Status Updates posted by Android25
Well... I finally did it.
When you're just trying to hide your WN8
I don't really play low tiers at all. My average tier is over 8 at this point... but I bought the black market mini maus in the heat of the moment and this tank is absolutely the most OP tank I've ever played...
In 5 games, I've averaged over 5 kills per game, have 3 top guns, 2 of which were 8 kill games, and 2 orlik medals... I feel dirty even owning this tank.
Finally got the 260 today... Glad grinding for that is over.
After almost 24k games I believe last night I had my first game where I was called a "total dumbass" and "pussy" by a player while pulling off over 1400 base xp and winning
Is fun game mode.
On 8/28/2019 at 1:30 PM, Deus__Ex__Machina said:
i also enjoyed it, one of the better special event mode they have done in recent times.
miles better than home front, and not as boring as FL.
only gripe id have is teaming in solo mode, and the retarded module dmg (fires and ammo racks are way to frequent due to them lowering module HP)
Yeah teaming in solo sucks ass. But man it felt good when I beat both their IS3 asses with my t54-55whatever tank.
Anyone who loves Myst/Obduction like puzzle games or is just really interested in more immersive VR world's. Cyan is running another kickstarter for their new VR/flat screen game Firmament. Campaign has another 9 and a half days and could use some funding help. https://www.kickstarter.com/projects/1252280491/firmament
When your team can't get a push together but you've learned all the objective sniping spots
Bought 75 boxes and got all the tanks (less the skorp but I already had it and the OP french arty :c), all the camo's, 40k gold, 3.5 million credits, and 74 days of premium...
I should really just try to get rid of my account... I never play this game anymore
I finally did it!
- Show previous comments 1 more
Blackmagic Disk Speed Test v8.0.1 (it's hard to find for windows)
It's actually a utility for determining your computer's ability to record and edit various sizes and qualities of desktop video. It tests speeds in file streams from 1 to 5 GB depending on settings, so its good at telling you your raw speed, but doesn't do any real good for testing everyday usage as it doesn't have any queue depth testing.
"When it takes too long to assign a player to a battle, the matchmaker may add extra Battle Modes to those you’ve marked. If the wait takes more than 50 seconds, you’ll enter the first available match in any mode."
Fucking hell WG... just let me play regular and grand modes in peace.
That ping though :3
New Soviet tier 10's and their comparisons
(clicking full size and zooming in will help with clarity)
Don't know if anybody cares about getting Dishonored 2 for PS4, but saw newegg has a combo deal for the same price as a game that includes a "free" $30 headset
Could someone who can read this tell me what exactly it is? I bought it at a poorly advertised estate sale for $15 because it looked cool (very old, very large book).
And if anyone can read it, what this says:
- Show previous comments 9 more
A very rough translation
Wednesday the 15th of January 1749 - Mark Gottlieb Otto(w?????) was born in/on "(??????) ...ütz"
You have to keep in mind that nowdays the german language is different. Many words are written differently, e.g. the german word for door - Tür - was written - Thür - until around 1900. Also many words arn't used anymore, different or used in an other form.
Add to that different font, especially when it comes to handwriting. Learning the old printed font is relatively easy for a native speaker, picked that one up myself as a teenager because I read some old books.
In the case of the name "Ottow" I'm not sure, if that translates to the modern name Otto, if the "w" was pronounced at all back than in spoken language, or if that might indicate eastern european ancestry.
Wikipedia suggests that it was a german family name, though I personally never stumbled over it until now.
Even English has changed quite a bit over the past 20 years
Yes, we now have beautiful words and phrases like:
Not to mention older phrases that make no sense to the non-native speaker, like "To ride shotgun", "Down the tubes", and my personal favorite, "I could care less"
Win with 0 seconds left on the clock
When people say "But you don't need 32GB of ram"
Could someone on EU do me a favor and in-game look at the service record of Amphibiios
It's the only account I've found while doing API stuff that has battles but no tanks. It was breaking my program until I added an exception for it.
A screenshot of the main service record and then one of their tank list (even if it's empty) would be most helpful.
ASIA user data from API finished populating (that was fast):
ID Space: 2000000000 - 2016656820 and 3000000000 - 3001186525 (17,843,347 ID's)
4,148,955 Confirmed User ID's
356,809 Active Users (60 days)
1,611,785 Users with no battles on WoT (Either signed up and never played, or played other games WoWP, WoWS, WoTB, etc.)
