ok, this is it! first beta out...

it would be great to have some feedback on the map.
especially on fps in larger servers with +24 people!

big thanks to wile e coyote for providing me with lots of feedback and the cool briefing-sounds you hear from time to time ingame!

ok, enjoy :)

map download as .zip
main download (http://pedroleum.dod-federation.com/maps/dod_plateau_b1a.zip) (tnx to dod-federation!)
mirror 1 (http://files.filefront.com/dod_plateau_b1azip/;5527224;;/fileinfo.html)
mirror 2 (http://rapidshare.de/files/33782829/dod_plateau_b1a.zip.html)
mirror 3 (http://www.filefactory.com/dlf/f/d4e22e/b/0/h/4e4d2c69f15f1f7c)
mirror 4 (http://www3.rapidupload.com/d.php?file=dl&filepath=20892)

map download as .bz2 (for server admins)
main download (http://pedroleum.dod-federation.com/maps/dod_plateau_b1a.bsp.bz2) (tnx to dod-federation!)
mirror 1 (http://files.filefront.com/dod_plateau_b1abspbz2/;5527244;;/fileinfo.html)


http://img116.imageshack.us/img116/7668/dodplateauscreensvol031ib9.jpg http://img300.imageshack.us/img300/9721/dodplateauscreensvol032ds3.jpg http://img300.imageshack.us/img300/5740/dodplateauscreensvol033zi8.jpg http://img133.imageshack.us/img133/8871/dodplateauscreensvol034ub5.jpg http://img133.imageshack.us/img133/9417/dodplateauscreensvol035wl9.jpg http://img133.imageshack.us/img133/3844/dodplateauscreensvol037mj2.jpg

Nice map. Added to Home of the Testicle.

Is this map suppose to be timed?? I just played it on our server and it looked awsome but as axis all u do is just defend. Only the allies can win??

i didn't include a timer yet. i wanted to see how it plays and how hard it is to capture the objectives. that would have been very hard to tell if the timer's to short. maybe i'll add one later. but until now, the axis only have to prevent the allies from bombing everything up.

Pedro,
You want the Federation to test?
Please go PM Cranbarry at http://www.dod-federation.com/forum/forumdisplay.php?f=18
We can set up a time that is good for you.

Awesome! I like the use of the new models. The cornfields turned out great!

Tested your map today. It looks quite good but I think it'd look even better if you add better lightning to it. Now it's just plain "grey" and no bigger visible shadows.

Also, it performs pretty poorly even on my high-end machine so I think fog/viewdistance trick could be working perfectly here. With new lightning and fog it could be a winner :)

What have you done ! Brilliant work !

Very nice looking work.
Look forward to giving it a spin on a server!

this map is a blast.

we just had a little run on it and found (besides some fun) a huge amount of bugs :( i'm almost ashamed!

the worst of them all: after you destroyed the barbwires or the churchwall/churchdoor and the round restarts, the playerclips/churchwall/churchdoor aren't reset. so on second round, everybody just can jump the barbs and skip the blowing. veery bad bug.

second is a player_hurt which i forgot to delete. it's near the sandbag which you get to when you exit axis spawn straight forward. the one near the other barbwires is it.

third are the teamwalls near the barwires. they're not logic to be there and only allies will feel them. those teamwalls should really be triggerable!

and fourth are a lot of collision issuses, where you can get stuck and so.
i really should have found these before.

and fifth the hrd is messed up. need to check that out aswell...

dradz:
i'll check out the errors firsts i think. after that, it would be nice to have a test on beta2 with some skilled people!

skdr:
maybe i have to try some different settings out before i decide on the lighting and how it looks best. but fog is my enemy.. i really don't like it at all. i'll add that only in case of emergency :P

I don't think you should add fog for fps related stuff, but I think you should add it because it makes sense.

I just ran through your map and when looking across really large areas there seems to be a lack of depth (even though spacially it seems like there should be alot of depth... bold powerful layout by the way). Even a map like bruegal, as good as it looks, the lack of fog entity has an unrealistic lack of depth... in real life all atmoshphere is visible in some way even if it isn't a "foggy" day.

we just had a 20 minute run of the map on "band of beavers" with 32 players. it was very intense and it played completely different than i thought :P axis didn't have a chance!
i think the bombtimer are much to short for a map of this size. you can't make it to the bomb even if you're "close" to it. really have to change that.

also, framerates were to low (on my rig and on some other peoples aswell)
well... i'll see how i can improve fps again.

but before i start to rework everything, i'll have some more runs on the map.

i'm already reloading the serverlist every minute ;)

nave: you're right, i'll try that. there is always a little distance-fog. even if it is only in the skybox.

...and even if you don't use fog or use just a little, you can still set the far z clip distance which should help boost frame rates.

good work thats all i can say i love this map cant wait for the final

it is a total new style GREAT MAP!!!



