87 replies to this topic

[50 50]

Why can't it be more dynamic with the group sizes?
If there are enough players to get a 4 v 4, that is what the match is within a certain time frame.
But should there be enough for an 8 v 8, that is what happens.... or 12 v 12.

If we have a massive number of players on then the games a full.
But during periods where it might be quieter, why not just have it time out and start with the 4 v 4 or 8 v 8 if that is what is available?
It's not even separate queues.

[MrMadguy]

Wildstreak, on 21 August 2017 - 08:57 AM, said:

Seen comments where some people think you just flip a switch and boom, it works.

Nope.

Here is my list of what actually MUST be done to make 8v8 work in QP Queues:

1 Shrink Maps - some existing maps are large so boundaries need to be shrunk. List includes
Alpine Peaks
Forest Colony
Frozen City (not as much but does need be done)
Grim Plexus (not as much but does need be done)
Polar Highlands
River City (not as much but does need be done)
Terra Therma

I would even say cut down all maps larger than 10x10 to that size, same size as Crimson Strait, Tourmaline Desert and Viridian Bog.

2 Redesign Content on Maps - Now that some maps are smaller, they need to redo how things work. Spawn Points, Conquest Nodes, Assault and Incursion bases, Escort Paths and Domination points. Look at non-shrunken maps too like Crimson Strait needing spawn fixes in Domination.

3 Escort Fixes - Might as well cut down Escort Paths per map to 3 maximum instead of 4-5 like now. Too many times players go one way, Escort goes another especially on bigger maps. Rarely does that work for Defender. Might as well also look at Escort fixes like either it shoots back at nearest target or replace it with a player.

While looking like 3 easy steps, this involves some work time. You just do not want half the maps being bigger than they need be.

You mean return all old maps, yeah? Do you know, that we had perfect 8vs8 maps in the past?

[El Bandito]

50 50, on 22 August 2017 - 11:14 PM, said:

Why can't it be more dynamic with the group sizes?
If there are enough players to get a 4 v 4, that is what the match is within a certain time frame.
But should there be enough for an 8 v 8, that is what happens.... or 12 v 12.

If we have a massive number of players on then the games a full.
But during periods where it might be quieter, why not just have it time out and start with the 4 v 4 or 8 v 8 if that is what is available?
It's not even separate queues.

Tweet it to Russ. Dunno if PGI can even make that work.

[MrMadguy]

I just don't think, that 4x4 would be enjoyable in 1/1/1/1 conditions.

[Almond Brown]

Bigbacon, on 22 August 2017 - 05:00 PM, said:

Im telling you 8v8 will greatly limit the gameplay to two murberballs thag meet at a single grid. Lights will have no place, mediums will barely have a place.

No more flanking in groups, no more light rushes to catch lone mechs. You leave the murderball you gimp your team immediately. You take a non perfect build meta mech you will gimp your team. No more fooling around with builds.

When old 8v8 was around the game wslas way different than it is now. It may have seemed awesome then but with current tech and fast ttk, forget it.

But keep thinking it will make it all better because it wont.

Sad thing is they will do this...and kill it even more. Maybe that is what they want.

This back and forth cause we cant make up our damn mind is pathetic.

Let us hope to our Deity of choice that PGI modifies the Match Earnings tables accordingly as well, and everyone likes the change. LOL! LOL! lol...lo..l

[Quicksilver Aberration]

MischiefSC, on 22 August 2017 - 03:26 PM, said:

To put it in context, you need to carry 33% harder than you did in 12 v 12.

We know this from practice to be false. Why? Because mistakes cost more in 8v8, and a good player capitalizes on those mistakes (so while their team can make the same costly mistakes, a good player gets more return out of capitalizing on the enemies as well).

MischiefSC, on 22 August 2017 - 03:26 PM, said:

need to keep an Elo for each chassis and weapon system.

That will never nor should it ever happen, because there is too many variables involved. Not all weapons are equal across all mechs or even the same mech. You essentially want a BV system which we all know is flawed for obvious reasons.

Edited by Quicksilver Kalasa, 23 August 2017 - 08:33 AM.

[MischiefSC]

Quicksilver Kalasa, on 23 August 2017 - 08:29 AM, said:

We know this from practice to be false. Why? Because mistakes cost more in 8v8, and a good player capitalizes on those mistakes (so while their team can make the same costly mistakes, a good player gets more return out of capitalizing on the enemies as well).


That will never nor should it ever happen, because there is too many variables involved. Not all weapons are equal across all mechs or even the same mech. You essentially want a BV system which we all know is flawed for obvious reasons.

Are you saying 8 v 8 is more forgiving than 12 v 12?

In 8 v 8 you have 33% fewer teammates while the player skill curve doesn't change. The result is that the corresponding carry expectation for whoever the highest skill level on the team is goes up. So if you're on the upper edge of the skill curve the "carry weight" that you're responsible for goes up. Less hands to share the load and as the population of your team goes down the uneven skill distribution will result in more weight going up the curve.

