Author
|
Topic: Net Melee Balance Mod: ALL ABOARD (Read 40652 times)
|
psydev
*Many bubbles*
 Offline
Gender: 
Posts: 136
Why don't all races have point defense lasers?
|
Shiver:
I'm not trying to "buff up the cruiser", we talked about the cruiser before and I agreed that in general, its weaknesses were appropriate. All I am noting is that the cruiser used to be a decent matchup against the Ur-Quan, since the damage it dealt was worth more than the ship value, thus making it a "risky soft counter" (cruiser can get caught in the fusion blast early w/o doing any damage).
I see that making the PDL a value of 1 would be making the cruiser too strong for its current point value. One possible offset of this would be to lower the firing rate, making the weapon not very strong but more useful vs. fighters.
With all that said, yes, the Earthling is one of my favorite ships (not my favorite in melee... tha'ts probably the Vux) but I'm trying not to be biased. I like ships that are challenging to use effectively but rewarding if used well or with luck.
I'm just noticing a very big difference in its usability vs. the Ur-quan now. The only reason I described the Cruiser as "nerfed" is that it was a pretty decent matchup vs Ur-Quan before, and now I think I would never pick Earthling vs. and Ur-Quan, even to finish him off if he was weakened. I also wonder if the cruiser is really worth 12 points...
Instead of a "buff-up", I wouldn't be necessarily opposed to a "re-imagining" of the cruiser. And yes, that's because I like it. I don't want to make it unfair, though. I guess that's for another patch, though.
|
|
|
|
« Last Edit: June 04, 2009, 08:36:39 pm by PsyDev »
|
Logged
|
|
|
|
Death 999
Global Moderator
Enlightened
Offline
Gender:
Posts: 3874
We did. You did. Yes we can. No.
|
This sounds very similar to an exchange Shiver and I had at SCDB...
|
|
|
|
|
Logged
|
|
|
|
Shiver
Guest
|
I'm not upgrading the Umgah cone. Unlike the ZFP tongue, the cone scales almost perfectly in size between zoom levels, so if I alter the size of every cone image at one zoom level then I will have to do them at all three. The cone animates between three images at any given zoom level and angle, so there are 144 cone images total. Although I would not need to manually redraw every image, I would have to redraw 27 of them and do the rest by rotation. Then I would get to do hotspotting on these images so they appear on the correct part of the drone. It's too much work. The Drone will remain as a 6 point cheapo for the sake of my sanity.
|
|
|
|
« Last Edit: June 07, 2009, 11:02:55 pm by Shiver »
|
Logged
|
|
|
|
Death 999
Global Moderator
Enlightened
Offline
Gender:
Posts: 3874
We did. You did. Yes we can. No.
|
Seems like some procedural generation would be in order for the cone. If not on the fly, once... Hmm. Specify a shape, and I'll see what I can do.
|
|
|
|
|
Logged
|
|
|
|
Shiver
Guest
|
Death999:
The cone on the right is Star Control 2's default. The cone on the left is the shape I'm going for; noticeably longer, a touch wider, and the exact same dimensions as the old cone at the very base. If it turns out that the new shape is too long, I can always go through and strip off a row of pixels myself from each image. Good luck.
|
|
|
|
|
Logged
|
|
|
|
|
|
jaychant
*Smell* controller
Offline
Gender:
Posts: 432
Please visit my homepage
|
I've noticed it. It usually happens when the photon shard (not usually the shrapnel) kills the last of the Dreadnought's crew. It never seems to happen unless the last of the Dreadnought's crew are being killed by the Broodhome.
|
|
|
|
|
Logged
|
|
|
|
Mormont
*Smell* controller
Offline
Posts: 253
I love YaBB 1G - SP1!
|
Wow, I wasn't expecting the cone to get that big a length increase. Are you planning to keep it at 6 points (if you find an easy way to do the new cone)?
|
|
|
|
|
Logged
|
|
|
|
Shiver
Guest
|
I really don't get how Ur-Quan vs. cyborg Chenjesu matches crash on a pretty regular basis for me (20% or so), and yet no one else notices it.  It seems to only happen when one of the ships loses its last crew and is just about to blow up.
The only way I can get that crash on my computer is to set two AIs against each other and leave my computer for half the day (AI vs AI games restart automatically after a few seconds). Meep-Eep and Fossil have already looked at my code back from when I was trying to nail down the Ur-Quan desync bug and they didn't notice anything. How should I, a non-programmer, fix this problem? Supposedly the next UQM release has some kind of error report tool built into it, so that's really the only chance I have.
