125 replies to this topic

[IceSerpent]

Joseph Mallan, on 22 September 2014 - 08:12 AM, said:

Though I am not a fan of Elo... BV2 was found to be exploitable by a few Commandos/Agents for TT. It is Better than BV1 was though from what I heard through the grapevine.

Joe, Elo and BV are two separate beasts. In TT there's no metric for player skill at all, "pilot" is essentially a part of the mech similar to tank crew in WoT. There's no real difference (aside from fluff) between better pilot and better targeting computer. Both are just doodads that change mech stats.

Regarding BV schemes being exploitable, they are no more exploitable than current 3/3/3/3 scheme - you can "exploit" former by using the most effective config that fits into chosen BV limit just like you can "exploit" latter by using the most effective config for a chosen weight class. With BV it's much easier to counter those exploits though, because it only requires changing BV of the element that makes that particular config so effective. With weight class you don't have the luxury of "fine tuning" JJ-capable mechs or PPC boats, which results in things like ghost heat.

[IceSerpent]

EgoSlayer, on 22 September 2014 - 09:49 AM, said:

Except it isn't an 'AND' it's two separate complaint items. When the match wait times were long the complaint was that it was a long match. Most people said the match itself was well matched when there was a long wait.

They are not separate at all. The fastest possible MM is what we had in CBT - single game mode, everybody in the same queue, no MM checks whatsoever. Nobody really liked that and people were more than willing to sacrifice that speed for more game modes, separate solo queue, private matches, etc. In other words, we already have ample evidence of players willing to trade time for some other benefit and not willing to trade time for something they don't think is beneficial.

Quote

Except we do have an idea how long it will be, based on past match maker changes. The 12-man queue was killed by match wait times due to only 12-man vs 12-man matching. 8 Mans where hard to match because 4 mans were allowed in the public/solo queue - resulting most groups being 4 mans and 2 man because that is the fast way to get a match. The players *have* voted by their behaviors, and all the evidence is there. PGI has even more data because they see what is in the pool and what the edge cases really are.

No, we don't - you can't extrapolate like this from specific number vs. specific number (i.e. 12 v. 12 or 8 v. 8) to combinations of various group sizes.
Besides, the biggest problem that killed 12v12 queue was not the wait time, it was the requirement to have exactly 12 players online. You have less - you can't play there at all. You have more - extra people are "on the bench", get bored and go do something else.

[Threat Doc]

Ice Serpent is absolutely right in this. I believe, as well, with the advent of BVMWO, it would make ALL 'Mech chassis and variants viable, again, as people think about how they would better use their 'Mechs, and I think those 'Mechs that have gone into obscurity would have new life, again.

[Joseph Mallan]

IceSerpent, on 22 September 2014 - 09:51 AM, said:

Joe, Elo and BV are two separate beasts. In TT there's no metric for player skill at all, "pilot" is essentially a part of the mech similar to tank crew in WoT. There's no real difference (aside from fluff) between better pilot and better targeting computer. Both are just doodads that change mech stats.

Regarding BV schemes being exploitable, they are no more exploitable than current 3/3/3/3 scheme - you can "exploit" former by using the most effective config that fits into chosen BV limit just like you can "exploit" latter by using the most effective config for a chosen weight class. With BV it's much easier to counter those exploits though, because it only requires changing BV of the element that makes that particular config so effective. With weight class you don't have the luxury of "fine tuning" JJ-capable mechs or PPC boats, which results in things like ghost heat.

I personally don't know if its gameable or not. But some of the guys/gals I know and respect on the TT forums say otherwise. Its a seniority thing for me Ice.

You understand right.

Kay Wolf, on 22 September 2014 - 10:03 AM, said:

Ice Serpent is absolutely right in this. I believe, as well, with the advent of BVMWO, it would make ALL 'Mech chassis and variants viable, again, as people think about how they would better use their 'Mechs, and I think those 'Mechs that have gone into obscurity would have new life, again.

