Quantcast

Jump to content


Photo

[beta] Attic Restock Analyser


  • This topic is locked This topic is locked
83 replies to this topic

#51 Scot

Scot
  • ≡^ᴥ^≡

  • 3935 posts


Users Awards

Posted 06 December 2013 - 06:16 PM

Need more people to do this! Participation is of utmost importance!



#52 Kat

Kat
  • KatDog 5ever

  • 2098 posts


Users Awards

Posted 06 December 2013 - 06:26 PM

Yes, we do need more people to help. The more data we have, the better.



#53 L33T

L33T
  • 862 posts


Users Awards

Posted 06 December 2013 - 08:06 PM

 Would help if I had any attic shells that weren't frozen <_<



#54 Dan

Dan
  • Resident Know-It-All

  • 6382 posts


Users Awards

Posted 08 December 2013 - 01:17 PM

OK guys, v1.6 released!

 

http://speedy.sh/hcv9P/Neocodex.AtticAnalyser.v1.6.zip


#55 Chameleon

Chameleon
  • vampire or somesuch

  • 627 posts


Users Awards

Posted 08 December 2013 - 02:26 PM

Run it an hour and send you the log each time, or run it an hour several times before sending the log?



#56 L33T

L33T
  • 862 posts


Users Awards

Posted 08 December 2013 - 02:43 PM

Run it an hour and send you the log each time, or run it an hour several times before sending the log?

 

One hour at a time.



#57 NapisaurusRex

NapisaurusRex
  • 🍴Aioli-American🍴

  • 9425 posts


Users Awards

Posted 08 December 2013 - 02:47 PM

One hour at a time.

I think he only said an hour at a time because the logs were too big to be maintainable. With the most recent version, it separates the time information into it's own .txt. so it'd be ok to run for longer if you're comfortable with it.



#58 Chameleon

Chameleon
  • vampire or somesuch

  • 627 posts


Users Awards

Posted 08 December 2013 - 02:49 PM

I think he only said an hour at a time because the logs were too big to be maintainable. With the most recent version, it separates the time information into it's own .txt. so it'd be ok to run for longer if you're comfortable with it.

Ok, thank you.



#59 L33T

L33T
  • 862 posts


Users Awards

Posted 08 December 2013 - 02:50 PM

I think he only said an hour at a time because the logs were too big to be maintainable. With the most recent version, it separates the time information into it's own .txt. so it'd be ok to run for longer if you're comfortable with it.

 

That makes sense, yeah.

 

P.S If anyone wants to sell me an attic shell for pure.. :p



#60 Dan

Dan
  • Resident Know-It-All

  • 6382 posts


Users Awards

Posted 08 December 2013 - 03:00 PM

Run it an hour and send you the log each time, or run it an hour several times before sending the log?

I think he only said an hour at a time because the logs were too big to be maintainable. With the most recent version, it separates the time information into it's own .txt. so it'd be ok to run for longer if you're comfortable with it.

 

 

Spot on.  ;)



#61 L33T

L33T
  • 862 posts


Users Awards

Posted 08 December 2013 - 03:10 PM

Spot on.  ;)

 

How many more logs do you think you need? :p



#62 Dan

Dan
  • Resident Know-It-All

  • 6382 posts


Users Awards

Posted 08 December 2013 - 03:12 PM

How many more logs do you think you need? :p

 

I think we'd like to get to Tuesday PM / Wednesday AM before we call to stop the efforts :) 

 

We're sitting at about ~670 data points since last Monday.



#63 L33T

L33T
  • 862 posts


Users Awards

Posted 08 December 2013 - 03:14 PM

I think we'd like to get to Tuesday PM / Wednesday AM before we call to stop the efforts :)

 

We're sitting at about ~670 data points since last Monday.

 

Are you guys close to cracking the pattern?



#64 Dan

Dan
  • Resident Know-It-All

  • 6382 posts


Users Awards

Posted 08 December 2013 - 03:19 PM

Are you guys close to cracking the pattern?

 

I'd say we're most of the way there, but there's a certain aspect of the pattern that's causing us issues in reliable predictability without utilising more data points over a much, much longer period.

 

I think there will most likely only be a certain level of accuracy we can provide until the point where we have a considerable (28 days perhaps) amount more information regarding restock timings.



#65 L33T

L33T
  • 862 posts


Users Awards

Posted 08 December 2013 - 03:22 PM

I'd say we're most of the way there, but there's a certain aspect of the pattern that's causing us issues in reliable predictability without utilising more data points over a much, much longer period.

 

I think there will most likely only be a certain level of accuracy we can provide until the point where we have a considerable (28 days perhaps) amount more information regarding restock timings.

 

Damn, that sucks. Well patience will pay off, I suppose.



#66 L33T

L33T
  • 862 posts


Users Awards

Posted 14 December 2013 - 10:17 AM

Progress update?

 

If it matters, please merge this with the above post.


Edited by L33T, 14 December 2013 - 10:18 AM.