Wow, I wasn't expecting the cone to get that big a length increase. Are you planning to keep it at 6 points (if you find an easy way to do the new cone)?
Going for a value of 8 points. The reverse zip is getting an overhaul too. If the new Umgah isn't fun or balanced, it's going back to its default characteristics.
|
|
|
|
|
Logged
|
|
|
|
meep-eep
Forum Admin
Enlightened
Offline
Posts: 2847
|
The only way I can get that crash on my computer is to set two AIs against each other and leave my computer for half the day (AI vs AI games restart automatically after a few seconds). Meep-Eep and Fossil have already looked at my code back from when I was trying to nail down the Ur-Quan desync bug and they didn't notice anything. How should I, a non-programmer, fix this problem? Supposedly the next UQM release has some kind of error report tool built into it, so that's really the only chance I have. We didn't see anything suspicious based on just the source code. A more effective way of debugging such errors is to look at the program while it is running. Which is where what you call the "error report tool" comes in. What it actually is, is some code to dump a lot of debug information, so you can easier find where exactly the sync loss occurs. In addition, with a little change to the code, you can force one side to always be one frame ahead of the other side, so that when one side detects the loss of sync, the simulation can be stopped, and you can step through the program on the other side, where the problem has not occurred yet.
It's still not going to be easy for a "non-programmer" to use, though.
An easier way to debug this code is to remove parts of your changes, and see if the problem still occurs. But that requires the problem to be reproducable, which it doesn't seem to be when you play on a single computer. Perhaps you could try something like VMWare to pretend that you have two computers. As I understand, you can even set things like latency and packet loss. Also, there's a program TMNetSim which looks like like it could be used to simulate less than perfect network conditions, by acting as a proxy. That should be easier to set up.
|
|
|
|
|
Logged
|
“When Juffo-Wup is complete
when at last there is no Void, no Non
when the Creators return
then we can finally rest.”
|
|
|
Shiver
Guest
|
It's a rare (in Resh Aleph's case, not rare) crash bug so I don't need an opponent to test with. The desync bug has miraculously disappeared through me arbitrarily rewriting bits of code until it went away. I am not keen on repeating that process again right now.
|
|
|
|
|
Logged
|
|
|
|
Death 999
Global Moderator
Enlightened
Offline
Gender:
Posts: 3874
We did. You did. Yes we can. No.
|
Hmm. I'm analyzing the distribution of colors... ah screw it... it'll look different but possibly cooler. Or really lame. Anyway, I have a plan. Like the cylons.
|
|
|
|
|
Logged
|
|
|
|
|
|
Death 999
Global Moderator
Enlightened
Offline
Gender:
Posts: 3874
We did. You did. Yes we can. No.
|
I have the code completed to recolor the cones, and it seems indistinguishable from the original. Now all I need are 2 ingredients:
1) I need to know what image size to use. Should there be some consistent scheme here, or should I just have each of the 3 distinct rotations have a particular size and you'll take care of the offsets when the time comes?
2) a painting program that does good pixelwise rotations. I've tried two programs which were simply inadequate.
Fortunately, we only need two rotations - 22.5 and 45 degrees. The rest can be obtained trivially by flipping.
I have another more promising one that I think will get the job done, but I have to stop on this for now.
One last thing. The umgah cone contains black pixels. Should I make these transparent so you can see shades of the covered ship, or should they remain black? I think transparent would be cool.
|
|
|
|
« Last Edit: June 10, 2009, 04:05:46 pm by Death 999 »
|
Logged
|
|
|
|
Shiver
Guest
|
1) I need to know what image size to use. Should there be some consistent scheme here, or should I just have each of the 3 distinct rotations have a particular size and you'll take care of the offsets when the time comes?
Hotspotting the images so they line up in the right place is awful, but that is better left to me. I've given you approximate dimensions for the new cone already. Remember that you need to do three animations of each zoom + angle combination.
2) a painting program that does good pixelwise rotations. I've tried two programs which were simply inadequate.
Fortunately, we only need two rotations - 22.5 and 45 degrees. The rest can be obtained trivially by flipping.
I have another more promising one that I think will get the job done, but I have to stop on this for now.
Try Gimp I guess.
One last thing. The umgah cone contains black pixels. Should I make these transparent so you can see shades of the covered ship, or should they remain black? I think transparent would be cool.
Use black pixels.
|
|
|
|
|
Logged
|
|
|
|
|