DOWNLOAD HEAR (http://www.oneworldsyndicate.com/owsdownloads/index.php?act=view&id=215) DOWNLOAD

You can still use fog and z-clip to your advantage and not ruin the look. Keep experimenting. BTW...the best way to do that is in game w/sv_cheats turned on. You can then use the cvars to move your fog up and back and change the z-clip distance. Do a "find fog" cvar and you'll soon figure them all out.

The real deal is with large open displacement maps you have the potential for lots of player models to be in view at the same time (which is the biggest problem w/fps on open maps). There are only several ways to combat this...z-clip and/or hard-core portal and hint work. Anyone who tells you not to worry about fps doesn't understand it's gameplay first, looks second.

FuzzDad:

Thats what i tried yesterday. the thing is, if i do a far z-clip, half of the level gets cut of at the worst part of the map. its the one where you can see from the church to the lake. that view is almost the whole level distance. far z-clip only would work together with fog and the darn fog is nothing of much beauty on an sunshine day map ;)

as i said, i'll add fog only if everything else failed. first i'm gonna remove some of the foliage and put some simple woodwalls there or something in that direction.

after that failed, i'll create a hedgemodel with half the polygons.

after that failed, i'll remove even more unnecessary stuff.

and after that failed, i'll ask the parrot what to do next!

and only after the parrot failed, i'll do a rainy, foggy and very muddy map!

Just ran through it, looks sweet, fix the performance up and you got a winner ! Fog if done properly doesn't look bad, in fact I think a small amount makes a map looks better, softens up lighting.

:)


P.S and O.T - I just got my computer back 10 mins ago! WOOHOO !

Just did a quick test map, and I must say this is the best map I've played yet for DoD:S, even if there were no others in the server. The sheer brilliance of the environment was awesome. The only time I lagged at all (went down 10 fps for a sec, then back up around 7 fps, so I only lost 3 overall) was when I was walking down the hill towards the lake. Other than that, it was sweet. Great damn job, man!

Pedroleum,

From Furyo over at the Federation, to tweak far z in game....

r_farz xxxx : Changes in game the far z clip distance as set in the env_fog_controller entity. Helpful to tweak said distance without time consuming compiles. XXXX is the number of Hammer units set.

http://www.dod-federation.com/forum/showthread.php?t=508

Its a great looking map, but you can do some work on the performance. There is a lot being rendered when you can't see it on screen. Some strategic hints and occluders should help things a lot.
I cant wait to try it out!

hey pedro, couldnt sleep so i used the time to play with your plateau screenshots.
i don't like fog neither but just for consideration if everything else fails...
(i don't hope so because i still like your original version the most)

anyway, have a look here:
http://www.dayofdefeat.net/forums/showthread.php?s=&postid=987147#post987147

:)

Originally posted by MrGrubby
Its a great looking map, but you can do some work on the performance. There is a lot being rendered when you can't see it on screen. Some strategic hints and occluders should help things a lot.
I cant wait to try it out!

actually, i thought i did a pretty good job on the hinting/leaf visibility. there is i.e only a small part of the other side (allies side/axis side) visible to each other. if you use sv_cheats 1 and then mat_wireframe 3 you can see which leafs (and it's content) is currently beeing rendered.. (use r_3dsky 0 and r_drawstaticprops 0 if it is to confusing)

but you're right. there is a lot of stuff visible at the same time. thats the problem with open field maps ;)

the problem with all the optimizing tools are that they're not made for open field maps. i can use occluders, but mostly, they make the whole thing slower because they take up cpu power during gameplay. the areaportals only work if two areas are sealed off completely. where in hells name should i seal off a whole area? i could do that on the open buildings, but there's nothing in them atm to seal off. and thats about it with the optimizing tools. isn't it?

the rendering distance of the models are mostly set. that helped the most until now. another heavy improvement would be to do r_detailprops 0 which turns off the grass. i'ts a boost of almost 10-15 fps. but it doens't look that good :P (i already lowered density of the grass, so there's a little boost already)

far_z and fog on a distance of 4000 only gives a benefit of about 4-6 fps on my rig (in the bad fps areas) maybe somone could test out aswell?

for performance testing, type:
sv_cheats 1
r_farz 4000

add these for the look:
fog_color 153 170 170
fog_start 500
fog_end 3800
fog_colorskybox 153 170 170
fog_startskybox 500
fog_endskybox 3800


as i said before, im kindof confused by the source engine. yesterday i removed half the polygons (for testing issue) from the hedge model (used maybe around 90 times) and the bushes model (used maybe around 70 times) and the performance gain was only about 3fps in the problem areas....

it seems to me the engine got more problems with the amout of objects and not the amount of polygons used per object. thats also what +showbudget sais. to many static props.

of course, if i add all of these things together, i get a performance boost of about 20 fps. but then it really looks <3<3<3<3ty!

We played this tonight with ten people or so...a blast and a visual feast. Very natural terrain and displacement work.

Reminded me of dod_campagne, my all time favorite dod map. But this map is so much more. Well done.

