647 replies to this topic

[Voidsinger]

Sagamore, on 13 March 2013 - 06:52 PM, said:

I have to say I did notice it more after the Host State Rewind. I may as well shelf my three mastered Commandos for the moment and lumber around in an Atlas.

Actually, at this point Host State Rewind hasn't been implemented for ballistics and missiles. When it is, expect an extremely low life expectancy on the battlefield in any light other than a Raven (due to it having broken hitboxes).

I guess my point is that the armour allocation system already takes splash damage due to distance into account by making armour points not dependent on area. So, this is a doubledip on damage for smaller mechs.

[MischiefSC]

Tested it for myself for giggles and yeah, this is exactly correct. What's strange is that in thermal you can see it happen - you'll see the flashing 'heat' spots on the points that take the splash damage. Catapult seemed to suffer from this - I also think it's why the new Artemis seems so nasty. Is splash damage cumulative?

For example if I put 10 LRMs into a single point does the splash damage radius expand? Perhaps the targeting for Artemis works a bit more like SSRMs in that it focuses the damage towards the mechs center mass and thus the splash damage is more likely to reach appendages?

[HRR Insanity]

Amaris is my hero.

To PGI:

1. Remove splash damage from missiles.
2. Build a testing suite to confirm that all weapons are doing the expected damage vs. all 'Mechs as they are implemented.
3. Run the test suite after any changes to weapons, mechs, or damage code.

[Felicitatem Parco]

I'm moving this thread to Patch Feedback so it may receive more direct Dev attention.

[M4rtyr]

Well I was reading the forums while the game patched, guess I'm not going to play after all until this gets fixed.

To be honest for something to this extent to go live, even in a beta, is a joke.

But I suggest the removal of the splash mechanic as well. There is no reason to have a splash mechanic with mechs, as one person stated earlier, exposions and shrapnel do nothing to hard targets. What does damage is a directed warhead strike.

But if the devs just have to have the stupid splash damage then besides just fixing this bug they also need to reduce missile damages and change it so smaller mechs aren't at a disadvantage to splash damage compared to larger ones (smaller hit boxes mean more splash).

On a side note, still need to make streaks lock alot harder to get/maintain, they aren't meant to be easy hit weapons they are just not supposed to fire unless you do hit so you shouldn't have a lock unless you're reticle is on the target.

[shabowie]

HRR Insanity, on 13 March 2013 - 07:01 PM, said:

Amaris is my hero.

To PGI:

1. Remove splash damage from missiles.
2. Build a testing suite to confirm that all weapons are doing the expected damage vs. all 'Mechs as they are implemented.
3. Run the test suite after any changes to weapons, mechs, or damage code.

Plus we need detailed logs that show way more information for us to actually beta test.

[Eddy Hawkins]

Prosperity Park, on 13 March 2013 - 07:07 PM, said:

I'm moving this thread to Patch Feedback so it may receive more direct Dev attention.

could you put this back into general? the comunity needs to read this and moving it to the Patch Feedback will kill the thread.

[WardenWolf]

Eddy Hawkins, on 13 March 2013 - 07:31 PM, said:

could you put this back into general? the comunity needs to read this and moving it to the Patch Feedback will kill the thread.

We'll just have to keep it alive with further content, suggestions, data, etc - until we get a Dev response

Since I didn't know how to embed video an hour ago, and my post with the video evidence link is now a bit buried, here is the proper embedded video showing this bug in action on live servers:

[stjobe]

Eddy Hawkins, on 13 March 2013 - 07:31 PM, said:

could you put this back into general? the comunity needs to read this and moving it to the Patch Feedback will kill the thread.

I was tempted to post this very thing, but then I thought "what does this need more; dev attention or player attention"?

The bug has been thoroughly described, tested, re-tested, confirmed on live, a dozen test scenarios have been shown both in text form, in screenshots, and in videos.

I think this one is better off with more dev attention than player attention at the moment.

Having said that, I'd also like to say: Well done, community. It's absolutely lovely to see the passion to have this game work properly.

WardenWolf, on 13 March 2013 - 07:35 PM, said:

Since I didn't know how to embed video an hour ago, and my post with the video evidence link is now a bit buried, here is the proper embedded video showing this bug in action on live servers:

I just minutes ago updated my OP in the other thread with a link to your video in this thread, so now it's right in the OP there.

Edited by stjobe, 13 March 2013 - 07:42 PM.

