The efficiency of scopes/lasers - Statistics/Discussion

Since the results we seem to keep getting appear ambiguous, I thought we might be barking up the wrong tree. While everyone seems to be focusing on HA multipliers, I thought I would do a small experiment to see if skillgains are the actual intended effect of lasers and scopes. Many people have suggested this, but I have seen no actual tests yet.

The setup: Opalo +A101
Target: Foul Young only.

I did not track the hits or misses, did not track the loot, in fact I did not even track the number killed. I went out with 20 PED ammo for each run, approximately 1 hour of hunting. I checked my skills before the hunt and noted them. I recorded my skills after the hunt and recorded them. I did 10 hunts one after the other, alternating between the naked Opalo +A101 and one loaded with a Seizzt 1950RS and two Seizzt 2000L+ for a 15% skill bonus. Each run with the scope and sights added 0.65 PED to my decay bill. My Opalo has been maxed for years, my laser sniper (hit) is 26. I used the chip optimizer tool to determine the PED value of the skills earned for each run. I believe that regardless of level the PED value of skills on identical runs should be equal as the optimizer tool accounts for the varying value of skills at differing levels. I used auto-aim only. Here are the results:

Run #1 (With attachments)
9 PED of skills

Run #2 (Naked)
6 PED of skills

Run #3 (With attachments)
8 PED of skills

Run #4 (Naked)
4 PED of skills

Run #5 (With attachments)
9 PED of skills

Run #6 (Naked)
6 PED of skills

Run #7 (With attachments)
7 PED of skills

Run #8 (Naked)
7 PED of skills

Run #9 (With attachments)
9 PED of skills

Run #10 (Naked)
7 PED of skills.


Naked total = 30 PED of skills

Attach total = 42 PED of skills.

Now, granted, the optimizer tool includes chip prices, so that would have to be subtracted out somehow. 200 PED for a test may not sound huge, but trust me, I killed a rather obscene number of foul youngs. The other problem is the rounding to the highest PED by the optimizer, it is not an accurate enough tool to get exact stats. I do know I spent an additional 3.25 in attachment decay in return for an additional 12 PED towards my skills (should I have purchased chips instead). That sounds like a pretty good deal, and does not count the agility bumps I picked up along the way either.

Perhaps this is why they raised the curve for hunting related skills... now you need uber scopes and sights to get past 5.0 HA before you retire.

You make the call. One critter only tested, one single setup. I wonder if the results would vary with different guns and different targets. Or if this is simply a statistic anomaly.

PS: selling foul bone buttons :)


BIG apologies for the off topic here, but, MY GF has created a new ava ;) and we went through over 15k of ammo (150 ped) with an opalo and she's only got 320 rifle skills !! I have no idea how you have gained so much skills !

back on topic :)

PS, I keep seeing statements like, well as theres definate evidence for that, we'll exclude it as wrong ! Surely someone stated the earth was round before they had definate proof. Little arrorgant in my opinon to say that.

Good work though otherwise
 
Actually, I just wrote down the skills in pen in the composition book I use for keeping track of EU stuff, no file exists.

I included all of the skills involved in hunting, including such innoculous things such as perception, evade, and alertness to get my PED values. I do not need a fap to kill foul young so there were no fap-related bumps.
 
It requires a python interpreter and isn't very user friendly. You still want it? :laugh:

I hoped to get the functionality included in NRF or a similar program, but didn't get any bites from Squee or see enough community interest to turn it into a finished program on its own.

davidfalkayn on gmail could help you out if you want to send in the NRF file Rival.

I haven't messed with Python much but I might be able to get it to work for me. :) So if you don't mind sending it...

And it would be sweet if it were in NRF.
 
To make it all a bit easyer, I heard there used to be an OCR program that reads the chat and can record the damage you do. What happened to this program? Or did it never exist? If not, maybe its a idea to create a tool like this. With help of Squee or Dutchy that should be easy to do right?

I can just set my anti RSI software on 2 seconds so the cersu doesnt die, and can let it do runs of 2000 shots or somthing, without any need of interaction.

Sarah did create an application like this a long time ago, but it stopped working after the interface update in the game, and as far as I know, the source code for it has been lost since a long time.

It would be great if someone out there could create a real working OCR application which just grabbed a screenshot of a small part of the screen and stored the text in a document file. Some features could make it even better, but isn't there already OCR applications out there which grabs screenshots, or at least, reads screenshots of your skill values?
 
