Jump to content
PA-Sniv

IL2 server stats (unofficial)

Recommended Posts

Hey Vaal, Stats are awesome--- thanks!

 

I was hoping you could clarify if there is a way to fix the way the statistics are scoring a specific type of event.  

 

I was hitting some ground targets on WOL, when my aircraft was hit by AAA.   I lost complete elevator control, so the game flagged me as "Destroyed by AAA" in the chat.  Even though the game thought I was destroyed, I was still able to control my aircraft's pitch with elevator trim.   Carefully, I flew my plane back to my home airfield and performed a perfect gear down landing without crashing.  

 

The statistics marked my aircraft as "destroyed" and "shotdown" in the log, and counted my perfect gear down landing as an emergency landing, preventing me from receiving the 100% bonus for "Landed on Airfield" even though I did so successfully.  Instead I received a -20% for being shotdown.  

 

Here is a link to the sortie log:  http://il2stat.aviaskins.com:8008/en/sortie/log/633270/?tour=9

Here is a twitch highlight video of the damage, rtb, and successful landing: https://www.twitch.tv/pand_twb/v/63305651

 

Hope this helps in troubleshooting.  On other flights, I've made it home with much more damage than this and the game/stats have counted it as a successful landing, simply because the excessive damage wasn't marked as "critical".   II believe that if you are able to fly your aircraft back to base and land successfully, regardless of damage taken, it should be considered Landed on Airfield.

 

Thank you!

Share this post


Link to post
Share on other sites
Events mission are written to a log in a very simple form.
Event "damage" is just a number that indicates the level of damage. That is, using the logs I can not conclude what aircraft part has been damaged.
And there is an event, "Kill", which indicates that the object 1 is kill the object 2.
 

 

 

Instead I received a -20% for being shotdown.  

About it. This is the changes made by third parties (WOL). I can not guarantee that it works correctly.

Share this post


Link to post
Share on other sites

 

Events mission are written to a log in a very simple form.
Event "damage" is just a number that indicates the level of damage. That is, using the logs I can not conclude what aircraft part has been damaged.
And there is an event, "Kill", which indicates that the object 1 is kill the object 2.

 

That's what I was afraid of .... I had a feeling it was dserver/log dependent which is disappointing.

 

Thanks for the quick response and efforts.  :salute

Share this post


Link to post
Share on other sites

When processing todays FNBM I got 2 messages about "unexpected AType ...". Additionally guys driving tanks are not listed in the stats.

 

stats.PNG

AType:22 is the Bf110 Botgunner in the database.

 

I added a zip archive with the message, the mission and the logfile.

stats_message.zip

Edited by [DBS]El_Marta

Share this post


Link to post
Share on other sites

Checking in from Menacing Ferret-land.  Been running this stats system for 6 months... 

 

 

@[TWB]Pand I've done what you've described may times, and in-game it shows destroyed by XXX in the buffer....  and crash/ emergency landing on the mission end screen... yet in the stats page I am given full credit for landing on airfield. I've even ended up on my canopy upside-down on runway, heck sometimes not even too the runway yet, but because I was close enough I am still given a landed at airfield.

 

Now the mission log will still show 100% damage for the plane, but it is now counted as a lost plane. The pilot is alive I still get the bonus. Now if you are captured (wrong side of the front lines) in-game and crash land, it'll show up kind of like a death in the il2stats page. Even though it doesn't count as "dead" in the pilot stats it does count as a death in the K/d ratio and also erases any streak you had going. (this is true even if pilot and plane are 100%)

 

I've noticed something strange going on with the assisted stat. I try to do ALL of my own gunning (from non-pilot positions) but sometimes the AI gunners get a chance to fire before me.   If I kill a plane and jump back into the cockpit before the enemy has crashed or bailed, I'll get an assist not a kill even if my AI gunners fired.  This was not always the case.  When assisted first showed up in the stats build it seemed to work great. Now I get assists on my own kills sometimes, if I land then re-up and finish off a plane.  That also didn't happen when assisted was first added. Not sure what exactly is happening there, just thought I'd mention it.

 

Love the hard work VAAL and isay give putting this together and maintaining it, and am very greatful for this tool!!

 

wish list....

 

-some additional % added to winning coalition based on secondary objectives completion

-Ground vehicle statistics (yes I do mean the player tanks)

 

otherwise thanks so much for the stats and support!!

Edited by MF-Belgarufo

Share this post


Link to post
Share on other sites

Hello, i have a Problem to install the Server stats. I get always the message to see in the bottom of the Picture. The install.cmd can't find the debug-toolbar to install it to Python.

