ITalianBadBoy

tool
Web Player's Statistic Calculator

74 posts in this topic

Yes, this is a mistake XD... (If you could tell me the colors that you prefer it would be appreciated)

The space occupied by the site has gone from 52,6MB to 0,3MB. (-99,4%)

The SQL queries for to load the page were 61 and now they are 0. (-100%)

The score given by PageSpeed Insights (Google Tool) has changed from 31/100 (mobile, desktop) to 86/100 (mobile) and to 93/100 (desktop). 

For me it was a remarkable success, however I'm aware that there are still other things to implement. (we are working for you)

 

0

Share this post


Link to post
Share on other sites
2 hours ago, ITalianBadBoy said:

Yes, this is a mistake XD... (If you could tell me the colors that you prefer it would be appreciated)

Mh.. I just noticed that yesterday it was so extremely hard to read since Red Shift was active, with normal colors it better readable.

But I would probably go with a similar color as normal text, since it is obvious that is a link so doesn't really have to be so special.

2 hours ago, ITalianBadBoy said:

The space occupied by the site has gone from 52,6MB to 0,3MB. (-99,4%)

You mean on your server? since the site pulls alone 0.5MB JS in.

2 hours ago, ITalianBadBoy said:

The score given by PageSpeed Insights (Google Tool) has changed from 31/100 (mobile, desktop) to 86/100 (mobile) and to 93/100 (desktop).

Imo PageSpeed Insight is more or less a useless nontransparent tool, I went testing both mirrors of my Stats Page and somehow my local server which doesn't support compression and other stuff scored higher then my github.io mirror (79 vs 83).

0

Share this post


Link to post
Share on other sites
1 hour ago, John161 said:

You mean on your server? since the site pulls alone 0.5MB JS in.

Yes, the are only 9 files in the server: 1 html, 4 js, 1 css, 2 php and 1 img. (skeleton, makeup, script and crypt)

The rest of the resources are provided by internet with some loaders. (font, chart, recaptcha, img and data)

 

PageSpeed Insight is influenced by so many factors, for example if you test while the band is used It'll indicate a slow server response...

However, it remains a useful tool for to improve the quality of the web ;)

0

Share this post


Link to post
Share on other sites

I guess you could use downloaded player data :D 

 

1

Share this post


Link to post
Share on other sites
1 hour ago, ITalianBadBoy said:

PageSpeed Insight is influenced by so many factors, for example if you test while the band is used It'll indicate a slow server response...

However, it remains a useful tool for to improve the quality of the web ;)

If it actually would give you any explanation why you got a certain rating.

https://developers.google.com/speed/pagespeed/insights/?url=NeoChontrix.github.io&tab=mobile

https://developers.google.com/speed/pagespeed/insights/?url=jzh.nerdpol.ovh&tab=mobile

Or can you say me why my router scored better then github.io.

 

There are imo better tools then googles out there.

0

Share this post


Link to post
Share on other sites
12 hours ago, MightyHoot said:

I guess you could use downloaded player data :D

Got all data in 0:21:29.053654

Sorry for my slowness, I will implement your advice as soon as I can  :3

 

12 hours ago, John161 said:

If it actually would give you any explanation why you got a certain rating.

https://developers.google.com/speed/pagespeed/insights/?url=NeoChontrix.github.io&tab=mobile

https://developers.google.com/speed/pagespeed/insights/?url=jzh.nerdpol.ovh&tab=mobile

Or can you say me why my router scored better then github.io.

 

There are imo better tools then googles out there.

Yes it's true.. But i don't know why your ruter is best of github...