#67 rhexis07

rhexis07
  • 205 posts

Posted 16 January 2014 - 05:47 AM

There's quite an easy way to do this and get all the necessary data points without worrying about restock bans or getting your account frozen for running the program for long periods of time.

 

1. Log out of Neopets, go to the AAA link: http://www.neopets.c...en/garage.phtml (you could see the shop and items in it while logged out and you won't get restocked banned this way).

 

2. Create a program to record all of the restock times, items that restock, etc.

 

3. Get all the data points you need.

 

4. Analyze data points.

 

5. ????

 

6. Profit.



#68 Waser Lave

Waser Lave

  • 25516 posts


Users Awards

Posted 16 January 2014 - 06:17 AM

There's quite an easy way to do this and get all the necessary data points without worrying about restock bans or getting your account frozen for running the program for long periods of time.

 

1. Log out of Neopets, go to the AAA link: http://www.neopets.c...en/garage.phtml (you could see the shop and items in it while logged out and you won't get restocked banned this way).

 

2. Create a program to record all of the restock times, items that restock, etc.

 

3. Get all the data points you need.

 

4. Analyze data points.

 

5. ????

 

6. Profit.

 

You can still get refresh banned without being logged in.



#69 rhexis07

rhexis07
  • 205 posts

Posted 16 January 2014 - 07:13 AM

You can still get refresh banned without being logged in.

 

Whoa didn't know that, thanks for the info. As for the new restock times, it seems fairly straight forward. It seems to have the potential to restock every 7 minutes from the previous restock with a variation of 0-10 seconds. 

 

Example: Suppose it restocked at 5:03:02, next potential restock would be sometime between 5:10:02-5:10:12. If that restock is skipped, next potential restock would be at 5:17:02-5:17:12. Now suppose the restock occurs at 5:10:09, we continue the process to get the subsequent potential restock at 5:17:09-5:17:19 and if that restock is skipped, the following potential restock would be at 5:24:09-5:24:19.

 

However, I am not completely sure on the variation of the seconds, it might end up being like 3-15 seconds instead of 0-10, need to gather a bit more data to be completely accurate on it. 


Edited by rhexis07, 16 January 2014 - 07:15 AM.


#70 Dan

Dan
  • Resident Know-It-All

  • 6382 posts


Users Awards

Posted 16 January 2014 - 07:44 AM

Whoa didn't know that, thanks for the info. As for the new restock times, it seems fairly straight forward. It seems to have the potential to restock every 7 minutes from the previous restock with a variation of 0-10 seconds. 

 

Example: Suppose it restocked at 5:03:02, next potential restock would be sometime between 5:10:02-5:10:12. If that restock is skipped, next potential restock would be at 5:17:02-5:17:12. Now suppose the restock occurs at 5:10:09, we continue the process to get the subsequent potential restock at 5:17:09-5:17:19 and if that restock is skipped, the following potential restock would be at 5:24:09-5:24:19.

 

However, I am not completely sure on the variation of the seconds, it might end up being like 3-15 seconds instead of 0-10, need to gather a bit more data to be completely accurate on it. 

 

It's not as simple as that unfortunately. Try analysing the data over an entire day and look for patterns there.

 

I think I'll be releasing the data that I (with the help of many others) collected over the space of a week, to see if anyone could help figure it out.



#71 rhexis07

rhexis07
  • 205 posts

Posted 16 January 2014 - 07:51 AM

It's not as simple as that unfortunately. Try analysing the data over an entire day and look for patterns there.

 

I think I'll be releasing the data that I (with the help of many others) collected over the space of a week, to see if anyone could help figure it out.

 

I've only analyzed 12 hours of data and it seems to hold up pretty well so far. 24+++ hours would be ideal. I'll keep collecting and analyzing data and get back to you. If you could release your data, that would be lovely :D



#72 Shane

Shane
  • Sad sack.

  • 1982 posts


Users Awards

Posted 18 January 2014 - 01:52 AM

Hopefully something comes out of this more sometime.



#73 rhexis07

rhexis07
  • 205 posts

Posted 18 January 2014 - 02:06 PM

After analyzing 3 days worth of data, a restock potential every 7 mins with 0-60 seconds variation seems to be correct. Even the RS Chat boards on the Neoboards confirm this. I don't know why you got contradictory data, maybe your data wasn't collected accurately.


Edited by rhexis07, 18 January 2014 - 02:09 PM.


#74 Waser Lave

Waser Lave

  • 25516 posts


Users Awards

Posted 18 January 2014 - 02:17 PM

After analyzing 3 days worth of data, a restock potential every 7 mins with 0-60 seconds variation seems to be correct. Even the RS Chat boards on the Neoboards confirm this. I don't know why you got contradictory data, maybe your data wasn't collected accurately.

 

That's not quite the whole story.



#75 rhexis07

rhexis07
  • 205 posts

Posted 18 January 2014 - 02:20 PM

That's not quite the whole story.

 

Would you like to PM additional details? I don't know exactly where I am going wrong, pattern seems to fit so far...




1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users