P.U.R.E. RC4.2 - Page 11

P.U.R.E. RC4.2

WolfeGames and projects headed by Argh.

Moderators: Moderators, Content Developer

User avatar
1v0ry_k1ng
Posts: 4656
Joined: 10 Mar 2006, 10:24

Re: P.U.R.E. RC4.2

Post by 1v0ry_k1ng » 17 Mar 2009, 23:24

your copying LoTR2 resource model?! that game was fucking terrible!
nobody wants to spend their game building and rebuilding mines in every square inch of the map, its utter tedium! the reason spring style metal went out of fashion and we went back to spots is BECAUSE placing metal extractors on every inch of the map is time consuming and about as fun as sanding bleach into your fingers.
[23:24:07] <[RoX]1v0ry_k1ng> pintle
[23:24:13] <[RoX]1v0ry_k1ng> you should dl pure
[23:24:47] <[RoX]pintle> its too big a dl
[23:24:49] <[RoX]pintle> cba
gg
0 x

Saktoth
Zero-K Developer
Posts: 2665
Joined: 28 Nov 2006, 13:22

Re: P.U.R.E. RC4.2

Post by Saktoth » 18 Mar 2009, 07:17

One of the problems with mechs is the AoE. They can take out 2-3 smaller units per shot, any kind of shell etc.

And infantry outside of buildings are almost useless, you can flatten a whole squad in 1 shot due to AoE.

Worst of all, the mech doesnt reflect its actual AoE with its FX size.

In fact, i think AoE may be a problem in general. Id suggest only putting heavy AoE on dedicated anti-spam units.

Seriously though, if you nerf mechs, it will become assault wheel spam.
0 x

Google_Frog
Moderator
Posts: 2442
Joined: 12 Oct 2007, 09:24

Re: P.U.R.E. RC4.2

Post by Google_Frog » 18 Mar 2009, 09:58

All I did in those 2 games is rush an inf squad to stop you making defence as Sak and I found that overmind can win if they contain res long enough to spam flamers. Then after the inf rush I made nothing but Heavy Mech.

With the Heavy Mech at least nerf missile range to normal gun range and fix the +60E.

Some more notes:
* Robot Snipers in buildings shoot a shot intead of a line. The different weapon graphics inside buildings was really confusing at first and makes snipers harder to spot, also it looks bad.
* Most of the time it's better for units to shoot enemies instead of neutral buildings. Make allied units not fire at enemies that can't be hit because a building is in the way as units can waste their long reload weapons before enemies can be hit or waste dps hitting a building.
* Increase the custom mex radius shower view distance as it's really annoying not being able to see which spots are covered when zoomed out.
0 x

User avatar
Argh
Posts: 10920
Joined: 21 Feb 2005, 03:38

Re: P.U.R.E. RC4.2

Post by Argh » 18 Mar 2009, 11:23

* Most of the time it's better for units to shoot enemies instead of neutral buildings. Make allied units not fire at enemies that can't be hit because a building is in the way as units can waste their long reload weapons before enemies can be hit or waste dps hitting a building.
Can't be done in the engine atm, so far as I've been able to determine. There really aren't any Lua workarounds for this issue so far as I can tell. If something's Neutral, and avoidNeutral == 1, then it's avoided. Changing the Neutral state of World Builder objects would suddenly make them targeted by every nearby Unit, which would lead to all sorts of interesting exploits ;)

Maybe next version of Spring, basically.
0 x

User avatar
thesleepless
Posts: 417
Joined: 24 Oct 2007, 04:49

Re: P.U.R.E. RC4.2

Post by thesleepless » 18 Mar 2009, 13:45

any ideas on resistance building crash?
kinda makes it unplayable, quite a showstopper.
0 x

User avatar
1v0ry_k1ng
Posts: 4656
Joined: 10 Mar 2006, 10:24

Re: P.U.R.E. RC4.2

Post by 1v0ry_k1ng » 18 Mar 2009, 14:09

theres about 6 versions of pure rc4 around, all have the same name/internal name but all of them are diffrent sizes, none of them sync with eachother and most of them crash
0 x

User avatar
Argh
Posts: 10920
Joined: 21 Feb 2005, 03:38

Re: P.U.R.E. RC4.2

Post by Argh » 18 Mar 2009, 22:48

Well, I guess it's a good thing that RC5's almost done, then ;)

Changelist, thus far:

1. Nerfed recharge on health for Assault Shells a bit. Dunno if it's balanced yet, me and Eman will test tonight before releasing anything.

