OFFICIAL: Combat Discussion

[markb50k]

two tidbits:

 

I had a code bug that prevented ships on ATTACK, DEFEND, or PATROL order from moving towards their targets, they just sat there.  if they were in range they would fire.  thats why the Hammerhead killed some TDs over the last few turns.  The TDs and ACTISs were just sitting there until i fixed the code then they immediately jumped into battle and the ACTIS were just in a better position to start with.  Sorry about that.

 

Also, remember the usefulness of Interdictors.  It will slow down battles a bit because it will prevent tactical microjumps.  It also provide a nice defense barrier for when you go to sleep.

[Turgidson]

the accuracies were pretty bad already for the HTLBs, so their improvement still makes them alot worse than TLC and LCs, but maybe we should run some tests.  Position MNFRGs at around 28KM from each other, not in ATTACK mode so they dont close the gap, and let them fire at eachother.  See how often they hit.  IMHO if they hit around 50-75% of the time then the accuracy is good.

Considering the relatively small size of the target, and this maximum distance, 50-75% accuracy is quite in the high zone IMHO. Considering what I understood of your hit model (random angle offset depending on accuracy - note: the BIG difference to HW2 being there's no minimum), a 20% hit average against a small target might still mean a very high rate (100% ?) against a larger one - assuming linear distribution of the offset, it'd mean a 5 times larger target would be hit 100% of the time.

 

 

As for fighter weapons, I saw how quickly those TDs destroyed those SACHN's earlier, like a couple of cycles, so I went with this.

When I saw the TD's firepower rate, I couldn't resist buying myself quite a lot of them.  ;D

 

However, it still seems that for many of the frigates, the balance between firepower and healthpoints is pretty much shifted towards the earlier, so that even these won't take very long to destroy their opponents. For larger capships, the balance moves more towards the healthpoints (even though the durations aren't that long...), at least relatively...

 

 

Now, lets test it out to see how it goes.  Your TDs are making mincemeat of the ACTISs in that last turn, so they can be effective anti-starfighter platforms.

They aren't bad, but the question is: wouldn't a LFRG be much better ? I agree that LFRGs are slower, and that since there's no/little warheads in use the interception of bomber runs hasn't really occurred yet. The tactical jumps still allow the LFRGs some speed compensation (despite higher jump calc time)... and anyway, with jumping fighters and bombers, how can an interception attempt even occur ? The bombers will be instantly in range, at any case way before fighters (or any other unit...) get the opportunity to shoot at them... and for a close range "air" superiority/control, fighter's patrol speed advantage over a capship is negligible.

 

Edit: OK, with interdictors this will change quite a lot. Thing is, still... we haven't seen that many interdictors in the movies, and even a certain amount of EU battles didn't explicitly involve interdictors... so battles still occur without them, and people don't jump out immediately. Maybe it takes a lot of time to ready a capship to jump, maybe they're much more sensitive to line of sight, gravity environment, whatever... Or maybe it'll just be impossible to jump when too close to a planet, which is perhaps the way things should/did happen.

 

I welcome your thoughts.  The penetration idea has merit but I would like to get more combat reports.

No problem.

Edited May 29, 2008 by Turg

[Turgidson]

Hadn't seen the later logs... wow, seems the ACTISes haven't scored a single hit since. I'm quite surprised by this: their turn rate is similar, and other differences don't seem sufficient to explain why the ACTISes wouldn't even have been able to line their weapons once.

 

So, my question is: how is movement computed, and especially targeting ? In case the units move one by one, sequentially and before the "firing" calcs are done... then if the ACTISes are unlucky to start first, they'd line the TDs properly, then the TDs would move, disalign themselves from the ACTISes' gunsight, while getting a nice spot on them... that'd explain these results, and also how TFs were able to defend pretty well against TDs (CJ's experience, IIRC).

 

Dunno if my guess is right... the result of the last two turns seems just "too big to be believable".

 

In case my guess is right: I'd suppose we'd have to use a 'trajectory prediction" by saving the last speed vector(s), have each unit target this predicted spot, then do the movements. That way, unit's movement is fully independent of the processing sequence. If that's already how it happened, I really wonder how the ACTISes got slaughtered in the last turns...