More so in PUG Que I would think. The comp crowd will pick BV apart and exploit its known metrics.

[IceSerpent]

Joseph Mallan, on 22 September 2014 - 11:05 AM, said:

I personally don't know if its gameable or not. But some of the guys/gals I know and respect on the TT forums say otherwise. Its a seniority thing for me Ice.

You understand right.

I understand completely. The big question is what those guys/gals call "gameable" and how is it different from current scheme in MWO. In other words, what possible advantage one can get from BV that doesn't exist in our current "stock Locust is equal to fully pimped Raven because they belong to the same weight class" situation.
From my perspective (and it's entirely possible that I might be missing something), the only way to "exploit" TT BV is to create a build that doesn't use known "bad" weapons (i.e. AC2) and go vs. someone who uses those. Yet, that's the issue with the actual values (numbers) and not the scheme itself.

Here's an interesting observation: I've seen a lot of people claiming that BV is exploitable on the forums (implying that concept itself can be gamed), but I haven't seen a single example of how it can be done. Which leads me to believe that folks who say that are referring to some specific values in TT instead of the idea in general.

Not really disagreeing with those people you mentioned (yet), just want to figure out whether there actually is a point of contention or is it just a case of not everyone being on the same page.

[EgoSlayer]

IceSerpent, on 22 September 2014 - 10:02 AM, said:

They are not separate at all. The fastest possible MM is what we had in CBT - single game mode, everybody in the same queue, no MM checks whatsoever. Nobody really liked that and people were more than willing to sacrifice that speed for more game modes, separate solo queue, private matches, etc. In other words, we already have ample evidence of players willing to trade time for some other benefit and not willing to trade time for something they don't think is beneficial.

They are separate complaints in the MM discussion. They are certainly related, but most people don't say 'I was matched too fast and got a crappy match!' They just say 'I got a crappy match' and sometimes 'I waited x minutes and got a crappy match!' because all the balance locks came off.
With the 100% rando MM we had before the biggest complaint was 'the evil pre-made boogy man' because voice coms in 8v8 matches is a huge advantage.

IceSerpent, on 22 September 2014 - 10:02 AM, said:

No, we don't - you can't extrapolate like this from specific number vs. specific number (i.e. 12 v. 12 or 8 v. 8) to combinations of various group sizes.
Besides, the biggest problem that killed 12v12 queue was not the wait time, it was the requirement to have exactly 12 players online. You have less - you can't play there at all. You have more - extra people are "on the bench", get bored and go do something else.

Now you're just contradicting yourself. The 12 man queue died because of wait times to find a match, good or otherwise. To get a "good" match in 12 man *requries* another 12 man, by virtue of unified voice coms alone. A 10+2 on seperate coms isn't a match for a 12-man, nor is an 8+4, 6+6, etc. The 12 man group drop is back now, with the advent of private matches and now their inclusion into the general group queue. It didn't die because of lack of 12 mans, it died from lack of matching 12 mans in the queue - a long wait.

People stopped playing most groups except 4 or less because of match wait time. It's a fact, one that leads to statictics like 'only 20% of the players play in groups'. Which is a number forced by a match maker that wouldn't allow more than 33% of the player base to play in groups.

IceSerpent, on 22 September 2014 - 11:55 AM, said:

I understand completely. The big question is what those guys/gals call "gameable" and how is it different from current scheme in MWO. In other words, what possible advantage one can get from BV that doesn't exist in our current "stock Locust is equal to fully pimped Raven because they belong to the same weight class" situation.
From my perspective (and it's entirely possible that I might be missing something), the only way to "exploit" TT BV is to create a build that doesn't use known "bad" weapons (i.e. AC2) and go vs. someone who uses those. Yet, that's the issue with the actual values (numbers) and not the scheme itself.

