|
|
| Author |
Thread Statistics | Show CCP posts - 1 post(s) |
Aeon Amadi
Chimera Core
9001
   |
Posted - 2015.03.13 08:10:00 -
[1] - Quote
Holy hell, this took -way- longer than I honestly wanted it to...
So the general theory is that Myrofibril Stimulants are not working as intended. Some of you have theories that the stacking penalties aren't being applied, some of you are saying that it's using the Melee Modifier instead of the intended Jump Height increase, etc. etc. The list goes on and on about the current theories as to what is going on but one thing is for certain: Community is mixed as to whether or not they're working as intended; let alone if they're balanced, I personally think they are.
But my personal bias isn't up for debate here. The science behind the modules are and I decided to put them to the test with the help of Sgt Kirk.
How We Tested It
The general idea is that Bacon + Module = Flying Bacon. Cool. But we want to know how far the module will go in making the bacon fly. So, to do that, we had Sgt Kirk design four fits that were exactly the same apart from one very important fact: The amount of myrofibrils he had equipped.
The first suit had none and this was considered our control. It would show us what the base height was whenever he jumped.
The second suit was loaded up with one complex myrofibril, the third suit with two, so on and so forth.
Once that was accomplished we needed a way to figure out exact metrics: How high he'd jump so we could effectively measure the output of each module on an individual basis. To do this we decided to find a slate wall for him to put his back to and repeatedly jump after changing suits to see how high, using a Mass Driver as a form of 'scale'. Later on, after reviewing the video, I realized this was wholly unnecessary as I decided to take a page from Game Theories' book and utilize pixel mapping from base screenshots. Given that I didn't move (at all) this would allow me to compare the pixel differences between each jump against his base jump height with little to no interference on my end.
Here's a screenshot explaining a bit more of how this works: http://i.imgur.com/BQIG6xM.jpg
With that said, we went into a Domination match for optimal red-line abuse (we didn't want to get shot while testing) and REALLY lucked out when we got the socket that we did in the redline as it offered a great comparison point when compared to pixel height and the line measurements of the Mass Driver.
How It All Turned Out
Here's the video if you want to watch for yourself: https://youtu.be/4WPYQlDoieg
Surprising, to say the least. Based on number differences between the pixel distances, we can look at the numbers themselves and compare them to the intended ratios. The amount of difference in the pixel distance from the control line to the top of Sgt Kirk's head would give us true values and percentages and we could compare them against the intended percentages with his base height.
And here's what that looks like:
Pixel Locations of base line and top of the head
Test 1: 517 - 349
Test 2: 517 - 309
Test 3: 517 - 250
Test 4: 517 - 157
Bonus Round: 517 - 62
Base Jump Height: 168 pixels
1 Complex: 208 pixels
2 Complex: 267 pixels
3 Complex: 360 pixels
Bonus Round 4 Compelx: approximately 455 pixels (based on where Sgt Kirk shot)
Myofibril Module Jump Height intended increases:
PRO: 25% (+50% skill efficacy = 37.5%)
As a result, at least one of the myths was completely outright -BUSTED- in that without stacking penalties and an applicable skill bonus from the Hand-to-Hand Combat skill (50% at level 5) the numbers were -WAAAAY- off. So off that I'm absolutely 100% confident in saying that: Yes, without a shadow of a doubt, the stacking penalties are 100% being applied even if the module receives a skill bonus.
Without Stacking penalties and skill bonus: BUSTED
1 Complex: 37.5% (231 pixels) --- (difference of 23)
2 Complex: 89% (318 pixels) --- (difference of 51)
3 Complex: 160% (437 pixels) --- (difference of 77)
Here's the following tests that are a little bit more in the realm of believable.
Without Stacking penalties and NO skill bonus:
1 Complex: 25% (210 pixels) --- (difference of 2)
2 Complex: 56% (262 pixels) --- (difference of 5)
3 Complex: 95% (328 pixels) --- (difference of 32)
With Stacking penalties and NO skill bonus:
1 Complex: 25% (210 pixels) --- (difference of 2)
2 Complex: 52% (255 pixels) --- (difference of 12)
3 Complex: 74% (292 pixels) --- (difference of 68)
With Stacking penalties and skill bonus:
1 Complex: 37.5% (231 pixels) --- (difference of 23)
2 Complex: 82.32% (306 pixels) --- (difference of 39)
3 Complex: 121.29% (371 pixels) --- (difference of 11)
So, in closing, in a purely number-based scenario, the most likely situation that is going on with the Myrofibrils in my amateur opinion is that the modules are -NOT- receiving a skill bonus but at the same time -ARE NOT- receiving stacking penalties. However, it stands to reason that even with stacking penalties being applied, the difference in jump height would be marginal at best until you start getting extreme scenarios like four or five modules.
