier to use. The zoom buttons move in larger steps than the mousewheel zoom and are not animated, as I find the zooming animation a little slow on my older computer (it is framerate-dependent).

Lastly, I added a main menu option, allowing the user to return to the login screen gracefully.

I discovered that logging in and out caused a lot of problems with the management of system resources. Models, images, interface elements and such needed to be unloaded and reloaded again, something which I could not find an "easy" solution for. In the end, I traced the loading of all the resources and added a corresponding unload command upon logging out.

Regarding the linux installer, I edited the file so that the client installs in the same manner as the wxWidgets client. With that, the debian maintainers for Thousand Parsec will be able to package the client into a .deb file, although I'm considering whether to include the binary version of Python-Ogre inside as well, so that the package can run standalone.

This week's focus has mostly been on the user interface, and I think any additional features will have to be put on hold for now, so that I can tie up the loose ends. For now, I will be writing the documentation and working on the mac installer, which is a giant time sink :)

Sunday, August 3, 2008

Sounds and Battle

This week I took a look at introducing sound into the client. I found OgreAL a snap to use, it's API is modelled after the built-in Ogre managers such as material and resource manager, so not only was it familiar but also cleanly designed. Unfortunately, technical problems (crashes without any indication of the error - my favourite!) arose when I tried running the code, which I attributed to driver issues, as switching to "generic software" audio drivers - done by passing in a parameter during creation - solved the problem. This parameter is not available for Python-Ogre 1.1 which uses an older OgreAL API.

So all I had left to do were the sound effects.

As I mentioned, I found a program for generating ambient soundtracks last week. It's really a sine wave generator and I'm currently trying it out as a background track. I haven't thought much about what sort of music would be suitable for a game of Thousand Parsec, perhaps I will have the client play any .ogg files it finds in the directory, similar to how Oblivion does it. This way, the user can specify his own soundtrack, although it would be nice to have some original music of our own too.

I spent a while trying to create sounds for other parts of the client. Right now, I've put in a "engine rumble" and clicking sounds for buttons on the gui. Once again, I found a useful utility for these sounds, called sfxr. Most of the sounds produced are pretty lo-fi, but it's suitable for stuff like laser noises and the engine rumble was made from this too (plus some processing in audacity). Hm, there should be some way to embed sounds in blogs too...

I think that the background track does add a lot of atmosphere, but my taste in music is pretty much like that anyway (see Drone Zone on SomaFM) so hopefully it won't annoy others. The sound effects will require a bit of tweaking as well, as their volume varies based on the user's location (automatically handled by OpenAL) and they can sound too loud or too soft at times.

One of the primary motivations for adding sound was to supplement the Battle module, a standalone program for viewing BattleXML files.

Currently, it does not process the actual battle, but it does create and show the initial locations of the fleets. I'm still figuring out how to define weapon locations and attach them to the ships, which should be fun once it is done. Also, there seems to be some texturing issues with the models as well, hopefully I can fix them in Blender.

An additional windows installer script has been created as well, which actually just places the compiled client into your program files directory and create start menu shortcuts. The process of creating the installer actually exposed some bugs - although everything works, on shutdown the client outputs an error log, saying that some background threads had been forcefully shutdown, which made me realise that perhaps I hadn't been closing the client library threads properly, something that I have yet to look into.

Next week, I will be finishing up the battle module and tying up loose ends:
- the starmap could use a navigation bar for zooming and centering
- the title screen should have the TP logo
- using theora to display the intro movie
- a minimap for lost players
- writing up documentation
- packaging linux and mac binaries (big problem here)

Sunday, July 27, 2008

Ship designs and order queues

After chatting with mithro this week, I found that the ship designer in the pywx client was an old part of the code (untouched in a year), and that ship design was not considered a priority (since only MTSec supports it). What I did instead was a viewer for ship designs in the game:

The design list on the left is colour coded by player, and the information pane is currently showing etower333's scout design. Wasn't that guy in a previous post as well? How long has that game been going on anyway?

Order queue management has also been added.

After selecting an order from the queue, you can use the delete button to remove it or the edit button which will bring up the arguments window to edit the existing values. The new button also serves to create a new order after selecting from the available orders list on it's left. This change led me to venture back into the rather untidy order handling code, which I have cleaned up a little.

