### Trending Games | Skyforge | Elder Scrolls Online | World of Warcraft | Firefall

Members:3,024,512 Users Online:0
842  6,480,788
Cloud view
List all forums
General Forums
Developers Corner General Discussion
Popular Game Forums
Click a status to find game forum
Game Forums
Click a letter to find game forum
A-C
D-F
G-L
M-Q
R-S
T-Z

# General Discussion

## News & Features Discussion  » [Column] General: Becoming a Designer - Part Two

 3 Pages 1 2 3 » Search
57 posts found
SBFord

Associate Editor - News Manager

Joined: 6/28/10
Posts: 16259

OP  4/28/13 4:36:11 PM#1

In the second part of Becoming a Designer, Matt Miller talks more about what aspiring developers need to do to make their dreams come true. Read on and then offer your ideas in the comments.

A couple people asked in the comments about the time restriction. I’ve seen design tests at other companies that are rigidly timed, with websites that track the minute you download the test to the minute you submit it. We didn’t do that at Paragon. We simply asked the candidate when they felt they would have enough time to devote to the test, and scheduled email delivery of the test to that time period. They would then complete the test and return it. When the HR department delivered the test for “grading” (not a real thing), we were told how long the applicant had the test for.

Associate Editor: MMORPG.com

## Login or Register to never see this panel

User Deleted
4/29/13 12:24:20 PM#2
And that factors in probability, you can pick that up in a math statistics course, my point .... drum roll please..... do your math homework kids and you can become a video game designer.
sumo0

Novice Member

Joined: 9/18/10
Posts: 122

4/29/13 12:27:29 PM#3
B-but... I never did my math homework. Guess ill stick to playing them games instead.

FrinkiacVII

Novice Member

Joined: 1/23/13
Posts: 45

4/29/13 12:32:07 PM#4

Here's my shot it:

1. Average damage is 100, assuming the target lives to see all 5 ticks (with no hit point info, I think we have to assume target has more than 100hit points and I will assume this from now on).

2. Average is 50 + 5(0.5)10 = 75

3. Average is 50x0.5 + 60x(0.52) + 70x(0.53) + 80x(0.54) + 90x(0.55) + 100x(0.56) = 58.1

That's what I get anyway.  It took me like 15 min to do that and I screwed it up at least once.  I know I screwed it up at least once because the first way I did it was different, and then I changed it.  It's entirely possible that both my first approach and this one are wrong, but I feel better about this one.

"Well sure, the FrinkiacVII looks impressive - DON'T TOUCH IT - but I predict that within 100 years computers will be TWICE as powerful, ten THOUSAND times larger, and so expensive that only the five richest kings of Europe will own them." -Prof. Frink

FrinkiacVII

Novice Member

Joined: 1/23/13
Posts: 45

4/29/13 12:33:08 PM#5
Wait, was I supposed to do that with or without Enhancement Diversification? ;)

/em holdtorch

"Well sure, the FrinkiacVII looks impressive - DON'T TOUCH IT - but I predict that within 100 years computers will be TWICE as powerful, ten THOUSAND times larger, and so expensive that only the five richest kings of Europe will own them." -Prof. Frink

Shaigh

Hard Core Member

Joined: 3/07/13
Posts: 526

4/29/13 12:56:26 PM#6

100

75

50+5+2.5+1.25+0.625+0.3125= rounded to 60

maplestone

Novice Member

Joined: 12/10/08
Posts: 3109

4/29/13 1:03:56 PM#7

I would have said "just under 60" for the last one, then asked if they needed more precision.

( the average number of ticks would be 1/2 + 1/4 + 1/8 + ... which converges to 1, so if there were an infinite number of ticks it would be 60, maxing out at 5 ticks would be a little less than that; since there aren't a lot of situations where you'd need an exact answer, so I'd start with the ballpark figure in a conversational interview and only break out the calculator only if a more exact answer was requested )

maplestone

Novice Member

Joined: 12/10/08
Posts: 3109

4/29/13 1:11:21 PM#8
( alas, my social anxieties always trip me up when getting the pressure-test question from people I don't already know; my ability to answer these sorts of questions on the fly changes dramatically after a few weeks of knowing the people I'm talking to ... unfortunately, one just doesn't have that luxury in an interview )

Quizzical

Guide

Joined: 12/11/08
Posts: 14684

4/29/13 1:22:30 PM#9

On the math question, it depends on when the mob dies.  For example, if the initial 50 damage kills it, then the attack does 50 damage in total, as the later ticks will never happen.

Assuming the mob survives to take all of the damage, parts (1) and (2) are trivial, so I'll ignore them.

For part (3), the answer is clearly 50 + 10 * (expected number of damage ticks that occur).  So the real question is computing that expected number of damage ticks.

