FYI: Incredible Facts!!! % Crit Chance vs. % Increased Crit damage

Hi entropians,

Bellow you will find some incredible facts, backed up by my log files that can't be disputed

The facts concern what is better to have

% Crit chance or % increased crit damage?

Find the answer bellow:

Apart from my tests in Merry Mayhem wich I posted here about % crit chance

I have done yesterday the tests again to include all posibilities:

Test 1

1 hour in MM no enhancer no amp no buff, just the gun, never fapped shoot constant

Facts (based on chat.log)

Total shoots: 3220
Hit Shoots: 2815 - 87.42% rate
Evaded shoots: 352 - 10.93% rate
Critical shoots: 53 - 1.65% rate

Total damage: 243015.2 HP

Test 2

1 hour in MM no enhancers no amp Accuracy buff ( 1% critical chance) never fapped shoot constant

Facts (based on chat.log)

Total shoots: 3205
Hit Shoots: 2755 - 85.95% rate
Evaded shoots: 366 - 11.41% rate
Critical shoots: 84 - 2.62% rate

Total Damage: 242680.3 HP

Test 3

1 hour in MM no enhancers no amp Kill Strike buff ( 10% increased critical damage) never fapped shoot constant

Facts (based on chat.log)

Total shoots: 3211
Hit Shoots: 2853 - 88.85% rate
Evaded shoots: 299 9.31% rate
Critical shoots: 59 1.83% rate

Total Damage: 248103 HP

Let's start with the simple thing first:

- as you see there is almost no difference in damage output (or DPS) in normal vs 1% crit chance
- % crit chance buffs only play with your slider between crit shoots and missed shoot and hit rate, they don't produce more DPS in my case I got 1% more crit but I also got ~0.5% more misses and 1.5% less hit rate

On the other hand, the % increased critical damage buff produced about 5000 HP more damage in same time interval and it seems it has slight increased the hit rate and reduced miss rate

But the only most important fact (that I think no one can dispute) is that:

% increased crit damage (nanobots Kill Strike or Omegator Laser sight) produces more DPS resulting in increased ECO and in better points in events

And if you want the real shit, combine them both ( Laser+Scope or nanobts accuracy+kill strike) and will get the best results, I estimate double , in my case another 10000 HP more damage in same time

I have the Log files prepared for who wants them, I have everything prepared, just ask, I don't know how to attach them here in forum.

Also, please don;t say that 1 hour test is not relevant, 5000 HP more damage, even if it;s in just 1 hour test vs normal is quote a difference to be just a random thing and also I have another 3 hours test from MM and also I can do this again and again to prove me or you this is right

SO, if any MA official wants to congratz me for this ( or ban me for telling the truth) please do so
Also I can take a ATH in-game for being smart anytime MA

!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
~!!!!!!!!!!ˇˇ

Following text is only about statistics and data, it has nothing to do with the theory. If you do not read about boring numbers, then skip this post, if you want to read about statistics it is your choice.






















1h tests is far from valid data for statistics.
I can guarantee that if you did those runs all three runs same settings your results would be differing atleast as much.
Do something like 2 weeks tests (5 h a day, every day different hours), then you might get something.. But atm it is just 3 hunting runs with no statistical value at all.

It is like saying.... Today it took me 10minutes to get home from work, BUT yesterday i wore green hat and got home in 11 minutes.. So green hat reduces my average speed. And oh no!!! One day i lent my friends porsche and it took me 14minutes... that means faster car results in less average speed..

Get some data before creating statistics.


PS! If it was all sarcasm.. then i salute you, you got me. Yeah some ppl are really that crazy that they believe that and shout it in forums.

Interesting data. Small sample size but i wouldn't dismiss it so easily - the compliance with the stats on the buffs is too obvious.

Now, if the natural Kill Strike would behave like this Kill Strike buff it would be really worth achieving...

whiteknut,

The fact that you disclose those fact so easily makes you unbelievable, this is not a loot test to be done in days week..etc

5000 HP damage is to much to be just a random thing

Please do the tests for yourself if you like, keep in mind that 1 kill strike nanobots costs 40 peds

I was expecting harsh comments but not the first one and definetly not from you and I bet you haven't even tested this

It's not that easy, you have to be focused, not to have lags, not to miss any fire bla bla for the test to be conclusive

and I also got 3 hours vs3 normal hours in MM that proved the same thing about % crit change

And if that damage is wrong, as you say, how come I haven;t had any difference when used % crit chance buff, those test produced almost same damage every time, shouldn't they fluctuate also?

bettlejus said:

whiteknut,

The fact that you disclose those fact so easily makes you unbelievable, this is not a loot test to be done in days week..etc

5000 HP damage is to much to be just a random thing

Please do the tests for yourself if you like, keep in mind that 1 kill strike nanobots costs 40 peds

I was expecting harsh comments but not the first one and definetly not from you and I bet you haven't even tested this

It's not that easy, you have to be focused, not to have lags, not to miss any fire bla bla for the test to be conclusive

and I also got 3 hours vs3 normal hours in MM that proved the same thing about % crit change

Click to expand...

