NullpoMino Options

[Rosti_LFC]

This isn't TGM where we have to use IRC or a forum to communicate. The avg players on Blockbox in the early days of HD and before HD is what nullpo could achieve right now and that wasn't 15 years ago that was a little over two years ago. We don't need the distraction of some HD chatroom.

You missed my point. Entirely. People are on the chatroom to chat. If they want to chat, then removing the chatroom is not suddenly going to make them just hang around in the Nullpomino lobby, because they want to chat, not play Tetris. And sure, people can chat in the Nullpo lobby, but why would they do that when there are better alternatives? Very few people ever used the Blockbox lobby to just hang out and chat in - if people were idling it was mostly because it was easy to just leave the browser tab open as it was a browser based game, which is a luxury Nullpomino doesn't currently have.

Rosti however doesn't ever play multiplayer and could care less.

For one, I do play multiplayer, and for two, if you don't think I care then you're clearly either retarded or don't bother viewing any posts on this forum which aren't direct replies to your own. I've hosted leagues and tournaments, I basically designed the whole Blockbox multiplayer system (though I'd never claim to have done more work than Deniax) and put a sh**load of time into beta-testing it. The main reason I play TGM is because the community there is so much stronger and enjoyable, and because the multiplayer community is fragmented, largely uncommitted, and apparently more receptive to asshats such as yourself. Plus when I want to play TGM, it's always there and can always provide me with a decent game. I don't have to go around asking people to play it with me to get good games.

If I don't spend all day on Nullpomino then I have my own reasons for that, but I laugh in your face if you think it's because I don't care.

[xlro]

I implemented time statistics for every player on nullpo.it.cx: http://nullpo.it.cx/stats/
Now we can actually see each week / month,

Who is the most active on NullpoMino?

... might it be Corrosive afterall?

The stats are tracked seperately for actually playing Multiplayer (more than 2 people seated in the room), Singleplayer, and as a bonus category: just loyal idling / spectating

Let the competition begin!

[xlro]

The first snapshot of the NullpoMino online statistics for the last 2 weeks is in:

Multiplayer Ranking

1. dslite [15.0 h]
2. CorrosiveX [14.9 h]
3. Mr. Salamander [12.2 h]
4. qmk [12.2 h]
5. Mike [8.5 h]
6. a [8.3 h]
7. Profane [7.6 h]
8. Caithness [5.8 h]
9. Paul676 [4.9 h]
10. mame [4.3 h]

In total 269+ unqiue players spent 332.6 hours batteling on multiplayer modes!

Singleplayer Ranking

1. DarthDuck [13.7 h]
2. CorrosiveX [11.6 h]
3. Minase [10.2 h]

In total 93+ unique players spent 108.7 hours setting new records on singleplayer modes.

Idler Ranking

1. CorrosiveX [36.0 h]
2. qmk [7.9 h]
3. dslite [7.6 h]

In total 329+ unique players spent 209.1 hours hanging around the lobby and populating the server while not actively playing.

Congratulations to dslite for being the most active player on MP (with a very close lead of a couple minutes) and CorrosiveX for being the overall most active player on the server in the last two weeks.

A thanks to all active players for supporting NullpoMino!

Visit the Nullpo server statistics page for more details and live-updated ranking tables + graphs.

[xlro]

as a reply to the problems some people have with freezing / lag on netplay:

I'd like to investigate at some point if this is really the connection of some people to the server (i.e. the particular routing to the datacenter), that is messed up, or if it might be actually a problem with the netcode of the nullpo server / client itself.
I can say that any time there are big freezes for people (as you can see on rostis stream for example, no data arriving for some seconds), the bw of the server is nowhere near maxing out, neither is the load at all. This same machine runs other game servers btw, for alot more demanding games, like FPS games, where I never saw complaints. And as you know for these kind of games with precise aiming, any form of lag is a total game killer.
This last aspect makes me think there is some problem with the netcode itself, where some players are just being stalled - especially, when people say in the chat the opponents field just froze for them, and I on the other hand as a spectator in the same room see the other party just playing fine (e.g. the bot on the server), but _their_ field actually froze for me, i.e. not only do they not get updates of the game state, but their packets are not arriving at the server either! so the route between them and the server seems to be broken (the tcp connection / socket stalled?), while the server itself just works fine for others at the same time. Then again, I never saw these kind of freezes myself on my home (or work) connection, which makes an error in the netcode alot less likely.
Let me know if someone experiencing these freezes wants to investigate some more, involving other network analysis tools, so prefarably someone with a bit of pc/network knowhow.

[Alexsweden]

Cant help you with the issues but I just wanted to say that I love the statistics thing. It feels like people are more active now, me included, becouse of it

[xlro]

just fyi: the netplay server has some troubles atm, mainly due to excessive RAM usage - it might be restarting couple times

activity stats are reset (I got a backup of the old table), 1vs1 bot is removed for now, because of memory leak

[meow]

I ran a more thorough traceroute. There seems to be a lot of packet loss towards the end. Not sure if it helps.

On another note, it takes a long time for the rule data to be sent, whereas it's almost instantaneous on Blink and Hebo's server. Could this be due to the packet loss?

