10417 replies to this topic

[somenothing]

Thanks for the pointer. But actually I miscommunicated my original question.. I'm actually more interested about the effect on ELO from all these crashes... I was getting 5 in a row and locked up all my favourite mechs :/

Nick Makiaveli, on 04 January 2014 - 06:43 AM, said:

No but nobody really cares about KDR in this game. It's a team based game, not an individual one.

For example, I am not a great pilot, but I don't totally suck either. Last I looked (just to make a point similar to this, so been a while) it was .33.

Which is low obviously. Yet that's mainly because if you die with 10 assists and 0 kills, your KDR goes down. If you die with 1 kill and 0 assists, it stays even, yet who would you rather have on your team? Some guy who holds back so he doesn't die, or rushes in to face hug the wounded enemy mech so as to block incoming fire so he can get the kill, or a guy who focus fires and helps win the match even if it means he dies or doesn't get any kills?

[Nick Makiaveli]

somenothing, on 04 January 2014 - 06:53 AM, said:

Thanks for the pointer. But actually I miscommunicated my original question.. I'm actually more interested about the effect on ELO from all these crashes... I was getting 5 in a row and locked up all my favourite mechs :/

Well first off, since you seem like a reasonable person and all that, I apologize for the implied snarkiness in my post. Not a huge fan of KDR whores.

Basically if you make it into the drop despite being a discon, and thus your mech is on the field to be destroyed, then yes it will count as a win/loss (if your XP changes by 100 or 300 you will know) and thus could affect your Elo.

But I wouldn't sweat it too much. The more you play, the less effect any game ( or series of 10 games) affects your score. Also, if your team was predicted to lose anyway, then you not showing up won't affect your Elo.

Hopefully though, the game detected you dropping and replaced you. Seen a lot of games with a last minute addition. Sadly even more with no one added and dropped 1 or 2 short.

[Redshift2k5]

Well, some basics on Elo: Elo looks at wins vs losses, and does so in a specific way: It compares your team's elo to the other team's elo, decides who should win (the higher Elo team), and compares this decision to the actual outcome- If the high Elo team won then there is no change because the prediction was correct. If the low Elo team wins then the prediction was wrong and the team's aggregate Elo is adjusted.

so, if you lose a match, this will only affect your Elo score if your team had a higher Elo. Which is all invisible anyway so there's really no reason to worry about it.

If your Elo dips, then you'll be put into matches more favourable for you until you begin to lose, then when you lose when you had higher Elo your Elo will drop again; in this way it fluctuates until it finds an equilibrium.

[Koniving]

Ideally with an ELO system you should lose 50% of your matches. This really isn't the case though.

There is something else to note: It keeps a separate ELO score for you for each mech weight classification. Light, Medium, Heavy, and Assault.

If you try a mech weight class for the first time, you're bound to be pit with the new players again until it settles you somewhere on the score for what it believes is your 'level'.

[Robert Montjoy]

Hi, it's me again. This Time not with a philosophical Question but one I hope an experienced player can easily answer.


Q: Is there something not right with the Damage dealt?


Explanation:
I managed to get behind a not moving Blackjack and released an Alphastrike from 100m Distance. I had 5x ML and 2x SRM6-Artemis, all carefully aimed at the center of the backtorso. The Blackjack didn't move, took the whole volley and engaged me. I mean how likely is it that someone in a 40 Tonnes Mech do have at least 49 Armor Points on his rear Center? Do I count somethin wrong?

Next Example I was shhoting at a Commando. The other Pilot was afk and again... I could not manage to put the Commando down with on Alphastrike (same conditions as above)

Next Example was an overheated and powered down Spider... surely I hit from the Front... but again... 49 Armor?

The Question is: Did I something not right, is there something I don't know? Is there something not right with my Hunchback? Is the Ratio 1 Damage equals 1 Armor not correct? Is the Damage mentioned with the Weapons not correct? Why could above mentioned Mechs have survived. Any Ideas?

Thanks in advance.
Robert

P.S. Even if the Artemis guided SRM did not have hit the same spot as the Lasers... isn't 25 Damage enough to kill a least a not moving Light Mech from behind with one Alphastrike?

Edited by Robert Montjoy, 04 January 2014 - 09:29 AM.

[Koniving]

Robert Montjoy, on 04 January 2014 - 09:26 AM, said:

Hi, it's me again. This Time not with a philosophical Question but one I hope an experienced player can easily answer.

Q: Is there something not right with the Damage dealt?