I saw the other test also, wanted to comment there also. I am not saying there is nothing here, but really 1h test is not enough (there might be hundreds of factors unknown to us, and by expanding the volume of the test, the data could change a lot). It is like almost as testing crafting success rate vs crafting terminal used (even 5h test wouldnt be enough in that case).
I agree there might be something here, but it is not a "INCREDIBLE FACT" it is incredible theory or something like that. I did look at my Hitrate in MM area also, and it changes quite a lot (more than 3% per run) so it might just be coincidene.
I am not saying your wrong, but you sure havent proved you are right.

Take this:

- this is not loot test
-there is only damage difference (when using % more crit damage) of 5000 HP, which is a lot, hard to believe is random
- there is NO damage difference when using % crit chance ( 4 hours here) which is also strange, this should have fluctuate, don't you say so?

Regarding your fluctuation of crit you said earlier, I think beside MM...it's kinda hard to do this test, you have to shoot constant, not to have lag, not to fap and not to have only 1 huge HP mob, so pretty hard to do it

Ok, maybe Incredible facts is exaggerating, but don't deny my results!

And don't deny the only important fact that damage difference was high ( to be random) when using % increased crit damage and almost 0 when using % crit chance..

If you can repeat each test two more times and every time you get the same results, you can begin to draw a conclusion.

At the moment you have seen some behaviour from 1 test round with each setup and you don't know if you can reproduce it or not.

I wasn't much into it in EU, but from other games i know that the right balance between increased Crit Chance and Crit damage is often key to a good dps increase. I would apply that here as well.

sitram said:

If you can repeat each test two more times and every time you get the same results, you can begin to draw a conclusion.

Click to expand...

I can reproduce this anytime if you give me a Liakon set (F) and Kill Strike nanobots (3)
I had to rent a set yesterday and buy Kil Strike at 40 and 45 ped to test this

or you can read the facts closely and have in mind this is not a loot test and the only difference appeared when using % more crit damage, almost no difference with the others, do you really don't see this as something that can be thought as fact?

bettlejus said:

I can reproduce this anytime if you give me a Liakon set (F) and Kill Strike nanobots (3)
I had to rent a set yesterday and buy Kil Strike at 40 and 45 ped to test this

or you can read the facts closely and have in mind this is not a loot test and the only difference appeared when using % more crit damage, almost no difference with the others, do you really don't see this as something that can be thought as fact?

Click to expand...

it is not important if it is loot test or not, as dmg inflicted and hitrate is also dynamic (little bit less, but still dynamic). You could do the same test without Kill Strike nanobots and if you get one hunt with results close to Kill Strike nanobots runs you can be sure that the data is not enough yet (as it might just have been coincidence)

Hey,

I extracted some data from my mayhem solo runs (Thx bettlejus for the idea with the chat.log)

I was doing my normal MM runs, so candy cranes and kill strike plus arso chip. The tracker splits the chat.log after 10 mb or so and i took one that went on for 4 days.

The problem was, that i used excel to get the numbers. I managed to exclude the arsonistic hits and crits by only using 3 digit dmg numbers (as my weapon does not do less than 3 digit dmg hits) But then I still had the evades in there and i couldn't get rid of those.

So what i did then was, i assumed the arso chip would on avg hit 10 times. So I added 1/10th of the 2 digit hits to the hits and included the crits, as they only show once on first hit.


Using that method I got the following numbers:

Total Shots: 82593
Hits: 72099
Crits: 2235
Evades: 8259

That results in:

87.3% - Hit %
2.71% - Crit %
10.00% - Evade %


I think the arso chip still screws with the results a little but i minimised it as best i could think of.

With this sample I can not confirm that it screws with the evade percentage but rather as expected a larger part of the actual hits are crits.

