Time for a quick little GTA:CW review. What I really liked about the game was that for the first time in a long while it felt like a full blown original game on the system. Most other games either boil down to simple puzzle mini-game stuff or cheap rip-off of titles from the big consoles. GTA:CW on the other side, while not 100% original, knows that its not running on a big console and doesn't try to emulate GTAIV. Instead it goes back to a more cartoony style of the original GTA PSone titles and for most part success quite well with it.
Car chases are over the top ridiculous with a new mechanic that lets you crash policecars into buildings and objects to get rid of them, giving it a nice Micromachines/Matchbox feel. It never feels realistic, but its always fun. Fighting with weapons is just as over the top, with flamethrowers, miniguns, molotov cocktails and chainsaws being at your disposal. The AI never looks good in those fights, but it doesn't have to, because the game does a good job of just throwing enough of those dumb enemies at you. Mission design provides quite a bit of variety and a drug trading system gives you something to do between the missions.
What I however probally like the best in the game is how it uses the lower screen. There are a few useless mini-games on there, but they are used rarely, most minigames are ok and some minigames are actually kind of cool. I really like how you can pay the toll by throwing coins while driving, its a really nice way to give a game more interactivity without interrupting the core gameplay. Most of the time however the lower screen isn't used for minigames, but as a fully functioning PDA/GPS system. You can read mails on the thing, define a routes you want to drive, look for people to trade drugs with, order weapons in a online shop and all that stuff. You also can switch weapons on it and throw grenades. It is great to finally see a game that uses the screen for extending the core game, instead of trying to replacing it. Thinking about it, the whole use of the lower screen is actually quite close to what I experimented with in my WindstilleDS prototype thing, so no supprise that I really like it.
But with all things pretty, there are of course a few ugly parts too. The most ugly being the story. While I like the way the story is told technically, via decent looking static-image cutscenes, the dialog is just awful. I don't mind profanity in a game, but here they are just trying to hard, every sentence is filled with it and you don't really meet a single friendly character in the game (well, there is tha girl, but that gets killed like 15 minutes into the game...). Which of course kind of makes you question why the hell you work for those guys in the first place. On top of that the plot just isn't interesting either. So it boils basically down to doing random jobs for random people for random reasons, to bad that they didn't come up with something more interesting.
Another thing that feels badly broken in the game is the virtual economy. Drug trading can make you thousands of dollar in a single deal, while solving missions hardly pays anything. When you can get a hundred times the money for simply driving from A to B, instead of solving a complex mission, it just doesn't feel right. On the positive side of things, I however had quite a bit of fun with it, in the beginning I once ran out of money, leaving nothing for drug trading and me stranded driving a taxi around to get some money in the pocket again, followed by me buying all the safe houses I could find, since drug traiding really pays once you have enough money to invest.
The last three issues aren't GTA:CW specific, but apply to all GTA titles and its annoying that they are still not fixed. The first one is the lack of reset points in the missions. You can easily restart missions and skip the cutscenes by just pressing a button, but you don't have any reset points in the mission itself, which gets really annoying when you die close to the end of it. The other big issue I have with the series is a direct consequence of former one. As you don't have any reset points, all the missions are all extremely simple and short. Its annoying to have such a huge world, but then only missions that can be solved in 5 minutes and are completly local in scope. It would be much more fun to have a proper quest system and some persistantce in the game instead of just missions and no persisntance. The final issue I have is simply that the game doesn't give you anything to do once you are done with the main quest. You can still drive around, find some hidden stunt jumps and trade drugs, but there really isn't much of a reason to do so. Once the story is over you won't receive any new mail for new missions and the whole world just runs kind of dry. It would be nice if the game would auto-generate new missions or something to keep things interesting, but it doesn't.
Overall, is not a perfect game, the story is just to bad for that, but its a really fun game for the 12 hours it lasts and for the first time since Mario64DS it felt like a full game making it to the DS, it makes great use of the hardware without feeling gimmicky or ugly.
Thursday, April 30, 2009
Wednesday, April 29, 2009
Interlude
Currently playing a bit around with the Twitter clone identi.ca, not sure if I'll stick with it or not, but so far it looks quite usable, especially the ability to post and read messages via Jabber/XMPP is pretty cool.
Also did a redesign of the SuperTux webpage, its now a good bit less ugly then before.
Also did a redesign of the SuperTux webpage, its now a good bit less ugly then before.
Saturday, April 25, 2009
Quick Windstille Update