The probability that tick i occurs is 2^(-i).  By a well known series,

$$\sum_{i=1}^{\infty} 2^{-i} = 1.$$

If there are n ticks possible, then we only want the first n terms from this series.  But the series can also break down as:

$$\sum_{i=1}^{\infty} 2^{-i} = \sum_{i=1}^{n} 2^{-i} + \sum_{i=n+1}^{\infty} 2^{-i}$$

from which

$$\sum_{i=1}^{n} 2^{-i} = \sum_{i=1}^{\infty} 2^{-i} - \sum_{i=n+1}^{\infty} 2^{-i}$$

$$= \sum_{i=1}^{\infty} 2^{-i} - 2^{-n} \sum_{i=1}^{\infty} 2^{-i}$$

$$= 1 - 2^{-n}$$.

So if the mob has h health and does not heal or take damage from other attacks, we let

$$n = \min\{ \max\{0, \lceil {h - 50 \over 10} \rceil \}, 5 \},$$

which caps the number of ticks at whichever is less of 5 or the number it would take to kill the mob.  Then the expected value of damage that the mob takes is $60 - 10 * 2^{-n}$.

FrinkiacVII

Novice Member

Joined: 1/23/13
Posts: 45

4/29/13 1:24:21 PM#10

My initial answer was wrong for part 3.  It should be:

Average is 50x0.5 + 60x(0.52) + 70x(0.53) + 80x(0.54) + 90x(0.55) + 100x(0.55) = 59.68

If the tick that does the 10 points that get you to 90 hits, you get a swing at another tick, if that last tick misses, you still did 90 points, and if it hits, you did 100 and stopped.  My weighted average failed to take that into account the first time.  Sorry about that.

Doing the math, you get caught up in exponents and sums and indices,etc.  I had to visualize it by drawing a numberline and breaking it up into chunks to get it right.  If you didn't account for/use the whole number line, you did it wrong, or such was my thinking anyway.  My original post assumed there was a 6th coin flip (and what's worse, if you failed it, my original math assumed that you did a total of 0 damage for missing the 6th flip, which would have been a disaster had there been fewer flips in the first place).

Probability and statistics questions are always harder than you think.  I TOTALLY won a goat that first time.  Thanks Monty -- er, I mean Matt.

"Well sure, the FrinkiacVII looks impressive - DON'T TOUCH IT - but I predict that within 100 years computers will be TWICE as powerful, ten THOUSAND times larger, and so expensive that only the five richest kings of Europe will own them." -Prof. Frink

mmoguy43

Spotlight Poster

Joined: 3/31/09
Posts: 2437

4/29/13 1:42:59 PM#11
I probably would have failed the last question were it a phone interview. But I could explain that the average would be a continuingly decreasing average between the 75 and 50.
User Deleted
4/29/13 2:00:06 PM#12
Yeah but .....the armor reduced the damage done. ... Booyah!!!!!! Better rework your equations oh and the mob was also 2 levels higher, and had a racial that reduces nature damage and ............ waaaaaahhhhh bangs head off desk smash smash smash ( blood covers pc screen brain matter flings across room )
Quizzical

Guide

Joined: 12/11/08
Posts: 14684

4/29/13 5:54:05 PM#13
Originally posted by KingofHartz
Yeah but .....the armor reduced the damage done. ... Booyah!!!!!! Better rework your equations oh and the mob was also 2 levels higher, and had a racial that reduces nature damage and ............ waaaaaahhhhh bangs head off desk smash smash smash ( blood covers pc screen brain matter flings across room )

I'd hope that the damage numbers were after accounting for such effects.

easy-r

Novice Member

Joined: 11/09/12
Posts: 40

4/29/13 6:09:53 PM#14
Math is nice, having got an 800 on the SAT and discussing complex physics equations with a fine woman being my preferred method of pre-intercourse foreplay I'd know, but things like this aren't going to make or break a new game anyways.  The idea about creating a world from a list of items is closer but I think the main problem with the 'industry' is that games have gone from the dreams of talented few to the product of a company, and with all that is required to create a quality game these days, it's necessary, and creates the catch-22.  Companies create games that appeal to the lowest common denominator, whether DF:UW or WoW, and the age of smart and challenging games created by MIT students in their garage is gone.  Big companies don't have the balls to create something challenging, hard, rewarding and complex, and it seems nobody but big companies is capable of pulling off the quality product MMO, with Eve perhaps being the last exception although one could go into its many, many flaws as well.  In any case, as the gaming industry has shifted from the dream of creating an RP world to pumping out the next Madden 14, quality, innovative, diverse games have been lost.  Hopefully some good ones will be released soon but it is looking doubtful.

Apprentice Member

Joined: 7/31/06
Posts: 2115

4/29/13 8:55:45 PM#15

design.  Knowing what will be fun, gratifying, immersive, engaging.  It's one of the paradigms of the "promote to incompetence" phenomenon in business.  Usually even when people are promoted for doing a good job in one area, the next one up is at least related to what they used to do.  Game design is an amalgam of creative thought, formulas, and intuition that only certain people possess.  Qualities that aren't usually learned through education or even experience in the field, but through introspection and philosophical dialogue.

We have good games coming out all the time, but look at all the mmo failures.  It's because people doing design simply don't possess the right qualities, people who do possess them would be considered as line-jumping, or undeserving, or not having paid their dues, because those people don't do programming or asset modeling, and they don't have BAs in finance or business management.