Do with this what you will, I just wanted to provide some more numbers, I will not upload the unedited chat.log so if you don't think the numbers are viable that's cool too

Regards
Thark

bettlejus said:

I can reproduce this anytime if you give me a Liakon set (F) and Kill Strike nanobots (3)
I had to rent a set yesterday and buy Kil Strike at 40 and 45 ped to test this

or you can read the facts closely and have in mind this is not a loot test and the only difference appeared when using % more crit damage, almost no difference with the others, do you really don't see this as something that can be thought as fact?

Click to expand...

I think you can reproduce easily first test scenario without any investment in nanobots. You can also hunt other mob if you can't find a suitable armor for MM mobs.

Can you tell me for sure you will get the same precentages every time you repeat Test 1?

What if you get similar precentages with those from Test 2 if you do another Test 1?

Nowhere near enough data.

As someone said, you haven't even proved that these results are repeatable - so your method isn't validated.

Considering you have an increased critical rate with the 1%+CR, and an increased damage output with the 10%+CD, shows plainly that the boosts work. You cannot however compare accurate values of CH, HR, Evade until you've at least repeated the test once (where you will see wild variations in HR and evade).

Plus don't forget there's variability in damage output too.

Thark said:

Hey,

I extracted some data from my mayhem solo runs (Thx bettlejus for the idea with the chat.log)

I was doing my normal MM runs, so candy cranes and kill strike plus arso chip. The tracker splits the chat.log after 10 mb or so and i took one that went on for 4 days.

The problem was, that i used excel to get the numbers. I managed to exclude the arsonistic hits and crits by only using 3 digit dmg numbers (as my weapon does not do less than 3 digit dmg hits) But then I still had the evades in there and i couldn't get rid of those.

So what i did then was, i assumed the arso chip would on avg hit 10 times. So I added 1/10th of the 2 digit hits to the hits and included the crits, as they only show once on first hit.


Using that method I got the following numbers:

Total Shots: 82593
Hits: 72099
Crits: 2235
Evades: 8259

That results in:

87.3% - Hit %
2.71% - Crit %
10.00% - Evade %


I think the arso chip still screws with the results a little but i minimised it as best i could think of.

With this sample I can not confirm that it screws with the evade percentage but rather as expected a larger part of the actual hits are crits.

Do with this what you will, I just wanted to provide some more numbers, I will not upload the unedited chat.log so if you don't think the numbers are viable that's cool too

Regards
Thark

Click to expand...

The same buffs apply to all the weapons you use.

If you are maxed on both weapons I would include in the calculation all damage, misses and critical hits.

sitram said:

I think you can reproduce easily first test scenario without any investment in nanobots. You can also hunt other mob if you can't find a suitable armor for MM mobs.

Can you tell me for sure you will get the same percentages every time you repeat Test 1?

What if you get similar percentages with those from Test 2 if you do another Test 1?

Click to expand...

I still need Liakon to test that, but I can tell you I won't get any different results and in no way close to test 2 percentages and also need to have same results because I would need to compare them with my initial ones

Why I need Liakon, because i only want to test this with Kerberos S3 where I am able to shoot constant and not fap

Thark, you need to be very exact with the chat log file for the results to be conclusive

I must find public ftp or something to upload my chat log so any non-believer can check for himself

Adi, do you really want to make a bet with me for this? ( remember, I have 4 hours with accuracy where I saw that same DPS i produce only the balance between miss and crit and hit rate is changed)

oh and btw, I had 2 hours with Kill strike but I haven't said about the first because I had 1 minute lag, but even so I have with 4,1K more damage then normal or with Accuracy

bettlejus said:

I must find public ftp or something to upload my chat log so any non-believer can check for himself

Click to expand...

I think it is not about nonbelievers but about not enough data (so those logs wont help us at all, unless you do some more testing).
I also believe any kind of incresed hit or increased dmg should increase dps, but in this test you havent proven it.. yet....

One more interesting thing:

when excluding the arso chip completely (as stated above - easy for my sample), and ingoring evades then:

Normal hits = 67465
Crits = 2090
Total Hits = 69555

Crits/Total Hits = 3% (3.00482)
Normal Hits/Total Hits = 97% (96.99518)

Which in theory means that the 1% increase is a perfect match! Sure there are evades in there as well BUT they shouldn't interfere with the ratio of hits and crits. It's just hard for me to exclude the arso evades from my data sample.

So there you have it. I think 70k hits spread over 4 days is a fairly large sample and the randomnes shouldn't play as much of a role anymore.

Also i'm to lazy to go through the log manually and i dunno how i could automate it so i dunno how many hours of hunting that was, but i didnt loose much dps so for 30h or mayem my weapon would have done 57*60*30 = 102600 uses, so maybe this sample is over a period of 20.5h of concentrated MM solo hunting.