Final Statements:
- Myofibril Stimulants DO NOT give the melee bonus to jump height, for a certain fact.
- Myofibril Stimulants MAY NOT receive a skill bonus.
- Myofibril Stimulants MAY NOT have stacking penalties applied.
- IF THEY DO NOT HAVE STACKING PENALTIES, than we can say without a shadow of a doubt that they do not receive skill bonuses from Hand-to-Hand Combat.
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
Aeon Amadi
Chimera Core
9007
|
Posted - 2015.03.13 08:32:00 -
[2] - Quote
Boot Booter wrote:Hmm interesting... Not so certain on the value of the results, we would need more tests to gather a larger sample size.
One thing I am certain about is... Blue Gatorade is awesome and definitely not OP. I'm in favor of keeping whatever the current numbers are.
Nice work on the science though.
Ps you should try it with other suits and take the average of percent change with varying levels of myos across the different suits. If that makes sense.
Maybe when I have more time.
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
Aeon Amadi
Chimera Core
9007
|
Posted - 2015.03.13 08:38:00 -
[3] - Quote
Fizzer XCIV wrote:Wouldn't it be more accurate to measure from the ground go their feet, instead of from the ground to the top of their head?
Kirks body length might be messing with the ratios...
..........
Did we forget how to basic math..?
It wouldn't matter if the distance from the base control line (517) was to either the head or the feet. The distance from it's original position would still be the same in either case.
Let's say the feet were 100 pixels away from the control line. If he jumps, we'll say 50 pixels high for the sake of reason, then the distance would be 150 pixels.
If his head was 200 pixels away from the control line then it'd be 250 pixels high because the jump height would still be the same in either case.
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
Aeon Amadi
Chimera Core
9009
|
Posted - 2015.03.13 08:48:00 -
[4] - Quote
Fizzer XCIV wrote:Aeon Amadi wrote:Fizzer XCIV wrote:Wouldn't it be more accurate to measure from the ground go their feet, instead of from the ground to the top of their head?
Kirks body length might be messing with the ratios... .......... Did we forget how to basic math..? It wouldn't matter if the distance from the base control line (517) was to either the head or the feet. The distance from it's original position would still be the same in either case. Let's say the feet were 100 pixels away from the control line. If he jumps, we'll say 50 pixels high for the sake of reason, then the distance would be 150 pixels. If his head was 200 pixels away from the control line then it'd be 250 pixels high because the jump height would still be the same in either case. Fine. I'll do the math then. It just makes more sense to me to measure jumping by how much you can clear, not by how high your head goes, like in the Olympics. Do you have the stills on Imgur or something?
It doesn't make any sense at all because it doesn't matter. The distance from Point A-to-Point B is still the same regardless. Measuring from the top of the head is better because it stays relatively in place during the jump animation whereas the legs bend and move around during the animation.
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
Aeon Amadi
Chimera Core
9011
|
Posted - 2015.03.13 09:13:00 -
[5] - Quote
Fizzer XCIV wrote:Well it depends on what you define as "jump height".
Personally, I define it as how high your feet come off the ground, since that it sorta the main point of jumping up anywhere. Getting your feet to a higher elevation.
I couldn't care less how high the top of my head goes in most cases.
I really want to visually explain this, because I can't quite articulate it with words, but I can guarantee you that Kirk's body length is messing with the results...
No, it isn't. It's physically and mathematically impossible. I went ahead and grabbed the numbers from his toes just to clarify this.
Base Jump Height: 517p - 438p = 79p
1 Complex Myofibril: 517p - 397p = 120p
Assuming that from the base height of the original test was 168p (distance from top of the head to the control line) than we can assume that he has a body length of 'x'
168p - 79p = x
x = 89
Thereby, we can derive he has a body height of ~90 pixels give or 5p in either direction to account for changes in the jump animation. That being said, with 1 Complex Myofibril he jumped to a height of 208p at the top of the head. This would mean that, subtracting his body height, his feet should be at 118p. Considering that the above statement said it was 120p than the difference is 2p.
Considering that our leeway was 5p in either direction, than 2p is well within the range for changes and thereby proving that it is freaggin -impossible- for the height of his body to be a factor in the difference between jumps if the same gauge is applied. The gauge in this case being the top of his head, it wouldn't matter where the top of his head was in any of the tests in comparison to where his feet were. This is very, very basic math.
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
Aeon Amadi
Chimera Core
9011
|
Posted - 2015.03.13 09:42:00 -
[6] - Quote
MINA Longstrike wrote:Interesting finds.