[OneEyed Jack]

p4r4g0n, on 13 March 2013 - 04:44 PM, said:

Interestingly enough, PPCs which also have splash damage (or are supposed iirc) only do about 50% of the listed base damage of 10 from 100m on a Comm 1B' CT. One salvo to the Comm reduced overall health by 5%.

Davers, on 13 March 2013 - 04:53 PM, said:

PPCs are not supposed to have splash damage. They are supposed to deal their entire damage to one location.

PPCs do not have splash damage in either intent or practice. I don't know where that fallacy started, but I assume it's from people using more than one in different locations hitting targets obliquely, so that they are impacting on separate locations due to convergence.

Selfish, on 13 March 2013 - 05:26 PM, said:

However, Testing Grounds is NOT indicative of in-game mechanics. Lasers start recycling once fired in TG, they recycle once the beam dissipates in game. It's not crazy to suppose some of the damage mechanics aren't working as intended as well.

I haven't been in-game today besides the testing grounds but, last I knew, the bug/change to cooldown was working in live as well, and also impacting missiles. It's not a testing grounds bug.

Gandalfrockman, on 13 March 2013 - 06:34 PM, said:

Honestly, I think they just need to tone the splash radius on missiles WAY down, I think people are overcomplicating how this is calculated, in the crytek engine there just isnt that much abstraction going on

When the missile hits, each section That intersects the 1.5 meter sphere centered on the point of impact takes the full damage.

On small mechs like the commando, that means it catches several sections at once.
On larger mechs it might catch two from time to time. I'm really only worried about the small mechs issue.

Make the radius for missiles tiny, and see if it works, Im fairly certain we dont care about the splash on missiles, it isnt a large enough space for near misses to even be tactically relevant.

You haven't been paying attention. The missiles are actually doing more damage than they are supposed to do when splashing. Not just doing max damage to all affected locations, but doing more than max damage. A COM-1B has 12 armor on each front side torso, stock. I fired 4 total missiles and had done internal damage on both side torsos. That means missiles that were intended to do a combined 10 damage did at least 13 total in each 2 different locations, in addition to the damage they did elsewhere. 13>10

IceFyre, on 13 March 2013 - 06:42 PM, said:

Absolutely great work, seriously. But here's the thing...Streaks, LRMs & SRMs are missiles, specifically guided and unguided missiles with warheads that detonate on contact. They are not penetrators with a delayed fuse. These weapons (missiles) do and will cause splash damage in the real world. End of story. I personally would like MWO to stay rooted in the physics of the universe I live in.

Shaped charges are used to penetrate armor for a reason and do not do equal damage in all directions. Conventional explosives without a shaped charge would do very little to even modern tank armor, let alone the ablative armor of a mech, without a much larger yield.

[Deamhan]

I think the best they should be able to do is make it so that the missile produces an 'explosion sphere'. This sphere spreads out extremely rapidly and is not capable of passing through hit boxes. Therefore, where this sphere comes in contact with a hit box, it will create a contact area on that hit box. If it comes in contact with multiple hit boxes, it will create a surface contact area for each hit box.

Take the contact surface area of each hit box, divide it by the total contact surface area, and multiply by 100 to determine a % for each hit box. That becomes the % of the total damage that is applied to that particular hit box.

Eg. Missile hits the RT. The explosion sphere spreads out and comes up against the right arm, right torso and center torso. Combined surface area of 4 square meters. The arm is 1 sq m, the RT is 2 sq m and the CT is 1 sq m.

That would mean that the arm takes 25%, the RT takes 50% and the CT takes 25%.

With the missile doing 2.5 dmg, the RA and CT take 0.625 dmg each while the RT takes 1.25. Total damage applied to the enemy mech...2.5 dmg as it should be.

[Maple Nachiman]

how much of the results are wonky due to ammo explosion damage?

[stjobe]

Maple Nachiman, on 13 March 2013 - 07:44 PM, said:

how much of the results are wonky due to ammo explosion damage?

None, since the bug is evident without breaching armour.

[krolmir]

Just got done with more testing, this glitch is actually making the tougher mechs tougher, and further weakening the lighter chassis mechs of every weight class.

I.E. shot a Jenner D at 50m with a 2 Large Lasers shots - 6% health loss (18 damage); shot same Jenner D at 50m with 1 volley SRM6 - 14% health loss (15 damage (that did 2.3 times the percentage of damage that was done by 18 laser damage?))

