So, Hit Reg
#1
Posted 30 March 2015 - 01:18 PM
On that evening, I dropped in a couple of PUG solo queue matches, with the expected above results.
Then I decided to do a drop in CW (before the event) and suddenly... Hit reg was working incredibly perfect. Lights zooming by at maxmimum speed? No problem, one alpha strike, dead light mech. Storm crows have difficult hit boxes? Not at all!. The weapons absolutely hit perfectly where I aimed at, and did apply full damage.
The result:
Now, I didn't reconnect or changed anything inbetween the pug solo drops and this CW match. The ping and tracert was the same. And I did not even tried very hard to play good in this CW match, it was a "fool around" drop.
So I start to suspect that hit reg actually is not affected at all by your ping or your route to the game servers (as I thought all the time). It seems purely depend on which server from PGI your match is hosted on and how much the overall load is on that cluster.
Meaning that PGI seems to intentionally run MWO on too low resources to save money, keeping it at the bare minimum of being playable.
Call me paranoid, but I got no other explanation...
#2
Posted 30 March 2015 - 01:23 PM
how many C-LPL's can a orange Grasshopers CT eat facing your frontally? answer: 6 and and still running.
GG hitreg.
there were for me soem matches where mechs ate 3x the firepower they should have. how many CT hp has a TDR? dunno ran around with orange armor afetr eating 3 clear alphas non moving. so 150 armor at last,.
Other matches then were just fine.
I really wonder if this is a bit luck dependend on which of the slices or wherever mwo runs onto. gett he wrong core or such and boom hitreg gone. Play on the other core, hitreg = fine. It's like having no real pattern.
#3
Posted 30 March 2015 - 01:42 PM
The other day I was in my Battlemaster- 6MPLs and 1LPL, and alphad a Kitfox as it was standing still, straight to the face.. barely scratched it. The Kitfox turned to run away, and I got off a second alpha dead center of its back, and it ran off whithout hardly taking any damage. Two Alphas, one to the front and one to the back and not a single red hit box with all of its limbs intact.
I can go 4 or 5 maps in the same mech wondering why Im not doing any damage, then suddenly Im blowing limbs off left and right and getting 4 kills and 800 damage in a single match with the exact same mech.
Could it just be me? Or is it the servers Hit Reg thing?
#4
Posted 30 March 2015 - 01:50 PM
#5
Posted 30 March 2015 - 02:07 PM
Lily from animove, on 30 March 2015 - 01:23 PM, said:
how many C-LPL's can a orange Grasshopers CT eat facing your frontally? answer: 6 and and still running.
GG hitreg.
there were for me soem matches where mechs ate 3x the firepower they should have. how many CT hp has a TDR? dunno ran around with orange armor afetr eating 3 clear alphas non moving. so 150 armor at last,.
Other matches then were just fine.
I really wonder if this is a bit luck dependend on which of the slices or wherever mwo runs onto. gett he wrong core or such and boom hitreg gone. Play on the other core, hitreg = fine. It's like having no real pattern.
Is it fixable? Or maybe a work around?
#6
Posted 30 March 2015 - 02:14 PM
Shadow Magnet, on 30 March 2015 - 01:18 PM, said:
Meaning that PGI seems to intentionally run MWO on too low resources to save money, keeping it at the bare minimum of being playable.
Call me paranoid, but I got no other explanation...
I have no evidence that suggests this is an intentional thing - but I also get a similar take when it comes to the hit reg situation.
One thing I do know, having worked extensively with computers, for decades, is no two machines are the same - and no matter how well built or maintained, computers suffer from accuracy issues when it comes to handling and storing data. It is also true that systems tend to develop patterns to how they mishandle the data, and frequent cold booting of these systems tends to minimize these effects.
So the problem could be in the code, the wires, the machines, the environment - isolating the issues can be very difficult.
Of course, PGI could also be shirking on the resources to save money:)
#7
Posted 31 March 2015 - 12:26 AM
JaxRiot, on 30 March 2015 - 02:07 PM, said:
Is it fixable? Or maybe a work around?
that was a guess what may cause hitreg issues, because hitreg seems to be matchwise, either its totally screwed, a bit screwed not not screwed at all.
But currently I have not experienced something like: beginng of a match it was bad, at the end it was good. So even with many dead emchs or such and then less pew pew happening did influenced it. it seems for a specific match be constan over the entire mech.
The only pattern indeed seems to be: hotreg is cosntant during a match, no matter if it started horrible, good or somewhere inbetween. This makes me think serverload has not much to do with it. Core related? connection related? Maybe,
#8
Posted 31 March 2015 - 02:54 AM
Lily from animove, on 31 March 2015 - 12:26 AM, said:
that was a guess what may cause hitreg issues, because hitreg seems to be matchwise, either its totally screwed, a bit screwed not not screwed at all.
The only pattern indeed seems to be: hotreg is cosntant during a match, no matter if it started horrible, good or somewhere inbetween.
I'll watch for this also - it seems a correct observation to me as well, but I haven't specifically been watching for this -
#9
Posted 31 March 2015 - 04:00 AM
I find it pretty suspicious that PGI basically never comments on hit reg. It should be a major issue for them to fix as it does massively influence the game quality and the fun of the players.
#10
Posted 31 March 2015 - 04:13 AM
#11
Posted 31 March 2015 - 07:32 AM
#12
Posted 31 March 2015 - 07:34 AM
Lily from animove, on 31 March 2015 - 12:26 AM, said:
that was a guess what may cause hitreg issues, because hitreg seems to be matchwise, either its totally screwed, a bit screwed not not screwed at all.
But currently I have not experienced something like: beginng of a match it was bad, at the end it was good. So even with many dead emchs or such and then less pew pew happening did influenced it. it seems for a specific match be constan over the entire mech.
The only pattern indeed seems to be: hotreg is cosntant during a match, no matter if it started horrible, good or somewhere inbetween. This makes me think serverload has not much to do with it. Core related? connection related? Maybe,
Connection related was my initial thought when I first noticed it. But my Ping is pretty much a constant 100 although Im not sure how much influence its having on the Hit Reg. My Ping is steady but my Hit Reg seems to be all over the place and acts differently each map.
Where are the servers located anyway?
#13
Posted 31 March 2015 - 07:41 AM
Shadow Magnet, on 31 March 2015 - 07:32 AM, said:
Actually missile hitreg improved greaty when they moved away from splash damage to impact damage. But yeah, when there's a lot of weapons fire going on, hitreg suffers because of how the cryengine handles things (prime reason why you shouldn't fire more than 3 SRM6s at the same time, or you will have a lot of them not do damage as their collision is queued and possibly dropped when hitting that upper limit). There is unfortunatly a hard upper limit to very certain things that often can't just easily be shifted upwards.
Edited by SethAbercromby, 31 March 2015 - 07:43 AM.
#14
Posted 31 March 2015 - 07:44 AM
JaxRiot, on 31 March 2015 - 07:34 AM, said:
Connection related was my initial thought when I first noticed it. But my Ping is pretty much a constant 100 although Im not sure how much influence its having on the Hit Reg. My Ping is steady but my Hit Reg seems to be all over the place and acts differently each map.
Where are the servers located anyway?
No idea, but connection is not equal to connection.
what if the other machine/slice/instance your match runs has a different port after data get distributed by any kind of laodbalancer, what if on that port, for some reason 20% packets get lost. May be, may not, IT is sometimes just strange.
1 user(s) are reading this topic
0 members, 1 guests, 0 anonymous users