2. GUI has been rebuilt, with a bigger button area, as requested.

3. You can now give a Guard order to nearby infantry, and expect them to put themselves into a building, to make things a bit more intuitive. Warning- due to some engine limitations atm, if you do this at a distance, you will see them just walk there and guard.

4. Resistance MultiTrucks made a lot more vulnerable on the move, and the AA variant will cost more Power.

5. Resistance HeavyMech rocket barrage nerfed in terms of range. Damage and AOE remain unchanged for now. Again, this will be tested, we're not sure if it's still OP.

6. Much smaller download size for people who already have the game installed.

7. Should perform a little bit better on low-end graphics cards, due to switch to DXT3 compression. May even work on lower-end ATi, dunno yet.

8. Major fixes for a number of sound-related problems that were probably the cause of crashes during play.

9. Fixed display of mined area.

10. T.R.O.L.L. weapons altered.

11. Ancients buffed a huge amount on standard Spring maps, so that they're (vaguely) playable. Not promising I'll spend any more time on that one, I'd rather refine the official gameplay.

12. New building death sounds.

13. Overlord finally gets a halfway-decent death sequence.

At any rate, I need to get back to finishing all of the sound-related stuff, I'll get that done and then go play some with Eman and try to find anything else that's really abusive tonight before uploading it.

One thing we'll test is maybe doubling the Power cost for mines, but giving them twice the area on official maps, to reduce the micro-suck a bit.
0 x

User avatar
1v0ry_k1ng
Posts: 4656
Joined: 10 Mar 2006, 10:24

Re: P.U.R.E. RC4.2

Post by 1v0ry_k1ng » 19 Mar 2009, 00:44

cut your losses, its a dumb resource model.
switch to some kind of lua territory or back to metal spots ftw
0 x

Andromeda
Posts: 60
Joined: 01 Oct 2008, 03:59

Re: P.U.R.E. RC4.2

Post by Andromeda » 19 Mar 2009, 02:42

I agree with 1v0ry_k1ng.
0 x

Google_Frog
Moderator
Posts: 2442
Joined: 12 Oct 2007, 09:24

Re: P.U.R.E. RC4.2

Post by Google_Frog » 19 Mar 2009, 07:15

Can you explain the difference between your model and the TA model as long as the mexes are evenly distributed around the map?
0 x

User avatar
Argh
Posts: 10920
Joined: 21 Feb 2005, 03:38

Re: P.U.R.E. RC4.2

Post by Argh » 19 Mar 2009, 10:11

Can you explain the difference between your model and the TA model as long as the mexes are evenly distributed around the map?
There are several key differences.

1. For the most part, maps don't have "evenly distributed" metal spots. Go look at the maps people actually play, and you'll see that they have clusters for easy base-building, then have strings leading to more clusters. It's a very artificial model, and it puts far too much responsibility into the hands of mappers as to how gameplay works out. Hence, there are literally hundreds of maps, probably at least a hundred that are pretty, maybe fifty that are pretty and have game-balanced heightmaps... and most people play most of their games on three. This doesn't seem like a good model to me. It seems like a massive waste of human effort.

I've been seeking out a better way to handle these issues for quite some time now. And I haven't been very happy with the fact that most games played using Spring are on two maps, both with very specific designs. I wanted to break that mold, and see if there was a better way- one that you could even play on those heightmaps even... but with different results, and a very different feel. After something like a year of thought and a lot of testing, this is the model that I think works.

2. This model requires you to hold square "feet" (or "meters" if you prefer) of actual land. It's the closest thing we have to a real sector-based take-and-hold, but without the major issues of a sector-based system, which I explored at one point. Among other things, since where exactly you place your mines is up to you, you're allowed to place them inefficiently (for maximum metal gain) in order to play more effectively (in terms of tactical use). For example, you may place a mine in a sub-optimal spot that's behind a terrain feature, simply to get it out of the line of fire of enemy units.