 

With debug=false the message is (last two lines):

 

post-86104-0-97244100-1462292281_thumb.png

 

 

 

With debug = true:

 

post-86104-0-14468100-1462292047_thumb.png

 

What can i do next, to fix it?

Share this post


Link to post
Share on other sites

unexpected AType is some new events, these messages can be ignored because Statistics do not use these events.

 

I./ZG15_Brecher

You need to check out all that is connected with the database. Created the base or not, the correct username and password.

 

Share this post


Link to post
Share on other sites

2 MF-Belgarufo

 

We do not plan to add statistics for tanks. Perhaps we will review this decision if devteam add more tanks, and will be more such players.

 

-some additional % added to winning coalition based on secondary objectives completion

 

Statistics is designed for simple dogfight servers. If add a "bonus" for the subtasks - will be a problem that the  subtasks have different "weight", etc.

 

On other issues, me it difficult to understand what it was about.

Need a link to a  sortie and a brief description of the error in the statistics. Also, may require the mission log file.

Then I can say this is a bug or expected behavior. And  can we fix it or not.

Share this post


Link to post
Share on other sites
I./ZG15_Brecher

You need to check out all that is connected with the database. Created the base or not, the correct username and password.

We had a writing error in the database name so it works now.

 

In Germany, the law forbids to show a swastika in a Webpage. So your Axis.png is a bad think in Germany. We edit the Picture and change it but then the Axis.png didnt show on the webpage.

To edit, change the Picture and restart the Webpage Builder didnt work. How we can change or edit something to get it to work ?

 

Change the German translation "Beste lebende Piloten" in "Beste überlebende Piloten" if you mean the best(highest score) survived pilots.

In English "Best live pilots" to "Best survived Pilots"

 

Thanks for your great work and help!

It`s a huge advantage for the Community.

Edited by A./ZG15_dasSofa

Share this post


Link to post
Share on other sites

Once you have changed the files in the il2_stats\src\stats\static - you need to run collectstatic.cmd

Share this post


Link to post
Share on other sites

Hey Vaal,

 

How do you do your Kill per Death calculation?   It appears you are calculating just Kills per (divided by) Deaths, rather than the more popular Kills per (divided by) Deaths + 1 (to account for your current virtual life).

 

For example:

     Lets say I fly my first sortie of the month and get 4 kills and make it home safely.

          - Current K/D Ratio is 4/0 (which ZERO is non divisible, so I assume you are substituting 1 to achieve) making it a 4:1 ratio Kills per Death.

     

     My second sortie, I don't get any kills, but instead get shot down and killed.  

          - In this case, the current K/D is 4:1, and the current scoring maintains the same Kills per Death as my first sortie, even though I died.

 

     My third sortie, I get 2 more kills and make it home safely.

          - In this case, the new K/D is 6:1, and I have improved my K/D without a penalty for the original death.

 

While the calculation matches the Kills per (divided by) Death ratio explanation, I propose implementing the K/(D+1) which would account for current  life, because this includes the total amount of virtual lives required to achieve the total number of kills displayed.   With this new system, at the end of my third sortie above, the K/D+1 would be 3:1 instead of 6:1.

 

 

This is purely a suggestion, as there appears to be no penalty on the pilot's K/D ratio for having their first death.  

 

Here's a comparison example that might make more sense why I believe this is necessary.

 

     Under the current scoring calculation with K/D:

          - If pilot #1 has 10 kills and 1 Death, his K/D would be 10:1.         Calculation: 10/1= 10

          - If pilot #2 has 10 kills and 0 Deaths, his K/D would be 10:1.       Calculation: 10/0= not divisible, but 10 if you substitute 1. 

     IMHO their ratios above should not be the same. 

 

     Utilizing the K/(D+1) system this is accounted for:

          - If pilot #1 has 10 kills and 1 Death, his K/D+1 would be 5:1.       Calculation: 10/(1+1) = 5

          - If pilot #2 has 10 kills and 0 Deaths, his K/D+1 would be 10:1.   Calculation: 10/(0+1) = 10

     Here Pilot #2's ratio is appropriately displayed for not dying, versus Pilot #1 who died.

 

 

Hope this helps, and thanks for your consideration!   :salute!

Share this post


Link to post
Share on other sites

Works fine now !

 

Is it possible to show up the Number of Players on the Server ? I know the Stats Script just read the new logs at the end of a mission.

For this it needs a new webbuilder script.

 

I read with php the number/name/ping from the remote console, cache them and show it on a extra page. It would be nice when the stats shows also the Players on the Server.