Depends on the armor allocation (for example, could it have been a turtle back? A Turtle Back is a mech whose armor is allocated very heavily on the rear to handle both lack of situational awareness and lengthy reload times on high powered weapons. It used to be popular with Hunchbacks and Centurions in closed beta).

The rear armor can be anything from 0 to 44 points and the structure at 22 points on a side torso and on a center from 0 to 56 and the structure is 23 points.

The typical numbers are 8 to 20 points of rear armor for sides and center. Some less, some more.

Let's say it's the minimum typical number, 8 on the rear center. Now add the 23 CT structure. That's a total of 31 points of damage you must deal to make the kill as a typical minimum. Also do not forget that SRMs have bad hit registration especially at ranges of less than 50 meters. The more SRMs you fire at once, the worse the registration gets.

At 100 meters, there's also the spread -- which also gets worse with more SRMs fired (so that they wont' go through each other they spread more).

Also note that on a Blackjack the following is generally true unless there was recent hitbox changes already made to them.
Their rear "CT" is narrower than their pelvis. It's about as narrow as their cockpit, and much narrower than their front CT. This is why I run my Blackjacks with lots of front CT and rear ST armor.

Spiders are also an interesting case. Note that it's easier to leg a spider than to kill it by the torsos. Even so if you get one overheated aim for the head instead. All spiders lower cockpit armor. It's practically a meta trade secret.

If you must go by the torso, stay within the area beside the 'laser' in the stomach. That laser marks the edge or border of CT and ST. Also note that the STs were bloated in size to make hit detection easier, thus reducing efficiency of CT shots despite making the CT wider as well. STs are easier to take out anyway.

The rear of the Spider is so ridiculously narrow that damage gets heavily spread even when you try to focus it. In fact, on the spider you can accidentally hit the front while hitting the rear with virtually any weapon as you 'graze' across it. Especially true near the head and the arm-pits.

Edited by Koniving, 04 January 2014 - 09:43 AM.

[Robert Montjoy]

Thanks for the quick answer.

Just to be sure:
=> Do you mean I am I better of using Streaks or LRM in my Hunchback instead of the 2x6SRM-Artemis ?
=> Or do you have some advice of using the SRM corectly so that registration and split is optimal ?

[Redshift2k5]

I would suggest trying a pair of streaks and see if it works better for you. You won't even have to remove Artemis. SRMs are not performing as intended, and playing from Germany may not be helping your hitreg issue.

[Junkman7mgte]

Robert Montjoy, on 04 January 2014 - 09:55 AM, said:

Thanks for the quick answer.

Just to be sure:
=> Do you mean I am I better of using Streaks or LRM in my Hunchback instead of the 2x6SRM-Artemis ?
=> Or do you have some advice of using the SRM corectly so that registration and split is optimal ?

Set the srm6's to chainfire, and see if that helps. The fewer you fire in one volley, the better they cluster which = better hit detection also.

[Koniving]

Robert Montjoy, on 04 January 2014 - 09:55 AM, said:

Thanks for the quick answer.

Just to be sure:
=> Do you mean I am I better of using Streaks or LRM in my Hunchback instead of the 2x6SRM-Artemis ?
=> Or do you have some advice of using the SRM corectly so that registration and split is optimal ?

Refer to both of these.

Junkman7mgte, on 04 January 2014 - 10:51 AM, said:

Set the srm6's to chainfire, and see if that helps. The fewer you fire in one volley, the better they cluster which = better hit detection also.

Redshift2k5, on 04 January 2014 - 10:02 AM, said:

I would suggest trying a pair of streaks and see if it works better for you. You won't even have to remove Artemis. SRMs are not performing as intended, and playing from Germany may not be helping your hitreg issue.

Ages ago I was explaining it to another player on teamspeak during the events of this Challenge Build.



Comparatively when firing twin LRM-15s at the same time in three sets (LRM 30, LRM 30, LRM 30) and firing 1 LRM 15 at a time (LRM 15, LRM 15, LRM 15, LRM 15, LRM 15, LRM 15) I would consistently perform better damage even when dying sooner by firing fewer LRMs at once.

The same is also true for SRMs. Now, streaks will always hit. But you have to get a lock, and it's a good idea to also equip a "Beagle Active Probe." LRMs 'mostly' register but truth be told there's always some that will fail to register, too.