This is something that should be a lot more clear when I get the next map done, which will be a hills-and-valleys thing, with some fancy effects I have planned.

3. The micro involved in dealing with this model is neither trivial nor silly, and I certainly didn't put it into the game because I hate players.

That said, I want players to have to balance between getting mines placed, repairing damage, building defensive webs, and using attack units. That's a level of management where you're actually having to do something significant, and you can't just micro some Flashes around- you're going to have to develop and grow and repair things a lot.

Having to stop what you're doing attack-wise, and repair your economy, because somebody used a strategic attack (ZAPs, Maxims, ArtilleryCannons, SPCs, and certainly not least, air) on your economy is a major feature, not a bug. Costing opponents micro-time is a major part of any good RTS, and this is meant to make losing chunks of economy a problem that you must sort out asap- it's a fairly major part of the final tipping-points for victory. It's also the reason that while we have MMs, we don't have dedicated storage. Not that you should ever have a lot stored up, unless you're trying to tech up.

4. The fact that the Ancients aren't super-powerful, but are instead about double the output of a mine on most maps, means that if you want to reclaim them for a sacrifice strat to rush, that's a valid strat, and isn't crippling you later. But in large games that go over 15 minutes, losing that output over time may actually be a key factor, as you get into the 50-60 range of Materials, where if you can eat that much, you're building a massive army or are at Tier 3 and having to support high-end units with their massive associated costs.



I hope that answered the question adequately. I also hope that people will respect that this is not something that I decided to do on a whim, and that I don't intend to change the basic game mechanic.

RC5 is very nearly done. I fixed the infinite-heavytroopers bug, did a few last (minor) balance changes on the T.R.O.L.L. and on the overall feel of Air (which is almost certainly not well-balanced yet, that said) and a few other minor things, and it appears that all of the new stuff is stable (no crashes at all in 5 hours of testing) so I'm going to upload this stuff tomorrow after I've had a chance to get some sleep and do some minor repairs on the maps (start-positions, mainly). If anybody has some last-second balance stuff they want me to look at, please post by 3PM EST, because I will probably be uploading by then.
0 x

User avatar
zwzsg
Kernel Panic Co-Developer
Posts: 7017
Joined: 16 Nov 2004, 13:08

Re: P.U.R.E. RC4.2

Post by zwzsg » 19 Mar 2009, 10:44

I tried Pure on another PC (Fedora 10, nVidia Corporation NV34GL Quadro FX 500/600 PCI, Intel Pentium 4 2.8Ghz, 1 GB ram), and spring exited without error during loading.
infolog wrote:LogOutput initialized.
Available log subsystems: ArchiveScanner, VFS-detail, VFS, unit, CollisionVolume, mapinfo
Enabled log subsystems:
Enable or disable log subsystems using the LogSubsystems configuration key
or the SPRING_LOG_SUBSYSTEMS environment variable (both comma separated).
Using script /home/stagiaire/.spring//script.txt
using configuration source "/home/stagiaire/.springrc"
OS: Linux
Using read-write data directory: /home/stagiaire/.spring/
Using read-only data directory: /usr/share/games/spring/
Using read-only data directory: /usr/lib/spring/
Scanning: /usr/lib/spring/maps
Scanning: /usr/lib/spring/base
Scanning: /usr/lib/spring/mods
Scanning: /usr/share/games/spring/maps
Scanning: /usr/share/games/spring/base
Scanning: /usr/share/games/spring/mods
Scanning: /home/stagiaire/.spring/maps
Scanning: /home/stagiaire/.spring/base
Scanning: /home/stagiaire/.spring/mods
Video mode set to 800 x 600 / 32 bit
[ 0] SDL: 1.2.13
[ 0] GL: 2.1.2 NVIDIA 173.14.16
[ 0] GL: NVIDIA Corporation
[ 0] GL: Quadro FX 500/FX 600/AGP/SSE2
[ 0] GLEW: 1.5.1
[ 0] Connecting to local server
[ 0] Starting demo recording
[ 0] Using map Mesa_City.smf
[ 0] Recording demo demos/20090319_113850_Mesa_City_0.78.2.sdf
[ 0] Using script Commanders
[ 0] Using mod P.U.R.E. RC4 0.9
[ 0] Using mod archive PURE_COMPLETE.sdz
[ 0] Became player 0 (team 0, allyteam 0)
[ 0] OpenAL: 1.1
[ 0] OpenAL: ALC_EXT_capture AL_EXT_capture AL_EXT_vorbis AL_LOKI_quadriphonic AL_LOKI_play_position AL_LOKI_WAVE_format AL_LOKI_IMA_ADPCM_format AL_LOKI_buffer_data_callback ALC_LOKI_audio_channel
[ 0] OpenAL: sounds/button9.wav has data length 291939 greater than actual data length 16388
[ 0] Parsing unit icons
[ 0] Loading all definitions: 0.381000
[ 0] You are missing the "ARB_shadow_ambient" extension (this will probably make shadows darker than they should be)
[ 0] Opening map file
[ 0] Loading Map
[ 0] Loading detail textures
[ 0] Creating overhead texture
[ 0] Creating ground shading
[ 0] Loading tile file
[ 0] Reading tiles
[ 0] Reading tile map
[ 0] Creating projectile texture
0 x