Here's an interesting observation: I've seen a lot of people claiming that BV is exploitable on the forums (implying that concept itself can be gamed), but I haven't seen a single example of how it can be done. Which leads me to believe that folks who say that are referring to some specific values in TT instead of the idea in general.

Not really disagreeing with those people you mentioned (yet), just want to figure out whether there actually is a point of contention or is it just a case of not everyone being on the same page.

Things like this:
http://themittani.co...ech-brotherhood


Although in MWO you can't get a numbers advantage, a group of low BV heavies and assaults will beat a group of high BV lights and mediums because of the durability factor. Also, the published BV's don't mean anything in the MWO game and would have to be redone because weapon BV only seems to account for damage, range, and to hit modifiers (reverse engineered BV calculator).
It's not an accurate balance of how weapons work in MWO since damage per hit fluctuates based on range/burn time for lasers, Clan UAC being burst vs IS which are single shot, accuracy is totally dependent on real pilot skill, missile hit percentages are all over the map, etc. A new system would need to be created for MWO, and likely see it's own balancing passes ad infinitum.

Edited by EgoSlayer, 22 September 2014 - 07:49 PM.

[Threat Doc]

Joseph Mallan, on 22 September 2014 - 11:05 AM, said:

More so in PUG Que I would think. The comp crowd will pick BV apart and exploit its known metrics.

That's no different than gaming the extant moving parts of the current Matchmaker system. Without hacking the game, which is illegal, anyway, the choices people make of what goes on their 'Mech and, overall, how it's configured, are as good for the goose as they are for the gander, don't you think?

IceSerpent, on 22 September 2014 - 11:55 AM, said:

Here's an interesting observation: I've seen a lot of people claiming that BV is exploitable on the forums (implying that concept itself can be gamed), but I haven't seen a single example of how it can be done. Which leads me to believe that folks who say that are referring to some specific values in TT instead of the idea in general.

I wanted to make sure I clarify something so it's understood where I am coming from... when I am talking about using Battle Value in MWO, I'm not talking about using a direct translation of BV1, 2, or 3 from the tabletop to MWO. Rather, I'm talking about PGI developing a system very close to those that can be used specifically for this game, so that if an Inner Sphere ER Large Laser has an Offensive BV of 250 for Joe Snuffy, it has that same BV for me. If, on the other hand, Joe adds a range extender to the weapon, the BV for the extender would be added to the weapon BV, so if the weapon goes down, not only does the enemy team get the points for destroying the ER LL, but they also get the points for the range extender.

Capiche?

[Dracol]

Russ Bullock, on 21 September 2014 - 10:23 PM, said:

Honestly we can make the MM amazing if players will accept the trade offs. For instance allow us to put 1 mech per weight class into even smaller groups so for instance a group of two can't be two heavies etc. This alone would mean the MM could fit the jig saw pieces together really easy. Put that on top of the game mode selector being a vote and were in business.

Would limiting groups to only 1 of each class per 4 pilots be an option?
Would the player base like this? (poll maybe?)

Personally, I'd rather work with my group to meet that criteria instead of having a chance to drop in a game mode I do not wish to play. And I can imagine the nightmare the MM has with the 2 heavies or 3 heavies mech group.

[Threat Doc]

EgoSlayer, on 22 September 2014 - 12:52 PM, said:

It's not an accurate balance of how weapons work in MWO since damage per hit fluctuates based on range/burn time for lasers, Clan UAC being burst vs IS which are single shot, accuracy is totally dependent on real pilot skill, missile hit percentages are all over the map, etc. A new system would need to be created for MWO, and likely see it's own balancing passes ad infinitum.

The idea is not to get super-accurate. The BV of a weapon is not going to change based on the hit rating/gunnery skill of the pilot, as the numbers are based off maximum potential while on a 'Mech without a pilot. The weapon, or the 'Mech's performance, are not based on the weapon or 'Mech, as they are only potential, only tools, to be used in the game. Rather, it's the pilot, with game-determined statistics for piloting and gunnery skills, as based on their past performance to-date, and their performance within the game itself, that grants a modifier to the overall BV of the 'Mech. If the pilot is crap, and consistently fights on the low end of the PS/GS spectrum, their 'Mech BV will be modified.

