76 replies to this topic

[Mcgral18]

Am I not understanding how Weapon variables are stored and pulled from? Because that could be a fair explanation due to my ignorance.

My understanding is that the variables are adjusted from a single location, the Weapons.XML file, which makes changing weapons a simple task, from a single location without risk of forgetting to change something in multiple folders, which bad coding can allow for.


If that isn't the case...care to explain what weapon changes actually entail?

If that is the case...why have we gotten year long "Metas" where there was a single SIGNIFICANTLY better weapon system without changing a variable?

I remember reading about being able to change variables and test them in the Testing Grounds (local system, not on the live servers) before it was patched out. Was it a rapid fire SL in a video? Simply editing the appropriate Weapons.XML variable.

Then there's the Former Secret Squirrel with his STD400, AC40 LOLcust with 20some JJs. This one's more in depth, and not quite the same thing.



Is there more than updating the Weapons.XML on Patch Day? Because the fact that Flamers are still worthless after 3 years, SRMs are Terribad for over a year, and MGs are still nerfed 20% also for roughly a year, upsets me.
Those could potentially be fixed within 10 minutes, every two weeks, because you'd adjust something each patch in moderate increments to ensure you don't need a Hot Patch.


Technical reasons, or You-Know-Who reasons?

[Deathlike]

Damn it, you're trolling meeeee!!!!!1111

"Because PGI" is the only answer I can come up with, because that Dartboard of Balance™ has to be replaced soon™.

[Tyler Valentine]

Because Paul says!!!

[Andi Nagasia]

it should be an easy fix, but i think why it hasnt happened is because PGI is working on "other' Things,
the problem is because their Focus isnt Sololy on Weapon Balance right now but on-
WeaponBalance & TechBalance & InfoTech & CWPhaze3 & Ect,

[Davers]

Andi Nagasia, on 23 September 2015 - 04:13 PM, said:

it should be an easy fix, but i think why it hasnt happened is because PGI is working on "other' Things,
the problem is because their Focus isnt Sololy on Weapon Balance right now but on-
WeaponBalance & TechBalance & InfoTech & CWPhaze3 & Ect,

Well it would be nice if we could get a heads up on ANY of those things.

[Wintersdark]

What's more, every time they start "rebalancing" they say they're going to keep making incremental adjustments. Those incremental adjustments are made 2-3 times(optomistically), then *poof* it just stops, and nothing more is said.

I sincerely don't understand why this is, why there aren't more frequent, small weapon stat adjustments to get sub-par weapons more usable.

[LordKnightFandragon]

Deathlike, on 23 September 2015 - 02:40 PM, said:

Damn it, you're trolling meeeee!!!!!1111

"Because PGI" is the only answer I can come up with, because that Dartboard of Balance™ has to be replaced soon™.

Yeah, the dartboard was proving to be to good, they had to go with a more abstract method.

Pin the Tail on the Donkey board of balance. Blindfolded and the whole 9 yards.

[El Bandito]

This guy. This effing guy.

[MischiefSC]

It's not that it's any easy fix - it's a lot of work and it will be unpopular. People tend to avoid unpopular decisions.

There legitimately are a good sized group of players who will threaten to leave and some likely will if Clans are not better than IS. When the top performing Clan mechs they bought for money get nerfed down to IS levels and the "bad" ones that are still better than their equivalent is bad chassis get brought down too, they will throw a fit. Absolute frothing fits. These are people who are RPing being Clanners and have some self-identity tied up there. Go to the Clans forum and poke around and you'll see what I mean.

Then you've got the inherent outrage from playing favorites. Even after the rebalance there will be good and bad mechs and when the rebalance is done there will be favorites left bottom tier and people will be outraged.

People dislike change. Would rather have something they dislike but are used to than potentially better but new. We say we want "justice" but really mean vengeance or mercy. We say we want "balance" but mean advantage, as in leave the stuff I like or buff it and change or remove what they don't.

What we need is to get moving and iterate quickly. Like every two days. That way change becomes the norm and the expectation becomes problems being fixed. You iterate too slow and you don't have an expectation of improvement.