Google_Frog
Moderator
Posts: 2442
Joined: 12 Oct 2007, 09:24

Re: P.U.R.E. RC4.2

Post by Google_Frog » 19 Mar 2009, 11:12

I was referring to the difference between complete metal coverage and single mex spots equally distrubuted around the map. Point 3 and 4 are completely irrelevant to metal distribution. 1 is irrelevant because you are already making the maps so you could place metal spots evenly around the maps. People play a wide range of maps now, when last did you check?
You're allowed to place them inefficiently (for maximum metal gain) in order to play more effectively (in terms of tactical use). For example, you may place a mine in a sub-optimal spot that's behind a terrain feature, simply to get it out of the line of fire of enemy units.
This is the only bit that really answers my question. I suspect the extra choice of lose vs packed hexagonal mex placement doesn't have enough efficiency difference to justify the extra micro in the model although placing behind buildings might.

Also unless units shut down when you E stall I don't see how taking out E Immediately stops someone's attacking power. Actually wouldn't it be a good idea to attack with your current army so some of them die to free up your remaining E income for remaking your power plants?
0 x

User avatar
1v0ry_k1ng
Posts: 4656
Joined: 10 Mar 2006, 10:24

Re: P.U.R.E. RC4.2

Post by 1v0ry_k1ng » 19 Mar 2009, 12:02

ok, bear in mind that the pool of players that only play BA on three maps, will NEVER play your mod. they will look at it, declare it rubbish and imba and slink straight back to ba delta, that is assuming they leave it long enough to try it.

the pool of intelligent players that trys all mods plays on very nearly every decently made map- thats a pool of well over 100 maps.

your system is simply the equivalent of there being a mex spot every foot. if you play comet 1v1 a few times you will quickly come to realise HOW TEDIOUS IT IS PLACING AND REPLACING METAL EXTRACTORS CONSTANTLY. the REASON maps have small numbers of important spots, not loads of crappy ones, is because the more spots, the less important each spot is and the more annoying it is placing mines.

seriously, modders need to think their resource models through.
gundam and nanoblobs with metal maker economy (no offense smoth, those were early dayz) were the first generation of failed model.
this model has already been tried in spring, its called spring style metal, it can still be seen on maps like egyptian desert and brazillian; but players hate it
0 x

Google_Frog
Moderator
Posts: 2442
Joined: 12 Oct 2007, 09:24

Re: P.U.R.E. RC4.2

Post by Google_Frog » 19 Mar 2009, 12:16

The metal placement is sort of ok on small maps. The extraction radius is so large that there's a similar number of mex spots on normal spring 1v1 maps. On large maps it's annoying, especially since buildings block mex placement so mexes can't be easily placed by draging a line of them with appropriate spacing.

Argh, when units that can garrison are given a right click command on a neutral building the command given should be capture, not attack like it is atm.
0 x

User avatar
lurker
Posts: 3842
Joined: 08 Jan 2007, 06:13

Re: P.U.R.E. RC4.2

Post by lurker » 19 Mar 2009, 17:03

Most maps I see have metal spots nicely sprinkled over the entire thing, the only exception being a triple-spot at base points.
0 x

User avatar
Pxtl
Posts: 6112
Joined: 23 Oct 2004, 01:43

Re: P.U.R.E. RC4.2

Post by Pxtl » 19 Mar 2009, 20:55

It sounds like this is mostly about making PURE work with DSD, since DSD doesn't really spread the mexes evenly - they're all pushed out on the edges of the planes.