A minor change is that double-clicking on an object will focus and zoom in on it.

One thing I have been looking into is introducing sound and battle scenes into the client.

Battle scenes in thousand parsec are defined by a BattleXML file, which is a file format detailing how many sides there are in a battle and the makeup of the individual fleets together with a turn-by-turn breakdown of how the battle went. A sample pygame viewer as well as documentation on the format can be gained from the battleviewer git repository, although I have not run the viewer yet. One problem with BattleXML is that currently it is only supported by tpserver-py and not tpserver-cpp, which is the more popular server. Another issue is that it has to be implemented by the ruleset specifically and requires a media definition file to map 3d models to ingame objects as well as any weapon effects. Perhaps there could be a random battle generator so that support can be easily added for all rulesets?

Sound support is already available in Python-Ogre via OpenAL, a cross-platform api for 3d sound. So it's really a matter of finding the right sound effects. I've found a program called song04 which generates ambient soundscapes, it sounds quite alright and I'll be trying it out as a background soundtrack next week.

In somewhat related news, my old computer is probably on it's last legs as it frequently hangs upon startup (or the kernel starts to panic), and it has no internet access as the network adapter refuses to connect despite using the same settings which worked before. I guess it's rather fortunate that I got a new PC recently.

Sunday, July 20, 2008

This week in TP

This week, I finally figured out how to do proper camera rotation in 3d. The actual code turned out to be simple. At first, I wanted to use spherical coordinates to position the camera, as I had success with using polar coordinates for the radial menus. By storing the vertical and horizontal angles and treating radius as the zoom distance, I could convert them into 3d coordinates by applying some formulas. This worked only for rotations along the Z-axis (meaning that the world was "rolling", done by moving the mouse left and right). However, for rotations along the X-axis (pitch) it would refuse to rotate properly - upon reaching the horizon angle it would "bounce" back again.

In the end, I put the camera inside an OGRE scene node, and made it the child of another scene node which would act as the focus. Hopefully this tilted diagram isn't too confusing:

In the diagram, you can see the camera floating at a distance above the map, which contains several planets. The focus node is always level with the other map objects, while the camera node can zoom back and forth. As I mentioned, the camera node is a child of the focus node and by rotating the focus node directly, OGRE will apply any transformations to it's child nodes, causing the camera to rotate around the focus node.

Allowing the camera to look around freely posed a problem, mainly with the starry background as it was a flat field of stars. Here's a picture breaking the illusion:

Thankfully this was actually quite easy to fix. OGRE uses particle emitters, which spew out particles in various arrangements, and there was a Hollow Ellipsoid emitter which I could use to create a starfield surrounding the camera. The manual is not very clear on how to use the Hollow Ellipsoid, the inner_ parameters are actually numbers with a range from 0.0 to 1.0 specifying the percentage of inner volume which should be hollow.

There always seems to be a cluster of stars at roughly around the same location, I'm not sure why. Letting the camera roam also let a few more issues crop up, such as the planet rotation. Previously, they were facing the player who was looking from overhead, since otherwise all the player could see would be their polar caps so I had to rotate them back again. Another stickier problem is the panning of the camera. Since the camera can pan in any direction, it screws up the zoom if the player pans downwards or upwards.

Finally, the messages window layout has been changed to include a panel showing all the message headers. As I said in the previous post, the layout looks like it could do with some tweaking.
Next week, I will implement the ship design window, although I've not been able to create ship designs using the existing client yet, I should take a look at at least letting the player view the ship designs, or else he won't know how badass his fleets are. :)

Another thing will be the order queue management, which refers to deleting and editing existing orders.

There will be a turn update window, showing new events received upon a new turn.

Finally, I will create setup packages for Windows, Linux (with the help of fellow Thousand Parsec developers) and Mac (which I will be working with my mentor for - this may take longer due to issues with Python Ogre). Although not a real release yet, getting the build process done will be a load off my mind. No use working on a project if nobody can run it after all. :)

Interface comparison

This post is mostly a screenshot comparison of the two different interfaces of the original wxWidgets client (tpclient-pywx) and the OGRE client (tpclient-pyogre). Okay, maybe they are not so different... but it would be interesting to see the layouts in different environments anyway.

