Waterfiends are weak to crush... Rapier time!

[Hedgehog]

I really dislike that everyone's just counting crims per trip. That's a really inaccurate way to gather data. If anything, we should work to find their defense rating.

[Powerfrog]

I really dislike that everyone's just counting crims per trip. That's a really inaccurate way to gather data. If anything, we should work to find their defense rating.

I acknowledged that it wasn't the best method for determining the effectivness of either weapon and recommended others to use alternative methods, i only counted crimsons as it was more of a small personal test than a quest to find a definite answer and counting crimsons was far less effort on my part. I know nothing about how defense ratings work and how accuratley they represent runescape but that sounds like a good idea.

[Lose No Hope]

I really dislike that everyone's just counting crims per trip. That's a really inaccurate way to gather data. If anything, we should work to find their defense rating.

I couldn't agree more. I think Levon had gathered some data a while back, but not enough to draw a definite conclusion.

[Quyneax]

If everyone counts crimsons/trip it'll all work out given sufficient tests... you need kills/hr to work out xp/hr, effigies/hr etcetera but charms/hr is perfectly relevant to relative kill speeds. A larger problem is time zones (crowded/uncrowded worlds & accompanying spawn rates) and picking up drops or not. Those things vary from person to person instead of being steady.

[Hedgehog]

Defense ratings would still be much more accurate.

[Quyneax]

Defense ratings would still be much more accurate.

More widely applicable and requiring less data-gathering maybe, but there's already much more data on charms.

[Hedgehog]

Defense ratings would still be much more accurate.

More widely applicable and requiring less data-gathering maybe, but there's already much more data on charms.

Fair enough, Diane.

[LitterBug]

Can't you just count hit/misses with the weapon?

If the hit of the weapon is a uniform distribution and we know the speed of the weapons --> won't we have accurate information?

 

Of course, we're not accounting for things like overkill, etc

[Hedgehog]

Overkill isn't hard to account for.

[Quyneax]

Can't you just count hit/misses with the weapon?

If the hit of the weapon is a uniform distribution and we know the speed of the weapons --> won't we have accurate information?

 

Of course, we're not accounting for things like overkill, etc

Counting hits/misses works, you just need to get quite a lot of counts done. There already is a lot of data on crims charm rates though, as that's the only thing most people cba writing down. Overkill isn't hard to account for provided you're only comparing two setups, as you have to run quite a lot of calculations for each setup.

[Hedgehog]

 

Gui, Add, Button, gHit x2 y2 w70 h50 , Hit
Gui, Add, Button, gMiss x72 y2 w70 h50 , Miss
Gui +alwaysontop +ToolWindow
Gui, Show, w144 h54, Hit/Miss
return

F1::
Hit:
Hit++
Total++
return

F2::
Miss:
Miss++
Total++
return

F10::

Gui 2: Add, Text, x2 y2 w30 h20 , Hits:
Gui 2: Add, Edit, x38 y2 w40 h20 , %Hit%
Gui 2: Add, Text, x2 y22 w34 h20 , Misses:
Gui 2: Add, Edit, x38 y22 w40 h20 , %Miss%
Gui 2: Add, Text, x2 y42 w30 h20 , Total:
Gui 2: Add, Edit, x38 y42 w40 h20 , %Total%
Gui 2: Show, w112 h75, Data
return


GuiClose:
ExitApp

 

 

Whoever wants to collect hit/miss data can use this. It'll make it much easier.

[Lose No Hope]

I'll end up finding their slash and stab defense rolls when I get back to my slayer data collecting. That may be awhile before I do it though.

[Jeff Delu]

Why would you collect slash rather than crush? If we had stab and crush we would be able to compare zs, ss, maul, and rapier. Unless there is a slash weapon that could compare with those I don't see the point.

[Lose No Hope]

because I'm collecting slash and stab data on all slayer monsters.

[Jeff Delu]

because I'm collecting slash and stab data on all slayer monsters.

 

Well that makes sence.. for monsters that you slash and stab against. Wouldn't it make sence to make an exception for waterfiends?

[Lose No Hope]

fair enough

[pulli23]

Can't you just count hit/misses with the weapon?

If the hit of the weapon is a uniform distribution and we know the speed of the weapons --> won't we have accurate information?

