Tremulous Forum
General => Announcements => Topic started by: Norfenstein on March 06, 2010, 08:45:17 pm
-
Time for some changes!
Reading the graphs: Each cross is one game, the distance from 0 on the vertical axis indicates how close a game was to the mean duration for all games on the graph (the weighting for how much this affects the lines is half as much as we used in phase 1). The numbers in the bottom right are the min/median/max/standard deviation for stage up times and game duration.
Both servers combined
(http://imgur.com/vtDGQ.png)
Euro server
(http://imgur.com/YiOnH.png)
US server
(http://imgur.com/9OTzi.png)
Compared to phase 3: worse ratio, better lines. I want the game to be able scale up well enough to higher playercounts, so I'm actually fairly pleased with this. It's probably safe to say friendly fire being on helped somewhat, but overall I feel the difference between the settings is minimal enough not to warrant specific changes for either. In other words: balancing with one is equivalent enough to balancing with the other.
Marauder (again)
While I don't agree with a lot of folks that the zap is an issue, I do think it's obvious that the air acceleration increase has had an effect on the balance. I was prepared for this however, and had planned on increasing the marauder's bounding box in response (since we want to keep the speed for the sake of fun). It'll still be smaller than the bounding boxes in 1.1.
Flamer (again)
The usefulness of the flamer seemed to take a dive with marauders being overpowered, but I'm still satisfied with its damage and everything else that's changed about it recently. The marauder change should help make it a little more attractive again, but I think its niche as a defensive weapon can still be improved by increasing its knockback. I don't know if x2 will be enough to really matter for anything, but I don't think it will hurt.
Other Stuff
Amid the pages of forum discussion about phase 3 were a handful of minor requests that I found myself amenable to.
- Shotgun being weak against dretches - I think the shotgun's role is as an inexpensive, short-range weapon against alien players, and not being so great against dretches is kind of bad for that. I'm not sure how much of a difference this will make, but I believe adjusting its pellet count and damage-per-pellet can make it better against dretches while not affecting much else.
- Teslas - The usefulness of teslas was actually something I wanted to address anyway, as it's become clear they're just not worth using instead of turrets in the majority of cases. I'm going to try spliting the difference between their current range and their 1.1 range.
- Basilisk gas repeat - The repeat rate on this makes it so the gas can't really be used in combination with the slash unless you're only fighting one target. There should still be a choice to make about how you attack, but I think a slight reduction in its repeat could make that decision more interesting.
Dretch?
Gotten some feedback about the dretch being underpowered. I'm not going to change anything with it at this time, but I'll be paying close attention in this phase to see what, if anything, should be done about it.
The Changes
- Marauder bounding box increased 44x44x34 -> 46x46x36
- Advanced marauder bouding box increased 48x48x38 -> 50x50x40
- Basilisk gas repeat decreased 2500 -> 2000
- Shotgun damage/pellet count changed 4/14 -> 5/11
- Flamer knockback increased 1.0 -> 2.0
- Tesla range increased 150 -> 200
-
This is live on both official servers now.
-
Thanks for the shotty change. In 1.1, I used it almost exclusively until s3. In 1.2 beta I had rarely used it at all due to the ineffectiveness against dretches. I'm hoping the 1 damage total decrease doesn't effect much against larger aliens though.
-
This looks good. I can see you are really working, and I think your changes for next phase are very viable. Keep up the good work. :)
@equi: It's all a balance and trade off.
-
Flamer (again)
I don't know if x2 will be enough to really matter for anything, but I don't think it will hurt.
I'm quite sure the flamer will still cause damage ;)
-
Nice changes, especially the tesla.
-
And especially the proximity nade - we had a hoot with that today! Too bad it was a bug :(
-
And especially the proximity nade - we had a hoot with that today! Too bad it was a bug :(
i was wondering what the grenade used when i saw that code was removed :P anyway its a feature till its fixed(which it has been now) :)
btw changes sound good don't know about increasing mara bbox ill have to see how it goes ingame atleast its only by a few points
-
After playing a few games, I am much happier with the shotgun now. I'm actually buying it in s1 instead of lasgun now.