This is not based on luck, or touchy-feely emotions, or whether an individual is having a great night or a crap night, these are simple solid numbers. The system I'm proposing builds BV in the MWO MechLab, just like many of the BattleMech Designers, including Heavy Metal Pro, of the past fifteen years have done, and then when the 'Mech is saved in the Lab, it has a solid BV to it that's not going to change unless/until the pilot changes the 'Mech's configuration. However, there's one last trick to it... the metrics already available in MWO for measuring pilot stats and abilities can be used to help determine Gunnery Skill for damn sure, and Piloting Skill can be a matter of how well a person's 'Mech has fared at the end of multiple hundreds of matches, a full history if PGI could pull that off. This PS/GS is converted into a percentage modifier for the BV determined in the MechLab, and appended at the time of launch.

If a person is great with Heavy's, but really horrible with Light's, their overall PS/GS is going to go down the longer they play that Light, which would be accurate to how fighting these 'Mech's works. Now, if PGI wanted to keep the by-weight class modifier, that could work, too. Match-to-match, if a MechWarrior improves, their PS/GS improves, the modifier for BV is dropped into the bucket when the player hits the Launch button.

Edited by Kay Wolf, 22 September 2014 - 01:14 PM.

[Johnny Z]

This game has one of the best match queues out.

Its missing a newbie queue still.

Any and all changes and additions they are making have to be taken in the context that they are making parts of it for the upcoming faction wars, which offers alot more problems with synch dropping etc than the game has right now. (synch dropping is basically rigging match outcomes by a group dropping onto both teams.)

For both the group queue and regular queue, making it an equal chance to drop into all game modes for all players would be a good direction for the queue system to take.

Edited by Johnny Z, 22 September 2014 - 01:33 PM.

[Dracol]

Kay Wolf, on 22 September 2014 - 01:13 PM, said:

-snip-

Unfortunately, the 40+ person dev team available to make such a system has to content with the 1,000+ players that will work to "game" the system. There is a segment of our player base that will make sure their mech counts for the lowest BV but can still perform above what is expected so that they are dropped against an easier opponent. And the build plans are released and BV for the most part will become meaningless.

Johnny Z, on 22 September 2014 - 01:20 PM, said:

Any and all changes and additions they are making have to be taken in the context that they are making parts of it for the upcoming faction wars, which offers alot more problems with synch dropping etc than the game has right now. (synch dropping is basically rigging match outcomes by a group dropping onto both teams.)

A post by Russ earlier in this thread said that this is not really the case. CW match making will be taking on a different form then the current MM.

[Johnny Z]

BV isnt an option, just like it isnt an option for prize fighting. They went the right route to create interesting mechs that are greater then the sum of their parts.

The game does use a form a battle value, but in the form of weight classes. This puts the balancing in the hands of the guys building the game, and the players can only hope they do a good job. If not, start a topic about how clans are OP for instance. Seriously, start one now haha I need help complainingg about them

[wanderer]

Quote

This is a valid point - although player counts have been consistent pretty much since the beginning.

So, what you've just said, much to our likely collective horror is that at best, the player count has been stagnant- any gains being washed out by loss of older players?

Perhaps you wanted to say that differently.

[IceSerpent]

Kay Wolf, on 22 September 2014 - 01:01 PM, said:

I wanted to make sure I clarify something so it's understood where I am coming from... when I am talking about using Battle Value in MWO, I'm not talking about using a direct translation of BV1, 2, or 3 from the tabletop to MWO. Rather, I'm talking about PGI developing a system very close to those that can be used specifically for this game, so that if an Inner Sphere ER Large Laser has an Offensive BV of 250 for Joe Snuffy, it has that same BV for me. If, on the other hand, Joe adds a range extender to the weapon, the BV for the extender would be added to the weapon BV, so if the weapon goes down, not only does the enemy team get the points for destroying the ER LL, but they also get the points for the range extender.