Friday, April 17, 2009
Braid Review (PC)
Just finished the PC version of Braid, so here a quick little review. The 2D graphics are pretty cool, Braid does some great things with particles, color and light that give it a very dynamic quality, haven't seen anything quite like it. Music is pretty good too. The story is far out in the weird artsy territory, every now and then it seemed to make a little sense, but then it lost me again. The story doesn't really have any impact on the gameplay, so its no big deal, but a little more coherence would have been nice. The way it is it leaves a bad "Now what was that about?" taste in the mouth.
In terms of gameplay Braid is obviously quite innovative, it does quite a few things with its time travel mechanic that one hasn't seen before and for most part its successes and is great fun. However not all puzzles are fun to solve, quite a few result in a huge number of trial an error attemps as the game mechnics aren't always obvious. Another issue is that the time rewind mechanic isn't usable for 'undo' in some puzzles, which adds another layer of trial an error. The game is also pretty much puzzle-only, it doesn't waste any time with game'y levels in which you use already familiar mechanics, its pretty much a run from one puzzle to the next, which gives the game a bit of a stop&go feel, as no real gameplay flow will settle in. I would have prefered a bit more classical jump'n run action, which thanks to the endless rewind is great when when it actually happens. Speaking about time rewind, I would have prefered it if the game would allow dynamic rewind via the analog triggers, the way it is implemented you have to hold X for rewind and can toggle with the triggers between 1x, 2x, 4x, 8x speed, its very easy to overshoot the target that way. I missed a way to fast forward the game (outside of timerewind) as well, as there are quite a few puzzles where you have to stand on the spot and wait for an enemy, fast forward would have been welcome there. Overall however a pretty great little puzzle game with some pretty cool crazy ideas.
In terms of gameplay Braid is obviously quite innovative, it does quite a few things with its time travel mechanic that one hasn't seen before and for most part its successes and is great fun. However not all puzzles are fun to solve, quite a few result in a huge number of trial an error attemps as the game mechnics aren't always obvious. Another issue is that the time rewind mechanic isn't usable for 'undo' in some puzzles, which adds another layer of trial an error. The game is also pretty much puzzle-only, it doesn't waste any time with game'y levels in which you use already familiar mechanics, its pretty much a run from one puzzle to the next, which gives the game a bit of a stop&go feel, as no real gameplay flow will settle in. I would have prefered a bit more classical jump'n run action, which thanks to the endless rewind is great when when it actually happens. Speaking about time rewind, I would have prefered it if the game would allow dynamic rewind via the analog triggers, the way it is implemented you have to hold X for rewind and can toggle with the triggers between 1x, 2x, 4x, 8x speed, its very easy to overshoot the target that way. I missed a way to fast forward the game (outside of timerewind) as well, as there are quite a few puzzles where you have to stand on the spot and wait for an enemy, fast forward would have been welcome there. Overall however a pretty great little puzzle game with some pretty cool crazy ideas.
Monday, April 13, 2009
Windstille Development Update
As previously mentioned, focus is now moving a bit away from the editor and back to the content side of things, as I have to get more of an idea what I want before I can figure out how to implement it. Anyway, there has still been a bit of editor progress, the navgraph tools got a little more improvments, a little compatibilty fix for older Gtkmm was added and some classes where split into seperate files. Most importantly however the layer tree got rolled back to a flat layer list, as the tree was just making things to complicated and error prone without actually providing any real advantage. This breaks the file format a bit, but the sectors in SVN have already been properly converted. Another new thing today is improved drag&drop, you can now drag stuff directorly from the ObjectSelector into Gimp, reload in the ObjectSelector is hooked up as well, meaning you no longer need to leave the editor to update graphics or mess around with a file select dialog.
I also replaced the COPYING file, it no longer contains the full GPLv3 license text, but instead an explanation which files fall under which license along with links to the licenses, namely GPLv3 and CC-by-sa.
I also replaced the COPYING file, it no longer contains the full GPLv3 license text, but instead an explanation which files fall under which license along with links to the licenses, namely GPLv3 and CC-by-sa.
Grub2 on an old Asus A7V
I happen to own an old computer with a Asus A7V mainboard (state of the art back in 2000), which happens to have a Promise PDC20265 controller on it. Said controller provides a few additional IDE ports for additional harddrives, on one of them hangs my Linux drive. So far so good, that setup works quite fine in Linux, the trouble however is booting it with Grub2.
First problem is that the Bios has a different idea about the drive ordering then Linux has, so editing /boot/grub/device.map is needed to swap the two drives around.
Second problem is a little more tricky, since not only is the ordering wrong, the second drive is completly missing in the first stage of Grub2, as can be seen by typing "ls", so neither the second stage nor any files can be loaded from it. So booting will fail.
Solution: Mount the Windows drive, the one tha Grub2 can see (sdb), to /mnt, then do a:
grub-install /dev/sdb --root-directory=/mnt
This will give you a /boot/grub/ directory on yoru Windows drive and a working Grub2, but without the menu. To fix this up I simply created a minimal menuentry that boots into the real Grub2 menu on the second drive, which Grub2 luckly can see without problem once Grub2 is fully booted.
$ cat /mnt/boot/grub/grub.cfg
set default=0
set timeout=0
menuentry "Real Grub" {
configfile (hd1,1)/boot/grub/grub.cfg
}
Some notes:
First problem is that the Bios has a different idea about the drive ordering then Linux has, so editing /boot/grub/device.map is needed to swap the two drives around.
Second problem is a little more tricky, since not only is the ordering wrong, the second drive is completly missing in the first stage of Grub2, as can be seen by typing "ls", so neither the second stage nor any files can be loaded from it. So booting will fail.
Solution: Mount the Windows drive, the one tha Grub2 can see (sdb), to /mnt, then do a:
grub-install /dev/sdb --root-directory=/mnt
This will give you a /boot/grub/ directory on yoru Windows drive and a working Grub2, but without the menu. To fix this up I simply created a minimal menuentry that boots into the real Grub2 menu on the second drive, which Grub2 luckly can see without problem once Grub2 is fully booted.
$ cat /mnt/boot/grub/grub.cfg
set default=0
set timeout=0
menuentry "Real Grub" {
configfile (hd1,1)/boot/grub/grub.cfg
}
Some notes:
- With Grub1 I had similar trouble, but I seem to have somehow managed to work around that, not sure how I did that, as there ain't a Grub1 install on sdb.
- grub-install creates a /boot/grub/device.map with hd0 and hd1 in it, that can't be right, since Grub2 starts numbering with 1, not 0. (Edit: Only partitions start at 1, disks still start at 0)
- I remember that there once was a grub-install --config-file option or something similar to make the grub.cfg hack unneeded, but that seems gone.
- sudo wodim /usr/lib/grub-rescue/grub-rescue-cdrom.iso, as you don't want to break your booting without a rescue disc
Wednesday, April 08, 2009
Windstille Editor - Day 22

