43 réponses à ce sujet

[UnknownMercenary]

Seems to corroborate what people expected a year ago.

Good work nonetheless.

[Cyonan]

Shampoohorn wrote...


Nice. 64% is close to the 77% that Cyonan pointed out.

There does seem to be a small difference in the distributions -/+ Hunter Mode. It's my understanding that it would affect the refire delay even on single fire weapons. There could be some inherent bias in your finger, heh. The fourth condition -- HM and RC -- might shed some light on that.

Not may replies yet. Math posts tend to get the best response on Mondays, when most of the engineers have had their coffee and become bored with actual work. Saturdays, not so much.

As it turns out, if I adjust the number for standard human reaction time(about 0.15 seconds) then I get a theoretical 64.86% increase in overall DPS.

Modifié par Cyonan, 29 juin 2013 - 06:46 .

[Dunvi]

OniGanon wrote...

Why on earth would you expect the 15% ROF increase to result in 15% faster reload? ROF boost doesn't boost reload time, it only reduces the delay BEFORE the reload.

Claymore has a ROF of 64RPM, adding 0.9375s delay before reload.

Boosting that by 15% means a ROF of 73.6, which lowers the delay to 0.8152 (rounded to 4dp)

Difference between the two is roughly 0.1223s

Also... Something strange is going on here... Your RC numbers look fine, but your non RC numbers are faster than expected.

Theoretically, with a listed ROF of 64RPM and reload time of 2.57s, without RC the Claymore should fire once every 3.5075s (vs your recorded 3.33s)

You forgot the time speedup. There's a reason that timing tests should be done on bronze. Accounting for the time speed up brings the theoretical best time value back in line.

(Incidentally, I believe this is why some of you may feel you reload cancel better on higher difficulties.)

PS: still haven't really slept
attempted efficient complex python and hated myself
attempted bad bull**** pointer arithmetic and hated myself
attempted java optimization and hated myself
attempted shell scripts and hated myself
attempted python shell scripting and hated myself
attempted python multithreading and hated myself
attemped hacky distributed computing and hated myself
attempted debbugging useless debugging messages and hated myself

at least i wasn't the one trying to learn common lisp overnight

Modifié par Dunvi, 29 juin 2013 - 07:27 .

[ComradeShepard7]

Dunvi wrote...

OniGanon wrote...

Why on earth would you expect the 15% ROF increase to result in 15% faster reload? ROF boost doesn't boost reload time, it only reduces the delay BEFORE the reload.

Claymore has a ROF of 64RPM, adding 0.9375s delay before reload.

Boosting that by 15% means a ROF of 73.6, which lowers the delay to 0.8152 (rounded to 4dp)

Difference between the two is roughly 0.1223s

Also... Something strange is going on here... Your RC numbers look fine, but your non RC numbers are faster than expected.

Theoretically, with a listed ROF of 64RPM and reload time of 2.57s, without RC the Claymore should fire once every 3.5075s (vs your recorded 3.33s)

You forgot the time speedup. There's a reason that timing tests should be done on bronze. Accounting for the time speed up brings the theoretical best time value back in line.

(Incidentally, I believe this is why some of you may feel you reload cancel better on higher difficulties.)

PS: still haven't really slept
attempted efficient complex python and hated myself
attempted bad bull**** pointer arithmetic and hated myself
attempted java optimization and hated myself
attempted shell scripts and hated myself
attempted python shell scripting and hated myself
attempted python multithreading and hated myself
attemped hacky distributed computing and hated myself
attempted debbugging useless debugging messages and hated myself

at least i wasn't the one trying to learn common lisp overnight

I was originally going to do it on bronze because of the speed up on higher difficulties but decided against it since I am too used to reload canceling on gold and I probably would have screwed up a bunch on bronze. For the next one I might do it on bronze so that I can get more data points of me screwing up to see if it is consistantly still faster than going without.

[Credit2team]

ComradeShepard7 wrote...

k1ngl1ves wrote...

Wait...

So shooting faster = shooting faster?

*mind blown*

Everyone knew that going in. The point of this was to QUANTIFY how much faster it actually is in practice.

qft, some people are satified knowing they get more of a boost no matter how big or small it is, others are number crunchers and like to calculate and measure every little detail before they decide what kit/weapon/equipment to use

[Dunvi]

thewalrusx wrote...

ComradeShepard7 wrote...

k1ngl1ves wrote...

Wait...

So shooting faster = shooting faster?

*mind blown*

Everyone knew that going in. The point of this was to QUANTIFY how much faster it actually is in practice.

qft, some people are satified knowing they get more of a boost no matter how big or small it is, others are number crunchers and like to calculate and measure every little detail before they decide what kit/weapon/equipment to use

so... where do you stand, given that you didn't want numbers?

[UnknownMercenary]

Cyonan wrote...

