PDA

View Full Version : Is it simple math



Slickster
05-19-2012, 10:53 PM
Ok brilliant minds of crime city, just have a simple question; how, why is it possible as I have both the diamond sig +1 respect points and also have the last event gun which I believe is called Midas gun or something which boost your respect points per fight by 40 % but at least 30% of my fights I continue to only get 2 respect points for the fight is anyone else experience this wonderful experience, shouldn't at minimum I get at least 3 wonderful respect points...?

PawnXIIX
05-19-2012, 10:56 PM
My thoughts on that matter is that instead of rounding the number they just simply trucate the number and call it a day.

Therefore 1RP base + 1RP sig = 2RP
2RP + (2 * .4) = 2.8

That 2.8RP is then truncated to 2. If it is calculated in the other order:

1RP + (1 * .4) = 1.4RP + 1RP sig = 2.4 truncated to 2

Either way it is added up you're still going to get 2RP min :(


If you think about it, they planned it that way so that way you weren't going to get 3RP min. I wondered why it was 40%, I thought that was a very arbitrary number. But now I realize this is their way of, if you have the sig, keeping the min respect points at a low number. Very clever.

ShawnBB
05-19-2012, 10:57 PM
Exactly buddy, the latest Midas gun keeps my RP range at 1 to 9 per fight, where the heck is the BS 40%?

Funzio serious need to fix this scam, some ppl tosses gold in this please be responsible for them.



The rounded theory is not valid, I discussed it in another thread, because 2*1.4 is 2.8, 3*1.4 is 4.2.
But during my PvP, there is 3 RP gain sometimes which can't be explained by round theory.

PawnXIIX
05-19-2012, 11:02 PM
Exactly buddy, the latest Midas gun keeps my RP range at 1 to 9 per fight, where the heck is the BS 40%?

It's not a scam, the system can't give you the .4RP so therefore it turns the 1.4RP you gained into a integer. It's not a scam is the programming language at work. When you take a double and turn it into an integer it slices off the decimal point. If Funzio rounded up then that is basically a sig in itself given to everybody.

Here's a few confusing examples of how computers and programming language do math:

1 / 5 = 0
1. / 5 = .2
1 / 5. = .2

3 / 2 = 1
3. / 2 = 1.2
3 / 2. = 1.2

the 0 is implied after the decimal therefore it omitted. In division if the two divisors are integers then the answer is provided in an integer. If one of the divisors in a decimal, then the language translates the answer into the more complex form. This is how C++ does it, i'm not sure about other languages but that's to my understanding and in my theory of the game how it is done :/

Slickster
05-19-2012, 11:02 PM
Ok brilliant minds of crime city, just have a simple question; how, why is it possible as I have both the diamond sig +1 respect points and also have the last event gun which I believe is called Midas gun or something which boost your respect points per fight by 40 % but at least 30% of my fights I continue to only get 2 respect points for the fight is anyone else experience this wonderful experience, shouldn't at minimum I get at least 3 wonderful respect points...? thanks for the response that makes sense, sucks but makes sense, well if I ever need to take another math course in college, you'll be hearing from me :)

PawnXIIX
05-19-2012, 11:09 PM
If you get into the right course you won't even be dealing with numbers. Senior year math was so full of Integrals and Derivatives...I can tell you that if the problem involved numbers it was a godsend.

ShawnBB
05-19-2012, 11:19 PM
If you get into the right course you won't even be dealing with numbers. Senior year math was so full of Integrals and Derivatives...I can tell you that if the problem involved numbers it was a godsend.

Hey pawn:), are you CS major or math? I'm taking statistics about distrubution and hypothesis stuff, also having high order differential equation courses which is a completely mess!

Alice
05-19-2012, 11:25 PM
^

Hey dude, Have you got to Ito calculus yet? I wanted a ST minor, only my adviser asked me to sit in lab:(

PawnXIIX
05-19-2012, 11:31 PM
Right now I'm a CS major and took no math classes this year...oddly enough. Just writing and bio and other nonsense classes that you need to get out of the way before you can really get into the good learning ^^

Senior year I took honors calculus as well as AP stat and my SAT math was 740 so that cleared me of taking any placement tests

<3 Hz
05-19-2012, 11:37 PM
From what I have seen personally it's a true 40% because the max # of respect I've gotten from an opponent in one hit so far is 15. I have the Midas & the Sig. 10+4 for the midas & +1 for the Sig. It just doesn't seem like 40% since the numbers you're dealing with are fairly small

ShawnBB
05-19-2012, 11:44 PM
Lol Alice, you just poped out from no where.
Well, it depends on your current major, st minor is easy to get for science focused dudes anyway. I guess I'm pretty much done with calculus.