[Mcgral18]

MischiefSC, on 23 September 2015 - 05:06 PM, said:

What we need is to get moving and iterate quickly. Like every two days. That way change becomes the norm and the expectation becomes problems being fixed. You iterate too slow and you don't have an expectation of improvement.

How about we start with every 2 weeks, as opposed to maybe once every second Month.

[Deathlike]

You know...

Way back when, I think there were old questions on those ATDs (Ask The Devs) and to my recollection... Paul's answer to this was something along the line of "we need 1 month to vet any changes to balance".

Mind you, i thought that was a crap answer, but had to begrudge the idea that you know it takes some time to balance, and any sort of "random change" would be chaotic and actually needs time to "set itself" (changing multiple variables increases the testing time)... so it is what it is.

Unfortunately, seeing we had "aggressive balance changes" before the MWO launch (LPL normalization being at the forefront along the list of bad balance adjustments) AND then the biggest debacle in the form of "an immediately reversed" CERLL change (aka CERLL duration of 2.0s, and then some... and that was before the recent PTS)... well, I'm pretty sure something was totally lost in translation.

Now, let's talk a bit of accountability. Who's responsible for these things? Sure we know whose "job description" entails this (it goes w/o saying), but then again, is there any promise to address this going forward (along with other PGI related apologies)?

We have too many "sources" of potential influence that we don't even know where to begin.

Let's start with the Secret Squirrels. What do they actually do? Do they have any influence? Do they actually play the game worth a damn?

It's hard to know whether they have had impact.. there's lots of rumors and stuff... but my understanding is the "squirrels" aren't involved ultimately (the rumors themselves don't paint a pretty picture).

Then there's the "media arm" of PGI. There's plenty of rumors that there's some indirect meddling... at least involving what was dubbed "community builds" when the Champion system was started... after the very first Heavy mech design contest. While ultimately we don't know who or where the terribad mech designs came from (writing about them is a chore in itself), it's not totally unreasonable to wonder how much influence that the "media arm" had anything to do with balance... if not just being the messenger, but also the originator of some ideas...


Then there's the comp community. There is some open admission of submitting their opinions, but having read "the apology" written, you can see where they are annoyed, and rightly so. While some are more vocal than others regarding it, it is unfair to just "throw them under the bus" as the "apology" said and when it comes to accountability... we kind of end up in the old saying "he said, she said" when it comes to someone's side of the story. It kinda makes me wonder if they are under any sort of NDA...


Sometimes I wonder myself if "balance" is in some feedback loop, where the Spider-5V and the Mist Lynx must continue to toe the line of "being irrelevant indefinitely", but it's honestly hard to understand any underlying logic of balance when it honestly doesn't get changed for extended durations... leaving weapons in the state of limbo like it was back in Open Beta (Flamers and LBX are top of the list).

Balance cannot be randomly done at a whim.. it must ALWAYS be an iterative process until things settle in nicely, and ALL weapons and mechs are viable/useful in some form... and not a random occurrence because they were overquirked.

Maybe someday we won't make jokes about the AC2. Today, we'll just keep posting "steering wheel" jokes referencing the LBX10. We can't have nice things in MWO.

Edited by Deathlike, 23 September 2015 - 05:24 PM.

[MischiefSC]

Mcgral18, on 23 September 2015 - 05:17 PM, said:

How about we start with every 2 weeks, as opposed to maybe once every second Month.

No less frequently than 1 time a week. The point is momentum. Your brain changes how you think about stuff after 72 hours. At 10 days you start to lose perspective on what prior iterations were like and initial impressions on the new one, which are the important indicators you need. A week is still too long but is frequently enough to avoid perception of "new normal" on changes. You change stuff and let it be "normal" before you change it again and you'll frustrate people.

weekly works because not everyone plays every day. Five days would be ideal but the first few updates need to be really fast, back to back. It's about managing expectations and setting the expectation at "this will change".

Currently we have "this is what you will have for 18 months"

[Deathlike]

MischiefSC, on 23 September 2015 - 05:30 PM, said:

Currently we have "this is what you will have for 18 months"

That sounds like UI 2.0, like all other things.