Kay, that goes without saying - verbatim copy of BV values from TT was never even on the table. There are 3 proposals that have been suggested at various times:

1. Keep the classic scheme where each component has value, total value of the build is simple sum of its component values. Component values are determined by PGI.

Pros: very easy claculation (straight up sum).
Cons: lots of numbers to type into the database

2. Base some values on mech characteristics instead of components, i.e. assign value to speed instead of each engine, assign value to heat dissipation instead of each heatsink, etc. (suggested by me a while ago)

Pros: less numbers to type, although some components (ECM, BAP, etc.) still need their own values.
Cons: more complex BV calculation that involves all sorts of modifiers.

3. Dynamically base values on how popular something is. I.e. the more people use ECM, the higher its BV becomes. (suggested by Roland)

Pros: self-balancing system, virtually no numbers to type in.
Cons: tricky BV calculation that involves statistical analysis

None of these three ideas had anything to do with BV values as they exist in TT.

[IceSerpent]

EgoSlayer, on 22 September 2014 - 12:52 PM, said:

They are separate complaints in the MM discussion. They are certainly related, but most people don't say 'I was matched too fast and got a crappy match!' They just say 'I got a crappy match' and sometimes 'I wanted x minutes and got a crappy match!' because all the balance locks came off.

Which brings us back to people weighing cost (in time) vs. benefit. You don't see people complaining about waiting extra 30 seconds for a good match (which was not really good eighter, as we don't have proper Elo matching - so, it's was more of a "better match than before"), but you do see them complaining about having to wait 10 minutes for that. You don't see people complaining about addition of new game modes, even though it slows down MM. You don't see people complaining about addition of private matches, even though it also slows down MM (decreases number of players in the public queue).
It all boils down to how much time we lose and what exactly we gain out of it.

Quote

With the 100% rando MM we had before the biggest complaint was 'the evil pre-made boogy man' because voice coms in 8v8 matches is a huge advantage.

Which is just another way of saying "match is unfair".

Quote

Now you're just contradicting yourself. The 12 man queue died because of wait times to find a match, good or otherwise.

Those wait times didn't come from any kind of balancing checks. It was a simple matter of getting 12 players online being difficult -> lots of units not bothering with that -> units who did bother not getting anybody to play against.

Quote

To get a "good" match in 12 man *requries* another 12 man, by virtue of unified voice coms alone. A 10+2 on seperate coms isn't a match for a 12-man, nor is an 8+4, 6+6, etc.

I beg to differ - disadvantage of 10+2 vs. 12 is not that huge. I agree about going lower than that though. 8+4 would probably need some assistance (better Elo, better mechs, etc.)

Quote

People stopped playing most groups except 4 or less because of match wait time.

That's total BS and we both know it - people COULDN'T play anything except 1,2,3,4, and 12 until recently. They "stopped playing most groups except 4 or less" because it was technically impossible to play them.

Quote

Things like this:
http://themittani.co...ech-brotherhood

That's an issue with TT numbers (Savannah Master having lower BV than it should have) and an issue with map/tactics (in that example mechs going "back-to-the-wall" would have much easier time dealing with the swarm).

Quote

Although in MWO you can't get a numbers advantage, a group of low BV heavies and assaults will beat a group of high BV lights and mediums because of the durability factor.

Depends on what exactly is "low BV assault" and "high BV light". You can easily end up in a scenario where in order to match BV of a Jenner your Atlas will have to have 200 engine and single small laser for weapons.

Quote

Also, the published BV's don't mean anything in the MWO game and would have to be redone because weapon BV only seems to account for damage, range, and to hit modifiers (reverse engineered BV calculator).