-
had a game then dunno if its just me but somethings wrong any alien class i used felt slow and laggy in controlling it didnt seem to affect others
-
Proximity nade is not fun. In a tight entrance where only one person can fit at a time there is absolutely no chance in hell that an alien is getting through when there is a human camping that same entrance waiting to asplode you right in the face. What I'm referring to specifically are the main entrances to a base.
Give aliens granger cannon to counter this.
-
It was actually pretty fun to explode Meisseli's face yesterday :>
-
I like all these changes. COOL!
-
It was actually pretty fun to explode Meisseli's face yesterday :>
:(
After playing a few games, I am much happier with the shotgun now. I'm actually buying it in s1 instead of lasgun now.
I still find the lasgun a lot more useful though.
-
What happened to sprint? It used to move out, but now normal run is 320 and sprint is only 384. Feels like mole asses.
-
The sprint modifier has always been 1.2.
-
Protection against idiots-TKers-TeamBaseKillers is VERY URGENTLY necessary!!!
-
After playing a few games, I am much happier with the shotgun now. I'm actually buying it in s1 instead of lasgun now.
I still find the lasgun a lot more useful though.
I'm someone who likes getting in the mix of things, so a shotgun is more useful to me since I am often fighting aliens up close. Before these new changes, the shotgun was not effective against dretches, so the lasgun was better in most situations. However, the shotgun is now kills dretches consistently from close ranges, so its usefulness increased exponentially in my opinion.
-
After playing a few games, I am much happier with the shotgun now. I'm actually buying it in s1 instead of lasgun now.
I still find the lasgun a lot more useful though.
I'm someone who likes getting in the mix of things, so a shotgun is more useful to me since I am often fighting aliens up close. Before these new changes, the shotgun was not effective against dretches, so the lasgun was better in most situations. However, the shotgun is now kills dretches consistently from close ranges, so its usefulness increased exponentially in my opinion.
I guess it's just that I fail with weapons such as massdriver and shotgun a lot with unlagged turned on. Lasgun and chain make up for it though and they are good weapons to chase aliens with.
-
nice, but, tt seems to be widely used, and mara still appears to be the strongest class ???
-
Good work yet again Norf, I just wish the other devs were as transparent as you.
-
Anyone else have problems with grenade with vms-gppr1980.pk3? Exploding on impact?
-
Anyone else have problems with grenade with vms-gppr1980.pk3? Exploding on impact?
Read the above posts. :)
-
Ah.. Proximity... my english.
That means it is not fixed yet?
-
Anyone else have problems with grenade with vms-gppr1980.pk3? Exploding on impact?
my main issue with phase 4 is moving feels slow i cant play at all now the game doesnt feel right gonna try recompiling the vm's later hopefully that will fix the issue i have already checked my settings/deleted extra/old files out of gpp folder
-
I've gotten a few errors in my console lately in the middle of games saying something about me having compiled ioquake3 with a buggy compiler.
Umm....?
-
I've gotten a few errors in my console lately in the middle of games saying something about me having compiled ioquake3 with a buggy compiler.
Umm....?
Happens to a lot of people's clients from what I understand. Dev told me to ignore it.
-
I have a couple of ideas to improve the dretch, nothing serious:
Increase its left/right strafe speed? That way we can keep our eyes on the MD and still be sufficiently hard to target...
Reduce size...
Increase HP to 30....
or
Bring value from 150 to 120 or 100...
or
give it a dash ability (basically a clone of human sprint, just less of it)
I think it needs something, because for a starting class it doesn't do as much damage to advanced enemies as human starting classes do to advanced aliens. (=imbalance).
-
Fixed QVMs, courtesy of Lakitu: http://downloads.mercenariesguild.net/gpp/vms-gppr1982.pk3
-
I'm sure someone has asked this question, but I don't really have the diligence to go look for it. With that being said how many "phases" are there going to be?
-
How ever many are needed.
-
>9000
-
Since Norf is free to make any amount of changes till 1.2 is locked down and released I suspect that 1.2 will have many, many, many phases. Since the rest of trem seems to be locked in a development cycle that it can't get out of. (No offense to you Norf)
-
You should probably omit the games from the April Fools map when calculating the new stats for this phase. It was pretty blatantly human biased.