First off, we have the bread and butter of any 4x strategy game... the map. The pywx client shows each system and which objects are in that system, indicated by the circle surrounding it.

The pyogre client has two views: an "iconised" view when the player is zoomed out, and a 3d view when viewing objects close up.

The pywx client keeps new order parameters and the order queue within the same area:

In the pyogre client, new orders are chosen from a radial menu, which produces the new order window.
There is also a separate order queue window for manipulating the orders of the selected object (not yet, sorry :()

The pywx system viewer shows the tree of all objects in the game.

The pyogre system viewer needs some work, once the client fully makes the move to CEGUI 0.6 I'll be able to include the new tree widget in.

Information window, showing some attributes of the currently selected object. I've always thought it looked a bit like debugging output. :)

The information window for pyogre shows similar details.

The message window for pywx, which supports html and can open your browser if a message contains hyperlinks.

The message window for pyogre. It looks a bit 'chunky', maybe I should have an option to toggle the message list panel?

The design window. I have been working on this feature, but it is not complete yet as I can't seem to create new designs in MTSec, even with the pywx client.

Sunday, July 13, 2008

Animated camera

This week, added some animation to the camera movement to make it more smooth. When zooming in or out, it will move gradually instead of instant 'steps' like before. In addition, there will be animated panning when a system is double-clicked on the system list, or when pressing 'c' to center on the selected object, or when pressing "goto" in the message window. Since it will move in steps every frame, it is somewhat slow when the frame rate is low. This could be solved by having calculating the step size based on the frame rate. A harder problem is how to rotate a camera in a spherical manner around an object, which is probably a staple of most 3d games.

I also added some movement for the ships which will occur after a turn update. This allows you to roughly see where they are going, particularly for enemy fleets.

I guess this week's changes are mostly animation based, so it's a little hard to take screenshots.

Some additional minor fixes
Message window parses html messages now. This was largely thanks to Tim Ansell's suggestion to look at the htmllib and formatter library that comes with python, which saved me from having to strip out the tags manually. I still can't believe the breadth of python's built-in libraries sometimes. Batteries included indeed. One possible addition is to add additional formatting according to the html tags, right now I'm just stripping them out adding line breaks appropriately:

The keyboard shortcuts for zooming in and out (the plus and minus key next to the backspace) is fixed now. Previously, it was stuck in 'icon view' all the time.

Saturday, July 5, 2008

Some interface changes

This week, I have added some additional elements to the user interface in order to make it more user-friendly.

The first would be the turn counter in the top-left hand corner, which is handy for knowing how long the game has been progressing for someone who is joining in half-way, and for knowing the current 'phase' in rulesets like Reach for The Stars, where there are three phases and certain orders are possible depending on the phase. However, it does not actually show the phase - the user will currently have to calculate it on his own.

Another difference is the End of Turn timer in the bottom bar, showing how many seconds are left until the turn is over. The 'End' button next to it is used to tell the server that you are done with your current turn, although this step must usually be taken by the majority of players before the server actually finishes the turn.

Next, when the user hovers the mouse cursor over any object on the starmap, a popup will display with some information about that object such as their name and owner (if any). In addition for fleets, the composition of the fleet will be shown, while planets will display their resources and star systems will just show how many planets they contain. One thing I'm trying to figure out is how to dynamically resize the popup in accordance to the text contained, to prevent text overrun such as in this case:

etower333's home planet - In the minisec ruleset, the home planet of a player is defined by having "Home Planets" in their list of planetary resources.

Another change made is to tweak the icon view so that it colours the icons according to their owner:

This makes icon view a lot more useful, especially as I'm still working out how to show ownership of planets in 3d view without colouring the entire planet.

A common issue with network applications is having the server down, or providing incorrect credentials, so an error message will be displayed when this situation occurs.

I'll be looking out for more areas where I can put information or error messages, but I probably won't post about it again as it's quite minor in implementation (though important).

