Kernel Panic Division Zero (PB v6)

[[XIII]Roxas]

I congratulate you on the Obelisk. I'm screwing around with it at the moment, and I must say it makes an impressive Area-Denial weapon.

[ILMTitan]

I think deploy and undeploy should have different hotkeys. Sometimes I want to select all of my units and get them moving, and I'm not sure that pressing x will do that.

[Saktoth]

If you want to get your worm to just not shoot unless ordered to, shouldnt noautofire in the units FBI do that?

[KDR_11k]

It's supposed to keep shooting once it starts shooting.

[Pxtl]

Finally got to try Div0. Have to agree with everything in zswsg's post - particularly the micro-cumbersome Worm. I'd switch it to standard decloak-to-attack behaviour with a long recloak-lag, and then default it to "hold fire". The hacker-side is already micro-intensive with the siege-tank Bugs.

That, and I want the old (Corruption) Security Hole back.

And the DOS attack needs a little bit of texture loving - the pure flat-shading makes it an amorphous blob most of the time. Still, I love the sparks while it's wheel spins.

Otherwise, though, DivO is pure awesome. Still figuring out how to properly use teh Obelisk, but it's special effect is pure win.

[Saktoth]

Cloaked units should generally default to 'return fire' over 'hold fire'.

Still, having micro units like this in a mod like kernel panic, which is all about purity and macro, is kind x_X

[ILMTitan]

1. I think hold fire for the worm would be better than return fire. Sometimes units can radar target it and you don't want it fighting back (yet).

2. The System still seems mostly macro. It's only the Hacker which is micro intensive. I think it is a good thing that helps to distinguish the two sides in ways that aren't commonly done. The risk is that at either high or low levels of play (or both), one side appears superior, but I think that is an acceptable problem.

I also agree that the DoS is the only unit that now looks out of place. It's the only one that isn't primarily black with team colored lines.

[SwiftSpear]

How many mods you made so far KDR? 6 now?

[KDR_11k]

Not sure, I haven't counted 'em .

[Pxtl]

How many mods you made so far KDR? 6 now?

I think the problem is that a lot of them feel 90% done. As in, they're gameplay-complete, but missing a little TLC to make them into a complete package - that is, a working AI, a proper gameplay guide, some recommended maps. As it stands, no matter how good the mods are, you have to learn them by trial-and-error. Not to say that the same isn't true of the major mods, but the major mods have active games online constantly that you can hop into and learn and play. Dozers, CvC, etc. do not.

I mean, I'd love to see a little pack of KDR's mods - they're consistently more fresh and easy to pick up than the major Spring mods. Perfect lan-party material. But the problem is that last 10% of the work needed to polish them up.

[zwzsg]

It seems that hacker is a bit OP at the moment. To check, I played three games in a row against Joanin, and indeed hacker always won over system despite we switching sides. (I'd link to them at www.spring-league.com however can't find the url to the page after player names has been inputed)


According to him, it's because worm builtime is too low.
According to me, it's because bug is too strong.

I looked into the FBI, and worm bt is 3200, compared to 3600 for byte.

Bug has only 400hp while bit has 600hp. However, bug does 130dmg twice a second while bit does only 80dmg twice a second.

So for next version, I'll increase a bit worm bt, and decrease bug damage. What do you think?

Edit: Just realised the bug BT is 448 while the bit BT is 240. It's odd, the bit should vastly overpower the bug with those values, and ingame I felt it was the contrary.

Edit: according to ILMTitan, it's because the Kernel, Hole, and Window, have a buildpower of 128, while the socket has a buildpower of 64, which make the CPU's minifac underpowered, and kernel bitspam over-effective.

So, maybe, I could double socket buildpower and double bit buildtime, so that kernel bitspam don't overpower hole bugspam. However, that would impact also eviless version, and althought I dislike the over effectiveness of kernel bitspam, maybe other people do. I could also make bug match bit more by halving window buildpower, and halve bug buildtime. Wait what? I was saying hacker was OP, and am now talking about nerfing kernel's bit or buffing hole's bug? lol I am so confused. Someone halp me balance KP!


Performed some test:

- I wanted to check if kernel wasn't too fast to build bits. According to the bit and byte BuildTime values, by the time the Kernel build 1 byte, it can build 15 bit. So I tested: when the 15 bits arrive one by one (spawned by a socket though) they get killed by the byte before being able to fire any shot. So at the end, the byte actually even had more HP! (from the XP, I guess). Then I spawned 15 bits at once, and told them to attack the byte. In group (somewhat spread by the travel but not that much because it was in major madness center flat area). The byte won against the 15 bits, and had 33% hp left at the end. Conclusion: despite bits being built very fast by the kernel, bits don't overpower byte. And that I need not increase bit buildtime.

- By the time a kernel builds 15 bits, a hole builds 14 bugs. I spawned 15bits, then 14 bugs, and made them meet. The bits won. With 11 bits left. Conclusion: When produced by kernel/hole, bit >> bugs.

