174 replies to this topic

[Joseph Mallan]

ReXspec, on 09 February 2014 - 11:16 AM, said:

When I finish collecting them, I'll run them by you first before posting'em, Joe. The issue right now is how many games to document. Right now I've documented about 265 games and the reaction to the discussion, but the rather daunting issue is what my target number for data collection should be... and right now I'm stuck between 1,000 - 5,000 matches.

Well you needa pool of at least 25 samples to start a SPC chart, and that remaing 240 would give you a solid shot at a good established Control range. 1,000 and you can be pretty safe calculating the CPK score and be fairly certain of how many fliers you can expect per Hundred thousand results or more.

[ReXspec]

Roadbeer, on 09 February 2014 - 11:30 AM, said:

And of course your compiled data comes with screen shots/video, and % of the 24 players/match polled who responded, as well as the % of players/match who even know about it, correct?

Well, to gather video from 1,000 - 5,000 matches would be arduous, however, it is far more plausible to collect e-signatures from players I DID discuss it with (google docs makes this quick and easy and follow-ups via PMs are easy as well). Creating a thread that can act as a "signature" thread is also an idea, but again, the primary concern is just collecting the mountains of data that I'll have to sift through. lol

[Joseph Mallan]

Roadbeer, on 09 February 2014 - 11:30 AM, said:

And of course your compiled data comes with screen shots/video, and % of the 24 players/match polled who responded, as well as the % of players/match who even know about it, correct?

He is saying he wants to step up and collect the data, Gotta give him credit for that, It is more than most of us are doing.

[Victor Morson]

Veranova, on 09 February 2014 - 06:46 AM, said:

For the record if they removed ghost heat, I would not hesitate to build a 6ppc stalker.

With the new (or old, given it was the original value) PPC heat, firing that would instantly murder you without Ghost Heat.

[Roadbeer]

Hey, I'm all for the attempt and all, but it's kind of like saying "Are you for or against genetically modified food?"
First, for those who don't know, you have to explain what it is, (without being biased in your description as it's going to skew your results), then collect your data.
I don't know a person who says "Hey, Ghost Heat is a great idea", everyone I know either dislikes it or (like myself) just doesn't care. The only people it truly impacts are the high alpha strikers, and it was put in as a direct response to the literally hundreds of "NERF (insert FOTM here)" threads.
Subsequently, since the implementation of Ghost Heat, I have seen dozens of threads created saying "Ghost Heat sucks, do this instead" and basically each of those threads are as hotly contested as Ghost Heat itself.

IMO, it comes down to what does the casual player REALLY need to know (since that is what this topic is about)? Which is more confusing for your player who just wants to jump in and play the game?
1. "If you put more than X number of Y weapons on your build, you will exponentially build heat IF they are fired at the same time UNLESS you do A,B,C.... Now here is a confusing chart to explain how it all works."
2. "Firing more than one type of weapon at the same time, MAY cause your heat to increase faster".

I'm fairly certain that the second option appears as a screen tip when you're dropping onto a map.

Now, maybe there could be a section in the tool tip for each weapon giving the # of weapons to stay under Ghost Heat, I'm cool with that.
But does the UI need to become the manual for Falcon 4.0? Nope. Because if we're going to use all the maths in the UI to explain how Ghost Heat works, we might as well do it for Terrain Navigation (how the mech you choose climbs a hill), and the voodoo that determines how quickly and how long missile lock retains it's target as well.

[Victor Morson]

Roadbeer, on 09 February 2014 - 12:08 PM, said:

IMO, it comes down to what does the casual player REALLY need to know (since that is what this topic is about)? Which is more confusing for your player who just wants to jump in and play the game?
1. "If you put more than X number of Y weapons on your build, you will exponentially build heat IF they are fired at the same time UNLESS you do A,B,C.... Now here is a confusing chart to explain how it all works."
2. "Firing more than one type of weapon at the same time, MAY cause your heat to increase faster".

I'm fairly certain that the second option appears as a screen tip when you're dropping onto a map.

What would work (on a similar vein) is this:

• Add a "Heat Scaling Limit" value to the weapons
  
• Have it impact the Heat graph.
  
• When you go to save give a notice similar to yours, but with more detail such as: "You have exceeded Heat Scale limits. Firing linked weapons in a .5 second period will generate <AMOUNT HERE> additional heat."

Where things get tricky in the UI design is indicating what weapons are linked. Like how the LRM/5 doesn't link to any other LRM, or the SRM2 on a similar note. That needs to be in there, but is very hard to explain in the tool tips.



Honestly this is why they just need a bloody Help Button that goes over all of this stuff. I'm sure ANYONE can code a quick reference help menu with screenshots and plain text, and almost all of the detailed stuff (including a visually upgraded copy of the Smurfy Heatscaling chart) could be dropped in there.