[Wintersdark]

MischiefSC, on 23 September 2015 - 05:30 PM, said:

No less frequently than 1 time a week. The point is momentum. Your brain changes how you think about stuff after 72 hours. At 10 days you start to lose perspective on what prior iterations were like and initial impressions on the new one, which are the important indicators you need. A week is still too long but is frequently enough to avoid perception of "new normal" on changes. You change stuff and let it be "normal" before you change it again and you'll frustrate people.

weekly works because not everyone plays every day. Five days would be ideal but the first few updates need to be really fast, back to back. It's about managing expectations and setting the expectation at "this will change".

Currently we have "this is what you will have for 18 months"

You'll frustrate people, but you need to have things be "normal" for at least a couple weeks so the impact of those changes can be sorted out. That simply takes time.

I'd argue that, to placate everyone involved, we should have a fixed interative balance interval: Say the first patch of every month containing small weapon balance changes. Have this every single month, without fail, until they can confidently stand and say "All the weapons are perfectly balanced" (hint: That will never happen)

Then they've got a full month to examine the impact of balance changes, and move from there. If they feel a month isn't enough time, they could alternate subjects. For example:
This month, PPC's have their velocity slightly increased, LBX's have their ammo increased, flamers get a damage buff.

Next month, SRM's get a damage increase, AC/2's get a heat decrease

The next month, LBX's are revisted and get a ROF increase, PPC's have their cycle rate reduced, flamers generate less heat.

This creates a situation where players expect balance changes at specific times, but PGI can take their time examining the impact of those changes - be it for a month, or two.

But most importantly, there's constant iteration. Players can see constant progress, and they don't get that sinking feeling that "This [weapon] sucks, and is going to suck for [18 months]/[forever]/[whatever]."

[Lyoto Machida]

Wintersdark, on 23 September 2015 - 05:45 PM, said:

You'll frustrate people, but you need to have things be "normal" for at least a couple weeks so the impact of those changes can be sorted out. That simply takes time.

I'd argue that, to placate everyone involved, we should have a fixed interative balance interval: Say the first patch of every month containing small weapon balance changes. Have this every single month, without fail, until they can confidently stand and say "All the weapons are perfectly balanced" (hint: That will never happen)

Then they've got a full month to examine the impact of balance changes, and move from there. If they feel a month isn't enough time, they could alternate subjects. For example:
This month, PPC's have their velocity slightly increased, LBX's have their ammo increased, flamers get a damage buff.

Next month, SRM's get a damage increase, AC/2's get a heat decrease

The next month, LBX's are revisted and get a ROF increase, PPC's have their cycle rate reduced, flamers generate less heat.

This creates a situation where players expect balance changes at specific times, but PGI can take their time examining the impact of those changes - be it for a month, or two.

But most importantly, there's constant iteration. Players can see constant progress, and they don't get that sinking feeling that "This [weapon] sucks, and is going to suck for [18 months]/[forever]/[whatever]."

Ahem...the guy who coded weapon value changes left the company. I think they only hire temps to make changes every 18 months (as per their budget).

[Mcgral18]

MischiefSC, on 23 September 2015 - 05:30 PM, said:

No less frequently than 1 time a week. The point is momentum. Your brain changes how you think about stuff after 72 hours. At 10 days you start to lose perspective on what prior iterations were like and initial impressions on the new one, which are the important indicators you need. A week is still too long but is frequently enough to avoid perception of "new normal" on changes. You change stuff and let it be "normal" before you change it again and you'll frustrate people.

weekly works because not everyone plays every day. Five days would be ideal but the first few updates need to be really fast, back to back. It's about managing expectations and setting the expectation at "this will change".

Currently we have "this is what you will have for 18 months"

Not sure if it would require a download/patch or not. Map rotations are server side, so the Weight on local files aren't necessarily the ones in use (as seen with the event) but weapon stats I'm not too sure about.

By 2 weeks, I do mean with Patches, which can be 3 weeks apart on the odd month. Download the new weapon stats at the same time, without causing much annoying on anyone taking part.


Winter also gives a nice point; you need time to judge weapon adjustments, as everyone flocks to them immediately, and that craze has to die done before "correct" numbers are found.