-
Aprils Fools day was not counted. Though now that I think about it, it's probably in stats.trem.net but we don't use those anyways :)
-
How about teamoverlay? It would be great to always see the available basilisks because those are the ones you usually need to know where they are.
Also, how about being able to customize what players you can see in your teamoverlay?
-
^That would be awesome to have multiple ways to show
Ex:
Way 1) Sort by Highest Score
Way 2) Show Lisks and Grangers First
Way 3) Alphabetical(?)
etc...
Or maybe allow Customization of what it sorts by.
-
As I recall it has little icons next to each person, you could have them colour coded so you can tell at a glance which they are.
I have team overlay disabled as it's massive, it'd be nice if I could have it bound to a key-hold.
-
It says name and class/armour/weapons....
-
I have team overlay disabled as it's massive, it'd be nice if I could have it bound to a key-hold.
seta teamoverlayon "cg_drawTeamOverlay 1;bind x vstr teamoverlayoff"
seta teamoverlayoff "cg_drawTeamOverlay 0;bind x vstr teamoverlayon"
bind x "vstr teamoverlayon"
cg_drawTeamOverlay 0
-
bind x toggle cg_drawteamoverlay simpler is better
-
The increase in the Tesla's range doesn't make it that effective in situations where it can still be sniped... Although it is useful if there is a wall of Teslas around a small base, as long as there are several people defending it to ward of the Adv. Marauders, it can last quite a while. (It did one time, all the way past the time limit.)
-
The increase in the Tesla's range doesn't make it that effective in situations where it can still be sniped... Although it is useful if there is a wall of Teslas around a small base, as long as there are several people defending it to ward of the Adv. Marauders, it can last quite a while. (It did one time, all the way past the time limit.)
My-style answer: Why was it built where it cannot fire and yet get sniped? (Think Acid Tube).
Dev-style answer: Why did you let aliens get to stage 2?
-
bind x toggle cg_drawteamoverlay simpler is better
Heh all right, forgot that toggle's in the new version too.
What about the teamoverlay, can it be done?
-
Made a new topic for brainstorming about teamoverlay.
http://tremulous.net/forum/index.php?topic=13199.0
-
Is the game data you use for your final stats when looking at each phase more strict that just tremstats? For example, the April Fools maps was extremely unbalanced and should be omitted. Also, more importantly, I noticed that there was a player that played on humans when no one else around and won a crap ton of games with no one else on the server by just killing the alien base. I don't know if he/she plans on doing this often, but if so, the stats for this phase could quite skewed. Perhaps you should only count if there is a certain amount of players on the server, such as 4.
-
They only count games with more than 6 players, a set list of maps, and a ton of other crap.
The April Fools stuff also had a different tag to prevent it being counted.
-
Cool beans.
-
Defence computers are still a laugh. 3 points per second is useless - people still do it the 1.1 way, 2-3 builders grab a conkit, repair, and the 3 point repair makes it 1/10th of a second faster.
Stacking the defence computers - I wonder what base has room for 1+ defence computers, and a tesla would benefit it way more. Why not remove the stack and increase the repair rate to be over the default human repair rate (12 was it?). 15-ish?
Basi gas could have more range to make it more useful.
That's all this time.
-
Defence computers are still a laugh. 3 points per second is useless
I disagree, the DC may heal buildings at 1/10 of the speed, but it heals *all* buildings at that rate, which makes it almost as fast as a ckit in the case of mara zap. Furthermore, building it frees a player (usually a half-decent one if they are bothering to repair) to continue to concentrate on fighting despite the mara zap. Clusterbases really need these if they can afford the space.
-
Also when some mara comes along and does some light damage, the DC will fix it before the next attack without you having to care. Thus stopping repeated attrition attacks while you are all out attacking. (Only really relevant if you are dominating preventing big attacks, but still handy).