I.E. shot 1 volley at 50m SRM6 into a Commando mech dropped to 53% health then exploded, (exited TGs and reload so I could test LLaser x 2); fired x2 at 50m Large Laser shots to the CT - 8% health loss

Both chassis took higher than expected damage from the SRM's, with the Commando getting the worst of it.

I.E. on an Atlas D at 50m 2 x Large Laser shots to CT caused 2% health loss, 1 volley SRM 6 at 50m caused 3% health loss

I.E. on the Awesome 8Q at 50m 2x Large Laser shots to CT caused 3% health loss, 1 volley of SRM6 to the CT at 50m caused 8% health loss


The story was the same with the Cicada and the Centurion, as well as the Cataphract and Catapult, thou not as drastic ( I feel that this is caused by the chassis's weight difference being less for the heavies currently in the TGs.

I think the only way to fix this is to remove splash damage until it can properly implemented.......

Edit: I also forgot to add that I only did these test on fresh armor, so no wonky numbers from ammo explosion etc.

Edited by krolmir, 13 March 2013 - 08:19 PM.

[Scop]

I tried my own test, taking my Founder's Atlas with its single Streak into Training Grounds and seeking out the AS7-D to see if I could repeat Amaris's streak-kitty results by shooting just two at a time. I set up 50m away from the dummy at a low angle and began the trial. The paperdoll indeed displayed splash damage over multiple sections, usually a torso section but the legs and head were also affected.

At Shot #25 (50 missiles launched), the mech overall was at 83%. CT was red-orange, RT was orange, LT goldenrod, both legs were yellow, and the arms were untouched. I then moved to behind the test subject, at 55m from a low angle again. (In an earlier, poorly-recorded test I discarded, I first noted the arms started receiving splash when shooting in the rear). From behind, 14 SSRMs stripped the RCT clean of armor. At that point the dummy read 77%, the RTR was red, the LTR goldenrod, and the right arm yellow.

I dropped again to repeat the test, fortunately landing on the same map, allowing me to duplicate the same distance and angle. This time, 48 SSRMs stripped the CT completely and left yellow internals, with an orange-red RT, goldenrod LT, and yellow head, right arm, and both legs. The overall percentage was 78%. For completion's sake I shot the last volley from the front, making the RT red and leaving the percentage still at 78%. I happened to observe in thermal vision and saw the heat splashes that seemed to correspond to damage splash. When I moved to the rear arc of the subject, 14 SSRMs left the RCT armor-free with darker-yellow internals, the RTR red, LTR goldenrod, and right arm yellow. The percentage at that point was 71%.

I only pursued my tests to armor removal because I'm leery of critical hits to the gyro, engine, or ammo explosions throwing off the percentage reading, and I can get a definitive measure of armor HP while I'm unsure of internal HP.

Next I'm going to create a paperdoll color grading graphic by whittling down an AS7's armor with an AC/2. This should help estimate remaining health of armor and internals in all cases, including what we're testing here. Comparison with it will hopefully allow more specific percentages than the overall we get in targeting info. I'll then repeat my tests and use it to estimate armor fractions.

[Deamhan]

Hmmm. I was looking over my weapon stat for the streak and......

502 shots fired with 333 hits.

Each time the weapon is fired (launching two missiles) it counts as a shot.
As long as one of those two missiles hit, it is considered a "hit".

So while this is not perfectly accurate, it would be a close proximation.

At 66.33%, 502 shots is 1004 missiles with just a 66.33% hit rate meaning that roughly 666 (rounded up from 665.9532) hit for a total damage of 1,041 dmg. which is 1.56 dmg per missile. Now if about half of these "hits" is because only one of the missiles hit and not both, then you can increase the dmg per missile by 50%.

.78 + 1.56 = 2.34

This brings it much closer to the 2.5 per missile.

But lets go extreme and assume that every hit was only from one missile. So instead of 666 missiles, it is 333 missiles for 1,041 dmg. That is 3.126 dmg per missile but with only one missile out of two per shot, that is only 3.126 dmg per shot. Not only is it still only .6 off from the 2.5 dmg per missile but in this particular case, it is less damage per shot than max. With the max being 2.5 dmg * 2 missiles for a total of 5 dmg per shot.

[Monky]

Deamhan, on 13 March 2013 - 08:19 PM, said:

Hmmm. I was looking over my weapon stat for the streak and......

