How reliable is draw damage?
blackholedreams
Join Date: 2004-02-04 Member: 26023Members
Join Date: 2004-02-04 Member: 26023Members
I feel like I've been having issues with hit detection at an abnormal rate in this game so far, and while it may just be my less than stellar FPS (20-50), I noticed something odd. Using a pistol on a whip today, I emptied the magazine and saw what seemed like the appropriate damage drawn. I then emptied two more magazines into the thing, but I noticed a drastic difference in the damage drawn. Firing full speed, I saw the occasional 20 and 40 interspersed with nothing. Normally the draw damage keeps a running total, but what I was seeing implied I was missing. Which should be impossible, as I was just outside of its attack range aiming center mass on a stationary target. What the hell is going on?
Comments
draw damage is 100% reliable as it is sent to you from the server.
What isn't reliable is hit reg on buildings that are flinching (there may be another hit reg reason, but this is the most likely one in this case). This is a known bug and you may very readily reproduce it yourself by shooting at an alien harvestor. I had no idea the same thing was occuring for whips. Where on the whip were you shooting? Base, middle, tip?
With the way ns2 handles instructions, i don't think bullets can disappear into a 'network void'. FPS volatility might cause slight input delays however which cause the timing of your stamps to go out but thats really my own conjecture and is probably wrong. This would only affect moving targets anyway and not against stationary targets. So it is most likely the structure flinching rego bug thing. You can type collision in console in your own server with cheats 1, dev 1 on and you'll be able to see the hitboxes for structures if you want to test/experiment more yourself.
The workaround that i've found is just to shoot the base of the whip in order to get reliable rego. Shooting the base of any structure seems to minimize the amount of damage loss from structure rego.
But all in all i think that the marines ones are pretty spot on.
But all in all i think that the marines ones are pretty spot on.<!--QuoteEnd--></div><!--QuoteEEnd-->
This is because NS2's armor and damage type design is horrible (sorry). In no other game I have ever seen is effective health scaled by damage type instead of the damage being scaled. It should always take the same amount of damage to kill something, especially if you decide to make the damage numbers accessible to the player.
This means that to kill 570 armor you need to do 1140 damage with a weapon that does "normal" damage (which are all weapons except for pistol, mines and miniguns), but 2280 damage with a weapon that does "light" damage. This is extremely confusing and leads to people not trusting the damage numbers and hitreg. E.g. it takes 90 rifle damage to kill a skulk, but the 25 damage per shot pistol needs 5 shots since the skulk has more effective health against light damage.
More about damage types in this post here (maybe I should do a sticky post about this?): <a href="http://www.unknownworlds.com/ns2/forums/index.php?s=&showtopic=123563&view=findpost&p=2011762" target="_blank">http://www.unknownworlds.com/ns2/forums/in...t&p=2011762</a>
draw damage doesn't seem to be very reliable to me. I have also shot a player, had no damage draw, but had the player confirm that they did take damage.
That would be the most obvious problem ever. This is an intended mechanic.
If you don't hit your bite on the center of your screen you will do less damage (either 50 or 25). This was introduced shortly before release and without this mechanic your 50 and 25 damage bites you have now would do <b>no damage at all</b>.
Anecdotal evidence ("I feel like I should have hit") is also of no use at all.
It only takes several hundred instances of -150, -190's on non-carapace skulks that still run away to realize damage numbers are broken or something much more important is broken.