Obviously we're at opposite ends of the spectrum on whether or not we feel that this is balanced, I feel that it isn't for a number of reasons (namely dust was not designed with this in mind, the module being too powerful for its cost and also it really messing with my suspension of disbelief) . Even if it makes portions of my ideas wrong it is still nice to have correct answers.
I think the only person who can say what Dust 514 was designed for is CCP Rattati but even then, it's a Science Fiction game and we honestly need more Sci-Fi and less 'reality'. We're super soldiers and it's very, very rarely we ever have anything that makes us -FEEL- like super soldiers. Before the Myofibril changes, we struggled to get over simple things that were completely unacceptable in a game set 23,000 years into the future where there are ships that are 20km+ in length with FTL travel that can take the ship through entire planets.
Very early on in Eve Online's lore they stated that they don't operate on Newtonian physics and from that point on were exempt from any 'reality' arguments. I think the same can apply for the sake of making a player legitimately feel as though they are capable of anything that they couldn't do in Battlefield or Call of Duty.
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
Aeon Amadi
Chimera Core
9011
|
Posted - 2015.03.13 10:06:00 -
[7] - Quote
Fizzer XCIV wrote:Thank you for at least giving me the height of Kirk's body. I can work with that...
Now, if we go on to subtract 89p(with a 5p variance) from all of the jump heights you listed, so that we can effectively remove Kirks body length, we can extract his ground clearance from the pictures, measured in pixels.
No Myos: (79)74-84p
1 Myo: (129) 124-134p
2 Myos: (179) 174-184p
3 Myos: (271) 266-276p
4 Myos: (375) 370-380p
Let's say the base jump height of ~79p is considered 100% jump height(ground clearance) for the next part.
No Myos: 100% clearance
1 Myo: 163% clearance
2 Myos: 227% clearance
3 Myos: 343% clearance
4 Myos: 475% clearance
As you can see these numbers show a completely different story, and in my opinion, a more important one.
After all, these modules were designed to get places, not get your head to places, right? These numbers more accurately displaying the boosts to mobility these modules are providing.
4 Myofibrils are allowing you to jump up onto things 475% better than you used to be able to.
Even just the 1st one is providing a 63% boost to our ability to jump.
In fact, the numbers appear to be all over the place. They appear linear at first, at least for the first two modules, but the 3rd and 4th are providing even larger boosts than the first two, and they seem to be increasing exponentially!
Something is wrong here.
The cause of this? I have absolutely no idea.
But something IS wrong.
The math.
Just ran the numbers against the screenshots again, basing from the bottom of the feet and came up with this:
517 - 437 = 80
517 - 396 = 121 (41)
517 - 339 = 178 (98)
517 - 247 = 270 (190)
Which is consistently 80(1.5)^x
Which would literally go to say that, if that were the case, that each myofibril attached would be giving a 50% increase per module (which is melee damage, no stacking penalties, no skill bonus) but there's a factor that isn't being considered (and the entire reason why I based my math on the top of the head) and that's the bend of the legs throwing off the results. The legs bend and move to different points in the jump animation and are unreliable as a controlled location to base off of, whereas the head stays relatively the same.
EVEN IF the math is correct here there is still the overarching problem: How the hell the percentages between the distance from the control line to the head are consistent with one scenario and the percentages between the distance from the control line to the bottom of the boots provide a completely different scenario.
It's something that I honestly can't be bothered to screw around with when I have to go to be at work in an hour.
Figure it out.
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
Aeon Amadi
Chimera Core
9014
|
Posted - 2015.03.13 10:22:00 -
[8] - Quote
Fizzer XCIV wrote:Actually, you are right! The 3rd and 4th module data almost perfectly lines up with the stacking penalty. Close enough that the 5p leeway could easily be responsible... The 1st and 2nd modules seem a bit off, but some of that could be leeway as well... The rest? Again no idea. These modules seem all kinds of messed up though, so it could be anything. However.. I think we might have just discovered what is actually happening. The Myofibrils are, by my observations: 1.) Providing the melee damage boost to jump height. 2.) Affected by the skill. 3.) Affected by stacking penalties. At least roughly...
Now you're making stuff up and grasping at straws.
You realize that if these things were applying the melee bonus AND the skill bonus itd be a 75% increase...? I -just said- that it was a 50% increase using your hackneyed 'bottom of the feet" math but again this does not explain the GROSS discrepancy of why the numbers originally posted were within reasonable parameters and not something so crazy as 50-75%.
Figure out why that is and I'll buy into this tin foil. If you can explain why the distance from the control line to the top of the head is within 25% margins but the control line to the feet is 50-75% then Ill start to listen. Until then, Im calling tin foil until I can get home and figure it out myself.
