[Unknown Target]

Here's a community project we could all do: Beef up FS2's now-ugly models. I've been playing some more recent space sims as of now (Like IWar 2) and I come back and look at FS2--it's horrible! I think the entire community should work on a restoration project, to double poly counts, make curved surfaces actually smooth, etc, to bring FS2's ships back in tune with the new DX8 code upgrade.

[Rictor]

i'm with you man, though thats not not very helpful i guess  :wink: :wink:

i think the textures need to be done first, like better testures for the existing models. i dont know what it is, but games like eve-online and earth and beyond (iwar2 too) just somehow look better, i guess its the lighting

c'mon guys, make us some better textures/models for some fs2 polulars (herc2, hecate etc )

[Galemp]

Something along the lines of the X-Wing Alliance Upgrade, you mean? With higher-poly models and bigger textures? Certainly a possibility, except that the only ship reference we have is the models themselves. With Star Wars you can get finer and finer detail until you've got a copy of the actual model used in the film, but an Orion is an Orion.

Larger fighter textures for the FS2 ships would be nice, though...

[RandomTiger]

I started a thread about hi res textures in the Art forum and everyone was all doom and gloom. No-one thinks they can come close to the quality of V's artwork. What I did discover is that there are already hi res textures for some models.

http://www.hard-light.net/forums/index.php/topic,10293.0.html

I think hi res textures would make the biggest visual impact of FS2.

But hold on a moment. The DX8 upgrade in itself may not actually provide much or indeed any speed boost. Im not actually changing the architecture of the engine so its still very restricted. However there are loads of things we can do to speed it up and DX8 in there will lead the way for stuff like that to happen so I think this is a very worthwhile project. Perhaps code and media would be ready at the same time.

Why Im I not making it better while Im upgrading to DX8, well one job at a time. Doing two jobs at once to not fully working code is the way to introduce bugs. The type that are really difficult to fix. However all this time in that code will make me more experienced for optimisations. I would like to have a stab at getting it to work in hardware T&L. At the moment the engine flattens the objects before rendering them. Its giving your PCU a lot of work that should be going to the GPU. And I only have a P350 so I would really like to see it go faster!

Im secretly (not any more I guess) hoping that textures might be a bit faster but we also have another problem, I believe that there are various limits set in the game that will get in your way so you will have to get on the coders cases (not mine, Im busy with DX8) and get them to sort them out.

Its actually been done before but crashed on one guys computer so it got dropped. But we are gaining more programmers all the time and if the tester list people are willign to put time in I dont see why it shouldnt all come together nicely.

I say again. Great project. If no-one gets off their ass and does this then all our coding is a waste of time. So go for it!

[Unknown Target]

Meh....erm, anyone going to help?

If you want some incentive, we could make this a campaign as well (Stealth, you could fill up your row!)

[RandomTiger]

I think you should stick the the original idea of making higher quality models and textures for FS2 campaign. Then everyone can make use of them.

[mikhael]

Originally posted by Unkown Target
Meh....erm, anyone going to help?

If you want some incentive, we could make this a campaign as well (Stealth, you could fill up your row!)

I wouldn't mind giving it a shot, but all my efforts are going for 158th stuff right now.

Of course, keep in mind that ships in Iwar2 aren't all that high poly. Mostly they're medium to low poly with good texturing. The player ships are the most high poly, and even they come in around a thousand.  Freespace, I'm almost certain, can at least 750. Except on capships, that's probably overkill, I would think.

[LtNarol]

Originally posted by mikhael
I wouldn't mind giving it a shot, but all my efforts are going for 158th stuff right now.

Heh, you better

Originally posted by mikhael
Of course, keep in mind that ships in Iwar2 aren't all that high poly. Mostly they're medium to low poly with good texturing. The player ships are the most high poly, and even they come in around a thousand.  Freespace, I'm almost certain, can at least 750. Except on capships, that's probably overkill, I would think.

FreeSpace2 does not like exceptionally high poly models, anything with more than 750 polys per subobject is subject to the Shards Of Death as well as any number of other problems, even some with less than 750 per subobject have failed.