Share this post


Link to post
Share on other sites

Works fine now !

 

Is it possible to show up the Number of Players on the Server ? I know the Stats Script just read the new logs at the end of a mission.

For this it needs a new webbuilder script.

 

I read with php the number/name/ping from the remote console, cache them and show it on a extra page. It would be nice when the stats shows also the Players on the Server.

 

No, we do not plan to add this functionality.

 

 

2 [TWB]Pand

 

Why do you consider only borderline cases?

 

For example  10 kills / and 2 real deaths + 1  = 3.33 

But my real efficacy in this cases 5 kills on 1 death = 5.0

 

I think it is a choice between the error only in borderline cases or error in most cases.

 

In any case, you can use your formula. For this it is necessary to make small chages.

 

in file \il2_stats\src\stats\models.py

change lines

self.kd = round(self.ak_total / max(self.relive, 1), 2)

to

self.kd = round(self.ak_total / (self.relive + 1), 2)

Player KD recalculated after the next sortie.

Share this post


Link to post
Share on other sites

I've seen it done both ways and just wanted to share a suggestion :)

 

Salute and thanks for the code!

Share this post


Link to post
Share on other sites

I find a wrong translation in your stats.

 

English: Tour "May 2016" ended. Axis victory

 

Your German translation: Luftschlacht geht in "Mai 2016" zu Ende. Siege Alliierte. It means Luftschlacht goes in future (Mai 2016 which is now past) to an end.

 

You should write this: Luftschlacht "Mai 2016" beendet. Die Alliierte haben gewonnen.

Share this post


Link to post
Share on other sites

Hi Vaal,

 

This project is totally awesome and I love it for FNBF. 

 

I was curious if the code was open sourced or if you had any plans to that end?

 

Thanks!

  • Upvote 1

Share this post


Link to post
Share on other sites

I think <<Luftschlacht "Mai 2016" ist beendet. Die Achsenmächte [=Axis Powers]/ die Allierten [=Allies] haben gesiegt.>> is even better.

Share this post


Link to post
Share on other sites

Ups i forgott the "ist", thank you for the correction.

 

In my mind i role the dice between "gewonnen" und "gesiegt" and choce the once where was at first in my mind which is most of the time correct.

 

I think "gesiegt" is right when the War is complete over. But its just a part/tour of it and therefore "gewonnen" fits better.

I personally fix the languagefile with: Luftschlacht "Mai 2016" ist beendet. Die Achsenmächte [=Axis Powers]/ die Allierten [=Allies] haben gewonnen.

 

The German language is very complex so it gets more ways to the same. So take what you want ;)

 

django.do at line 24:

change

 

#: .\chunks\models.py:7
msgid "A unique name for this chunk of content"
msgstr "Êindeutiger Name für dieses Inhalt"

 

to

 

#: .\chunks\models.py:7
msgid "A unique name for this chunk of content"
msgstr "Eindeutiger Name für dieses Inhalt"

Edited by A./ZG15_dasSofa

Share this post


Link to post
Share on other sites

Being "axis" we should be happy about every tiny little bit of "Sieg" we could declare (feels more immersive) :). Actually the point that made me reply was "Die Allierte[n] haben gewonnen." and the translative switch from "axis" to "Allierte".

The missing "n" was most likely a hasty mistake ("Flüchtigkeitsfehler" for the foreign speaking fans of the german Language), but did hurt my "gesundes Sprachempfinden".

Also "gesiegt" is better than "gewonnen" because it is shorter and therefore saves bandwith. The ideastic term of "Victory" opposed to the materialistic  "Win/ Gain" bla bla peng peng .... (you know the story).

 

P.S.: "diesen" instead of "dieses Inhalt". Der Inhalt (masculinum not neutrum). [Grammar Nazi off]

Edited by [DBS]El_Marta

Share this post


Link to post
Share on other sites

Hi Vaal,

 

This project is totally awesome and I love it for FNBF. 

 

I was curious if the code was open sourced or if you had any plans to that end?

 

Thanks!

 

 

The project is distributed under the MIT license. But no other plans do not.

Me have already been asked about the github repository for the project, but now I'm not ready for this. This will cause I need more time to support and develop the project.

Share this post


Link to post
Share on other sites

Great work Vaal, many thanks again.  

 

Do you plan to release a light version in order to parse the offline campaigns and quick missions ?

Would be a great addition to better analyse missions, damages, scores, pilot progression in training or campaign.

Share this post


Link to post
Share on other sites

No. And it can not be called lite version, because it will require a lot of work.