New thing today are two new tools for navigation graph editing, along with load and save of the graph to the sector file. Probably time to let the game engine catch up to the editor and then figure out how to implement all the other stufff that will be needed (animation support, particle system, characters, dialog, etc.).
Sunday, April 05, 2009
Windstille Editor - Day 19

All that thinking about the future of Windstille asside, I coded a texture packer today, so that images can share the same textures. This fixes some issues with blending artifacts in non-power-of-two textures and also allows to waste a little less space on larger textures.
Saturday, April 04, 2009
Windstille Editor - Day 18
Fixed a bunch of issue with the rotate and scale ControlPoints today, they are now properly placed on rotate or scaled quads. They also now work properly with scaled or rotated quads. Another new addition is pixel perfect selection of objects. Other then that there has been some more work on moving and improving graphics.
Friday, April 03, 2009
Windstille Editor - Day 17

Faking directional shadows via a simple black blob, scaled and rotated, works quite nicely (sink in the toilet room). Still missing a proper object property dialog that would allow adjust alpha of a decal properly. The mirror is actually the same object as the house in the city sector, just scaled down, while obviously not perfect, it nicely shows how reusable objects are even for completly different purposes.
Thursday, April 02, 2009
Windstille Editor - Day 17

A new feature is uv-coordinate based object flipping, which is nice and simple, as it doesn't change the object shape. It however is a little redundant, as the same could be done with rotation and scaling. One big issue right now is that its to easy to end up with holes in the map, either due to float impression or duo to scaled object that don't fall on clean integer coordinates, the hope is that flipping might help to limit that.
Wednesday, April 01, 2009
Windstille Editor - Day 16


On the level side of things I tried to replicate a few more of the previous artworks. Still just night/dark stuff with plenty of light added, have to try a daylight setting sooner or later. Color is also still mostly missing.
Subscribe to:
Posts (Atom)