[somenothing]

snarkiness noted but all cool, don't get me started on kdr whores haha

for myself, i couldn't care less.. i've been very careless (read: reckless) when i first started really playing about 6 months ago, and it was probably close to 0.1… it might be better since and close to 0.7 haha
but most of the time, doing lots of damage might not give you the kill.. or on the other hand, making 'good' hits should inherently give you less damage (pts) because you kill with less hits.. but i digress

anyhow, I guess I shouldn't worry about my elo getting affect that much, even though I get numerous crashes per night nowadays. it's just that the prospect of being stuck on a lower elo and hence having to play with "less skilled players", all because of my d/c's, doesn't hit right with me somewhere...

Nick Makiaveli, on 04 January 2014 - 07:00 AM, said:

Well first off, since you seem like a reasonable person and all that, I apologize for the implied snarkiness in my post. Not a huge fan of KDR whores.

Basically if you make it into the drop despite being a discon, and thus your mech is on the field to be destroyed, then yes it will count as a win/loss (if your XP changes by 100 or 300 you will know) and thus could affect your Elo.

But I wouldn't sweat it too much. The more you play, the less effect any game ( or series of 10 games) affects your score. Also, if your team was predicted to lose anyway, then you not showing up won't affect your Elo.

Hopefully though, the game detected you dropping and replaced you. Seen a lot of games with a last minute addition. Sadly even more with no one added and dropped 1 or 2 short.

[Mahws]

Robert Montjoy, on 04 January 2014 - 09:26 AM, said:

P.S. Even if the Artemis guided SRM did not have hit the same spot as the Lasers... isn't 25 Damage enough to kill a least a not moving Light Mech from behind with one Alphastrike?

Don't underestimate how much armor lighter mechs have, it's far higher than it seems.

If it was a stock armor Blackjack (a lot of newer players seem to leave their armor stock) to one hit kill it firing into the rear center you'd have had to do 18+28 = 46 points of damage. If it was me in one of my blackjacks you'd have had to do 38. Either way liable to survive your attack.

As a general rule when I get a sneak attack on the back unless the enemy is already damaged or is already fighting on two fronts I take a side torso if I have the firepower to one hit kill or do as much damage to a leg/arm as I can if I don't. After that first free hit you're not likely to get another clean shot at their back so it's best not to waste your shot weakening armor you won't get to fire at again. I've wasted more free back shots on rear center torso's where I didn't even have enough firepower to cut through the armor than I care to admit to.

Nick Makiaveli, on 04 January 2014 - 06:46 AM, said:

Yea completely useless on Assault, I mean who wants to win and get a bonus 75 XP for Capturing the base and another 50 XP for helping cap the base.

Let's just completely ignore the fact that on Assualt it's frequently more profitable to lose fighting than win by cap. What percentage of Assault matches you've played were won by cap, maybe 10%-5%? Now how many of those did you lose because the cap was stopped with less than about 10% left on the cap? Being generous, maybe one in ten?

So going by those numbers Capture Accelerator helps you in about 0.5% to 1% of Assault games. If you already have a capture accelerator and a free module slot, then yeah, of course it's worth taking. Otherwise you're sacrificing a module that might help you almost every match like target decay or seismic sensor and paying 15,000GXP and 6 million c-bills for the privilege.

Edited by Mahws, 05 January 2014 - 05:04 AM.

[Alaskan Nobody]

Mahws, on 05 January 2014 - 05:03 AM, said:

So going by those numbers Capture Accelerator helps you in about 0.5% to 1% of Assault games. If you already have a capture accelerator and a free module slot, then yeah, of course it's worth taking. Otherwise you're sacrificing a module that might help you almost every match like target decay or seismic sensor and paying 15,000GXP and 6 million c-bills for the privilege.

I for one am saving up for the cap accel - mostly because
1) I play mostly conquest
2) I seem to be one of the only ones willing to ever stand on the cap points
3) Conquest pays more than Assault does

Of course, exceptions to every rule still applies

[Nick Makiaveli]

Mahws, on 05 January 2014 - 05:03 AM, said:

Let's just completely ignore the fact that on Assualt it's frequently more profitable to lose fighting than win by cap. What percentage of Assault matches you've played were won by cap, maybe 10%-5%? Now how many of those did you lose because the cap was stopped with less than about 10% left on the cap? Being generous, maybe one in ten?

So going by those numbers Capture Accelerator helps you in about 0.5% to 1% of Assault games. If you already have a capture accelerator and a free module slot, then yeah, of course it's worth taking. Otherwise you're sacrificing a module that might help you almost every match like target decay or seismic sensor and paying 15,000GXP and 6 million c-bills for the privilege.

Losing is more profitable? I'm sorry, but proof or else that's BS. Salvage for 12 mechs is often only around 10-12K with occasional spikes if you leg mechs instead of blowing the engine.