-
Defence computers are still a laugh. 3 points per second is useless
I disagree, the DC may heal buildings at 1/10 of the speed, but it heals *all* buildings at that rate, which makes it almost as fast as a ckit in the case of mara zap. Furthermore, building it frees a player (usually a half-decent one if they are bothering to repair) to continue to concentrate on fighting despite the mara zap. Clusterbases really need these if they can afford the space.
3 points to each building does no difference whatsoever. What you want is to repair the armoury so it doesn't get destroyed and lose you the game, not heal random stuff for 3 points, which does no effect in the upcoming attack. 3 points is way too low.
Only situation I would see it handy is when a mara gets one chomp in the armoury, and you're too lazy to repair the building. Hardly worth using any build points.
-
As said it repairs everything at the same time at 3hp/s. Maybe it should just have a total repair rate, and that is divided by all buildings that need repairing in dc range? And at niveus default you can build a base with 5 reps, 3 nodes, 2 armos, 2 medis, 11 rets, dc (2 bps left over), or replace 1 node with 2nd dc (4 bps left over). So that base sure has room for it.
-
Maybe it should just have a total repair rate, and that is divided by all buildings that need repairing in dc range?
Neat idea.
-
Maybe it should just have a total repair rate, and that is divided by all buildings that need repairing in dc range?
Neat idea.
Indeed.
-
or we could have the DC have energy. it could heal other buildings by marking the DC and that specific building you want to be healed. The DC will generate energy overtime but if you want to heal, the energy will be used up. Once the building is fully healed, the DC will stop using its energy on that building and start storing up more energy. you could also set up the DC to heal multiple structures at one time but it would use up energy faster and at a proportional rate. I could see this being useful when being attacked by adv mara's where you can build 1-3 DCs and set them to heal turrets or teslas. idk just a concept.
-
Then you could camp to store more 'heal energy' and then rush...
-
what if we had upgrades that could be bought with evo's? lets say... increase dretch speed or increase secondary attack strength. something that we could buy with evo's that could last the entire game. but would take like 1 - 2 minutes to upgrade. maybe... 9 evo's from the team to get a dretch speed upgrade?
-
I remember someone was talking about it on TF forums. I think it's good idea, but for a mod, it doesn't feel like it would fit in "official" game.
-
I know the new phase isn't over yet, but can we see a new batch of stats for the current phase as it is so far? I am interested in how it compares to last phase.
-
what if we had upgrades that could be bought with evo's? lets say... increase dretch speed or increase secondary attack strength. something that we could buy with evo's that could last the entire game. but would take like 1 - 2 minutes to upgrade. maybe... 9 evo's from the team to get a dretch speed upgrade?
[cough]ns[/cough]
i think that if there were an upgrade system it'd be best to have it on an individual basis (since not everyone specializes in the same classes) with a small upgrade time (30 seconds?) that requires evos to be stockpiled and used all at the same time (to make you REALLY want to go goon, but also make you want to save up those evos for the 15% health upgrade). in NS they've got a gestation period where you're immobile while you upgrade ("you gaiz stole frum ns").
just my 1,233,564,432,788 yen.
-
Ideas are not protected by copyrights, so unless you take the actual code from NS that's no infringement. But don't you think that this system would encourage camping and killwhoring ?
-
in dom? no way. in vanilla dm trem? yeah, probably.
-
How about a new update? The game has some un-balancements still on some (most) maps.
-
Such as what? I have yet to see anything that really warrants further changing the balance.
-
Can dragoon repeat change to 600? (im pretty sure 500 is 1 per second and 1.1 was 1.5 times a second and 1.2 is 1 per second) because goon vs helmet is obvious human will win, expecially because all my strikes depend on 'luck'. No helmets is no problem, as soon as i strike BAM dead. :) and i like this feature, it's challenging to be human...
-
500 would be 2x a second, and 600 would be 2x in 1.2 seconds....
-
Can dragoon repeat change to 600? (im pretty sure 500 is 1 per second and 1.1 was 1.5 times a second and 1.2 is 1 per second) because goon vs helmet is obvious human will win, expecially because all my strikes depend on 'luck'. No helmets is no problem, as soon as i strike BAM dead. :) and i like this feature, it's challenging to be human...
Try pouncing them instead.