One suggestion was to have a way of selecting between objects that are clustered closely together when the view is zoomed out. In the wxwidgets client, clicking on a star system will cycle through all the objects in orbit around that system, so I have attempted to make something similar - when clicking multiple times on the same spot, the client will cycle through objects overlapping that area.

Issues encountered this week
A new version of python-ogre was recently released (1.2 RC2), and contains a huge amount of changes from RC1. After downloading the windows build, I found that it wraps versions of ogre that I had never even heard of! (1.7 in this case - the latest stable version is 1.4.9 and the next slated release is 1.6, codenamed Shoggoth. Talk about bleeding edge. :)) Needless to say, it brought a few interface-breaking changes with it and I was considering whether to port the project over or not. One desirable thing about RC2 is that it supports CEGUI 0.60 which comes with a brand new tree widget - something which I'm sure would be very useful. One course of action might be to compile a custom version of python-ogre with just the new stuff that I need - something quite easy to do on linux as I just have to change some version numbers in the build scripts. On the other hand, that would make it more difficult for people to just download the source and run it when they already have python-ogre installed. I think that for now, the client will continue to target the previous version of python-ogre (1.1 since 1.2 RC1 has disappeared) until I can wrap my head around all the new stuff that's just come in.

Things to come next week
I will be working on animating the camera view to make it smoother (for instance, zooming in and out should be gradual and not abrupt) while making it possible to rotate around an object.

Also, I want to revamp the messages window to implement the great suggestion of having a two-paned system, and to handle messages with html tags in them, although I'm not very sure how I'm going to do that at this point.

Thirdly, I want to squish all of the bugs I've found so far. :)

Saturday, June 28, 2008

Progress bar and preferences

This week, I added a progress bar which will show the status of connecting and downloading objects from the server.

Having a progress bar is crucial as users will not be expected to look at console output to determine what is going on. This was implemented by using Ogre's built-in progress bar and tweaking the code as well as colours to match with the current gui scheme. The progress bar shows up when connecting to a new game, and also when updating objects in-game. The background is blanked out to prevent sending orders during updates, however it looks very empty right now, so I'm thinking of putting up some of the thousand-parsec art as the loading screens.

One bad point about the client is that it runs slowly on systems without a 3d card, although I'm not decided on what minimum requirements I should be targeting. One suggestion made was to make the icon view (the map with abstract icons) dependent on frames per second, meaning that when the performance gets too low, the client should switch to icon view. I tried this out and it gave me some issues, specifically with how to switch the client back into 3d mode. Since switching to icon mode speeds up the rendering time, the client will keep switching back and forth between the two modes making the user feel nauseous. As a compromise, I made a preferences window for users to set the zoom level at which the client will switch to icon view. The window is summoned by clicking the "Config" button on the Login screen.

The thousand-parsec project has kindly provided us with some hosting space, so I will be putting up windows binaries in order to get some feedback. I have been messing around with tools like Inno Setup, but for now it is simply a self-extracting file.

My objective for the coming evaluation is to make the client as playable with the Reach for the Stars ruleset as I can, and at first I wanted to focus on the ship design component, until I learnt that RFTS does not require the user to create their own designs, which leaves MTSec as the sole ruleset with player-designable ships. I feel there is scope for adding some cool stuff here - ship design has always been a favourite of many 4x games.

Currently, I have the following issues to fix:
Select objects which are very close to each other - It is hard to click on objects, especially when they are clustered together, so clicking on a cluster of objects should bring up a menu to choose from.

No good way to indicate which planets are yours, the enemy's, or neutral - Perhaps have a coloured rectangle somewhere around it to serve as a flag.

Having a turn counter, to keep track of the current phase in rulesets like RFTS. Related features to this also include an End-of-Turn counter keeping track of the time before the turn updates.

I will be working on these issues tomorrow, as soon as I compile the latest version.

P.S. (ugly personal confession) - I guess I should have seen it coming, but this week saw a lot of distractions due to my powerful new system and it's ability to play the latest games, not to mention the Euro 08 tournament. Am I allowed to call playing Sins of a Solar Empire as doing research? Really sorry... I will make up for lost time next week somehow.