Alice
05-19-2012, 11:48 PM
^

lol, never thought calculus could be that hard...I learned that by myself at high school

Yahkin
05-19-2012, 11:52 PM
Something still doesn't add up...as I posted in this thread: http://www.funzio.com/forum/showthread.php?29280-Average-Midas-Stats-2R-and-the-Economy-of-R&p=226005#post226005

Here are my findings from that thread:


Over about 100 attacks I have received RP rewards of 1-14, every single one. Which actually confuses the matter more...the math is fishy. Unless the original RP received is not actually an integer.

Here is 140% of various values:

1 = 1.4
2 = 2.8
3 = 4.2
4 = 5.6
5 = 7
6 = 8.4
7 = 9.8
8 = 11.2
9 = 12.6
10 = 14

With rounding I should never receive 2, 5, 9, and 12 RP. Yet I do.
Rounding down I should never receive 3, 6, 10, and 13. Yet I do.

PawnXIIX
05-20-2012, 12:02 AM
@Alice
Oh without a doubt. I personally enjoy math since it's very algorithmic and very structured. There's one answer.

Unlike every other class which is very abstract, which is why I depend on my math class to bring my grades up :P

@Yahkin
That's because the values are not rounded, the values are truncated. My theory on the matter is that it generates a base RP upon an algorithm, it never just generates an integer value between 1-14. Take for example, this algorithm:

int x = ((rand * 130) + 10) / 10;

That gives you a random value between 1 - 14

Now everybody knows that the respect is calculated based on who you are attack, so the random element is as dramatic, but what i'm trying to say is that you can get a value of 7.8 or 9.4 or 12.6.

Before it truncates the number though, it applies the 40% before it changes the value into an integer. Now that would need a good amount of trials to prove it is possible to get any value because now I find that 140% of 5 is 7 and 140% of 4 is 5.6 and you said you got 6RP once. So now I think that it generates a random number between 10 and 140 and then divides it by 10, thus providing those strange values that would allow you to get 6RP.

ShawnBB
05-20-2012, 12:14 AM
Something still doesn't add up...as I posted in this thread: http://www.funzio.com/forum/showthread.php?29280-Average-Midas-Stats-2R-and-the-Economy-of-R&p=226005#post226005

Here are my findings from that thread:

Perfect dude, that's what I was talking about, there is no round theory here, gotta be something else.

PawnXIIX
05-20-2012, 12:20 AM
int x = ((rand * 130) + 10) / 10;

That gives you a random value between 1 - 14

Now everybody knows that the respect is calculated based on who you are attack, so the random element is as dramatic, but what i'm trying to say is that you can get a value of 7.8 or 9.4 or 12.6.

Before it truncates the number though, it applies the 40% before it changes the value into an integer. Now that would need a good amount of trials to prove it is possible to get any value because now I find that 140% of 5 is 7 and 140% of 4 is 5.6 and you said you got 6RP once. So now I think that it generates a random number between 10 and 140 and then divides it by 10, thus providing those strange values that would allow you to get 6RP.

I edited that after you had already replied, I took forever to type it out D:

ShawnBB
05-20-2012, 12:41 AM
Oh, actually it could be explained like this.

They generate a random number from (0, 1) and multiply by 1.4 right after.
In the end, times 10 and rounding the number(not cut off,because you get 14 RP sometimes which can't be achieved by cut off)

PawnXIIX
05-20-2012, 01:05 AM
Well i think it's based off of this...It just calculates a min and max based on level. That max can be above 14 and can be below 0.

After everything it does one more simple thing like:
if(x > 14)
x = 14;
if(x <= 0)
x = 1;

Maybe? I get a feeling like we'll never know with any certainty :/

Max Power
05-20-2012, 06:18 AM
I think I remember hearing from CCMark that the sig adds the number off the fight screen, so if somebody want to babysit their totals, that might make more sense.

I can definitely see the 40%. I had a Pvp pay 17 RP recently. I have both guns, btw.

Edit: I just did a small sampling and what you see is what you get. Never mind and carry on.

Schtikman
05-20-2012, 09:01 AM
What about the possibility that the respect points gained before the calculations being done are actually not exact numbers. What if they give us very random rp of X.XXX and it's not rounded. Then that number is used to multiply by 1.4 for the Midas Gun then you add 1 rp for diamond sig. Is this possible? It would explain some of the issues discussed. I only have the Midas Gun and am ticked I still get 1rp per fight quite often. I will do some calculations based on my next 100 fights and report back.

