Jump to content

Round 11 = Turret 115 Hp

informative

45 replies to this topic

#21 Virlutris

    Member

  • PipPipPipPipPipPipPipPip
  • The Spear
  • The Spear
  • 1,443 posts
  • Google+: Link
  • Twitter: Link
  • LocationVery likely goofing off in a match near you.

Posted 29 October 2015 - 10:16 AM

View PostDave Forsey, on 29 October 2015 - 08:34 AM, said:

So I've changed the opening screen to allow the Player to change the starting Round using the up/down arrow keys.

This does away with the original reasoning for using an exponential function (which was for the player to not spend too long at the easier levels).

So under this new condition, which do you think the player base (not just you) would prefer?

1) A linear increase each round (2 or 3)

2) The n^1.7 (or whatever) suggested by Hornsby

I'm leaning towards linear because eking out an extra 2 points at higher damage levels can be tough

I'm voting Hornsby, because let's try it :D

Using Hornsby's math would give us an incremental change. It would still ramp up the diffuculty interestingly at the later levels, taking a bit longer to do so. It would also do it without requiring 50 levels (guesstimating!) to get there via linear change.

If it turns out that your (Dave's) suspicions are right that linear is bettah, then we've eased into it while taking an opportunity to try a little science and confirm/disconfirm the hypotheses of a player and a dev. It's win-win. :)

#22 BSK

    Member

  • PipPipPipPipPipPipPipPip
  • Mercenary Rank 7
  • Mercenary Rank 7
  • 1,040 posts

Posted 29 October 2015 - 11:06 AM

The linear increase is the most motivating and challenging in my opinion.

I have thought about the feedback for an uav training ground. Have you ever considered to change the hovering speed depending on the player tiers? Giving them lower moving speed at lower levels for example?

#23 pyrocomp

    Member

  • PipPipPipPipPipPipPipPip
  • Philanthropist
  • 1,036 posts

Posted 29 October 2015 - 01:57 PM

View PostDave Forsey, on 29 October 2015 - 08:34 AM, said:

So I've changed the opening screen to allow the Player to change the starting Round using the up/down arrow keys.

This does away with the original reasoning for using an exponential function (which was for the player to not spend too long at the easier levels).

So under this new condition, which do you think the player base (not just you) would prefer?

1) A linear increase each round (2 or 3)

2) The n^1.7 (or whatever) suggested by Hornsby

I'm leaning towards linear because eking out an extra 2 points at higher damage levels can be tough

Linear is a kind more intuitive. But since at higher tiers it's hard to deal even an extra point, I'd suggest a much flatter curve (N^something_less_than_1) at higher rounds. This will make it more attractive to try to go further than round N at which the damage needed to kill all the turrets exceeds the total possible damage you can deal within fixed time even at optimal range (which is ultimate level, but crits allow to go further). The addition of 0.1 hp to the turret makes it possible to kill it running closer to it, hence a little more challenge to optimize the route and oder (if turrets are fixed).

Edited by pyrocomp, 29 October 2015 - 01:57 PM.


#24 Dave Forsey

    Senior Designer

  • Developer
  • Developer
  • 198 posts

Posted 29 October 2015 - 04:19 PM

Good point about having fractional health values for the turret.

Seems that the curve should be more like a sigmoid (or logistic) curve, so the increment at the higher rounds tapers off

#25 pyrocomp

    Member

  • PipPipPipPipPipPipPipPip
  • Philanthropist
  • 1,036 posts

Posted 29 October 2015 - 05:06 PM

View PostDave Forsey, on 29 October 2015 - 04:19 PM, said:

Good point about having fractional health values for the turret.

Seems that the curve should be more like a sigmoid (or logistic) curve, so the increment at the higher rounds tapers off

Yep, error function (F) also fits. But there is always a possibility to overestimate of underestimate the max possible damage, so maybe 0.1*N+F(N/_something_) is safer against unexpected build after any rebalance. You don't want to test it to that level after each patch, don't you? :)

#26 Omi_

    Member

  • PipPipPipPipPipPip
  • The Blade
  • 336 posts
  • LocationWinnipeg, Manitoba, Canada

Posted 29 October 2015 - 09:42 PM

Wow, my post picked up some traction since I last checked. :D

So before I posted that graphic, I also considered linear and sigmoid curves. While I believe my own curve of n^1.7 would be a marked improvement over the current numbers in-game, I admit wholeheartedly that it only delays the problem of turret health growing too quickly. In fact, any exponential function will only delay the issue.

Linear would probably be ideal, since as you said it can be tough to squeeze out even 2 damage at the higher levels. I dismissed it because I figured it would take forever for the game to become challenging for the heavier classes and I wasn't considering design changes such as allowing the player to pick a starting level. So with that said, I would actually vote for a linear system over my own curve.