Edited May 29, 2008 by Turg

[markb50k]

As for movements it used to be linear, but i just put a code change to randomize what order stuff does its orders, targeting and attacking, so that should be minimized.

 

i havent done an indepth study of how the turnrates would affect things but my guess is that with dogfighting, it will have a major effect on who can get in a firing position.  i wonder what will happen if two TDs target eachother and get in a dogfight.  its a good test.  do they get into an infinite dogfight?

[Idle_Scholar]

Themicrojump does appear to be the tactic of the moment. In that combat the NFRG and CORHMR exchanged places about three times before I thought screw it and decided to engage. If I'd had anything to actually lose I'd have jumped the CORHMR streight out of there and waited for the shields to regenerate.

 

With regards to the ACTIS I had 74 at the start of combat, and I think they killed about 20 or so TIED's.

[Turgidson]

As for movements it used to be linear, but i just put a code change to randomize what order stuff does its orders, targeting and attacking, so that should be minimized.

 

i havent done an indepth study of how the turnrates would affect things but my guess is that with dogfighting, it will have a major effect on who can get in a firing position.  i wonder what will happen if two TDs target eachother and get in a dogfight.  its a good test.  do they get into an infinite dogfight?

The randomizing would be fairer, but, still... half the fighters just wouldn't have a chance to hit anything by engine design.

 

As for turn rates, TD and ACTIS have the same one if I'm not wrong...

 

 

Regarding the microjump: this does show that someone being active/online has an advantage over someone who isn't...

[Idle_Scholar]

Regarding the microjump: this does show that someone being active/online has an advantage over someone who isn't...

 

Which kinda sucks if it counts too much. I mean I'll probably be in full time employment in a few months and I'd hate to be playing a game rulled by 12 year old kids and unemployed insomniacs.

[EvilleJedi]

I'm in the same boat, I may have less than 10 minutes to actually check and play each day because of a very busy work schedule (and no way of getting around work firewall either)

[Deathwatch]

the accuracies were pretty bad already for the HTLBs, so their improvement still makes them alot worse than TLC and LCs, but maybe we should run some tests.  Position MNFRGs at around 28KM from each other, not in ATTACK mode so they dont close the gap, and let them fire at eachother.  See how often they hit.  IMHO if they hit around 50-75% of the time then the accuracy is good. 

 

I can't say much about this, but Turg and Skyblazer's ships have been trying to shoot my TDs from a rather long range for 2 turns now. They don't seem to be hitting them though (fortunately ), which would make sense considering the range. At point blank they seem to be happily hitting fighters now though, but I have to admit that that fight was at point blank.

 

Microjumps do sometimes seem to be a necessary evil considering the size of the sectors and they allow for the only way out of certain situations due to the movement being slow sometimes. But maybe to limit their effectiveness, maybe have a sort of "cooldown" for them on the larger ships.

[skyblazer]

Hmm, I'm about to Have an interesting engagement with 2 players. One Named Deathwatch the other Kobra and U know something I've noticed? They're not shotting at one another lol. Silly boys why go ally with each other when its more fun to stab in the back

 

Anyways this should be an interesting fight.

[Idle_Scholar]

I almost got one of their RGUN's and the other's on half shield so that should give you more of a some chance.

[skyblazer]

What did u engage them with?

[Turgidson]

I can't say much about this, but Turg and Skyblazer's ships have been trying to shoot my TDs from a rather long range for 2 turns now. They don't seem to be hitting them though (fortunately ), which would make sense considering the range. At point blank they seem to be happily hitting fighters now though, but I have to admit that that fight was at point blank.

IMHO, HTLBs and even TLBs should have a very low chance of hitting fighters. Even at point blank, as their mechanics aren't visibly fast enough to properly track fighters. Watching the movies, heavy gunnery seems rather inefficient at getting fighters down... as should be the case.

[Deathwatch]

Very true about the (H)TLBs - they aren't meant to be anti starfighter. You can also see it with ships made for certain roles (for example LFRGs). It's still funny to see them trying to shoot them at that range though.