LRM speed buffs brought that issue; hotfixed 15M/s almost immediately.


Monthly I feel is a tad too long, but if it's consistent I'll gladly take that. 2 weeks would give plenty of time for the craze to die down.

Edited by Mcgral18, 23 September 2015 - 05:52 PM.

[Jman5]

In my opinion, aside from a few outliers like flamers, balance is not terrible. You can do well with a lot of different mechs and a lot of different weapons.

That said, I have absolutely no idea why PGI is so effing slow to fix those outlier weapons which could be improved with simple adjustments.

• Flamers; Double the DPS and just see where it goes. It's not great, but at least you're close to medium laser DPS.
  
• LBX: pellet damage from 1 - 1.5
  
• LRM 20: Cooldown 5s - 4.5s

Boom. Not hard. They may not be perfect or uber competitive, but at least they are much better than before.

Edited by Jman5, 23 September 2015 - 05:55 PM.

[Mcgral18]

Deathlike, on 23 September 2015 - 05:19 PM, said:

Maybe someday we won't make jokes about the AC2. Today, we'll just keep posting "steering wheel" jokes referencing the LBX10. We can't have nice things in MWO.

Sadly, I'm not sure we can do that...

If the Nerfinator wants the LBx to be Crit weapons,

critDamMult="2.0" >> critDamMult="5.0"

That magical change right there, makes them glorious Crit Weapons.

Want them to hit harder?

damage="1" >> damage="1.4" 

That is, if it's actually that easy to adjust weapon stats. We still haven't gotten clarification on that part.

Then, someone on Reddit reminded me of something...remember when they touched the Ghost Heat limit on AC2s?
They changed this thing:

minheatpenaltylevel="4"

Or, removed it entirely...remember how it broke Clam Arm Actuators?


I guess some QA testing would be required, even for those simple variables...I had forgotten there was that side effect. Probably something unrelated, but still worth a note.

[Mcgral18]

Jman5, on 23 September 2015 - 05:54 PM, said:

In my opinion, aside from a few outliers like flamers, balance is not terrible. You can do well with a lot of different mechs and a lot of different weapons.

That said, I have absolutely no idea why PGI is so effing slow to fix those outlier weapons which could be improved with simple adjustments.

• Flamers; Double the DPS and just see where it goes. It's not great, but at least you're close to medium laser DPS.
  
• LBX: pellet damage from 1 - 1.5
  
• LRM 20: Cooldown 5s - 4.5s

Boom. Not hard. They may not be perfect or uber competitive, but at least they are much better than before.

Well, quirkless inter faction, XL disparities, PoorDubs, and missiles in general (specifically SRMs, or isSSRM2s).

As well as the MGs, Flamers, as you say LBx series.



It's what makes me wonder, exactly as you say, why don't those things happen?
It's possible I/we don't understand the system it had to go through...but come on, 3 years for the Flamer?

Edited by Mcgral18, 23 September 2015 - 06:05 PM.

[Deathlike]

Mcgral18, on 23 September 2015 - 05:58 PM, said:

Sadly, I'm not sure we can do that...

If the Nerfinator wants the LBx to be Crit weapons,

critDamMult="2.0" >> critDamMult="5.0"

That magical change right there, makes them glorious Crit Weapons.

Want them to hit harder?

damage="1" >> damage="1.4"

That is, if it's actually that easy to adjust weapon stats. We still haven't gotten clarification on that part.

Then, someone on Reddit reminded me of something...remember when they touched the Ghost Heat limit on AC2s?
They changed this thing:

minheatpenaltylevel="4"

Or, removed it entirely...remember how it broke Clam Arm Actuators?


I guess some QA testing would be required, even for those simple variables...I had forgotten there was that side effect. Probably something unrelated, but still worth a note.

If such a change breaks something else... then that's on whoever is coding the damn weapons.

Usually changing numbers around doesn't cause said problems, but sometimes using the wrong data types (like the 9 ERPPC Direstar that wouldn't die from Ghost Head) is problematic.

One would hope that you wouldn't code something that isn't representative of what is stated (like MW2 Mercs Titanium Pack Clan Medium Pulse Lasers).