A sigmoid would, at first glance, yield the benefit that the game provides some warm-up rounds and then accelerate towards the more challenging levels, eventually tapering off as the going gets tough. However, the sigmoid would naturally have to target the junction at which the greatest rate of increase between levels would be desired, depending on some amount of firepower and the average skill level of the userbase using it. As this is a moving target, the scale and offset for the sigmoid would need to be dynamic and adjust according to the weapons loadout of the player or else you'll overshoot/undershoot the ideal degree of challenge. If the goal of Rapid Fire is for the player to try to get a high score with respect to their chosen spread of weapons, a static sigmoid function cannot work well.

TL;DR:
I agree linear is better with the starting level being the tuning element that adjusts player challenge. The player will always feel like they can reach further and they won't need to dredge through rounds that are too low for their loadout. Sigmoid is a novel idea but won't work because you can't easily tailor the curve to every player who's going to use Rapid Fire, and there isn't really any benefit to trying.

Edited by Hornsby, 29 October 2015 - 10:13 PM.


#27 Medi0cr3

    Member

  • PipPipPip
  • Bad Company
  • Bad Company
  • 82 posts

Posted 30 October 2015 - 10:29 AM

View PostDave Forsey, on 25 October 2015 - 02:16 PM, said:

Congrats!

This is exactly how I was hoping players would use the Academy to challenge themselves and hone their skills.


Most ******** thing I've ever seen. If you want players to hone skills make the turrets in variable places instead of fixed easy to muscle memory places.............................. so lame............

View PostMedi0cr3, on 30 October 2015 - 10:27 AM, said:


Most r3t4rd3d thing I've ever seen. If you want players to hone skills make the turrets in variable places instead of fixed easy to muscle memory places.............................. so lame............


#28 Dave Forsey

    Senior Designer

  • Developer
  • Developer
  • 198 posts

Posted 30 October 2015 - 11:04 AM

Folks, only warning.

I'll leave mediocr3's post up as an example, but if you can't put forward your viewpoint without insults, the post will be deleted without comment or explanation.
If you want me to stick around in the forums, treat it as if you're in my living room talking to me face to face. I won't put up with childish behaviour in my own home, and I won't do so here.

#29 Dave Forsey

    Senior Designer

  • Developer
  • Developer
  • 198 posts

Posted 30 October 2015 - 11:09 AM

Yeah the sigmoid shape was more descriptive than prescriptive - just to capture the idea that the increment drops off as the rounds get higher.

The increments themselves would look more like an exponential decay.... and we're talking more like a max value of 5 and a min value of.... what.. 0.1?????

#30 Elizander

    Member

  • PipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 7,540 posts
  • LocationPhilippines

Posted 30 October 2015 - 03:06 PM

Would be nice if these little things could also be included in the profile info on the forums or made public somehow.

#31 pyrocomp

    Member

  • PipPipPipPipPipPipPipPip
  • Philanthropist
  • 1,036 posts

Posted 30 October 2015 - 03:28 PM

View PostDave Forsey, on 30 October 2015 - 11:09 AM, said:

Yeah the sigmoid shape was more descriptive than prescriptive - just to capture the idea that the increment drops off as the rounds get higher.

The increments themselves would look more like an exponential decay.... and we're talking more like a max value of 5 and a min value of.... what.. 0.1?????

Let it realy be 0.1. Some time ago there was a thread (do not remember much detail to find it) that within some units there go competion who will finish all mech in the training grounds faster. Bet those units will use this opportunity to compete also.

#32 Omi_

    Member

  • PipPipPipPipPipPip
  • The Blade
  • 336 posts
  • LocationWinnipeg, Manitoba, Canada

Posted 30 October 2015 - 03:49 PM

View PostDave Forsey, on 30 October 2015 - 11:09 AM, said:

The increments themselves would look more like an exponential decay.... and we're talking more like a max value of 5 and a min value of.... what.. 0.1?????

At this point, I think you're better off going with whatever feels good while testing. It all sounds good on paper.

While responding to this, I thought back to a game called Sin Episodes: Emergence, and how it had a skill-based survival mode that tracked the player's progress and matched it's challenge level according to their performance. If you're still considering design changes to Rapid Fire, have you considered letting the player jump ahead multiple rounds if they have a large proportion of time left at round end? With letting the player just pick a later round to start at, they might miss out on some needed warm-up depending on how slowly your curve decays.

If you do consider skipping rounds, maybe require that the player have many consistently great rounds before engaging round-skip to be sure they don't just suddenly overheat due to turret health skyrocketing unexpectedly.

#33 Virlutris

    Member

  • PipPipPipPipPipPipPipPip
  • The Spear
  • The Spear
  • 1,443 posts
  • Google+: Link
  • Twitter: Link
  • LocationVery likely goofing off in a match near you.

Posted 02 November 2015 - 02:21 PM

Okay, after watching you guys sort this out, I'm coming around to the benefits of a linear-ish progression with a tapering effect.