Your ruter provide the site in 700ms

Spoiler

Connection: close

Content-Length: 895

Content-Type: text/html

Date: Mon, 08 May 2017 06:45:42 GMT

Last-Modified: Sat, 06 May 2017 22:07:30 GMT

 

GitHub in 200 ms

Spoiler

Accept-Ranges: bytes

Access-Control-Allow-Origin: *

Age: 0

Cache-Control: max-age=600

Connection: keep-alive

Content-Encoding: gzip

Content-Length: 471

Content-Type: text/html; charset=utf-8

Date: Mon, 08 May 2017 06:47:32 GMT

Expires: Mon, 08 May 2017 06:57:28 GMT

Last-Modified: Sun, 07 May 2017 15:48:00 GMT

Server: GitHub.com

Vary: Accept-Encoding

Via: 1.1 varnish

X-Cache: HIT

X-Cache-Hits: 1

X-Fastly-Request-ID:xxxxxxxxxxxxxxxxx

X-GitHub-Request-Id: xxxxxxxxxxxxxx

X-Served-By: cache-fra1221-FRA

X-Timer: S1494226052.245920,VS0,VE0

I don't know why.. But i think that the answer for your question is in the header of the site response ....

 

However, you can earn more points with a few simple changes:

  • Embed CSS resources into the head of the page. eg:
      <head>
        <style>
          .blue{color:blue;}
        </style>
      </head>
    
  • Add async at script tag eg: <script src="demo_async.js" async></script>

The script is executed asynchronously with the rest of the page (the script will be executed while the page continues the parsing)

 

0

Share this post


Link to post
Share on other sites

Added some player's suggestions, a special thanks to MightyHoot for his participation.

Updates:

  • Added to the adv. text the execution time in 'ms' or 's'.
  • Inserted the radar chart.
  • Improved some details of CSS and JavaScript.
  • Improved the max score given by PageSpeed Insights 98/100 (mobile/desktop)

The radar chart will be viewable when the user seeks statistics for two players. It will not be printed for a player because it would be a perfect hex.

The maximum distance of a point from the center equals 1 and is equal to the higher value of that category, the lowest ones will be compared to the largest to establish its distance from the center.

0

Share this post


Link to post
Share on other sites

Just checked it out and noticed two bugs

a) it looks like if you look up 2 players of the same corp and the logo isn't cached already, it will be shown only for the 2nd player

b)your radar chart needs more space, the text is somtetimes overlapping

https://s17.postimg.org/qpyvwnovj/2017-05-10-134920_1920x1024_scrot.png

1

Share this post


Link to post
Share on other sites

Well done! Thank you :3

Adjustments:

  • Solved the non-display of the first player's corporation's logo for a double search.
  • Enlarged the radar chart of a third.

I'm still undecided about the values to put in the radar chart.

It is better to have few deaths and losses and this is not valid for other data...

Maybe is better add the cap rate and something else instead of this two... What do you think?

0

Share this post


Link to post
Share on other sites
23 minutes ago, ITalianBadBoy said:

Maybe is better add the cap rate and something else instead of this two... What do you think?

Like this:

khaFOf7.png

1

Share this post


Link to post
Share on other sites

I have some small questions to ask...

  • is DD the acronym of Dmg Dealt?
  • How the Survivability, the Support and the Experience are calculated?
  • What scale is used? And are pure numbers the values used? (Win ratio excluded)
0

Share this post


Link to post
Share on other sites
17 minutes ago, ITalianBadBoy said:

I have some small questions to ask...

  • is DD the acronym of Dmg Dealt?
  • How the Survivability, the Support and the Experience are calculated?
  • What scale is used? And are pure numbers the values used? (Win ratio excluded)
  1. yea
  2. iirc
    • Survivability - death/game
    • support - simple assist or combined some assist with heals ratio
    • exp - games
  3. Scale? 1-9 based on something(posted data somewhere - that ~20mb .txt) like that:

 

1

Share this post


Link to post
Share on other sites

Fantastic! I need read again some notions, of statistical inference and operational research, for to handle these data appropriately.

With this considerable number of data perhaps we could do an accurate simulations with the Monte Carlo method.

0

Share this post


Link to post
Share on other sites
8 minutes ago, ITalianBadBoy said:

Fantastic! I need read again some notions, of statistical inference and operational research, for to handle these data appropriately.