I want a basic BV because while it's never going to be totally accurate (no measure of humans skill ever is) the gap between good and bad mechs is huge now. The XL King Crab LRM boat is nowhere and nothing comparable to a laservomit MAD. If you're not accounting for that you're building a MM that's inherently biased toward good players always taking good mechs and bad players always taking bad mechs. That's going to reduce the accuracy of the MM.

Again, the MM is never going to be perfectly accurate. The goal however is to cover the big bases and get it as accurate as possible and reasonable. Mech choice is a big piece of that puzzle. Yes, someone who's terribly inaccurate won't benefit as much from direct fire meta builds but it's still going to be better than his STD heatsink lore build.

For the MM it's mitt about perfect precision, just allowances for the biggest factors.

[Quicksilver Aberration]

MischiefSC, on 23 August 2017 - 09:07 AM, said:

Are you saying 8 v 8 is more forgiving than 12 v 12?

The opposite, but that is why it is easier to carry for good players.

[MischiefSC]

Quicksilver Kalasa, on 23 August 2017 - 09:07 AM, said:

The opposite, but that is why it is easier to carry for good players.

You're conflating ability to carry with need to carry. It will be easier to carry so long as you're in full carry mode - the odds of the other team having someone better than you goes down by 33% as well, which helps -

But you really have to show up to win every match.

Make sense? In the end I absolutely agree that 8 v 8 will benefit top tier players the most and punish low/mid players, especially LRM boat and hide-in-the-back kill leeches. However any time you do not full on hero carry the odds of you still pulling a win drop in 8 v 8.

As to a mech BV of sorts, look at it like this. Yes, BV is not a good balance tool on its own. However take 2 options - in option 1, we both get 8 mechs, 2 of each weight class and a BV of X amount. In the other you just both pick 2 mechs per weight class. While BV isn't perfect, it's significantly better than nothing when used as a modifier to the players Elo/PSR/whatever.

[Quicksilver Aberration]

8v8 would be less forgiving simply because mistakes are more expensive to make and scouting matters a bit more since it is easier to get blindsided by virtue of less eyes on the field to spot all avenues of approach which compound things (8v8 makes flanking easier for faster units).

MischiefSC, on 23 August 2017 - 09:34 AM, said:

It will be easier to carry so long as you're in full carry mode

Sure, but that's kinda the goal. No one likes carrying and still losing. It is harder to hero carry in current matches than it is 8v8 simply because of how hard it is for one person to kill all the dudes compared to in 8v8 (time is a factor here, because you need time to burn down the enemy team before they overwhelm your team, which is much easier in 8v8 than 12v12 since each kill has a bigger impact).

Edited by Quicksilver Kalasa, 23 August 2017 - 09:47 AM.

[MischiefSC]

Quicksilver Kalasa, on 23 August 2017 - 09:40 AM, said:

8v8 would be less forgiving simply because mistakes are more expensive to make and scouting matters a bit more since it is easier to get blindsided by virtue of less eyes on the field to spot all avenues of approach which compound things (8v8 makes flanking easier for faster units).


Sure, but that's kinda the goal. No one likes carrying and still losing. It is harder to hero carry in current matches than it is 8v8 simply because of how hard it is for one person to kill all the dudes compared to in 8v8 (time is a factor here, because you need time to burn down the enemy team before they overwhelm your team, which is much easier in 8v8 than 12v12 since each kill has a bigger impact).

Yeah, but then how are you ever going to get to practice with your RAC2s, MRMs and frankenbuilds?

[Quicksilver Aberration]

MischiefSC, on 23 August 2017 - 09:57 AM, said:

Yeah, but then how are you ever going to get to practice with your RAC2s, MRMs and frankenbuilds?

I mean you will still use them in solo queue, but really group queue will still work for practicing builds as well.

[Daurock]

My Prognostications, for posterity

1) If the game moves to 8v8 in QP, the games length will get marginally shorter, with individual time to kills remaining about the same. The only thing that basically changes here is that that 3v3 that was happening away from the main group has less mechs and HP for the winners of the main brawl to mop up. (As it turns into a 2v2, or a 1v1.) (Exception: Big maps may occasionally get longer due to fewer scouts being able to find the other team.)

2) Builds will change, marginally lowering ammo counts in many mechs. My mauler carries roughly 1000-1200 dmg in ammo on a given day, depending on what weapon I'm running. If if this moves to 8v8, I'll likely drop that to 800-1000 dmg, spending that extra ton or two in extra HS, or maybe an AMS. Honestly, this is probably a fairly minor difference.