Regards
Thark

sitram said:

The same buffs apply to all the weapons you use.

If you are maxed on both weapons I would include in the calculation all damage, misses and critical hits.

Click to expand...

well yeah and no.

The chat log shows hits of the arsonistic as 10 different lines with the following format:


You inflicted 34.9 points of damage.
You inflicted 34.9 points of damage.
You inflicted 34.9 points of damage.
You inflicted 34.9 points of damage.
You inflicted 34.9 points of damage.
You inflicted 34.9 points of damage.
You inflicted 34.9 points of damage.
You inflicted 34.9 points of damage.
You inflicted 34.9 points of damage.


and when i count the lines i get all those single hits as seperate hits. But an evade is only shown once. so by design it's impossible for me to get an accuarte count, especially since they don't always show as 10 seperate dmg numbers but maybe only as 8 when there is lag or something. That's why, at least with my method of using excel, the arso screws with me.


But you are right, the arso chip will follow the same rules as all other weapons. But my way of evaluating the sample doesn't play nice with it

but ignoring the arso, the post above this has some nice info!

Regards

Thark, you don't have a proper chat.log for measuring data

whiteknut, are you bad at math, ignoring facts or what?

Why do you say I haven't proven an increase in DPS?

when I used Kill Strike (10% ore crit damage) I have 5K more HP damage in same time interval, same shoots

because this is vas very important for the test to have same time interval and same shoot just to be able to measure damage output

and if normal hour I had 243K damage and kill strike hour I had 248K, you just have to split 3600 second once by 234 and second to 248 and there you have better dps when using kill strike

Crucial thing about this test is the same time interval and same shoots, this allow me to measure damage output / DPS

bettlejus said:

Thark, you don't have a proper chat.log for measuring data

whiteknut, are you bad at math, ignoring facts or what?

Why do you say I haven't proven an increase in DPS?

when I used Kill Strike (10% ore crit damage) I have 5K more HP damage in same time interval, same shoots

because this is vas very important for the test to have same time interval and same shoot just to be able to measure damage output

and if normal hour I had 243K damage and kill strike hour I had 248K, you just have to split 3600 second once by 234 and second to 248 and there you have better dps when using kill strike

Crucial thing about this test is the same time interval and same shoots, this allow me to measure damage output / DPS

Click to expand...

- Actually I do, just think about it for a minute, i mean really think about the problem and the solution I used.

- What whiteknut is saying is that a sample of some 3k odd shots is too small to be absolutely certain that digits after the comma are reliable.

- It is actually not at all important in what time frame you shot the 3500 shots. All we are trying to figure out is if more of your hits are criticals when using a 1% more crit chance buff. It doesn't matter if you shoot the shots in 1 hour or over a week (assuming MA doesn't change anything inbetween). Once this is known, the dps can easily be calculated for any given gun etc.

Regards
Thark

bettlejus said:

whiteknut, are you bad at math, ignoring facts or what?

Why do you say I haven't proven an increase in DPS?

Click to expand...

Because you haven´t. Simple as that
You can only calculate statistics(which you are trying to do) if you have enough data. Until you dont have enough data you can calculate all you want, but you will never able to prove the theory.

I believe you yes, they do affect DPS and such, but so far there is too little evidence to support your theory (even if it is true, it is still a theory until proven).

So instead of accusing others for being bad at math and ignoring facts, please take a good look on your data and think it over.

You to are making my life harder because you don't know what you're talking about

4 hours with accuracy haven't change damage output but just percentages between crits and evade and hit rate

1 have only 1 hour with kill strike but even so there was a big difference in damage output for the same shoots

Thark I am try8ing to control time just in case there re other variables in equation, I know, all I need is same shoots but maybe if you take same number of shoots on different times maybe thing change, that's what I'm trying to avoid by doing same shoot and same time

But really now, how can you dismiss this:

quite a different damage output with Kill Strike
almost no difference with Accuracy

Same number of shoots and same time

Instead of keep telling me I have few data maybe think about this

You say I have way to small sample, ok, ok, but why there is almost no difference in DPS (DAMAGE OUTPUT) with accuracy and way to big difference with Kill Strike???

Damn i feel like I saw a ghost ( I know I did) and try to prove you people I did

I am at your disposal to test whatever you want but in the end I don't want to get my ped card to 0 ( lost 7K already past week)

If you all want to co-operate in this feel free to help, but lighten up a bit and see that huge difference in Kill Strike and no difference in Accuracy first

Different results should have been in accuracy also, but there weren't

one Question i might have missed it but what weapon did you use?