With this considerable number of data perhaps we could do an accurate simulations with the Monte Carlo method.

ye but to do that... limiting data to players which got at least (lets say) 10FS would be good idea 

also devs won't help with providing proper data to analyze :(

q7KQhQq.jpg

0

Share this post


Link to post
Share on other sites
35 minutes ago, MightyHoot said:

ye but to do that... limiting data to players which got at least (lets say) 10FS would be good idea 

also devs won't help with providing proper data to analyze :(

q7KQhQq.jpg

Just get something like sikuli for text from imagile recognition and extract as much data as tolerance allows.

 

0

Share this post


Link to post
Share on other sites

 

1 hour ago, MightyHoot said:

ye but to do that... limiting data to players which got at least (lets say) 10FS would be good idea 

also devs won't help with providing proper data to analyze :(

Good information. We will have to calculate the level of significance for to identify the "waste region" and the "acceptance region" for to reduce the error.
At the end of everything, It would be nice to answer the question: "Is He cheating?" Or "With which probability is He cheating?"

0

Share this post


Link to post
Share on other sites
22 hours ago, xKostyan said:

Just get something like sikuli for text from imagile recognition and extract as much data as tolerance allows.

 

That's too slow. Also what tolerance?

21 hours ago, ITalianBadBoy said:

 

Good information. We will have to calculate the level of significance for to identify the "waste region" and the "acceptance region" for to reduce the error.
At the end of everything, It would be nice to answer the question: "Is He cheating?" Or "With which probability is He cheating?"

Yeah, that's why I wanted ~10k players with FS above 10. Well, for two reasons:

  • 10FS is high T2/low T3
  • there are ~100k players with >10FS and 10% base would be quite good for creating some ratings

 

 

btw @ITalianBadBoy u wanted this:

Spoiler

29215 out of 29219 which took 0:00:00.340232
29216 out of 29219 which took 0:00:00.397271
29217 out of 29219 which took 0:00:00.342242
29218 out of 29219 which took 0:00:00.342242

Got all data in 0:31:10.488062
Players which didn't change nick: 27306

Done in 0:31:12.069177

 

raw.rar

0

Share this post


Link to post
Share on other sites

Thanks for the file but I can't download it :\

Error code: 2C171/1

"The page you are trying to access is not available for your account. "

0

Share this post


Link to post
Share on other sites
4 minutes ago, ITalianBadBoy said:

Thanks for the file but I can't download it :\

Error code: 2C171/1

"The page you are trying to access is not available for your account. "

Blame forums :D You can also try in my thread.

0

Share this post


Link to post
Share on other sites
57 minutes ago, MightyHoot said:

That's too slow. Also what tolerance?

Let's say it would take 5 seconds to process a page of 12 players from PvP rating and get to a start of next 12

Overnight, 8 hours -> 28800 seconds -> 5760s page processed -> 69120 entries pulled from PvP rating

Depending on an image resolution and desired processing speed lets build in 20-25% failure rate -> you are looking at ~50 000 names pulled overnight from pvp rating

0

Share this post


Link to post
Share on other sites
6 minutes ago, xKostyan said:

Let's say it would take 5 seconds to process a page of 12 players from PvP rating and get to a start of next 12

Overnight, 8 hours -> 28800 seconds -> 5760s page processed -> 69120 entries pulled from PvP rating

Depending on an image resolution and desired processing speed lets build in 20-25% failure rate -> you are looking at ~50 000 names pulled overnight from pvp rating

yea... look what happens when u refresh it few too many times:

fZWL4CQ.jpg

0

Share this post


Link to post
Share on other sites

well it may not even be 1 per 5 - really depends how many records returned per request and your processing speeds, but ehn I personally never had an issue for being kicked out from the game for scrolling ratings :)

0

Share this post


Link to post
Share on other sites
1 hour ago, xKostyan said:

refresh once in 5 seconds is not to many :D

ye, comparing to 16requests/sec while getting player data

53 minutes ago, xKostyan said:

well it may not even be 1 per 5 - really depends how many records returned per request and your processing speeds, but ehn I personally never had an issue for being kicked out from the game for scrolling ratings :)

I just unlocked scroll on mouse and kept scrolling and then got kicked out of server :( 

0

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.