3) Que times will get shorter. Pretty easy one here. Fewer players -> more games -> lower wait times

4) Steamroll games will get more common. The law of averages states that the larger a random sample gets, the "Tighter" the spread (in terms of % of the whole) becomes. In terms of skill level imbalances between teams, This means that the chance a skill/build/mech mis-match that leads to a steamroll gets higher, the smaller a match gets. (It's easier to carry 1 LRM assault on your team in 12v12 than 8v8) Essentially, a lot of those 12-7 or 12-9 matches that were teetering on a steamroll becomes an 8-2 stomping.

5) The "high skill" Players will generally reap the benefits of 8v8, at the expense of the average/bad player. Due to (4), a great player will have a bigger effect on the game in an 8v8.


Now, my two cents:
My honest opinion here is that I'd like to see it stay 12v12. The only notable change I can see is that the stomp games get more common, and more harsh.

I suspect that the main thing PGI is trying to do here is to reduce wait times in que, due to solaris coming out, diluting the que. Instead of moving to 8v8 for QP, I'd instead use the Skill MM to instead evenly distribute players based on their rank, not put an artifical barrier between the good and bad players. (For example, if there were 4 level 1 players in que, a half dozen 2/3s, and the rest 5s, I'd have the MM put 2 of each level 1 on each team, 3 to each team of the level 2/3s, and the rest filled with 4/5s.) I realize that some of the goods out there really don't want to play against (or with) bads and vice/versa, but I really don't think the MM has the population to support that separation at the moment, let alone if/when new modes come into the game. (Its my opinion that a small fish getting eaten by a single good player in his match is less dis-heartening than having his entire team ground to paste in an 8-0 steamroll.)

(As a side note: If/when solaris comes out, and becomes the place to go for 1/2/4 person groups, i would drop Scouting from the CW map. It becomes redundant at that point.)

[Alan Davion]

Daurock, on 23 August 2017 - 12:10 PM, said:

I'd instead use the Skill MM to instead evenly distribute players based on their rank, not put an artifical barrier between the good and bad players.

That is of course assuming there is an even distribution of players between each rank, which I can almost guarantee there isn't.

I would go so far to say that I'd bet 75% of the games population is somewhere in Tiers 1 and 2, with the remaining 25% spread through Tiers 3, 4 and 5, given that all new players start in Tier 3, and if played right, their Cadet Bonus will rocket them up into Tiers 1 or 2, when they really should not be.

What's the Cadet Bonus time period again? 20 games? No body should be getting into Tier 2, let alone Tier 1 in just 20 games.

[MischiefSC]

Tiers isn't going to work for 8 v 8.

[Brain Cancer]

Tiers don't work, period. Your point?

[Quicksilver Aberration]

Daurock, on 23 August 2017 - 12:10 PM, said:

My honest opinion here is that I'd like to see it stay 12v12. The only notable change I can see is that the stomp games get more common, and more harsh.

You will see games end in 8-2, but that doesn't necessarily mean they will be stomps. Due to individual skill mattering a bit more, games will swing a bit harder than they do in 12 v 12 depending on mistakes being made. So a match that starts out 0-2 could end in an 8-2 decision in the originally losing team's favor. That's the important distinction to make because the sway of a match makes them more exciting to play.

In other words, let's be careful just throwing around the end result as a sure sign of a stomp because that isn't necessarily a stomp. A stomp is where a team starts and ends with a serious advantage (for a variety of reasons).

Edited by Quicksilver Kalasa, 23 August 2017 - 01:30 PM.

[MischiefSC]

Quicksilver Kalasa, on 23 August 2017 - 01:28 PM, said:

You will see games end in 8-2, but that doesn't necessarily mean they will be stomps. Due to individual skill mattering a bit more, games will swing a bit harder than they do in 12 v 12 depending on mistakes being made. So a match that starts out 0-2 could end in an 8-2 decision in the originally losing team's favor. That's the important distinction to make because the sway of a match makes them more exciting to play.

In other words, let's be careful just throwing around the end result as a sure sign of a stomp because that isn't necessarily a stomp. A stomp is where a team starts and ends with a serious advantage (for a variety of reasons).

When you are a potato, all the numbers are dark sorcery. The difference between an enemy team that's all 1 touch and 8 people who rolled you without really even having to try is invisible. There's the numbers up in the middle and if one is way bigger than the other it was a bad match and a stomp.

All the rest of that stuff? Could be anything. Probably all haxx.

[Quicksilver Aberration]

MischiefSC, on 23 August 2017 - 02:18 PM, said:

When you are a potato, all the numbers are dark sorcery. The difference between an enemy team that's all 1 touch and 8 people who rolled you without really even having to try is invisible. There's the numbers up in the middle and if one is way bigger than the other it was a bad match and a stomp.

All the rest of that stuff? Could be anything. Probably all haxx.

It would be nice if percentages or something were visible at the end of round screen partially for this reason.

[STEF_]

It is amazing to see how many potatoes in the brown sea fear the coming of 8vs8...... and would like to be carried and hide in 16vs16, or even 24vs24