Kaza

Kazanori said:

one Question i might have missed it but what weapon did you use?

Kaza

Click to expand...

Maxed SIB

Maxed SIB what?

the difference between the number of shots that hit target and the damage of the gun can quite easy make up 5000 damage

Kaza

Kazanori said:

Maxed SIB what?

the difference between the number of shots that hit target and the damage of the gun can quite easy make up 5000 damage

Kaza

Click to expand...

Just to be clear, we are talking about hits here

When you over-damage a mob all the damage done is recorded not just what the mob had left

so, all the damage I done ( I could have shoot punys, same results, over-damage don't influence this) is 5K higher then normal only when I used Kill Strike, almost no difference with accuracy


I think what you're talking about applies to faps where they heal only the part of life that you have left, but with weapons all damage (including over-damage) is recorded

So, what were you trying to say?

its a simple question what gun did you use?

you used chat log i know how guns work, overkill included yes i know, just tell me what gun you used.

Kaza

Test 1
Evaded shoots: 352 - 10.93% rate

Test 2
Evaded shoots: 366 - 11.41% rate

Test 3
Evaded shoots: 299 9.31% rate


Here's the secret.
You have 2% hit rate difference between Test 2 and Test 3. With total dmg about 250k that's exactly 5000 damage. Are the nanobots responsible? They're not supposed to. They should only affect critical shots. So I believe this (quite insufficient IMHO) data is just a result of the normal game dynamics and not of the usage of nanobots.

In test 1 you had 87.42% rate

In test 2 you had 85.95% rate

In test 3 you had 88.85% rate

Don't you think this influence total damage?

Seems to me that in the test 2 where you were suppose to have higher hit rate because of the 1% increase in number of critical shots has a lower hit rate. You had with 1.47% less hits then in test 1.

In round 3 you had the highest hit rate of all.

How can you prove me that the more total damage you saw in test 3 is caused by the extra damage added to the critical hits and not because you simply had more normal hits. Compared to test 1 you made with 1.43% more hits.

Now you understand you me and others say you need more data?

For the sample size the numbers are valid.

Though as we all know the cycles of EU are less predictable and over time the averages are different.

I think your numbers could use some averaging...if all tests had the same amount of shots then we would have closer numbers but the variables in the formula throw alot of anything short term off. But I also notice some unstable variables, and it highlights the issue with chat logs, or anything over time that isn't over a very long time.

The biggest factor in your numbers is your hit shot %, and the damage correlates directly to those figures.
3% variance on hit% is alot of damage difference, and also makes for determining the effect of crits that much harder.

Your hit rate has the biggest and primary effect on your total numbers
87% 85% 88% getting...... 243015.2(mid damage), 242680.3(lowest damage), 248103(highest damage)


If you did longer runs this gets even more confusing, and hard to decipher.

Also, 100 crits for 250 damage each is only @1000 extra points of damage gained not the full 2500. Something to consider, often people look at the crit number and think thats how much more damage the crits are doing to the mob, its a much smaller effect, the base hit needs to be subtracted.


Overall the % to crit and 10% to damage on crit does not add a ton of damage in reality. The scope is better by itself. 2 sights will be fairly close to 1 scope.

For me for every 100 shots on my weapon it should work out to about the following: (theoretical fixed numbers are about as accurate as the small run chat logs unfortunately, both wont match the game exactly, but at least we can roughly derive an estimate with controlled numbers)

(gun used in figure 138 max hit 276 max crit.)
SIGHT (adds crit damage)
I would add 22-28 damage per crit for mk1 (10%)
If I hit 2 crits per 100 I get 44-56 extra damage,
mk1(10%) = I get 44-56 extra damage (2/100 crits)
mk2(20%) = I get 88-112 extra damage (2/100 crits)

SCOPE (adds crit chance)
If I hit 2 crits per 100 I get 440-560 damage, of this 160-280 is already base so we get a benefit of 280 damage for the crits (140 each crit).

mk1(1%) = I would get 140 extra damage approx(3/100 crits) (140 damage is extra from of original base hit)
mk2(2%)= I would get 280 extra damage (4/100 crits)


I think if you had a consistent hit % the numbers would show something similar when scaled up. The hit rate variance is a killer for logs though, need to have a ton of runs to get close enough hit averages for all 3 tests, and then I think we would see the effects better.