Either way, a "claim turf" model would be fine if it were low-maintenance... but I have to agree that building hundreds of mexes is freakishly tedious. The obvious solution would be to simply use mexes with larger extraction radii, but that causes its own problems (like the edge of the map, or on super-large and open/empty maps). An auto-mex-spam widget is kind of necessary at the bare minimum... but even that would be nasty, since it would require doing multi-unit travelling salesmen problems to implement.

Also, out of curiosity Argh, how would you feel if a map "fixed" your mod?
0 x

User avatar
1v0ry_k1ng
Posts: 4656
Joined: 10 Mar 2006, 10:24

Re: P.U.R.E. RC4.2

Post by 1v0ry_k1ng » 19 Mar 2009, 22:04

why dont you just make your maps have an even spread of metal spots across them in a grid.
that way, people can choose to play it lotr style or they can choose to use spring maps
0 x

User avatar
Eman
Posts: 37
Joined: 07 Mar 2009, 01:38

Re: P.U.R.E. RC4.2

Post by Eman » 19 Mar 2009, 22:24

I'll give a shot at Ivory and Googles questions.

One kinda hard thing to point at is the economic value of space; wind farms vs. fusion plants.

Another is the economic value of time; micro to move to moho mines.

Any resource model is two edged at the very least. Territory driven (metal spot or not) resource models shape the "burden of defense". Examples run a broad spectrum and create very different games.

Starcraft/Warcraft/CC - Trains of peons (annoying micro). Sites decay (creates burden of expansion and forces the endgame). Fixed sites based on map. Most of map is dead space without economic value. Defense is focused on the sites, and securing future sites.

The "daddies" of mainstream RTS's. Anyone who is on this forum prefers different dynamics. Still clearly a valid approach - even if you don't like Garth Brooks, his millions of fans mean you have to accept that there is some value to his product.

TA - "MEX" convention. Fixed sites based on the map. Most sites are in clusters (either by raw proximity or because of other terrain creating borders). Much of map is dead space without mining value. Metal maker economy is second option. Reclaiming the dead is a third factor. Defense is focused on defending the clusters.

Maps make exceptions. Take Comet Catcher - why is it one of the best 1 vs. 1 maps and why does it require more skill?

LOTRII - Mines without metal spots (not entirely true, but close to true). Even distribution over controlled territory is "encouraged". Very little of map is economic dead space. Defense is focused on choke points if available and response forces when unavailable.

One of the less typical dynamics in LOTRII is that on a fairly open map you cannot defend all your territory with porc even though porc is not crippled. This forces response forces as a means to defend territory.

PURE - Mines without metal spots, truly even distribution. Ancients create value to certain territory and add options - reclaim or keep. Metal maker economy is third option. Map specific buildings effect initial defensible positions. Shields, Bunkers, Turrets, and Build towers effectively reform terrain. Burden of defense is ... complex.

The intent is something along the lines of territory and space driven economy rather then control of fixed "victory points". Another is to move away from choke point based map design. The basic difference we have tried for is asking "What territory can I defend?" rather then "What metal spots can I defend?"

I am not going to say that I am sure it is working correctly, but I imagine you can see some of the intent; to change how territory matters and how you defend it. Part of the intent of PURE is to do something different - not to rebalance TA.

------------------------

Long and short, it is not simple. If you don't like it the way it is;

A. suggest a change

B. besides "make it like *A games"

Aside - Pointing out that Argh is bad at listening/kinda unapproachable is valid. The flip side is that a lot of the "feedback" here is not much better.
0 x

User avatar
Otherside
Posts: 2296
Joined: 21 Feb 2006, 14:09

Re: P.U.R.E. RC4.2

Post by Otherside » 19 Mar 2009, 22:40

just because people dont like writing wall of text's like argh doesnt mean points arent valid :]
0 x

Post Reply

Return to “Argh's Projects”

cron