29 replies to this topic

[Space Hulkster]

Which one worked for you?

[Carrioncrows]

In other random news....

[Levi Porphyrogenitus]

I'm unsure what your question actually is about, OP.

Assuming you mean, what brought you back to the game, the answer would be that I never left.

That said, I really like the mech+bay challenges PGI has been doing recently. They're great. I also love the srm changes that are already live, and eagerly anticipate the upcoming ones set for the 29th.

[Not Bob]

Short Range Missile Host State Rewind

Versus

Free Mech Rewards.


... Yes? Can you clarify what you asking or comparing?

Edit: Yes Free mechs happened, and SRM HSR works?

Edited by Not Bob, 18 April 2014 - 10:58 PM.

[Deathlike]

SRM hitreg is waaaay better for me, but still inconsistent.

Hopefully the next patch will correct that, but... we'll see.

[Divine Retribution]

The free mech reward. It took about 2 hours after my 5th win, but I got it. SRMs are still hit and miss for me. Hopefully next patch will take care of that.

[Bobzilla]

Both are good. I'd say the free bay Is better cause hsr still isn't 100%.

[Screech]

Re-log in and your free mech will be there champ.

[Coralld]

Sense I learned to lead the target while chain firing SRMs I have been able to get around the SRM hit detection bug for the most part and the 29th patch will hopefully make it better, but the hit detection over all still needs to be improved.

Free mech and bay is always awesome.

So, win, win for me?

[LauLiao]

Space Hulkster, on 18 April 2014 - 09:19 PM, said:

Which one worked for you?

I know that most of the community is in the Oranges camp, but I'm a firm believer in Apples. I've seen all the Orange arguments, but in my own experiences, Apples are clearly superior.

[Troutmonkey]

Levi Porphyrogenitus, on 18 April 2014 - 10:54 PM, said:

They're great. I also love the srm changes that are already live, and eagerly anticipate the upcoming ones set for the 29th.

What's happening on the 29th?

[Eddrick]

Troutmonkey, on 19 April 2014 - 06:41 PM, said:

What's happening on the 29th?

Launch Modual and 3/3/3/3

[Mad Porthos]

Also, specifically, they stated they should be removing a part of cry engine called "explosion throttling" which is causing some of these SRM and probably LRM misses when compounded with HSR code. Basically, explosion throttling was a system whereby the game engine tried to make sure that lower end computers would not be overwhelmed by too many explosions and impacts having to be "calculated" at once, so would move the "processing" / checking to see what is hit into several different instances over a period of time. So for example, if you shot 3 srm 6 at a target and it looks like 16 hit, explosion throttling might have decided that ten of these should be the maximum number "processed" at the instant of impact, and then in the next frame or moment of action, it would process the remaining 6 hitting. The problem is, if the target is moving at 140kph, even though all 16 should have hit, by the time the last 6 are being processed a split second later, the target has moved a few meters and now many of those that really SHOULD have hit, are now going to register as misses... false positives.

This explosion throttling, they described as being a recently discovered hidden cause for why we have had such bad registration on exploding weapons, SRMs and very likely on LRMS as well, though I can't confirm this. Likely one of the reasons LRMS became more effective with speed increase was that once they were faster, more of them which before were missing because of this "explosion" throttling process, could actually hit because even the new "location" that the target was moved to when the additional hits came in the next frame, was able to be hit by faster missiles whereas slower missiles would be considered invalid hits because they would not have had the speed to hit the target in the new location after it had been incrementally moved as the hits were "throttled".

I know, that sounded probably incomprehensible, but perhaps Koniving will get it, eh? Point is, come APRIL 29th, something built into CryEngine, which PGI was not aware of, but that was causing problems, should be altered or removed so that we have fewer if any of this type of problem. It may mean that suddenly LRMS are even more deadly and light mechs will start dropping like flies even, stripped of the advantage their speed gave in tandem with this weird phenomenon.

[Deathlike]

Mad Porthos, on 19 April 2014 - 07:08 PM, said:

I know, that sounded probably incomprehensible, but perhaps Koniving will get it, eh? Point is, come APRIL 29th, something built into CryEngine, which PGI was not aware of, but that was causing problems, should be altered or removed so that we have fewer if any of this type of problem. It may mean that suddenly LRMS are even more deadly and light mechs will start dropping like flies even, stripped of the advantage their speed gave in tandem with this weird phenomenon.

Well, that could also explain how legged light mechs take massive LRM damage...

[BatWing]

About the SRMs, after many experiments, I fiund a viabke solution to how to have them working correctly.


Just use SRMs ONLY in chain fire.

If you have 2 or more launchers, if you fire alpha some of them will overlap and will not register.

In single fire all of them will register correctly.

[Mcgral18]

BatWing, on 20 April 2014 - 01:26 PM, said:

About the SRMs, after many experiments, I fiund a viabke solution to how to have them working correctly.


Just use SRMs ONLY in chain fire.

If you have 2 or more launchers, if you fire alpha some of them will overlap and will not register.

In single fire all of them will register correctly.

So, the same as before the patch?

[101011]

That is because nothing was changed in this patch. The actual fixes come on the 29th.

[Mcgral18]

101011, on 20 April 2014 - 01:56 PM, said:

That is because nothing was changed in this patch. The actual fixes come on the 29th.

They changed it so missiles fly through enemies instead of falsely exploding.

You get to see just how poor hitreg is.

[101011]

Mcgral18, on 20 April 2014 - 02:06 PM, said:

They changed it so missiles fly through enemies instead of falsely exploding.

You get to see just how poor hitreg is.

Technically, they made the effects server side, so it only exploded if the server showed a hit, but yes.

[Mad Porthos]

BatWing, on 20 April 2014 - 01:26 PM, said:

About the SRMs, after many experiments, I fiund a viabke solution to how to have them working correctly.


Just use SRMs ONLY in chain fire.

If you have 2 or more launchers, if you fire alpha some of them will overlap and will not register.

In single fire all of them will register correctly.

Yes, this has been the solution for a long while, but I don't think most people get why it works most likely. If you are firing in chain fire mode, you are never actually firing so many that you are over the "limit" of how many explosions is too many. It's hitting that limit that starts making the additional miss, or at least be very off target. Once again, 3 srm 6 example... fired as 6, then 6, then 6 more... all will register, whereas if you fire 18, if the "arbitrary" throttling limit was 10... you might lose 8 of them. Admittedly, due to spread, NOT all the 18 would have hit... but now even fewer hit that way because you "saturated" the max number of explosions/missiles at one time.

I think this is probably why LRM5 flooders can see effectiveness at times. They are more efficient, accurate because none of them are being "throttled" to be processed after thier target has an additional chance to move. They don't lose batches of missiles because of "over saturation". AMS does a real number on them of course, though apparently not as bad a number as "explosion throttling" is doing on massed SRMs.

If explosion throttling is removed, we might be seeing both SRMs and LRMs all counting, even in huge groupings... i.e. 4 srm 6 all hitting at once for 24 srms, or 4xLrm15 for 60 lrms, all hitting at once, with little variance in damage based on the movement speed of the target mech, since the explosions will all be essentially at the same time, not as spread out and able to be interfered with/miss due to movement that actually shouldn't have happened. 50-60 damage will be nothing to sneeze at.

Edited by Mad Porthos, 21 April 2014 - 12:10 PM.