Battlesnake

Joined: 10/26/09
Posts: 6

4/29/13 10:28:36 PM#16

1: 100

2: 75

3: The first 50 points of damage is always going to happen.... so just move on to the ticks of damage to figure this out.  The 1st tick does an average of 5 points of damage, because it does either 0 points of damage or 10 points of damage [(10 + 0) / 2 = 5].  That 1st tick averages to 5, because it ALWAYS gets a chance to occur.  The 2nd tick of damage averages out to 3.333, because it has three different "states".  It could do 0 damage in the case where the 1st tick doesn't even happen, 0 damage because it got a chance to occur yet failed its own 50% chance, and finally a state where it does apply 10 damage.  So the average there is [(10 + 0 + 0) / 3 = 3.333].  For each tick after that, you just add on another "state" that could occur and divide from 10 to find the average damage that tick applies to the total.

50 + 5 + 3.333 + 2.5 + 2 + 1.667 = 64.5

Does that make sense at all?

TwiPhoenix

Aion Correspondent

Joined: 8/07/09
Posts: 19

4/29/13 10:53:21 PM#17

The first two are pretty easy.  1. is 100 and 2. is 75.  3 is tricky.  Is tick has a 50% chance of occuring, so each subsequent tick is half as likely to occur as the previous one.  So 50% > 25% > 12.5% > 6.25 > 3.125.  Luckily, it's easy to count from there since we're working with an easy number (10).  So simply multiply the % chance of that tick of occuring with 10 to get the average damage of that tick.

So, in the end, 3. comes out to 59.6875, which could be safely rounded to either 60, 59.7, or 59.69 depending on how accurate you wanted to be.  I find it a bit interesting that the average damage added beyond the first two ticks amounts to so little.

I'm not sure I could have figured this out in the middle of a phone interview.  I'm not entirely sure I'm correct either, but I'm confident in the answer.  You'll give us the correct answer in your next article, won't you?

Overkillengine

Apprentice Member

Joined: 4/29/13
Posts: 4

4/29/13 10:53:56 PM#18

1. 100 (assuming no other factors)

2. 75 (ditto)

3. Hoo boy. I figure I did this the hard way and made a copy+pasted array in Excel to generate the results 1323 times; with an average that hovered around 59-60 (truncated to the nearest whole number for the final average). I'm sure there are more mathematically elegant ways to accomplish what I did; I just did what I knew would work.

Prior disclaimers apply to this too.

Edit: extended excel model out to 15,876 rolls, for an average of 59.73419.

Another (simpler) way of doing it:

 base damage add dmg chance total dmg 50 0 50% 50 50 10 25% 60 50 20 12.5% 70 50 30 6.25% 80 50 40 3.125% 90 50 50 3.125% 100

Then you use a weighted average to get a result of 59.688.

maplestone

Novice Member

Joined: 12/10/08
Posts: 3109

4/29/13 11:59:31 PM#19
Originally posted by Battlesnake

The 2nd tick of damage averages out to 3.333, because it has three different "states".  It could do 0 damage in the case where the 1st tick doesn't even happen, 0 damage because it got a chance to occur yet failed its own 50% chance, and finally a state where it does apply 10 damage.  So the average there is [(10 + 0 + 0) / 3 = 3.333].

The problem with your reasoning is that the three states ( 0, 10/0, 10/10 ) are not all equally likely: no damage on the first tick happens happens half the time, 10/0 and 10/10 split the other half of the possibilities (making the average damage 2.5 rather than 3.333)

The average damage on each tick is half of what it was on the previous tick (think of the previous tick as just being a coin flip as to whether or not to proceed).

( I feel that the mechanic in question 3 is more natural to think of in terms of "what is the average number of ticks until damage stops" because the limit of 5 ticks, although in the same pattern as previous two questions, is just an artifical limitation in this mechanic - adding that 5-tick limit is actually more coding than simply having a "50% chance of expiring each tick" )

Felkin1

Hard Core Member

Joined: 3/19/10
Posts: 31

4/30/13 5:17:22 AM#20

If a power does 50 Points of damage, followed by 5 “ticks” of 10 points of damage, what is the average damage the power deals?

Since you're saying "Average" and not "Average per tick" or anything, I would ofc say 50+5*10 = 100 damage per power cast.

If a power does 50 Points of damage, followed by 5 “ticks” of 10 points of damage, but each tick only has a 50% chance of applying damage, what is the average damage the power deals?
50+0.5(5*10) = 50+25 = 75  Since its 50% chance, its 0.5 the normal dmg on avg.

If a power does 50 Points of damage, followed by 5 “ticks” of 10 points of damage, but each tick only has a 50% chance of applying damage, and if it fails to apply damage no further ticks occur, what is the average damage the power deals?

50+ 0.5*10 + 0.25*10 + 0.125*10 + 0.0625 * 10 + 0,03125 * 10 = 50+5+2,5+1,25+0,625+0,3125 = 59,6875 dmg.

I want to become a game designer myself and reading your column is so interesting for me, thank you :)

I'm the hardcore player, the one that rushes lvl cap before you even finish the starting area.

 3 Pages 1 2 3 » Search