Sarah did create an application like this a long time ago, but it stopped working after the interface update in the game, and as far as I know, the source code for it has been lost since a long time.

It would be great if someone out there could create a real working OCR application which just grabbed a screenshot of a small part of the screen and stored the text in a document file. Some features could make it even better, but isn't there already OCR applications out there which grabs screenshots, or at least, reads screenshots of your skill values?

I've been working on it but the biggest problem is the transparency of the chat window. Also it hasn't been an actual text window for may VU's.

I was using Fraps to capture every 5 secs. I decided to try; making sure the background was relatively dark but I was testing using a knife and some of the captures had the knife behind the chat which really messed up trying to pick out blue letters from the blue of the knife blade.

I haven't given up yet.....

James
 
I've been working on it but the biggest problem is the transparency of the chat window. Also it hasn't been an actual text window for may VU's.

I was using Fraps to capture every 5 secs. I decided to try; making sure the background was relatively dark but I was testing using a knife and some of the captures had the knife behind the chat which really messed up trying to pick out blue letters from the blue of the knife blade.

I haven't given up yet.....

James

There is a button by the side of the chat window that lets you switch between fully transparent, semi-transparent, and opaque.
 
There is a button by the side of the chat window that lets you switch between fully transparent, semi-transparent, and opaque.

No opaque on my system?? Something in the options I need to set??


James
 
No opaque on my system?? Something in the options I need to set??


James

Not as far as I know, but I could be wrong, I never use this feature myself, I have it permanently on semi-transparent.
 
Any updates? Any new tests?

Bump? :)
 
There are about 5 billion extraneous threads about this same topic running rampant in the forum right now, so i guess it's time to bump this one (which actually has hard data rather than testimonials).

One thing that i think is new to this thread that was pointed out by The Nightbird is that the old attachment description explicitly states that it increases the chance of hitting the target.

untitled_144680.JPG


Q.E.D.
 
There are about 5 billion extraneous threads about this same topic running rampant in the forum right now, so i guess it's time to bump this one (which actually has hard data rather than testimonials).

One thing that i think is new to this thread that was pointed out by The Nightbird is that the old attachment description explicitly states that it increases the chance of hitting the target.

Q.E.D.

It also says it is for energy weapons ;)

(Lasers sights were ingame before blp)
 
To contunue from the other thread:

I made a little tool that records damage you do. You can download it here: http://www.entropiawiki.com/EUOCR.rar.

Some notes:

*Only works when EU is in Windows 2k compatibility mode

*Only tested on 1024x768, although it should work on other reslutions too

*Only works in Fullscreen mode (stay fullscreen, else the program will get confused).

*It is fully on your own risk, so only download it when you trust me, and dont care your computer can crash etc.


I did a first testrun using the tool (mainly to test the tool):

Weapon: LR41+A105, no attachments, all maxed

The amp decayed 953.42 pec which equals 386 shots exactly.
The tool recorded 346 shots. I counted about 4 shots due to bad aim. So hitrate is:

(350/386)*100%= 90.7%

Avarage damage = 63.34. Expected was 64.5


The results are not very usefull, but at least it shows that that tool seems to work :cool:
 
I just gave it a try in both windowed mode and in fullscreen mode. It seems like it only captures the form where I enter my chat-messages, and not the actual chat-window.

I don't know what it captures when I launch it in fullscreen mode however, as I can't see the window then. I suppose it would start working if you managed to support it for windowed-mode as well however?

Edit... I just forgot to run EU in Win 2000 compatibility mode.

To contunue from the other thread:

I made a little tool that records damage you do. You can download it here: http://www.entropiawiki.com/EUOCR.rar.

Some notes:

*Only works when EU is in Windows 2k compatibility mode

*Only tested on 1024x768, although it should work on other reslutions too

*Only works in Fullscreen mode (stay fullscreen, else the program will get confused).

*It is fully on your own risk, so only download it when you trust me, and dont care your computer can crash etc.


I did a first testrun using the tool (mainly to test the tool):

Weapon: LR41+A105, no attachments, all maxed

The amp decayed 953.42 pec which equals 386 shots exactly.
The tool recorded 346 shots. I counted about 4 shots due to bad aim. So hitrate is:

(350/386)*100%= 90.7%

Avarage damage = 63.34. Expected was 64.5


The results are not very usefull, but at least it shows that that tool seems to work :cool:
 
Last edited:
I just gave it a try in both windowed mode and in fullscreen mode. It seems like it only captures the form where I enter my chat-messages, and not the actual chat-window.