I like the focus on objectives and blowing obstacles (even if some of them are a little buggy i.e. the det thing staying there after blowing church walls)

thanks :) i'll need to play it again somewhere, but it's never on anything with an acceptable ping ;)

the buggy objectives are gone in the next version. i'll made sure of that. (i didn't test enough for the first version)

Originally posted by pedroleum
as i said before, im kindof confused by the source engine. yesterday i removed half the polygons (for testing issue) from the hedge model (used maybe around 90 times) and the bushes model (used maybe around 70 times) and the performance gain was only about 3fps in the problem areas....

it seems to me the engine got more problems with the amout of objects and not the amount of polygons used per object. thats also what +showbudget sais. to many static props.

Hehe, I did the same on our hedge models a while ago. Replaced them with hedges of 500 polys less per section, so taking into account that most of our maps have a render distance similar to yours, that makes good 10000-20000 polys (just a guess) less to render. Yet no noticeable fps gain.

I'm planning to make a little stress test soon and place a hundred static props of one type into one room and compare it's performance to the fps I get when I combine those 100 props into one single prop.

It might help if you reduce the number of different props. It's easier to handle (clone) 100 hedges of the same mdl than 100 different hedes. But you're prolly already having it at a minimum.

You should also try to do a full compile with a z-clipping of 4000 instead of using the runtime cvar, which is maybe just a fast inaccurate clipping not similar to what vvis does (afaik it also takes account of the clipping distance when determining which leafs see what leafs). That's just an assumption, I've never really tested it. I can only say that I used r_farz on our maps and ranges from 3000 to 10000 didn't make much of difference in fps, which is just strange.

BTW I played your map, I really like what you've done. FPS are ok on my rig. Tho thats probably because I run at rather low settings (GF4TI with 64mb forces me to) 800x600 with med textures, high models and low shaders. I had absolutely no visual lag while playing it on a rather full server.

ranson:

i see you made the same experiences!

actually, merging each cell of grainfields into one big prop was the next step i planned... that way, i can do better lod aswell.

but i haven't started yet...
i'll let you know when i got it working.
if you're faster, let me know if it works ;)

Any idea when we can expect the next version of this map ?

oh, sorry i didn't see your post :(
i don't know yet. i think i need to redo a lot of stuff, so the performance gets better. the ultralow version i made is just plain ugly. its also a completely different map in terms of gameplay, so i don't want to release it that way.
give me a little time to figure out what i could optimize ;)

hi just want to let u know we played the map on our servers last night
and there are 2 things i didn't like 1. barricade that u have to blow up you can just jump over .
and the 2nd thing is there is no way for the axis to win the game .

i was thinking if u do it all in flags for both and make the barricade higher u wouldn't need a timer and both teams would have the same chance to win. just an idea .

still a great map.

hey coldair

thanks for feedback!
nr.1 is a known bug. i forgot to reset all the playerclips after the allies won. so in the first round, everything is fine. in the second, you can jump/walk through the obstacles. consider it fixed ;)

nr.2 i thought about a time limit, but i don't really like it.
i'll figure something out! promise!

I never made this suggestion, as we were more concerned with hammering out gameplay, but I always assumed there would be a timer. I know you stated you are against it, but given the amount of objectives already in place at the moment it makes the most sense.

I have one other rather unusual idea, as far as I know it's never been tried in DOD, I'll PM you later ;)

Running on our server ! Gj

Great map pedroleum :-D

This is a really great map, and I had a good feeling when I played it.
I know it is just a game, but as a player it just felt believable like a real place.

I think the gameplay is fun, and I realy enjoy a good objective map like this one. But it seems a bit unfair, that the Axis have no way to win. Sooner or later the Allies will get all the objectives and cap the last flag.

I personaly think that a timer will do good for the map.

You did a good job on the model of the fields... may I and other mappers use those? ... of course with credit given :-D

I can see you are using some models of foliage, that I have not seen before... where did you get those? I would like to ask permission to use them.


Keep up the good work :-D

tnx for feedback guys :)

i'll add a timer in the next version. unless wile e's unspoken revolutionary idea (which he didn't bring to my attention yet ;) is so sweet that it doesn't need a timer...

about the corn/grainfield models. i got tons of pm about them the last days. they're not finished yet! i'm not satisfied with them. they suck at lighting, the corn texture is ugly, ... in plateau they're used as unlit generics, which makes them look very flat. in dod_farmers_b1 they're used as vertexlit generics with the double amount of faces to get the lighting right. but they're still messed up. i need to think of a way to get them right ;) (same problem i had with a recompile of the hedge model. i still can't figure out how valve/doddevs does it)

but i'll release - as every model i make - a nice pack of them to the public.
atm they're made as "fields" of 128x128 units and i plan on doing some for 256x256 aswell as 512x512 and some other values. mebby a scarecrow aswell ;)

also, i have to try something ranson and i talked about a few days ago (merging smaller models into a large one - on page2 of this thread)
my first test didn't show any difference, but i'm not sure if it was a good test, so i'll do another one.

until then, please be patient... i'm a little short on time atm.