CODE

|------------------------------------------------------------------------------------------|
|                                      WinMTR statistics                                   |
|                       Host              -   %  | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
| ...
|                         203.208.131.245 -    0 |  108 |  108 |  171 |  199 |  350 |  195 |
|                          203.208.145.90 -    0 |  108 |  108 |  167 |  196 |  345 |  193 |
|   po9-0.palcr1.PaloAlto.opentransit.net -    0 |  108 |  108 |  170 |  197 |  351 |  186 |
|tengige0-1-0-8.paltr1.PaloAlto.opentransit.net -    0 |  108 |  108 |  178 |  204 |  360 |  199 |
|tengige0-12-0-0.nyktr1.NewYork.opentransit.net -    0 |  108 |  108 |  252 |  282 |  437 |  278 |
|pos0-1-5-0.madtr1.Madrid.opentransit.net -    0 |  108 |  108 |  351 |  378 |  526 |  369 |
|gigabitethernet3-1-0.madcr3.Madrid.opentransit.net -    0 |  108 |  108 |  342 |  372 |  519 |  367 |
|                          mad-1-6k.es.eu -    9 |   83 |   76 |  339 |  372 |  669 |  350 |
|              ldn-5-f10.routers.chtix.eu -    9 |   84 |   77 |  338 |  370 |  523 |  377 |
|                          th2-1-6k.fr.eu -    4 |   97 |   94 |  336 |  371 |  545 |  484 |
|                          th1-1-6k.fr.eu -   31 |   49 |   34 |  338 |  371 |  521 |  353 |
|                         rbx-g1-a9.fr.eu -    5 |   95 |   91 |  337 |  365 |  515 |  355 |
|                          vss-3-6k.fr.eu -   15 |   69 |   59 |  338 |  385 |  721 |  341 |
|                                xl.ro.lt -    1 |  104 |  103 |  337 |  365 |  520 |  355 |
|________________________________________________|______|______|______|______|__
____|______|
   WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider

[xlro]

Thanks for the traceroute meow, I will contact you later with more things we can maybe try.
Indeed the high packet loss after the node in Madrid is probably a good indicator for everybody that gets routed over this path. still a very long route, goes from west coast to NY then to madrid - hm, really looks ISP dependent. (will check with other ppl from US)
The traffic in outbound direction seems to be rather high on that link atm (at least on 2 of 4), should definitely not affect the connection as it does tho, since its a handful of small game-state packets.

also see here: http://weathermap.ovh.net/

edit: Additionally on the topic of long connection time (rule data), I'm not sure if its relatd with duration of netplay server uptime, but I strongly think it is (also as Aga has sugested already couple times). We definitely need to do some debugging together later, including nullpo server restarts.

ah I just realized, meow you are located in Australia iirc? (I thought US somehow before)
the first ip of your traceroute is marked as from Singapore (Asia Pacific Network Information Centre)

btw, example route a friend of mine just got from US (near NY):

CODE

...
4  tenge-0-4-1-0-nyquny91-rtr001.nyc.rr.com (24.164.161.18)  25.119 ms  28.002 ms
    tenge-0-6-0-10-nyquny91-rtr001.nyc.rr.com (24.164.160.21)  20.955 ms
5  ae6-nyquny91-rtr002.nyc.rr.com (24.29.148.254)  49.050 ms  28.347 ms  26.808 ms
6  107.14.19.22 (107.14.19.22)  19.340 ms  18.737 ms  14.801 ms
7  107.14.19.147 (107.14.19.147)  75.715 ms  18.128 ms  14.898 ms
8  vlan583.icore1.nto-newyork.as6453.net (209.58.26.109)  19.921 ms
    vlan576.icore1.nto-newyork.as6453.net (209.58.26.33)  18.904 ms *
9  195.219.243.21 (195.219.243.21)  92.681 ms  92.485 ms  92.592 ms
10  if-2-2.tcore1.l78-london.as6453.net (80.231.131.2)  86.401 ms  87.682 ms  96.998 ms
11  vlan704.icore1.ldn-london.as6453.net (80.231.130.10)  102.863 ms  117.670 ms *
12  ldn-5-6k.uk.eu (94.23.122.74)  102.443 ms  96.373 ms
    ldn-5-6k.uk.eu (94.23.122.121)  105.260 ms
13  rbx-g2-a9.fr.eu (91.121.131.177)  97.200 ms  96.614 ms  93.762 ms
14  vss-3-6k.fr.eu (213.251.130.77)  94.053 ms  94.262 ms  179.635 ms
15  xl.ro.lt (87.98.244.142)  102.384 ms  96.253 ms  96.087 ms

no packet loss, decent ping too - routed over London.

[Zircean]

Sometime soon I'll be contacting Myndzi and I'll see if we can set up the actual specifications for the protocol. In the meantime, though, the work that he's done with compression and such means that we are going to be able to send game data over UDP so we won't have to worry about TCP blocking and such. This will probably alleviate many of the issues people have.

[xlro]

I made some modifications to the NetServer to potentially reduce the lag/blocking problem that some players with not-so-perfect connections experience.

I also made a client patch that should improve network performance.
Especially players with these network lag issues (that are NOT related to rendering / low fps), please test this patch (for 7.5.0 release only):
http://nullpo.it.cx/files/lag_patch/NullpoMino.jar
(replace the NullpoMino.jar in your NullpoMino directory, make a backup copy of the original one first if you like)

...and report back, if it improved the lag. Thanks!