I don't know what it captures when I launch it in fullscreen mode however, as I can't see the window then. I suppose it would start working if you managed to support it for windowed-mode as well however?

I'd love to try it out, so I hope you'll be able to fix this. :)

It monitors the chat where it would be when running in fullscreen. To determine the position I take the resulution minus an offset. Although I am not sure the pixels scale 1:1 when resolution is changed. If not it would only work in 1024x768
 
I tried to run it in 1024x768. I also tested to run it using both 16bit as well as 32bit, as I had a simular program earlier which was unable to read the numbers when the game was ran in 16bit.

The logs are always the same, zero hits. So, I'm not really sure what's wrong. Would it be possible to make a modified version of the program which captured a screenshot and saved it to the harddrive a few seconds after pressing start? Then I guess you would be able to see if there's any difference between the pictures from my computer (Colours, position, etc), even if I don't think that the position should be wrong.

Edit... I just forgot to run EU in Win 2000 compatibility mode.

It monitors the chat where it would be when running in fullscreen. To determine the position I take the resulution minus an offset. Although I am not sure the pixels scale 1:1 when resolution is changed. If not it would only work in 1024x768
 
Last edited:
I tried to run it in 1024x768. I also tested to run it using both 16bit as well as 32bit, as I had a simular program earlier which was unable to read the numbers when the game was ran in 16bit.

The logs are always the same, zero hits. So, I'm not really sure what's wrong. Would it be possible to make a modified version of the program which captured a screenshot and saved it to the harddrive a few seconds after pressing start? Then I guess you would be able to see if there's any difference between the pictures from my computer (Colours, position, etc), even if I don't think that the position should be wrong.

Are you sure you run EU in win2k compatibility mode? You should have sharp edges on the text else it wont work. You can send me a screenshot so i can see whats wrong.
 
Ahh, I guess I was to much in a hurry to try it out that I forgot about your notes.

It seems to be working now however, and the logs are looking great. Thanks for making this, and thanks for helping me out. + Rep for it!

Are you sure you run EU in win2k compatibility mode? You should have sharp edges on the text else it wont work. You can send me a screenshot so i can see whats wrong.
 
That sounds like a bug to me. If you can specificly state when that happens (what chat message), I can fix it. PS, like suggested before, it might be an idea to continue discussion in this thread: https://www.planetcalypsoforum.com/...ency-scopes-lasers-statistics-discussion.html

it seems on of false hits is when mob inflicts crit on avatar (but in first log file I've ~770 hits while I shooted ~500-600 and those troxes don't crited me 170 times, so something is fishy, maybe normal chat ? as as far as I can tell you don't care for text color ?), altho value isn't correct (which isn't surprising with your method of parsing.
 
it seems on of false hits is when mob inflicts crit on avatar (but in first log file I've ~770 hits while I shooted ~500-600 and those troxes don't crited me 170 times, so something is fishy, maybe normal chat ? as as far as I can tell you don't care for text color ?), altho value isn't correct (which isn't surprising with your method of parsing.

Yes I realized a crit hit from a mob results in a fault, it start with exactly the same text as then you do a CH ;). About the extra entries, do you alt-tab in the meanwhile? If so, that might well be the cause, I had the same. I will take a look if I can make it less sensitive too that.
 
plus rep for a great program, and it works great in windowed mode as well. you just have to position the screen in the lower left corner. make sure it's exactly on the pixel though
 
Yep I alt tabbed all the time :), so that would be it (check for foreground window title, or window title from point you're screening would be great). can also confirm it's working great in 1280x1024. now to make it perfect it "only" needs to find out weapon/tool used, count number of shots and record claim size :laugh:
great work +rep (when I spread enuf elsewhere that is)
 
plus rep for a great program, and it works great in windowed mode as well. you just have to position the screen in the lower left corner. make sure it's exactly on the pixel though

Yup, thats how I test it, placing a screenshot at the right spot ;).

Yep I alt tabbed all the time :), so that would be it (check for foreground window title, or window title from point you're screening would be great). can also confirm it's working great in 1280x1024. now to make it perfect it "only" needs to find out weapon/tool used, count number of shots and record claim size :laugh:
great work +rep (when I spread enuf elsewhere that is)

I updated the critical hits, but im not sure if it works correctly now as I dont know the text when you get a crit ;). As for the other functions, the sourcecode is included so feel free to add them :D.