#34 Dave Forsey

    Senior Designer

  • Developer
  • Developer
  • 198 posts

Posted 02 November 2015 - 11:46 PM

I'm thinking of letting the player jump to any damage value at the end of each round. After all, it's not a contest.
(BTW - I don't have any user-input UI exposed to the Flowgraph, or I'd use the simple expedient of letting the player type in a value... :-) )

#35 Nightshade24

    Member

  • PipPipPipPipPipPipPipPipPip
  • Overlord
  • Overlord
  • 3,972 posts
  • LocationSolaris VII

Posted 03 November 2015 - 02:36 AM

I must say Dave Forsey. It is very very good to see you as a member of PGI and being a senior designer to be directly communicating with the players and community about things (in this case: A certain challenge in the current Mechwarrior Academy).

I can not simply express how happy I am to see this. It is really good to see this communication as well as seeing inputs on your ideas of it as well as your response and most of all looking at the communities opinions and responses to this.

Good work so far! I can not wait to see this event expand... however I think I will not be getting any high scores any time soon.

#36 Jay Z

    Member

  • PipPipPipPipPipPip
  • Deadset Legend
  • Deadset Legend
  • 436 posts
  • Twitch: Link
  • LocationAustralia

Posted 03 November 2015 - 06:13 AM

View PostDave Forsey, on 02 November 2015 - 11:46 PM, said:

I'm thinking of letting the player jump to any damage value at the end of each round. After all, it's not a contest.
(BTW - I don't have any user-input UI exposed to the Flowgraph, or I'd use the simple expedient of letting the player type in a value... :-) )


Awesome! An up/down arrow is fine. Adjustable time limit/number of targets would be nice to.

#37 Darth Futuza

    Member

  • PipPipPipPipPipPipPipPip
  • 1,239 posts

Posted 03 November 2015 - 01:23 PM

View PostDave Forsey, on 02 November 2015 - 11:46 PM, said:

I'm thinking of letting the player jump to any damage value at the end of each round. After all, it's not a contest.
(BTW - I don't have any user-input UI exposed to the Flowgraph, or I'd use the simple expedient of letting the player type in a value... :-) )

Should be nice, I never liked games that make me replay the easy level a whole bunch of times just cause I want to play that one hard level and beat it.

How's the progress for the whack-a-mole idea you suggested earlier?

#38 Omi_

    Member

  • PipPipPipPipPipPip
  • The Blade
  • 336 posts
  • LocationWinnipeg, Manitoba, Canada

Posted 03 November 2015 - 09:20 PM

View PostDave Forsey, on 02 November 2015 - 11:46 PM, said:

I'm thinking of letting the player jump to any damage value at the end of each round. After all, it's not a contest.
(BTW - I don't have any user-input UI exposed to the Flowgraph, or I'd use the simple expedient of letting the player type in a value... :-) )


Sounds like a solid plan! And I mirror others' thoughts that it is wonderful having you here on the forums. :)

#39 LORD ORION

    Member

  • PipPipPipPipPipPipPipPip
  • Heavy Lifter
  • Heavy Lifter
  • 1,070 posts

Posted 04 November 2015 - 06:53 AM

View PostDave Forsey, on 30 October 2015 - 11:04 AM, said:

Folks, only warning.

I'll leave mediocr3's post up as an example, but if you can't put forward your viewpoint without insults, the post will be deleted without comment or explanation.
If you want me to stick around in the forums, treat it as if you're in my living room talking to me face to face. I won't put up with childish behaviour in my own home, and I won't do so here.


You are not in your living room. You are polling ideas from the web.
You will never be a jedi until you can filter out extraneous content and get to the ideas within them...
Because what is not said can be as important as what is said.
I wonder how often the best ideas are discarded or overlooked because a person does not like words they are wrapped in.

Regardless, on topic, I'll add that I don't care for the challenge because I must constantly redo the lesser levels once I have a good build going to test my mettle at the higher levels.

Now, going back to being a jedi... what was not said in that content? Hmmm... You might come up with the thought "Maybe I could find a way to let players select the levels they have completed? That way when they are in the mood to play the challenge, they can try to blow away targets at their highest level, and then adjust their mech ideas, and repeat, instead of quiting it and never returning.

So now... on the subject of polling the internet for ideas. In a country that advocates free speech and equality (as in the government is legally not able to punish you for your words or life experiences overtly) are you getting into the spirit of being a member of the ideals of the country by being selective of posts you don't like? There are plenty of people who have good reasons for posting good ideas in uncouth ways.

It would be good if more people could put on the "mask of the institution" and realize it is nothing personal against them when negative comments are made about the institution. Alot of ways to improve "the institution" could be contained within them.

"The institution" in this case being PGI on MWO

#40 The Great Unwashed

    Member

  • PipPipPipPipPipPipPip
  • Giant Helper
  • Giant Helper
  • 919 posts
  • LocationNetherlands

Posted 04 November 2015 - 07:12 AM

If you need the free speech argument to defend what you're trying to say on a gaming forum then it is probably not worth reading and neither is nonconstructive criticism.





1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users