11 replies to this topic

[Hotthedd]

Matchmaking needs to be fixed NOW, before MW:O loses/chases off any more new players.

Please do not tell me to "wait" for ELO, we need something NOW.
Specifically, the matchmaker must try to even out guardian equipped 'mechs. Whether premade or straight PUGs, having one team with 4 or 5 ECM 'mechs to zero for the other team equals predetermined win or loss (I have dropped on both sides).

Nothing will chase away new or casual players faster than the experience of getting stomped repeatedly due to ONE PIECE OF EQUIPMENT.

[Syllogy]

Hotthedd, on 09 January 2013 - 07:51 AM, said:

Matchmaking needs to be fixed NOW, before MW:O loses/chases off any more new players.

Please do not tell me to "wait" for ELO, we need something NOW.
Specifically, the matchmaker must try to even out guardian equipped 'mechs. Whether premade or straight PUGs, having one team with 4 or 5 ECM 'mechs to zero for the other team equals predetermined win or loss (I have dropped on both sides).

Sure, we'll just snap our fingers, and it will magically appear.

Hotthedd, on 09 January 2013 - 07:51 AM, said:

Nothing will chase away new or casual players faster than the experience of getting stomped repeatedly due to ONE PIECE OF EQUIPMENT.

Since only 2 weapons are truly hindered by ECM (LRM's and Streak SRM's) try using something else.

Edited by Syllogy, 09 January 2013 - 08:49 AM.

[CrushLibs]

I would like to see more size balancing. Since you pick your mech before the match having a min and max mech sizing should be doable.

Example: min -- 1 of each class ,,, Max 3 of each class

3 assault , 3 heavies , 1 med , 1 light ---- 1 assault , 1 heavy , 3 med , 3 lights ------- 2A,2H,2M,2L etc etc etc

This would stop the 5 ravens or 5 atlas or 5 dragon weighted matches.

On a side note --- larger 4-5km wide maps and if possible 12v12 or 16v16 match sizes would be nice using the above ratios to keep things spread out over the weight sizes.

[Hotthedd]

Syllogy, on 09 January 2013 - 08:46 AM, said:

Sure, we'll just snap our fingers, and it will magically appear.



Since only 2 weapons are truly hindered by ECM (LRM's and Streak SRM's) try using something else.

No need to be snarky. The fix is pretty simple.
It isn't about the weapons that ECM counters, it is about the cloaking entire groups, coupled with the ability to use those weapons while the other side cannot.
Streaks and LRMs are serious weapons, so saying "only" those two are affected is like saying "The doctors ONLY removed one kidney."
Again, if you want to leave ECM alone, fine. As long as both sides are roughly equal with it. The problem is that the current matchmaking ignores ECM (The most powerful equipment in the game) completely.

[Syllogy]

Hotthedd, on 09 January 2013 - 09:14 AM, said:

No need to be snarky. The fix is pretty simple.

Here's a Non-Snarky answer:

If it's pretty simple, please program the solution.

Then PGI can just compile it and the fix can be put into next week's patch.

[Hotthedd]

Syllogy, on 09 January 2013 - 09:26 AM, said:

Here's a Non-Snarky answer:

If it's pretty simple, please program the solution.

Then PGI can just compile it and the fix can be put into next week's patch.

PGI does not pay me to write their programs, but do you really think the algorithm is that hard?

[Syllogy]

Hotthedd, on 09 January 2013 - 09:28 AM, said:

PGI does not pay me to write their programs, but do you really think the algorithm is that hard?

Have you ever tried to program a Matchmaking Algorithm?

[Runenstahl]

Hotthedd, on 09 January 2013 - 07:51 AM, said:

Matchmaking needs to be fixed NOW, before MW:O loses/chases off any more new players.

Please do not tell me to "wait" for ELO, we need something NOW.