502 shots fired with 333 hits.

Each time the weapon is fired (launching two missiles) it counts as a shot.
As long as one of those two missiles hit, it is considered a "hit".

So while this is not perfectly accurate, it would be a close proximation.

At 66.33%, 502 shots is 1004 missiles with just a 66.33% hit rate meaning that roughly 666 (rounded up from 665.9532) hit for a total damage of 1,041 dmg. which is 1.56 dmg per missile. Now if about half of these "hits" is because only one of the missiles hit and not both, then you can increase the dmg per missile by 50%.

.78 + 1.56 = 2.34

This brings it much closer to the 2.5 per missile.

But lets go extreme and assume that every hit was only from one missile. So instead of 666 missiles, it is 333 missiles for 1,041 dmg. That is 3.126 dmg per missile but with only one missile out of two per shot, that is only 3.126 dmg per shot. Not only is it still only .6 off from the 2.5 dmg per missile but in this particular case, it is less damage per shot than max. With the max being 2.5 dmg * 2 missiles for a total of 5 dmg per shot.

It is far more likely that damage reporting is simply bugged for missiles. We are able to see on live that an SRM2 objectively does more damage than a medium laser

Edited by Monky, 13 March 2013 - 08:25 PM.

[Jack Lowe]

and yes missles are screwy live game as well, however it's a little harder to notice. Was getting ready to start taking screen shot's and scream WTF. If I get another clear occurance I'll get a screeny to post here. I'm using a centi. I've seen some weird things. Getting one shot by a splat cat, not so bad right. Except in one salvo I got nothing left but a head and legs. Got shot in the front but front and rear armor on the torso's was gone just nothing left but head and legs literally. Couldn't of been an ammo explosion I don't keep ammo in any of those locations if I carry it. That's happened a couple times. Another instance I took lrms to the front armor I made CERTAIN of it cause my rear armor was got chewed up earlier on tangling with a raven. I died but I still had front armor spash damage must of transfered to the rear unarmored portions and poof dead mech. Yes I seen instances of it both on the recieving end and giving end. It does seem to be inconsistant in some way or another perhaps because of movement or the speed being traveled by the target or target plus shooter don't know. Either way I'd rather have it working properly.

[Gandalfrockman]

M4rtyr, on 13 March 2013 - 07:17 PM, said:

Well I was reading the forums while the game patched, guess I'm not going to play after all until this gets fixed.

To be honest for something to this extent to go live, even in a beta, is a joke.

But I suggest the removal of the splash mechanic as well. There is no reason to have a splash mechanic with mechs, as one person stated earlier, exposions and shrapnel do nothing to hard targets. What does damage is a directed warhead strike.

But if the devs just have to have the stupid splash damage then besides just fixing this bug they also need to reduce missile damages and change it so smaller mechs aren't at a disadvantage to splash damage compared to larger ones (smaller hit boxes mean more splash).

On a side note, still need to make streaks lock alot harder to get/maintain, they aren't meant to be easy hit weapons they are just not supposed to fire unless you do hit so you shouldn't have a lock unless you're reticle is on the target.

I think once they fix this so streaks are doing the intended damage, They will probably be fine, If You actually do the math streaks just arent that damaging, I never understood all the complaints, BUT I DONT TYPICALLY PILOT LIGHT MECHS, So Its unlikely I would ever have been on the receiving end of this.

When working correctly streaks will be Fire and forget, accurate, and remarkably easy to use, but they will hit about as hard as a wet napkin. (This is about how hard they hit heavies and assaults)

For Newer players, they will be a nice easy starter weapon.

But for more experienced players they will be an ideal weapon for those who want to focus on evasion and mobility to such an extent that they are willing to sacrifice much of their firepower.

[Felicitatem Parco]

- My Personal Observations -

I just tested this in my Atlas in the training grounds, comparing Streak SRM2 to normal SRM2 and a Gauss Rifle.

Against small Mechs, the Streak SRM2 and the normal SRM2 are causing more damage per volley than a Gauss Slug (approx 1.5-2x), HOWEVER, there is no discernible difference between the damage caused by Streak SRM2 and Normal SRM2.

What I observed is a case where ALL SRMs are causing too much splash damage transfer, and it adversely affects Light Mechs due to the proximity of their body parts. I'm sure this has been brought up before, but I just checked it myself and felt like spouting off a little.

Edited by Prosperity Park, 13 March 2013 - 08:38 PM.