Saturday, June 21, 2008

Icon view

For this week, I've been trying to implement a view where all the objects appear as icons when the user is zoomed out. The advantages of having an icon view are that the objects tend to get harder to spot when far, as well as harder to click on. Also, it's a good way to increase the frame rate when viewing many systems at once. At first, I tried making icons with letters on them, where an 'F' would represent a fleet, a 'P' would represent a planet and so on. However, the letters turned out to be too hard to read, so I went for pictures instead.

Hopefully what the icons represent is clear enough. Triangles for the fleets and circles for the planets and stars. I haven't coloured the objects according to which player they belong yet, but the colour scheme should be the same, except that I can now colour the planets as well. I'm not sure how to represent different ship types, perhaps I could scale the size of the triangle icon?

Another thing I've been working on is trying to create a starfield for the background. Since the current background code is slow, I tried a few different alternatives to the particle system approach.

The first thing I tried was to create a static image for the background, then make it into a skybox. At first, it looked okay:

However, at the corners of the skybox the stars appeared warped which made it unusable. I'm not really sure how to solve that issue; after some googling I found one method of using blender to map a sphere and then render the result, but I must have screwed up somewhere because the result looked even worse.

In the end, I went back to the same approach but reduced the "layers" of stars and replaced the star image with a pixel instead. This seemed to give me somewhat better performance. I also made the starfield stay at the same distance when the camera zooms in or out which should make it look more natural during movement. I'm still not too happy with the look of it, but I'm out of ideas for this at the moment.

This week hasn't seen too many updates, reason being I just got a new computer and I'm trying to install Mac OSX on it. With a little luck, I can try supporting the Mac versions of the client in the future.

Monday, June 16, 2008

Just playing around

Today, I decided to take a little break from coding and explore some of the other projects in the Thousand Parsec universe.

Quite a few of the GSOC projects (the two AI projects and the server configuration project) focus on allowing TP to be played by a single player. This is very exciting and I'm already envisioning creating a coffee-break ruleset similar to strange adventures in infinite space. :)

After reading through the list, I wanted to try out some of the rulesets, particularly Risk as it has a playable version available already. Unfortunately, due to my noobishness in gcc compilation I was unable to get the git version of tpserver-cpp (where all the rulesets are stored) working. What I tried instead was to copy the risk folder over to my working tpserver 0.5 directory and compile it instead. This seemed to work at first:

The screenshot shows the client connected to a server running the risk ruleset. Unfortunately, I could not seem to get the game working properly, as it did not assign me any planets or ships to start with. This problem also existed in the tpwx client, so I suppose the ruleset requires the latest tpserver in some manner.

Slight disappointment aside, I spent the latter part of the day looking at how I could improve the frame rate of the client somehow. The base fps on the client is very low, and even when the screen is not showing anything (all objects are culled) the framerate is clipped at around 40 fps. I found the bottleneck to be the starry background code, once I commented it out the framerate became about 99 fps (still without looking at anything).

Level of detail, in which objects reduce their polygon count as the camera becomes further, was another technique I wanted to apply. Thankfully, this turned out to be easy as I simply had to run OgreMeshUpdate (a program bundled with Ogre) on the models I wanted. After trying a few options, the framerate went up to about 15-16 fps. After a while I noticed some corruption in the model skin due to polygon reduction.

What I did was to tweak the distance levels so it only reduces when really far. I guess someone who can see at the pixel level might still notice it though.

After that, I rebooted in Windows to try creating a binary using py2exe. Since the script was already written, packaging it was not difficult but what I did notice was a tremendous boost in fps when I ran the client.

Here's a "action-packed" shot of the client connected to the demo tp server. With 6x anti-aliasing and 1280x1024 resolution, it still has better fps than the linux version. Wow, windows is great! Actually, the cause of the difference is probably my linux drivers. Since ATI has abysmal driver support, I'm running the open-source drivers which are slower in 3d performance but at least they work. (ATI drivers give me graphical corruption even when browsing the net or openoffice)

Here are my relevant system specs consisting pretty much of last-gen hardware.

Processor: Can't remember the model, but windows reports it as a 1.3 ghz Athlon.
RAM: 1 gig. I thought it was overkill when I first bought it.
Video card: Radeon 9800 Pro. I'll get Nvidia next time.
Motherboard: Nvidia nForce-2. Irony?