Share this post


Link to post
Share on other sites

Thanks for your answer Vaal. 

Sorry, I thought naively that the parse side of the project was re-usable .. but it seems more complex obviously.

 

Cheers 

Share this post


Link to post
Share on other sites

WATCH OUT !!!

FIRST INSTALL/UPDATE beta1.1.0 BEFORE YOU INSTALL/RUN beta.1.1.1!!!

 

it would be better, if vall write this into the readme/update.text

 

we spent a lot of time while we use the latest version and get a lot of errors.

 

The new Image Pillow for python did a mess.

After we change the header logo, run collectstatic, the whole page is "damaged."

it creats a css, but the css name in the html file and the new created css name missmatch ...

Also the Squad logo cache url is wrong and missmatch with the generated one.

Edited by I./ZG15_dasSofa

Share this post


Link to post
Share on other sites
El_Marta" post="361965" timestamp="1464805260"]

 

Also "gesiegt" is better than "gewonnen" because it is shorter and therefore saves bandwith. The ideastic term of "Victory" opposed to the materialistic  "Win/ Gain" bla bla peng peng .... (you know the story).

Even in the age of 32k modems the shorter word wouldnt count  ;)

I take "gewonnen" which was my first impression.After your discussion here in this thread about "gewonnen" and "gesiegt", i ask 21 People (age 30-60) which all chose "gewonnen". Not even one chose or like "gesiegt" in this matter.

If you look the stats translationfile, there is no right or big information in wich manner the translation word is used. Also some word sets are splitted and therfore it is not easy to do a first right translation which you can imagine with your "gesundes Sprachempfinden".

Even in Applications which costs hundreds of €, you finde some missspelling words and you beef about the stats translation where some guys spend their free time for the community :o:

Edited by I./ZG15_dasSofa

Share this post


Link to post
Share on other sites

I messed up my update. I try to reinstall the latest IL2_stats, but ger the following message. PostGres SQL version is 9.4.:

 

 

 

post-1277-0-63511400-1470853001_thumb.png

Edited by [DBS]El_Marta

Share this post


Link to post
Share on other sites

Check directory:

il2_stats\wheelhouse

 

there are these files?

Pillow-3.3.0-cp35-cp35m-win_amd64.whl

Pillow-3.3.0-cp35-cp35m-win32.whl

Share this post


Link to post
Share on other sites

Honestly I have no idea. Just I checked a fresh install, and everything works.
Alternatively you can try to edit the file install.cmd
 
 

cd ../
python -m venv --clear .venv
".venv/Scripts/python.exe" -m pip install -r requirements.txt
pause
cd src
"../.venv/Scripts/python.exe" manage.py collectstatic --noinput --verbosity 0
pause
"../.venv/Scripts/python.exe" manage.py migrate --noinput --verbosity 0
pause
"../.venv/Scripts/python.exe" manage.py import_csv_data --verbosity 0
pause
"../.venv/Scripts/python.exe" manage.py createsuperuser --verbosity 0
pause


In this case, the installer will download the library from the Internet, rather than using local archives.

Share this post


Link to post
Share on other sites

I messed up my update. I try to reinstall the latest IL2_stats, but ger the following message. PostGres SQL version is 9.4.:

 

We have the same pillow problems as it shows on you cmd picture.

 

The reason was that we update from 1.0.16 to 1.1.1 without 1.1.0 .

Maybe it works if you update after 1.0.16 to  -> 1.1.0 -> 1.1.1 ->1.1.2

Edited by I./ZG15_dasSofa

Share this post


Link to post
Share on other sites

 Thank you for the advice. Starting wit 1.0.16 and installing 1.1.0, 1.1.1 and 1.1.2  worked.

Edited by [DBS]El_Marta

Share this post


Link to post
Share on other sites

Yesterday I did the updates from 1.0.16 onwards up to 1.1.2 beta meticuously copying the files listed in the changelogs. After I had finished the system complained about a missing SuadManager iirc.

For the time being I prefer using 1.0.16. Does the 1.1.2 objects.csv work with 1.0.16?

Share this post


Link to post
Share on other sites

Hey Vaal!  :salute!

 

I have an idea if it hasn't already been mentioned... now that you're adding profiles, would it be possible for us to identify our current timezone.   This way times and dates for maps and sorties are calulated and shown accurately for the timezone we live in?

 

Thanks!

Share this post


Link to post
Share on other sites

Did not find it :huh:

Last log was 14-24-21(in attach). I remove this from log folder and run stats again. 

1.zip

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×
×
  • Create New...