V's fighters and bombers weigh in at 400-550 polys, which is a good number because large numbers of high poly ships will affect gameplay. Remodeling could bring out some of the finer details on the ships, but really isn't too practical in the FS2 engine.

Redoing some of the texture maps on the other hand can make some ships look much better, the Myrmidon, Tauret, Erinyes, and several other fighters/bombers look not so hot up close because of low res textures, if someone wants to work on those..

[RandomTiger]

If you work with the coders on this we can up the limits and try to speed up the engine for you.

[Nico]

Originally posted by LtNarol
Heh, you better

FreeSpace2 does not like exceptionally high poly models, anything with more than 750 polys per subobject is subject to the Shards Of Death as well as any number of other problems, even some with less than 750 per subobject have failed.

820 polys actually, never had any pbs with less than that.

V's fighters and bombers weigh in at 400-550 polys, which is a good number because large numbers of high poly ships will affect gameplay. Remodeling could bring out some of the finer details on the ships, but really isn't too practical in the FS2 engine.

lods are here to ease the cpu work. For exemple my OTT fighters often have 700 or more polys, still they work fine coz I've loded them heavily.

Redoing some of the texture maps on the other hand can make some ships look much better, the Myrmidon, Tauret, Erinyes, and several other fighters/bombers look not so hot up close because of low res textures, if someone wants to work on those..

you have no idea how difficult is is to make high res maps

[Unknown Target]

The reason I was actually bothering with this idea is because of the new DX8 upgrade, which would increase FS2's model polycount capacity (hopefully)

[RandomTiger]

As I keep on saying Im only changing it to use DX8 from DX5.
DX5 did not set the polycount limit. did.
Different jobs. However DX8 is designed for higher poly counts so I imagine it will help. You will have to push your case with the other programmers.

[Nico]

but still: polycount is not a pb, it's easy to get around the 820 polys limit with subobjects. pb is with too detailled models, fs2 will be slow. so if using DX8 improves ( or at least allows future improvements of ) the performances, such project could be nice.

[Unknown Target]

What's the point of having a DX8 engine if you're using DX5 graphics?

[RandomTiger]

One thing at a time. These (DX8 and hardware T&L) are the two most difficult jobs that are likely to get done. Trying to do them both at once would be a disaster.

We could have improved the engine in DX5 but then we would have to learn DX5 as well. Also DX8 has a lot of high level stuff that makes it easier to change the engine after. At the end of this there will probably be less code than before because DX8 makes things easier.

Having DX8 will give us access to new functionality like vertex and pixel shaders. I bet more people would be willing to code in DX8 than DX5.

I see DX8 as just one step of many we need to take to get make the FS2 engine respectable again and get you guys everything you want.

EDIT: I'll tell you what, I'll shut up until DX8 is finished and then we'll see what we can do.

[Killfrenzy]

*Dances around* HIGH POLYS! HIGH POLYS!

Cuz I can't LOD.....I dunno how....

[LtNarol]

More polys will not make models look better if the textures are as low res as some of these are, no matter how much detail you put on a myrmidon in terms of polys, if you use the same maps the end result will be so insignificant that it wont be worth the time it takes for you to reapply the textures.

[Unknown Target]

Then how about both????

[Rictor]

we need to find who made these,





THIS is what we are working towards, and can very closely resemble.

dont you think that having fs2 look that good is worth the effort

[aldo_14]

I don't see the point in redoing the existing models, when hardly anyone can really match their at present.

Besides which, it's sodding boring reworking existing stuff.  Even the less original designs I've done has some degree of uniqueness/individuality (for lack of a better word).  why waste the relatively few **** hot modellers we've got here on making rehashes of existing ships?

Besides which - low poly models = more ships per mission (potentially).  And I always work to a sub-3000 poly count (that's for ALL the LODs) on ALL my ships, anyway..... (as a point of fact, the only fighter I've made that's far over the :V: ships was the latest Claymore (about 720 IIRC), and that has an extra LOD to take that into account - so hi poly fighters within current limits are reasonably plausible)