Here are the results: all with Midas Gun only

1rp in 2 fights
2rp in 4 fights
3rp in 10 fights
4rp in 11 fights
5rp in 4 fights
6rp in 9 fights
7rp in 14 fights
8 rp in 4 fights
9 rp in 8 fights
10rp in 9 fights

450rp in 75 fights
Average of 6rp per fight
Maximum of 10rp per fight / Minimum of 1rp per fight

Slickster
05-20-2012, 10:05 AM
From what I have seen personally it's a true 40% because the max # of respect I've gotten from an opponent in one hit so far is 15. I have the Midas & the Sig. 10+4 for the midas & +1 for the Sig. It just doesn't seem like 40% since the numbers you're dealing with are fairly small so you have actually got a 15 respect point hit, max I have seen is 14 :(

PawnXIIX
05-20-2012, 11:19 AM
What about the possibility that the respect points gained before the calculations being done are actually not exact numbers. What if they give us very random rp of X.XXX and it's not rounded. Then that number is used to multiply by 1.4 for the Midas Gun then you add 1 rp for diamond sig. Is this possible?


...it never just generates an integer value between 1-14. Take for example, this algorithm:

int x = ((rand * 130) + 10) / 10;

That gives you a random value between 1 - 14

Now everybody knows that the respect is calculated based on who you are attack, so the random element is as dramatic, but what i'm trying to say is that you can get a value of 7.8 or 9.4 or 12.6.

Before it truncates the number though, it applies the 40% before it changes the value into an integer. Now that would need a good amount of trials to prove it is possible to get any value because now I find that 140% of 5 is 7 and 140% of 4 is 5.6 and you said you got 6RP once. So now I think that it generates a random number between 10 and 140 and then divides it by 10, thus providing those strange values that would allow you to get 6RP.

I like the way that sounds ;D Good to see i'm not the only one to think that. That's 2 votes for that theory ^^

Ramshutu
05-20-2012, 02:49 PM
I would expect it would be as follows:

1.) generate a random floating point number between 0 and 1.
2.) multiply it by a given multiplier.
3.) round to obtain RP.

With the 40%, you can add a 2.a) multiply by 1.4
For a sig, you add a 4.) add 1.

The above will generate any number from 0 to multiplier. They will probably add one to ensure that you always get RP, although if they did, the Midas would increase the multiplier by 40% but it's real impact would likely be a bit less (and that should be relatively testable.)

I can test the above, as it would mean that all numbers should be equally likely with the exception of 1, which should occurr 50% more often than any other number.

Giedrius
05-20-2012, 03:36 PM
What about the possibility that the respect points gained before the calculations being done are actually not exact numbers. What if they give us very random rp of X.XXX and it's not rounded. Then that number is used to multiply by 1.4 for the Midas Gun then you add 1 rp for diamond sig. Is this possible? It would explain some of the issues discussed. I only have the Midas Gun and am ticked I still get 1rp per fight quite often. I will do some calculations based on my next 100 fights and report back.

Here are the results: all with Midas Gun only

1rp in 2 fights
2rp in 4 fights
3rp in 10 fights
4rp in 11 fights
5rp in 4 fights
6rp in 9 fights
7rp in 14 fights
8 rp in 4 fights
9 rp in 8 fights
10rp in 9 fights

450rp in 75 fights
Average of 6rp per fight
Maximum of 10rp per fight / Minimum of 1rp per fight

You are low level I think, because I'm averaging nearly 8 with Sig and Midas. Your max 10rp is very low I'm getting 10-16 regularly. I think Max got even 17. Even before Midas 14 was common.

PawnXIIX
05-20-2012, 06:43 PM
I think it's safe to assume that we can all agree that the system picks a random double between a designated low and high based upon your level. It cannot be an integer because that would mean that with ONLY the midas it would be impossible to achieve a 6RP hit.

4RP base * 1.4 = 5.6RP -> 5RP
5RP base * 1.4 = 7RP

Therefore according to those statistics I think we are all in agreement that the number of RP you get in between an arithmetically obtained min and max based upon the player level, then the midas and sig are applied in a predefined order, then the number is converted into an integer upon which it is provided to the player as a nice whole number.

(int) 2.999999 = 2 <- Using casting to truncate the data into a integer, no rounding used because when converting a more complex data form to a simpler form everything to the right of the decimal is simply forgotten. This is not how us humans would do it but this is how programming languages deal with conversions from decimals to whole numbers :(

All in agreement?

Schtikman
05-20-2012, 07:00 PM
I am level 120
Now at level 121. Rp has jumped up to a maximum of 12rp per fight max and still 1rp minimum.