That makes 2,537,170 WoT player accounts on ASIA with 14.06% activity.
(NA, EU, and RU data in my last status updates)
On Thursday I will be writing the code to pull tank data based off the existing user ID's I discovered with all these API runs. Then it should take about 15 days to pull data on all ~9 million active accounts. Then I'll run the program in 15 more days for 30 day data on all players
Another fun fact. Just storing all the ID's that are in use (so I don't have to iterate over all the id's with no account data ever again) takes up 692 MB. This is just ID's, no account data/tank data/etc.
- Show previous comments 1 more
Well I noticed the differences between account ID's across servers. NA always starting with 10 and always being the same length, same with EU except always starting with 5. RU is a bit different as the ID's don't have any leading numbers standardized across RU accounts. So I assumed RU started at 0, EU started at 500000000, and NA started at 1000000000. Asia server was weird because I checked a few accounts there (I just mouse over the hall of fame on each server's website to see the URL which includes the player ID), and noticed account numbers starting with 20 but also 30. So I assumed they must have merged another server into SEA at some point, which they did, Korea.
I literally brute force the API with ID's from there.
The API allows you to ask for player personal data in batches of 100. So I just built a string of 100 user id's starting at x and ending at x+99 and dropped that into the URL asking for last_battle_time field.
The rest of the program parses and saves the API response. In each reply you get 100 player id's back. If the last_battle_time is null, there is no account at that id (skip it, don't save anything), if the last battle time is 0, there is an account but it has never played WoT, and if it's a standard unix timestamp, you can look at the current time and compare it to see if the player has played in the last x days.
Because I'm just doing this as an initial setup for now, and this isn't code that will be used in the main program because I have to GET this data first, I just had it save id's that have account data to 2 files, one for accounts played in the last 60 days and one for accounts that hadn't. If they had never played a game, I had it add an asterisk at the end of the ID just so I could keep track.
At first I wrote it so if the program had 200 consecutive null accounts, I assumed that was the end of the id space and moved to the next server. That ended up being a bad assumption as even on NA there are empty spots in the id space spanning for over 4000 accounts. I moved it up to 10k but even then ran into some issues where I had to sit there and help the program with the RU server, where there is a gap between 900,000 and 1,100,000 where almost every id in that 200,000 id gap is blank.
I ran all this through eclipse because it's not any code that will need to be run again. That made it a lot easier to catch errors and fix them.
Now that I have a list of player id's that actually exist as real accounts, I'll never have to iterate over all those blank ID's again, which will save around 66% of the time the first run of checking last played date took.
thanks for the detailed information, very interesting.
according to their doc... "The request per second limit is set for each application_id."
And if I remember right, you need an application id per region anyway (or maybe you used to), my app was only ever setup for NA.
So I think you can run your data grabbing in parallel on all territories.
You're allowed 20 requests per second per server application. But based on some of the other rules I can only infer they consider it cheating if you use multiple applications for the same thing and will act accordingly if you try to circumvent the limit. They've changed it so that application ID's work on all the servers.
I could still run the data in parallel and plan to on multiple threads because parsing and saving even the ID data at 100/request took about 200-300 milliseconds.
RU user data from API finished populating:
ID Space: 0 - 80152563 (80,152,564 ID's)
34,747,338 Confirmed User ID's
6,228,921 Active Users (60 days)
6,913,398 Users with no battles on WoT (Either signed up and never played, or played other games WoWP, WoWS, WoTB, etc.)
That makes 27,833,940 WoT player accounts on RU with 24.83% activity.
(NA and EU data in my last status updates - AISA being populated now)
EU user data from API finished populating:
ID Space: 500000000 - 547340230 (47,340,231 ID's)
19,725,657 Confirmed User ID's
1,941,916 Active Users (60 days)
7,638,197 Users with no battles on WoT (Either signed up and never played, or played other games WoWP, WoWS, WoTB, etc.)
That makes 12,087,460 WoT player accounts on EU with 16.07% activity.
(NA data in my last status update)
RU, then Asia. Aisa is weird in that it has user ID Space 2000000000 - 200000000 + x (x < 1000000000) AND 3000000000 - 3000000000 + x
I don't keep up with the goings on of that server but I assume they migrated an entire server (Korea?) Into the Asia server and everybody's ID's just got moved over
Idk yet if I'll use RU or Asia for WNR calculations but it will be nice to have a list of valid user ID's in case I ever decide to.
It may be interesting to have different expected values for each server, even if those calculated scores were only for xvm integration.