Edited by Victor Morson, 09 February 2014 - 12:18 PM.

[Sandpit]

ReXspec, on 09 February 2014 - 11:21 AM, said:

You're assuming that the data is being collected solely on the basis that I'm asking players whether they like ghost heat or not. Truth is, I'm also asking what their suggestions for showing it in the U.I. would be, and what their possible alternatives for ghost heat would be.

I've taken a Sociology class before, Sandy. Don't you worry your pretty, little head about "biased" data.

Point being that question is biased, you should know that if you've taken a soc class. Putting it in with a list of other questions isn't going to change that.
That question is completely irrelevant. Lots of people don't "like" Locusts, so if I asked them if they'd rather have had a javelin instead of the locusts I'm sure most would agree. It's a biased question.

Truth be told if you do this right you could get a half way decent sample (the other problem is WHO you sample. The better thing to do would be put it up in an open forum and do it double blind so that any and everyone from the community can answer it because no matter how much you try if you just sample those in your group of buddies it's going to be biased)
But I'm sure you already knew that. I'm sure you'll also take steps to ensure there aren't any double posts in the answers, alt accounts questioned, sample from new and old players, etc.

Just like I'm sure you'll take steps to ensure you have a control and such. I have the utmost faith in you. You'll probably do a great job pulling in the data. It will be interesting to see your testing process and the data collected from it

[Victor Morson]

Sandpit, on 09 February 2014 - 12:16 PM, said:

That question is completely irrelevant. Lots of people don't "like" Locusts, so if I asked them if they'd rather have had a javelin instead of the locusts I'm sure most would agree. It's a biased question.

PGI could have done things to the Locusts to make them far less of a joke than they are. That Minimum Heatsink issue outright murders the 'mech, and frankly, 'mechs under 30 tons should get an exception to that rule. Likewise their hit box is far worse than a Spider, and hitting a Locust is far, far, far too easy.

Long story short is the Locust was kind of doomed from the start but could at least justify it's place for it's tonnage if not for some serious handicaps like that.

[ReXspec]

Sandpit, on 09 February 2014 - 12:16 PM, said:

Point being that question is biased, you should know that if you've taken a soc class. Putting it in with a list of other questions isn't going to change that.
That question is completely irrelevant. Lots of people don't "like" Locusts, so if I asked them if they'd rather have had a javelin instead of the locusts I'm sure most would agree. It's a biased question.

Truth be told if you do this right you could get a half way decent sample (the other problem is WHO you sample. The better thing to do would be put it up in an open forum and do it double blind so that any and everyone from the community can answer it because no matter how much you try if you just sample those in your group of buddies it's going to be biased)
But I'm sure you already knew that. I'm sure you'll also take steps to ensure there aren't any double posts in the answers, alt accounts questioned, sample from new and old players, etc.

Just like I'm sure you'll take steps to ensure you have a control and such. I have the utmost faith in you. You'll probably do a great job pulling in the data. It will be interesting to see your testing process and the data collected from it

I detect a hint of sarcasm in this post...

You'll have to forgive me, it's kind of hard to read someone over the internet. lol

[Sandpit]

Victor Morson, on 09 February 2014 - 12:21 PM, said:

PGI could have done things to the Locusts to make them far less of a joke than they are. That Minimum Heatsink issue outright murders the 'mech, and frankly, 'mechs under 30 tons should get an exception to that rule. Likewise their hit box is far worse than a Spider, and hitting a Locust is far, far, far too easy.

Long story short is the Locust was kind of doomed from the start but could at least justify it's place for it's tonnage if not for some serious handicaps like that.

Still not the point. A biased question or presentation of information is biased and will slant and skew results to suit the agenda of the data collector

[Sandpit]

ReXspec, on 09 February 2014 - 12:30 PM, said:

I detect a hint of sarcasm in this post...

You'll have to forgive me, it's kind of hard to read someone over the internet. lol

No sarcasm. If you've got the time and energy go for it. I still don't think what you've described thus far is a good way to collect data though. You're going to have to come up with something that does away with your own bias regarding the issue so the data isn't skewed. If you're going to do it, just do it right.
If you really do understand stats and data collection then you know what I'm talking about and you should have ways to present the survey and data in an unbiased manner.
No small chore when talking about a community of gamers who have a history of slanting data, misrepresenting data, and providing small anecdotal evidence as "proof" of things.

Regardless of your personal attacks and vendetta against me I can check my bias at the door and am interested in looking through your testing methods and data collection.

[Roadbeer]

Victor Morson, on 09 February 2014 - 12:16 PM, said:

Where things get tricky in the UI design is indicating what weapons are linked. Like how the LRM/5 doesn't link to any other LRM, or the SRM2 on a similar note. That needs to be in there, but is very hard to explain in the tool tips.