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
Aeon Amadi
Chimera Core
9014
|
Posted - 2015.03.13 10:38:00 -
[9] - Quote
MINA Longstrike wrote:The thing is though aeon, that it doesn't really matter where the head goes in a jump because the 'clearance' of a jump isn't defined by head height, but rather by the height of your feet when trying to get onto something.
I think that investigating the position of feet at apex of jump might be the avenue that adds to some of the most correct answers.
What is so hard to understand about this? The distance between A and B doesnt change regardless of what part of the body it is. If the jump height is 50p than ANY body part would be elevated by 50. The only way that would not be the case is if the point on the body being measured is dynamically changing mid-flight and that would apply to ALL body parts unless the body is being distended and contorted itself.
It would literally say that one part of the body is moving independantly of the rest of the body to such drastic proportions as to say the legs are consistently moving 50% from the original position and the head is moving 25% of the original position, which would then mean that the legs would eventually outpace the head and we'd be seeig some REAL SCARY ****
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
Aeon Amadi
Chimera Core
9057
|
Posted - 2015.03.13 19:51:00 -
[10] - Quote
Avallo Kantor wrote:A question I'd like to ask since you seem so willing to do the research:
how does a running start affect overall jump height?
what 'real' abilities do these mods translate to? (For example, how many blues would you need to jump over a shipping crate, or over a wall? What is the highest object a full blue, running scout could jump onto?
I think one important factor in all this is what sort of actual mobility does this all translate to, as defined by what structures are now "jumpable"
Also: Very nice work. Thank you for your research and effort.
Can check running start when I get home but the others I can kinda sorta answer.
On A Gallente Assault it takes a -minimum- of x3 Complex Myofibrils to jump a large crate, this I know from personal experience (see my video 'Myofibril Madness') as well as the video provided here in which Sgt Kirk jumps (there's a crate nearby that we screwed around with in between takes).
Scout? Dunno. I dunno how much mass actually applies but a Scout would be a terrible choice because the only ones who could run the myrofibrils sufficient enough to out-class an Assault would be the Caldari (and maybe the Minmatar) but you'd be sacrificing so much tank to do so that you'd literally die from the fall damage you brought onto yourself from jumping so high (which I find absolutely hilarious btw).
As far as what is 'jumpable', check out my Myofibril Madness video on youtube. It shows a -LOT- of the new opportunities provided.
Thank you =3
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
|
|
Aeon Amadi
Chimera Core
9060
|
Posted - 2015.03.13 20:25:00 -
[11] - Quote
NINEinch WEAPON wrote:Has anyone tried to see if the hep metabolics effect this jumping business...? Seeing as how you can stack them on the other side..?
Not willing to risk losing them as long as there is no way to replace them 
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
Aeon Amadi
Chimera Core
9084
|
Posted - 2015.03.14 15:27:00 -
[12] - Quote
Vell0cet wrote:One point to consider is that there will be error due to the perspective of the camera. You would get more error at higher jump heights because the angle of the "eye" to the target is greater at higher heights. At least I think that's right. Thanks for putting in the considerable effort to do all of this.
I considered that and I'm working with some college friends who are better at Trig than I am to re-evaluate the testing methods and go at it again.
This time we're looking at recording from a side angle and measuring the distance from the wall to a known point and then calculating it all based on those referencings. But, by that point, a decision would have already been made as to whether or not they're going to be changed so I'm not sure we're -actually- going to go through all that work, but we'll see.
MINA Longstrike wrote:Another point to try and see what's going on would be to get data for basic & adv myo's
Maybe, if we have the time. As I said above, a decision has probably already been made as to whether or not they're going to be changed just based on community crying (for lack of a better word) and the general "feel" instead of the overall hardcore math behind it all.
Will likely have to test all of this on Monday as I work 12-15 hours/day on the weekends.
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
Aeon Amadi
Chimera Core
9093
|
Posted - 2015.03.14 22:41:00 -
[13] - Quote
Id like to take a moment, just sit right there, Ill tell all about how I became the prince of DERP.
So, talking with a friend (mathematician) who explained why this test was full of dumb and how we're going to completely redo it - correctly, this time. Judging distance based on pixel density is full of derp because the scale at which is applied in the eventuality of faults in scale. 10 pixels at 10m is not the same as 10 pixels at 100m.
I dont often admit to being wrong about something (im a perfectionist at times and am incredibly stubborn when it comes to my personal accuracy with these things) but in this case, I fully admit that I kittened up and whats worse is in doing so I led a portion of the community - who at no fault of their own didn't correct me or try to stop me - down the same derpy path of misguided intelligence and ill applied logic. For that, you all have my sincerest apologies and I will work to correct all of this tonight when I get off work and can apply correct testing procedures and valid mathematical application.
Im sorry.
Have a suggestion for the Planetary Services Department?
Founder of AIV
|
|
|
| |
|