 

RGUNs are slower than snails, they are nicely armed but they are just too slow using jumps in my opinion. But, they are quite strong, so that's ok.

 

BTW are any of you using scout vessels? I did get several of them and they do seem to make detecting things much better.

[Idle_Scholar]

What did u engage them with?

 

I jumped a pair of THRNTA in (the Hobgoblin and Sly Fox) and engaged at HTLB range. I basically did damage while they were attempting to jump out.

[Deathwatch]

Well, I got decimated there - although I have to admit I had no time to manage my fleet other than moving it slightly and setting up some priority targets.

 

From what I can tell, this is also where it went wrong - as I set the ships to fire at priority targets based on ship type (like the small cap, med cap, small craft for the CRRCK, etc), it ended up causing some chaos, where units ended up targetting (and moving against) units that were way too far away, such as Antilles' THRNTA's which were hundreds of kilometers away.

 

But well, I'll be back

[Idle_Scholar]

Did your RGUN fire at 'Blazers stuff with its HTLB?

[Deathwatch]

Yups, the RGUN did fire at the random ships (maybe due to arcs being able to target certain ships only, I can't tell right now as I don't have java installed on this pc), the LFRG never moved into range as I did intend for it to do, it just turned away to head towards the other ships.

 

I reckon it was the combination of using the Attack command with the priority targets that caused the behavior, so perhaps next time I'll end up having to use a more direct set of commands or perhaps Patrol or microjump them.

 

Overall, the CRRCK is a nice vessel - pretty decently armed for its price, reasonable speed as well. It does lack some more point-blank firepower though. Unfortunately, the RGUN was too slow (in my opinion) to be a good escort for it, if you rely on jumping it regularly.

[General_Antillies]

Yah I noticed the same thing. I had my ships target fighters and they went all over the place . Also, just for something new, would anyone want to do a team battle, something small like 3v3 or even 2v2? I just want to see if its possible to coordinate with there turns flying around so fast. Also, might be nice to not always see three times as many red ships as you have, even if they are all against each other.

 

Also, I'm confused as to how movement per cycle works. For sublight speed, if a ship has a max speed of 600 m/s can it go 600m in a cycle? Same for turn rate, 65 deg/s= 65deg/cycle? As for lightspeed, if a ship has a jump speed of 3xL what does that mean? Also, how does Jump Calc Time factor in?

[markb50k]

600m/s means 600m/cycle

same for 65 deg/s

 

jump speed is 10 sectors per cycle (thats for a 1.0xL speed ship).  3xL means 3 times faster

 

calc time happens before the jump.  so total time to get somewhere is calc time  + jump time

 

For those noticing crazy behavior for attack orders, ill see about implementing a nearest target first kind of targeting.  it will add more calculations and may slow down the turns but it has its merit.

[Turgidson]

For those noticing crazy behavior for attack orders, ill see about implementing a nearest target first kind of targeting.  it will add more calculations and may slow down the turns but it has its merit.

However, if units have already engaged a target, I'd suppose they should continue - at least as long as it is in range. If units keep switching back and forth on "closest" unit, this might become quite inefficient at the end...

[Idle_Scholar]

Maybe 1 cycle should equal 2 seconds. It just seems that ships are really slow in relation to the speed and range of firing.

[skyblazer]

Yeah In that round I preset targets for my units before I went to sleep or tryed to sleep anyways. The Carrack was at the top of the list of things to blow up.

 

From what I could tell my Ranger Gunship that went in with 5% shields was taken out in first round of combat with my Neb B frigate becoming the main target after that. My Neb B is still alive though with 19% hull left to it and now my Attack Drones which I had 72 of are attacking Kobra's Bwings and TD with a 5 to 1 kill ratio in Kobra's favor.

[markb50k]

Maybe 1 cycle should equal 2 seconds. It just seems that ships are really slow in relation to the speed and range of firing.

 

that is something I could tweak pretty easily, if it is agreed that it is a problem.

[skyblazer]

Something I've noticed lately is my fighters aren't moving to engage lately. They can be 5k of a enemy fighter or ship and they just sit there regardless of telling them to use attack, defend or patrol. its really annoying.