- By the time a socket builds 15 bits, a window builds 28 bugs. I spawned 15 bits and 28 bugs, and made them meet. The bug won. With 11 bugs left. Conclusion: When produced by socket/window, bugs >> bits.

Being bewildered by having 11 unit left in both case, I redid the last two tests, this time, the number of bit left was 9, and number of bug left was still 11.

Not sure how, but I got my number of units wrong.

- When a kernel builds 15 bits, a hole builds 8 bugs (not 14). The bits still win. But with 11 bits left as well? I'm cursed by the elven!

- When a socket builds 15 bits, a window builds 16.06 bugs (not 28). The bits win. With 4 bits left. When a socket builds 14 bits, a window build 15 bugs. The bits win. With 8 bits left. Wait, redid the experiment of 15 bugs vs 15 bit, this time the bug won, with 4 bugs left. lol, I'm so confused!

So bit >> bug. Especially when kernel-built, but even when socket-built.

Though I'm not counting bugs special ability there.

So if hacker is overpowered, it can't be because of bug, can it?

Worm is clearly OP: build faster than byte, and 1 on 1, or 5 on 5, worm win. Unless the bytes are really well micromanaged. Worm can kill whole batches of bits in one shot. And get virus by doing so!

[Jonanin]

I think that Bit vs Byte is irrelevant. While someone is building byte the other person could build pointers instead. Bit vs byte won't tell us much.

My suggested changes are:

• Worm BuildTime increased from 3200 to 3500 (100 less than Byte's)
• Window WorkerTime decreased form 128 to 64. This makes it so that Socket and Window WorkerTime's are equal. Otherwise a Window builds units twice as fast as Socket.
• Multiply Bit's BuildTime by 1.25

Relevant informaton (this does NOT include my suggested changes):

It is also interesting to know: Bug and Bit kill each other the same amount of shots, AND bug has bigger range. Still, 15 bit vs 15 bug wins.

[[BUG]]
Health = 400
BuildTime = 448
Damage = 130
Range = 320
ReloadTime = .5

// Damage Per Second
DPS = 260

// A pretty good measure of unit power
DPS * Health = 1040000

[[BIT]]
Health = 600
BuildTime = 240
Damage = 80
Range = 256
ReloadTime = .5

DPS = 160

DPS * Health = 960000

[KDR_11k]

I think one factor might be that the byte is fairly weak in the overall scheme of things. A worm can attack both small and large targets successfully while a byte only succeeds vs small targets and AFAIK is easier to overrun with a bunch of spam. I was thinking about an "AP shell" mode for the byte where it loses its AOE but does much more damage to a single target for a few seconds. Then again the system has pointers which provide a deadly amount of fire support vs large targets... I think the solution might be to make bytes win the worm vs byte battle so the hacker has to rely more on DoSes to down those bytes, pointers have enough other roles so losing a monopoly there won't hurt much.

[zwzsg]

Spawning 14 bug and 15 bit at opposite side of Major Madness plateau, and making them meet in the middle, it's not always the same side which win. I think it depends how they spread during travel, if their spread their shot on diff target or target the same, that sorta of stuff.

1 v 1, the bug win. But it may be only thanks to its bigger range. When I move them in range of each other, make sure they face the right direction, and only then let them fire, they are about the same power: during experiment, both the bug and bit died at once.

A worm can attack both small and large targets successfully while a byte only succeeds vs small targets

Oh, and to me, it seems that worm is much better against small target, as it kills a dozen at once.

[ILMTitan]

A worm can defeat a byte one on one. That is also a problem. A worm does over twice the damage of a byte, but a byte only has 25% more health. Combine this with the fact that the only thing that can detect a cloaked work while remaining outside its range is a builder (or another worm), and you will have trouble taking it out with pointers.

Solutions I have briefly considered are making bytes ineadable, and giving pointers a long radar range.

[Pxtl]

Perhaps buffing Byte's firepower is necessary - it kills most traditional units in one shot anyways, so buffing the damage output would really only affect heavily armoured things like the Worm.

Likewise, the Worm's firepower could be nerfed, since it, likewise, mostly needs to be able to kill smaller units in a single shot.

That being said, I still can't figure out how to use the Hackers effectively - mostly because I can't figure out how to bring out the DOS or Exploits properly. In a pitched battle, pointer/byte teams rape my worms.

[KDR_11k]

Nope, the byte doesn't oneshot anything. Sometimes the burst will kill a bit but usually it will just hurt them badly in one shot.

[Jonanin]

Well, nerfing the worm has proved harder than I thought.

Even when I lower the worms health from 12000 to 9000, and it's melee attack damage (the unit it directly hits, not the splash damage) from 2700 to 1000, the worm STILL wins, with two left over. I think one of the problems is that the worm gets lots of experience from killing the byte. I didn't want to lower the splash damage, because if I did it might not be able to kill a bit in one shot (one that might have gained some experience)

[Zoy64]

i think the virus should be a suicide unit that explodes into a cloud of infecting gas, in which produces more suicide units

[Pxtl]

Wait, there's experience in KP? Isn't that somethign that should be gutted out for the sake of "simplicity"?