Sure on rare occasions one or both sides rushes to cap and the match is over quickly, but that just means you can start a new match quickly and thus get more XP/Cbills per hour.

Your logic needs work as the second percentage you refer to is 0%. Yep, you heard me. Of all the matches I WON by capping, NONE of them were lost because the capping was stopped.

I assume you meant of the games where we attempted to win by cap, did we lose because we were stopped. In that case, yes some were stopped but that ignores a very salient fact. The vast majority of the time we attempt to cap win in Assault it's for one of two reasons. Either we are losing the battle, or the other side has decide to start capping. If it's late game, then often we are closer to the enemy base and thus it's easier to cap their base.


Finally, and the biggest mistake of your whole post. WTF are you doing playing Assault if you want Skirmish?

Seriously, WTF is the malfunction? You do know that if you select Skirmish it will add ANY players to the match right? So picking ANY doesn't really speed up the MM process. Don't want Conquest, don't play it. Don't want to defend your base? Go play Skirmish.

Some of us are preparing for CW and since we don't know what will happen, we like to play all types and work on builds, tactics etc for all modes.

If you want to play just pure team deathmatch, then set it to Skirmish. Problem solved.

One last comment. WTF!?!? You are saying the module is a waste because it's not always useful in one mode. What if I want to play Conquest on occasion, then it's very useful. Also, yes sometimes I have spare slot so I stick it in just in case.

See your problem is you think your way is THE WAY, handed down by Jeebus Himself. Which is stupid. Personally, I've been spending cash on mechs and upgrades etc so I only have the Capture Accel. and the Adv. Zoom. I have over 15K GXP sitting around for when I decide which one I want next. Right now I am still leveling up all the PP mech using x2 XP so I don't have a dedicated mech to play so holding out.

So now do you see that not everyone plays your way? Good, Now work on the idea that I can believe something different and it doesn't mean it's a personal insult to you and your MWO religion.

[Robert Montjoy]

Sorry to bend this discussion about the Hit-Registration of Missiles. But it leads to a maybe stupid question.


Q: Am I better off using a single 10 LRM or a single 20 LRM (or something else)


I mean, what good is a Volley of 20 if 10 Missiles don't have a chance to hit/register on the target while on the other Hand a Volley of 10 crashes home for a net damage output of both 10?
So If there is no (or little) Difference i would rather spare on the Weight and install a better Laser (for example). Is there some experience what the best single LRM is?

To make it simple let's assume the target isn't sporting a AMS.

Greetings
Robert

Edited by Robert Montjoy, 05 January 2014 - 02:21 PM.

[Koniving]

Robert Montjoy, on 05 January 2014 - 02:20 PM, said:

Sorry to bend this discussion about the Hit-Registration of Missiles. But it leads to a maybe stupid question.


Q: Am I better off using a single 10 LRM or a single 20 LRM (or something else)


I mean, what good is a Volley of 20 if 10 Missiles don't have a chance to hit/register on the target while on the other Hand a Volley of 10 crashes home for a net damage output of both 10?
So If there is no (or little) Difference i would rather spare on the Weight and install a better Laser (for example). Is there some experience what the best single LRM is?

To make it simple let's assume the target isn't sporting a AMS.

Greetings
Robert

In reality you get almost identical results. Although if you really want you could test it. Use the "Testing Grounds" option and then study a single LRM-20 versus two LRM-10s.

Here's what I can tell you.

• LRM-10s will fire faster.
  
• Even chain-fired they will fire more frequently.
  
• Two LRM-10s fired one at a time should have less spread than a single LRM-20.
  
• However, an LRM-20 will have less heat than two LRM-10s. Both due to less heat from firing and from firing less frequently (also reducing the potential for wasted shots).
  
• An LRM-20 can be more easily destroyed than two LRM-10s.
  
• An LRM-20 from a ten tube mech will easily have identical results to twin LRM-10s without increasing your hitbox size (example: Wolverine shoulder launchers) and without the heat build up.
  
• An LRM-20 with Artemis costs 1 extra ton for a total of 11 tons.
  
• Twin LRM-10s with Artemis costs 2 extra tons for a total of 12 tons.

There's more, but the best part of customization is discovery.

Personally I've used single LRM-5, 10, 15, and 20. It varies depending on how you plan to fight. The Stalker 4N is the best example I can think of. I'd do an LRM-15 in the odd launcher, and twin SRMs in the side torsos with lasers. It helps soften them.

