| View previous topic :: View next topic |
| Author |
Message |
larsivi
Site Admin
Joined: 27 Mar 2004
Posts: 453
Location: Trondheim, Norway
|
 Posted: Thu Jun 17, 2004 10:03 am Post subject: |
 |
|
| eye wrote: | | The capability is not purely theoretical, it is quite well reached by the landscape demo from DevLib, which uses one huge static VBO. Even despite of multiple overdraw, full screen rendering. |
My point exactly (unless I misunderstand you completely). This setup is nice for showing a landscape, but to get a game out of it you need dynamic data and dynamic data need CPU processing power some way or another.
| eye wrote: | | My card is the lowest performance of the curently produced cards, GeForce FX-5200. |
And my card went out of production a couple of years ago  It supports most of the new extensions through the driver (which runs on the CPU); making use of them slows down the application severly even though it looks like they might be supported in hardware.
| eye wrote: | | And while VBOs simply don't work on older hardware, or don't make much sense, there simply are fall-back paths to utilise more standard means. |
Needing the programmer to double up on code, or the driver must do the work (on the CPU).
| eye wrote: | | Only hierarchies have to be processed on the host. Even basic T&L hardware supports 2-matrix skinning. So CPU is really, really not a bottleneck. The GFX cards get faster at a much higher rate than CPUs, and even now are multiple times as fast at vertex processing. |
Yes, but the speed doesn't always help to offload the CPU, although it's getting better at it. (Note that most new features on the GPU helps with good image quality despite low triangle counts.)
| eye wrote: | | And i have yet to find a generic occlusion culling scheme. |
Quite right, that's the reason OGRE gives you the choice between several, depending on the type of scene. In general, portals can help distinguish different types, meaning that you can cull an outdoor and indoor environment differently in the same game. Whether OGRE can do this, I don't know yet.
Still, manual tweaking of the scene and how it's organized is the most important factor to get a good result.
In case you're interested, I have a master degree in computer science, specialized on 3D graphics and algorithms. I wrote my diploma on [disclosed subject] for http://www.falanx.no. |
|
| Back to top |
|
 |
csauls
Joined: 27 Mar 2004
Posts: 278
|
| Posted: Thu Jun 17, 2004 2:39 pm Post subject: |
|
|
Just wanted to say that larsivi is right in that I didn't look over every engine I probably could have. I'd toyed around with a few, including OGRE, CrystalSpace (another damn good one), and a couple others, and just decided that OGRE seemed like a good one for my idea of having a D engine. We could always have done one truly from scratch but... well I know I don't have that kind of time on hand, so I wasn't going to expect it from anyone else. 
_________________
Chris Nicholson-Sauls |
|
| Back to top |
|
|
sean
Joined: 24 Jun 2004
Posts: 609
Location: Bay Area, CA
|
| Posted: Sat Aug 28, 2004 1:08 pm Post subject: |
|
|
| Very slick. I haven't looked at engines in a long time, but OGRE seems incredibly easy to use. Is this project still on track? |
|
| Back to top |
|
|
larsivi
Site Admin
Joined: 27 Mar 2004
Posts: 453
Location: Trondheim, Norway
|
| Posted: Sat Aug 28, 2004 1:17 pm Post subject: |
|
|
| sean wrote: | | Very slick. I haven't looked at engines in a long time, but OGRE seems incredibly easy to use. Is this project still on track? |
Not very, I think. csauls is busy elsewhere and I'm just short on time. The idea is very good though, and some foundations exist. |
|
| Back to top |
|
|
csauls
Joined: 27 Mar 2004
Posts: 278
|
| Posted: Wed Sep 01, 2004 11:07 am Post subject: |
|
|
Its on track in the sense that, "YES I want this to happen!" and not on track in the sense that, as larsivi said, we just haven't been working on it. Feel free to grab a copy of OGRE and the skeleton of Sinbad that's in SVN (I think its half-outdated but won't swear to it).
A number of things have contributed to its being currently stagnant. Lack of time. Waiting for D to hit 1.0 so we have a stable language. (I still remember converting all the singletons to use a mixin... grand improvement, but I dread having to make similar broad changes later when the project has lots of code.) And the fact that us two were the only ones paying it much attention.
Now that Walter claims the language is more-or-less fixed for a while, it might be time to start work on this thing again -- maybe even from scratch now that we've all had time to learn more about the "Tao of D". I still have no time though... and for the time being, no long-term lab to work in, which means none of my projects are getting any attention. Unemployment stinks.
_________________
Chris Nicholson-Sauls |
|
| Back to top |
|
|
larsivi
Site Admin
Joined: 27 Mar 2004
Posts: 453
Location: Trondheim, Norway
|
| Posted: Wed Sep 01, 2004 11:55 am Post subject: |
|
|
| csauls wrote: | Its on track in the sense that, "YES I want this to happen!" and not on track in the sense that, as larsivi said, we just haven't been working on it. Feel free to grab a copy of OGRE and the skeleton of Sinbad that's in SVN (I think its half-outdated but won't swear to it).
|
Actually, I just started looking at it again. I'll slowly fill in blanks (stubs first, implementations later). I've moved a couple of things (and I might move some more), just read the SVN log if you wonder about something.
| csauls wrote: | | Unemployment stinks. |
Isn't that right. I do have a fair home computer and 2.5 Mbits of broadband, though.
PS: The codestyleguide is offline, could you mail it to me, post it here or put it in SVN? |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
Powered by phpBB © 2001, 2005 phpBB Group
|
FAQ
Search
Memberlist
Usergroups
Register
Profile
Log in to check your private messages
Log in