 

Of course, we're not accounting for things like overkill, etc

Counting hits/misses works, you just need to get quite a lot of counts done. There already is a lot of data on crims charm rates though, as that's the only thing most people cba writing down. Overkill isn't hard to account for provided you're only comparing two setups, as you have to run quite a lot of calculations for each setup.

Problem is: with the differences you're looking here you have to be carefull on "crimson rate / hour". It would only work if you are close to the "standard" with maul. - Or do the double test (maul + rapier) as 1 person.

[LitterBug]

Can't you just count hit/misses with the weapon?

If the hit of the weapon is a uniform distribution and we know the speed of the weapons --> won't we have accurate information?

 

Of course, we're not accounting for things like overkill, etc

Counting hits/misses works, you just need to get quite a lot of counts done. There already is a lot of data on crims charm rates though, as that's the only thing most people cba writing down. Overkill isn't hard to account for provided you're only comparing two setups, as you have to run quite a lot of calculations for each setup.

 

Wut?

 

I don't really know all that much about data calculators but, from a purely statistical standpoint -

Counting hit/misses are going to be much better then counting charms.

 

If we use the same stats - no pray/ovl/etc, and a predefined p1 (probability that a rapier hits) and p2 (probability that a maul hits) - you wouldn't need more then a couple of hundred of hits.

Although this may not be very fun to calculate - just one or two people collecting one sample of 500 each could easily draw conclusions.

 

For charms - you're going to count only crimsons I presume (or have a huge anova test for gold/green/crimson/blue - but this induces a lot more error)

a.) The difference in #of crimsons is going to be very minimal compared to the total number of crimsons

From Nattylight, the avg crimsons per hour seem to go from 200-250.

With a difference of 10-15 crimsons, you're only seeing a 5% difference

b.) A lot of the difference in charm drops can be attributed to proportion to charm drop

c.) You're assuming that the kill rate is constant per trip (you're equally focused, putting the same amount of effort, etc)

d.) You're going to need A LOT of different people's information

 

If somebody can provide me with rough estimates of how accurate the maul/rapier is - I can calculate some stuff to provide a # of hit splats that one should gather to draw an accurate conclusion.

 

EDIT:

Also, it seems that you'll only find the DPS rate which may be different from the actual kill rate.

I.e, you lose a lot less turns grabbing charms inbetween hits with a maul then a rapier

[Sir_Squab]

The point being made was that if you go looking for it, you can find a large amount of logged WF trips where the only data given is only really the charms from the trip and the time. Using charms requires only finding already existing data, whereas counting hits/misses require you to actually GO to WFs and count the hits/misses.

 

I don't think anyone thinks that crims is actually BETTER data then hits/misses, it's just that the crims data is already there.

 

I'd like to point out:

If somebody can provide me with rough estimates of how accurate the maul/rapier is - I can calculate some stuff to provide a # of hit splats that one should gather to draw an accurate conclusion.

 

I don't think anyone can actually give you that data; you can either hope someone is willing to go collect the data for you (in which case said person will probably do the calucations themselves) or go collect it yourself. The crims data is likely the only data that someone will have recorded and be able to give you.

[Quyneax]

Squab covered it, I know quite well that dps data would be preferrable but until such data is available you got to use what you got :). I'm pretty sure Nattylight didn't give you charms/hr but charms/trip there, for 2 hour (or 2 unicorn) trips.

[xat]

One 30 minute test is nowhere near reliable enough to say one weapon is better then another.

 

Make it around 10+ hour long test and you'll have conclusive results.

[Th e Doctor]

So at 90 att 92 str and 84 defense would a rapier definitivly be better than a sara sword/zspear?

[Issildur]

Is anyone collecting data on Goliath gloves vs WFs? I have the results of one 37 minute trip, but it's woefully skewed with dreadnip use.

[That_Brown_Kid]

i use rapier here only because i dont have a crush alternative that defeats rapier in dps. it works well for me :thumbsup:

[Powerfrog]

I think when i get 99 attack i'll try to get some actual data, using ovls and piety at all times. I need to get some charms anyway :D

 

I'm glad this thread got so much interest but it's a shame no one has tried collecting data to verify if the rapier is a contender or not.