The lower the number, the more concentrated the shot (note this is the number of missiles fired at once. Example, an LRM-20 firing from a 5 tube launcher is more concentrated and pinpoint than an LRM-10 from a 10 tube launcher, because only 5 missiles leave at once for the LRM-20 in this instance). Tubes are what really matter.

[Mahws]

Nick Makiaveli, on 05 January 2014 - 10:25 AM, said:

Losing is more profitable? I'm sorry, but proof or else that's BS. Salvage for 12 mechs is often only around 10-12K with occasional spikes if you leg mechs instead of blowing the engine.

Often more profitable.

Here's your source:
http://mwomercs.com/...-functionality/

No c-bill reward for win by capture, no c-bill reward for winning rather than losing. All you lose is the salvage bonus which is usually less than 1.5 kills or assists.

With the exception of double cap rushes or when you're winning and it's faster to cap than hunt down the last player it's usually better for your income to fight and die quickly whilst earning more than to run and attempt to cap, ultimately earning less regardless of whether you win or lose. It's a stupid system, but that's how it is.

Nick Makiaveli, on 05 January 2014 - 10:25 AM, said:

-stuff-

If you actually read my post again without inventing completely baseless opinions to assign me so you have something to argue against you'd probably realize that we actually agree with each other and you're arguing about things I've never said.

Where did I say that I don't like capping? Where did I say I only play Skirmish? Fun fact, I prefer to play conquest and assault because I feel Skirmish is just assault without the added tactical options that make assault fun.

Things I was saying:
For a new player that doesn't play conquest the cap accelerator module isn't as useful as many of the other modules they could get instead.
This is because relatively few assault matches are won by cap and of those that are even fewer are close enough that a 15% increase in cap speed would make any difference.

Things I didn't say:
Any of the stuff you addressed in your post.

To quote your own words;

Nick Makiaveli, on 05 January 2014 - 10:25 AM, said:

Now work on the idea that I can believe something different and it doesn't mean it's a personal insult to you and your MWO religion.

And if you feel the need to respond maybe you could address something I actually said or an opinion I actually hold.

Edited by Mahws, 05 January 2014 - 09:15 PM.

[Nick Makiaveli]

Since you are going to be all logical and ****, I will have to agree that it does appear I read more into what you said than you meant.

That said, let's clarify the capping issue. If we are talking about PURELY C-bills, then yes, you are better off fighting. If you die, quit the match and start a new one. Counting XP, unless you are a master at killing, a win is better than a loss.

So if you meant fighting is better than capping for cash, then you are right. If you meant XP wise, generally capping is preferable as you get a minimum 275 more (win difference plus bonus) plus an additional 50 if you help cap.

That said, I don't support capping in Assault unless we are down by 3-4 mechs with no sign of a reversal, or "they started it!!"

The exception is if we are all mediums and lights and I see a Steiner Scout company coming over the hill, then yea base rush FTW!!


Lastly, in case you took my first statement to be facetious, I did apparently read into what you said, having been poisoned by various [Redacted] folks on here, and so for that I apologize for being a Typical Intarwebz User toward you.

[IraqiWalker]

Mahws, on 05 January 2014 - 05:03 AM, said:

Let's just completely ignore the fact that on Assualt it's frequently more profitable to lose fighting than win by cap. What percentage of Assault matches you've played were won by cap, maybe 10%-5%? Now how many of those did you lose because the cap was stopped with less than about 10% left on the cap? Being generous, maybe one in ten?

So going by those numbers Capture Accelerator helps you in about 0.5% to 1% of Assault games. If you already have a capture accelerator and a free module slot, then yeah, of course it's worth taking. Otherwise you're sacrificing a module that might help you almost every match like target decay or seismic sensor and paying 15,000GXP and 6 million c-bills for the privilege.

You sir have not dealt with many caps, I am a light mech specialist (Commandos almost exclusively) and I will say this, cap accel on assault mechs is better than cap accel on lights. See we can get to the cap faster, every time, but if there was ever the situation where an assault mech hits the enemy cap, you WANT them to have a cap accel, it's harder for them to move, them having it makes up for the time loss. Also, they can hold a cap better than most other mechs.

Also, speaking as a light mech pilot, I win 60-80% of all "Assault" matches because of capping, either with a cap feint, or actually fully capping the enemy. Never under-estimate capping, or cap accels, which I would argue are even more important now that capping is harder.

[mikelovskij]

Hi, this question has probably been asked 1k times but it's quick to asnwer and I cannot find enoguh informations about it: if I remove the artemis from a mech, when I add it again do I have to pay for it another time?