Now you're becoming more reasonable, but again, Why does it need to be in there?

I'm fairly certain that 80% of the worlds population neither knows nor cares WHY the sky is blue, they're content just in knowing that it is. Those who want to know, are more than capable of researching it.
Same goes for Ghost Heat.
I'm surprised at you Vic, because in one thread, you complain about how buggy and counter-intuitive UI 2.0 is, and in this one you want to pile on to it. All of these things take system resources when you load the game, so I'm DEAD SET against adding more into it than necessary.

NOW, if you say "there should be a detailed manual of the game" I'll 100% back your play. They can create a section of the website, and they can create a Falcon 4.0 manual breaking down how each esoteric facet of the game works.
In fact, since it's such an issue for you, how about you write the section on Ghost Heat, and break it all down, and do it in an unbiased way so it can be pinned in the New Player section.

Because right now, if someone WAS actually curious about how Heat Scaling works, they have to wade through dozens of threads like this to find the threads in the CC section that I pointed out on the first page of this thread.

[Victor Morson]

Roadbeer, on 09 February 2014 - 12:34 PM, said:

Now you're becoming more reasonable, but again, Why does it need to be in there?

I'm fairly certain that 80% of the worlds population neither knows nor cares WHY the sky is blue, they're content just in knowing that it is. Those who want to know, are more than capable of researching it.

That is kind of a ludicrous statement, though, when it comes to something as important as knowing which weapons link to each other, and having a reference without external sources in the game.

This isn't a hidden mechanic (why is the sky blue). Explaining, say, the way beam lasers do ticks for damage and such - that fits what you are talking about. This is a major thing for 'mech design.

For example if I'm running LRM/30 I've been running:

1x 15, 1x 10, 1x 5. That way I have two launchers that link, one that does not, and I do not violate anything.

Configs I see pugs run all the time:

3x LRM/10
1x LRM/20 2x LRM/10

Now in both of those, the pug is taking Ghost Heat unnecessarily. In the first example, not much; they aren't even aware they're being hit by it, but they are running warmer than they should.

The second example is why this becomes important. They are taking 3x LRM/20 penalty, despite only having a single LRM/20.

Another example would be builds that don't know the SRM/2 goes outside of the rules; 4-6 SRM4/6 will cause huge problems, but some builds can be setup that run 3 "large" launchers and some small ones to get a bigger salvo; again, you wouldn't know this without Smurfy's.

You should never have to go to another website to get information on major systems.

Knowing which weapons chain and which weapons do not is not some nebulous behind the scenes mechanic; it's a vital bit of information about design. And vital stuff should be in the UI.

Edited by Victor Morson, 09 February 2014 - 12:42 PM.

[Prezimonto]

Roadbeer, on 07 February 2014 - 01:18 PM, said:

I'll just leave these here

http://mwomercs.com/...general-update/
http://mwomercs.com/...cale-the-maths/

Found on the front page of here.

http://mwomercs.com/...-command-chair/

Not disagreeing that it couldn't be better documented, But to assert that there is nothing, is entirely disingenuous.

They've talked about it, but a MUCH better way to describe it is here:
http://mwo-builds.ne...ons-table/1001/

They need a simple graphical tool tip when building a mech (or even just when creating a weapon group) that lets the player know that this grouping will suffer from some or LOTS of extra heat.

In the game this could be something simple like a BB warning a red flash on the weapon groupings bar when you get hit with a penalty.

It wouldn't take a lot to let players know something special, and bad, is happening.

[Victor Morson]

Sandpit, on 09 February 2014 - 12:30 PM, said:

Still not the point. A biased question or presentation of information is biased and will slant and skew results to suit the agenda of the data collector

To be fair, I don't know what you are trying to prove with the Locust example, honestly. If your point is that "bias" would result in the Javelin being far more loved than the Locust well.. you're probably right. Yeah, they should have probably put a Javelin that people would actually want on their team in over a 'mech that - due to a bunch of factors, including what I just talked about - is gimped and a detriment to the team.

It's less bias and more "Would you rather have had something practical?"

EDIT: And if you want to talk Bad vs Bad, I absolutely guarantee more people would have been interested in an Urbanmech than a Locust, too. There's such a thing as a reasonable bias, like not trusting a known thief with your wallet.

Edited by Victor Morson, 09 February 2014 - 12:46 PM.

[Roadbeer]

Victor Morson, on 09 February 2014 - 12:45 PM, said:

It's less bias and more "Would you rather have had something practical?"

Still biased, because in that very wording, you are stating that the locust is impractical based off of your personal opinion. I've seen some very effective locust pilots (no me, but I've seen them), at which point, you're already skewing the poll with your bias.