That's a given, nobody suggested to use TT values.

Quote

A new system would need to be created for MWO, and likely see it's own balancing passes ad infinitum.

As opposed to current system that is totally not seeing "it's own balancing passes ad infinitum."?

[EgoSlayer]

IceSerpent, on 22 September 2014 - 02:52 PM, said:

<snip>
As opposed to current system that is totally not seeing "it's own balancing passes ad infinitum."?

Don't entirely disagree with anything except this. The Elo system has seen very little changes. A few bug fixes, and modifications to the player's starting elo, otherwise it's unchanged.

[Gyrok]

IceSerpent, on 22 September 2014 - 07:54 AM, said:

Been "touching lines of code", as you so eloquently put it, for almost 19 years and counting.



You are contradicting yourself - if PGI hasn't messed anything up and the game is perfect, why are there "many things you would have done differently"? If the game is nowhere near being perfect, why do you think that it's not PGI who is at fault?

Not to mention that desyncs are a separate issue that has nothing to do with the topic.



Sure it is, bro. None of us had ever worked with any sort of queue algorithms before.



1. Where did you get that 20-30 minutes figure from?
2. What makes you think that we prefer fast queue with lousy matching to slow queue with good matching?
3. Being a game developer, you do realize that we can make that an player option (select slow/good or quick/bad) and essentially have both camps fairly happy, right?

1. Coding what? Java script?

2. I would have made different design decisions, but what they have now, while a bit buggy, is not malfunctioning...

3. If you agree it is more complicated than you anticipate, then WTF?

4. I realize that sacrifices are made for the sake of more of one thing and less of another. I pulled the 20-30 minutes figure out of the air, but considering the PTS times were often 15 minutes under some circumstances, I think you can imagine it might not be too far off. As for making everyone happy, if you really are a coder, then, you and I both know that is a pipe dream that sales people talk about while they stand around and sing kumbayah at a convention. Especially in the world of video games. You constantly piss people off, you are just making educated guesses about who is going to be the least detrimental to piss off.

As for this matchmaker crap...I can tell you this much, if they screw with the ability to select game modes, or the group queue size capabilities right now...it would literally kill a boat load of units. My entire active population would up and quit overnight if they did this...literally...our clan forums are full of comments like..."dealbreaker", "they gave us great groups, and now they kill them...done", "Deuces, call me when they unf*ck this sh!t"

[IceSerpent]

EgoSlayer, on 22 September 2014 - 03:06 PM, said:

Don't entirely disagree with anything except this. The Elo system has seen very little changes. A few bug fixes, and modifications to the player's starting elo, otherwise it's unchanged.

...and it remains broken. The size of a "bracket" is large enough to make match outcome luck-based, which means that rating doesn't accurately reflect player skill, which in turn makes the whole setup fairly useless.

On a side note, I just played a few matches with Savannah Masters in MegaMek. Single AS7-D vs. 9 SMs (1897 BV vs. 1935 BV). Atlas that can protect its back by parking on the map edge or next to a building/hill has a very serious advantage. SMs are very squishy and Atlas gets 4 to-hit dice rolls per turn while staying heat-neutral. Worst outcome so far is knocked out AC20 and some armor damage.

[Redshift2k5]

Yeah, actually, it is hard- balance team aggregate elo, balance team size cluster while retaining weight class balance (most of the time two 6s aren't gonna have exactly the same value of weight classes to fit together in 3333), all at the same time without making you wait more than 5 minutes. And god forbid you have two perfectly equal teams but they're in completely separate game modes.

Go write a better sorting algorithm and get back to us.

Edited by Redshift2k5, 22 September 2014 - 03:31 PM.

[aniviron]

Mawai, on 21 September 2014 - 12:36 PM, said:

Organize your team better ... it can be done.

Yeah, it's easy, just use the in-game VOIP and command wheel functions to keep everyone coordinated! Easy.