Edit: The program now detects EU, and made recognision more strict, so alt tab should no longer be an issue (not tested yet).

Edit2: After more shooting, I solved some more bugs(12:30). New version is uploaded.
 
Last edited:
Just did some tests together with Witte and it all works fine, +rep for a useful tool!
 
Last edited:
Some updates:
I added an option to filter out damage below a certain value, so finisher weapon hits are not logged. I also added an option to log recieved hits. Only thing that is not working yet are armor penetration hits (still need a screenshot of it).

I did 2 runs, both with CB19+beast (59dmg, maxed). I determined total numbers of hits from amp decay (I discovered the beast decay 0.006 PEC more as stated on entropedia, updated). I made sure I did not miss. I included the logs in the rar file of the tool.

First run:
Mob: Robots N of twin.
Shots: 998
Hits: 920
Critical hits: 14 (1.40%)
Hitrate: 92.2%
Avarage damage:45.1 (w/o CH: 44.2)

Second run:
Mob: Argos S of nymph
Shots: 572
Hits: 529
Critical hits: 11 (1.92%)
Hitrate: 92.5%
Avarage damage:45.6 (w/o CH:44.34)

(Min+Max)/2 would result in exactly 44.25 damage, so I think it is safe to conclude damage has a uniform distribution, as was already concluded from earlyer tests. The hitrate at 10HA seems to be either 92% or 92.5%, some more tests might be needed. I think 92.5% might be more logical, since damage increases 12.5% from level 0 to 100, asuming that at 0HA hitrate is 80%.

Next time I will test a weapon with lower HA.
 
What's next?

Will you test UL weapons with sights and scopes? I would also like to see some tests with maxed L weapons with lasers and scopes and see if you can't get a 100% hitrate! that would be blood awsome for PVP!

Good work and + rep to all who take the time and expense to gather data like this and share it with us lazy folk :)
 
What's next?

Will you test UL weapons with sights and scopes? I would also like to see some tests with maxed L weapons with lasers and scopes and see if you can't get a 100% hitrate! that would be blood awsome for PVP!

Good work and + rep to all who take the time and expense to gather data like this and share it with us lazy folk :)

The lazy folk ought to read the thread, at least. ;) At the beginning there are the results of my tests that indicate that one does get 100% hitrate in PvP with or without attachments if the weapons are maxed. I also tested a maxed weapon with and without attachments and didn't see a significant change in hit rate.

The thing that was not tested is the misrate against someone with extremely high dodger standing, but we had a report from someone else who said that they always had 100% hitrate when they tested in PvP against skilled players.
 
The lazy folk ought to read the thread, at least. ;) At the beginning there are the results of my tests that indicate that one does get 100% hitrate in PvP with or without attachments if the weapons are maxed.

:ahh::ahh: You know what is even worse! After reading this, I remembered I did read that already.... :ahh::ahh: About 6 months ago before I tried PVP.
 
Just did another run. This time with a m2910+105 @ 47.8(hit)/50.7(dmg)

Mob: Argos S of nymph
Shots: 504
Hits: 432
Critical hits: 6 (1.19%)
Hitrate: 85.71%
Avarage damage:46.8 (w/o CH:45.7)

I will do a couple more runs to get a somewhat accurate hitrate, and then I will start adding attachments.
 
Last edited:
Native HA: 5.35
Modification: 31%
Modified HA: 7.01

mob: ambulimax young-provider
hits: 7645
shots: 9419
hit rate: 81.17%
Expected native hit rate: 86.69%
Expected modified hit rate: 88.76%
(80% base HR, 1.25%/HA)

I knew i was getting ripped off. :mad:

Other info:
103 of 7645 hits were criticals (1.35%)
Total recorded number of taken hits: 717 of which 6 were critical: 6 (3 armor pen., 3 increased damage)
My average damage done isn't of interest because i changed amp magnitude in the middle somewhere.

Witte, the parser crashed somewhere at or after 5150 hits recorded. Is there an expected limit based on data structures? I didn't see any...
Anyway, I'm not sure if it missed saving some of them or not.

I couldn't get it to work in windowed mode, FYI.
 
Last edited:
This is really a great tool. I've been looking for an EU parser for a while.

+rep
 
Back
Top