[Imperius]

I think the UI should have a Warning box that is in every menu to provide help. In this warning box ghost heat warning should give info if the mech has the potential to fire a ghost heat alpha. Also in this warning box messages like no ammo/engine should be displayed. After all UI 2.0 was supposed to be made for the new user in mind. I have quickly thrown a build together before and forgot to change the ammo. (AC5 to Ultra AC5)

Edited by Imperius, 09 February 2014 - 01:09 PM.

[Sandpit]

Victor Morson, on 09 February 2014 - 12:45 PM, said:

To be fair, I don't know what you are trying to prove with the Locust example, honestly. If your point is that "bias" would result in the Javelin being far more loved than the Locust well.. you're probably right. .

That's exactly what I was getting at. It was in reference to respeco up there. The question he posed to collect data is biased the same way the question abotu Locusts and Javelins would be. If he's going to collect data and what's it taken seriously and wants it to be "legit" there's certain things that must be done during the data collection otherwise the results are biased and skewed from step 1 which would result in all of the data collected being unusable if you want it taken seriously and be considered credible.

Of course I'd rather have had a Javelin than a Locust, so would the vast majority of other players I'm sure, that's exactly why asking that kind of question in regards to something akin to judging the Locust's "worth" is bad and biased and won't give you accurate data. That's all the locust example was for though.

I personally think it's going to be near impossible to collect accurate and reliable data like what he's wanting though. You would have to come up with an unbiased survey and then get a few thousand responses to it minimum to have anything even close to a good sample size. I'm coming up with the ball park "few thousand" based on the largest polls seen on the forums. Whereas you typically have a few hundred visiting the forums on a regular basis over a few weeks, you had thousands pour out for 3pv poll.

Then you have to ensure that players aren't creating alt accounts to take the survey twice, then you have to collect the data. Without formulas in excel or using SPSS that's a very time consuming process.

There's a lot of stuff involved above and beyond "well I played 12 games last night and this is what happened". I'm not saying it can't be done (although there a lot of obstacles as I've stated) but getting the kind of data he's wanting is definitely going to be no small undertaking

[Victor Morson]

Imperius, on 09 February 2014 - 01:08 PM, said:

I think the UI should have a Warning box that is in every menu to provide help. In this warning box ghost heat warning should give info if the mech has the potential to fire a ghost heat alpha. Also in this warning box messages like no ammo/engine should be displayed. After all UI 2.0 was supposed to be made for the new user in mind. I have quickly thrown a build together before and forgot to change the ammo. (AC5 to Ultra AC5)

All good points; warning boxes for more things than Ghost Heat would be exceptional. "Warning: You have no ammo for <X Weapon>" would be a great thing for everyone.

I'm sure everyone here has done that before!

Edited by Victor Morson, 09 February 2014 - 01:31 PM.

[Sandpit]

Victor Morson, on 09 February 2014 - 12:45 PM, said:

There's such a thing as a reasonable bias, like not trusting a known thief with your wallet.

While that may be true, that doesn't apply to this kind of example. Those examples are purely subjective and based on personal opinion, not a fact such as the known thief example

LOTS of people say energy weapons are useless (along with SRMs, SSRMs, LRMs, NARC, etc., etc., etc.) but that doesn't make it true. It only makes it true for them personally. Those opinions will vary based on experience, macth records, loadouts (a loadout on a mech may not work for a player while another one will on the same mech)

This is why collecting meaningful data is so difficult, especially when you don't have factual data to work from. The whole "Every game ends in a stomp" thread is actually what kicked a lot of this off for me. I like to be an optimist. Therefore I like to think (until proven different) that most people simply don't understand how statistics and data collection work and why
40+ matches over 48 hours resulting in 4 stomps does not in any way shape or form prove anything other than the player was involved in 4 stomps.

So I posted information on how statistics actually work and why you can't perform data collection like that and have it taken seriously by anyone that is either unbiased OR understands how stats work. That's getting a bit off-topic though but I'm just trying to get more people to understand why it isn't personal and better yet why their data isn't always taken seriously

Imperius, on 09 February 2014 - 01:08 PM, said:

I think the UI should have a Warning box that is in every menu to provide help. In this warning box ghost heat warning should give info if the mech has the potential to fire a ghost heat alpha. Also in this warning box messages like no ammo/engine should be displayed. After all UI 2.0 was supposed to be made for the new user in mind. I have quickly thrown a build together before and forgot to change the ammo. (AC5 to Ultra AC5)

Victor Morson, on 09 February 2014 - 01:31 PM, said:

All good points; warning boxes for more things than Ghost Heat would be exceptional. "Warning: You have no ammo for <X Weapon>" would be a great thing for everyone.

I'm sure everyone here has done that before!

Agreed. I've completely forgotten to load ammo when putting a new build together before

This should get posted in the feedback thread. +1