As it turns out, if I adjust the number for standard human reaction time(about 0.15 seconds) then I get a theoretical 64.86% increase in overall DPS.

math owns

[Credit2team]

Dunvi wrote...

thewalrusx wrote...

ComradeShepard7 wrote...

k1ngl1ves wrote...

Wait...

So shooting faster = shooting faster?

*mind blown*

Everyone knew that going in. The point of this was to QUANTIFY how much faster it actually is in practice.

qft, some people are satified knowing they get more of a boost no matter how big or small it is, others are number crunchers and like to calculate and measure every little detail before they decide what kit/weapon/equipment to use

so... where do you stand, given that you didn't want numbers?

I was looking for a different set of numbers, but what the OP gave is better than nothing

[Ziegrif]

thewalrusx wrote...
I was looking for a different set of numbers, but what the OP gave is better than nothing

I don't think you'll get girls numbers on BSN hun.~
*Badumdish*

[solidprice]

did a solo w/ the marksmen, damn. the claymore is one of the fastest firing snipers in the game lol:lol:

Modifié par solidprice, 29 juin 2013 - 08:35 .

[Shampoohorn]

UnknownMercenary wrote...

Cyonan wrote...

As it turns out, if I adjust the number for standard human reaction time(about 0.15 seconds) then I get a theoretical 64.86% increase in overall DPS.

math owns

I think that is a bit of a conflation though. Optimal reload cancelling is a function of timing and precision rather than reaction to a stimulus. /nitpick

[ComradeShepard7]

Ziegrif wrote...

thewalrusx wrote...
I was looking for a different set of numbers, but what the OP gave is better than nothing

I don't think you'll get girls numbers on BSN hun.~
*Badumdish*

[img]http://thumbs.newschoolers.com/index.php?src=http://fim.413chan.net/fic/src/132597704176-Clap.gif&size=400x1000[/img]

[Dunvi]

Cyonan wrote...

Shampoohorn wrote...


Nice. 64% is close to the 77% that Cyonan pointed out.

There does seem to be a small difference in the distributions -/+ Hunter Mode. It's my understanding that it would affect the refire delay even on single fire weapons. There could be some inherent bias in your finger, heh. The fourth condition -- HM and RC -- might shed some light on that.

Not may replies yet. Math posts tend to get the best response on Mondays, when most of the engineers have had their coffee and become bored with actual work. Saturdays, not so much.

As it turns out, if I adjust the number for standard human reaction time(about 0.15 seconds) then I get a theoretical 64.86% increase in overall DPS.

I would expect those who have practiced reload cancelling enough to not need that reaction time adjustment, since it's not a reaction (you can predict it). The analogy that comes to mind for me is that, for fast tempos, you're looking at a quarter to a half of a beat late (yeah I'm a musician suck it) which is Not Acceptable... since it's prediction more than reaction you can avoid it though.

[landylan]

I don't get the purpose of this. We already know that perfect reload cancelling leads to approx. 77%(I think) on a claymore according to Cyonan (I think).

You're data makes every thing look to complicated, and you didn't label both directions of horizontal and vertical on the graph.

And it looks like hunter mode does improve single-shot weapon accuracy in your tests, but you say it doesn't?? No matter how negligible the increase, you can't neglect it.

[landylan]

So should I still be taking RoF?...

[ComradeShepard7]

landylan wrote...

I don't get the purpose of this. We already know that perfect reload cancelling leads to approx. 77%(I think) on a claymore according to Cyonan (I think).

You're data makes every thing look to complicated, and you didn't label both directions of horizontal and vertical on the graph.

And it looks like hunter mode does improve single-shot weapon accuracy in your tests, but you say it doesn't?? No matter how negligible the increase, you can't neglect it.

That 77% is based on what the claymore is theoretically capable of. This data is from actual testing so it shows an example of what can actually be done in game.

As for labels, histograms don't need a vertical axis label because the vertical axis always represents the number of instances where data falls into a certain category, in this case a range of time.

I assume you mean fire rate for your third point and not accuracy. While it does appear that HM does offer a bit of an increase, the amount of variation in the data makes it difficult to say just how much of an increase it provides. All that I can say with confidence from current data is that it provides an increase of between 1.9% and 3.61%. Once I take more data with a better resolution I should be able to narrow that down more. I shouldn't have said that it doesn't seem to impact fire rate until I had calculated the confidence interval.

[Credit2team]

Ziegrif wrote...

thewalrusx wrote...
I was looking for a different set of numbers, but what the OP gave is better than nothing

I don't think you'll get girls numbers on BSN hun.~
*Badumdish*

<_<

[OniGanon]

Dunvi wrote...

You forgot the time speedup. There's a reason that timing tests should be done on bronze. Accounting for the time speed up brings the theoretical best time value back in line.

Ohhh *facepalm*

Of course.

MasterReefa wrote...

asians and their math..

That is a dirty stereotype!

In my case it happens to be true... but still!