Uh... you DO realize that this is still a beta, yes ? I'm okay with your general statement that they should look into ECM and missile mechanics, but there's no need to scream "FIX IT NOW". A beta isn't supposed to be perfect, it's supposed to be playable (which it IS) and it's supposed to help the devs to get to the "final" product.

[Hotthedd]

Syllogy, on 09 January 2013 - 09:30 AM, said:

Have you ever tried to program a Matchmaking Algorithm?

There is already an algorithm in place based on class. All that needs to be done is classify 4 chassis differently.

[Syllogy]

Hotthedd, on 09 January 2013 - 09:42 AM, said:

There is already an algorithm in place based on class. All that needs to be done is classify 4 chassis differently.

For every filter that you introduce, you make the matchmaking process exponentially harder.

For example, if we go with your suggestion, an ECM Raven vs. an ECM Commando is not a problem.

Now, introduce the Cicada and the Atlas. Would you go with only Atlas vs. Atlas, or would you introduce the possibility of ECM mechs being matched regardless of their weight class?

This is also assuming that each variant is classified by whether or not it can use ECM. If each variant is only classified by weight without the actual use of ECM (as it is right now), for example: What happens when a Raven 3L that is notusing ECM? Is he classified as a Light, or as an ECM Raven?

What if there is a 4 man group that drop an ECM Raven, ECM Commando, ECM Cicada, and ECM Atlas? The fact that they are all grouped means that they have a definite advantage over a random Raven, Commando, Cicada, and Atlas that join, assuming that there are 4 of those variants out there searching for a group to begin with.

Also, what do you do with matches that are short? Currently it's possible to see 8v7, and (rarely) even 8v5.

This happens because one team populates before the other team (because one team's makeup is how the matchmaker pulls from the pool of available game seekers.)




In summary, please tell me how simple that change is

[Hotthedd]

Runenstahl, on 09 January 2013 - 09:42 AM, said:

Uh... you DO realize that this is still a beta, yes ? I'm okay with your general statement that they should look into ECM and missile mechanics, but there's no need to scream "FIX IT NOW". A beta isn't supposed to be perfect, it's supposed to be playable (which it IS) and it's supposed to help the devs to get to the "final" product.

And how do they do that?

Oh yeah, by feedback from the Beta testers.

That is all I am doing, and I am suggesting that it be prioritized.

[Hotthedd]

Syllogy, on 09 January 2013 - 09:51 AM, said:

For every filter that you introduce, you make the matchmaking process exponentially harder.

For example, if we go with your suggestion, an ECM Raven vs. an ECM Commando is not a problem.

Now, introduce the Cicada and the Atlas. Would you go with only Atlas vs. Atlas, or would you introduce the possibility of ECM mechs being matched regardless of their weight class?

This is also assuming that each variant is classified by whether or not it can use ECM. If each variant is only classified by weight without the actual use of ECM (as it is right now), for example: What happens when a Raven 3L that is notusing ECM? Is he classified as a Light, or as an ECM Raven?

What if there is a 4 man group that drop an ECM Raven, ECM Commando, ECM Cicada, and ECM Atlas? The fact that they are all grouped means that they have a definite advantage over a random Raven, Commando, Cicada, and Atlas that join, assuming that there are 4 of those variants out there searching for a group to begin with.

Also, what do you do with matches that are short? Currently it's possible to see 8v7, and (rarely) even 8v5.

This happens because one team populates before the other team (because one team's makeup is how the matchmaker pulls from the pool of available game seekers.)




In summary, please tell me how simple that change is

Good points.

If a Raven 3L is NOT using ECM (have you ever seen this since ECM was introduced?) The his team would be at a disadvantage by the matchmaker.

It is not perfect, but it will be a good improvement until ELO is sorted out.

Instead of: Light/Medium/Heavy/Assault
it could be: Light/LightE/Medium/MediumE/Heavy/Assault/AssaultE.

Yes it must be assumed that ECM capable 'mechs are using ECM.

That theoretical 4 man pugstomping group may have a long wait in the queue.