New (correctly) hitfixed matchservers
#31
(20 Dec 14, 01:48AM)ExodusS Wrote:
(20 Dec 14, 01:05AM)Vanquish Wrote: The fact is that many players simply prefer to play on servers without anticheat because they believe it drops less hits and that they perform better on them.

 Because, well you know, in the past, w00p private servers were helping players with Fra/w00p tags, and people believed it.

they still do.
Thanks given by:
#32
(20 Dec 14, 01:48AM)ExodusS Wrote: Placebo or reality?

I'd argue both.
The (possible) placebo is that people feel they hit better on servers without anticheat.
The reality is that they're performing better.

Even if the whole "hitting better" debate really is a placebo effect, you'll surely agree that such things impact one's performance. Do people really get greater awareness from texreduce, does modifying your hitsound help with knowing how much you've hit on your enemy, does playing with/without a hudgun help you to aim better?
If you don't feel comfortable with your setup it's always gonna be at the back of your mind and inhibit you from doing as well as you'd like to, especially if the thing you are being limited by is out of your control. I believe it's the same with "feeling" your hits in AC, from about 30 seconds of playing on a server I can instantly tell whether it's any good or not (although of course everyone's opinion on what quantifies a "good server" is different).
Thanks given by:
#33
(18 Dec 14, 10:48PM)Luc@s Wrote: the issue was fixed client side for 1.2 release.
(19 Dec 14, 01:20PM)Luc@s Wrote: the "old" patch fix is indeed useless

I had an opportunity to try Larry's patch on 5 different machines with different hardware configurations and data centres. The patch was applied to the 1.2.0.2 version of the server. The result is simple: in 4 cases the hit-drop counter showed a nonzero value. I had 0% of hit drops on the only server, the one with the best configuration (a 1 Gb/s port and an Intel Xeon E5-1660 v2 CPU). By the same token, the next configuration after the said one in a descending order was a 100 Mbit/s port with an Intel Core 2 Duo T7700 CPU. Maybe you indeed sorted the trouble out in your perfect world, but the reality gives an example of the contrary. The further word I hand to you.
Thanks given by:
#34
(20 Dec 14, 09:17AM)Sveark Wrote:
(18 Dec 14, 10:48PM)Luc@s Wrote: the issue was fixed client side for 1.2 release.
(19 Dec 14, 01:20PM)Luc@s Wrote: the "old" patch fix is indeed useless

I had an opportunity to try Larry's patch on 5 different machines with different hardware configurations and data centres. The patch was applied to the 1.2.0.2 version of the server. The result is simple: in 4 cases the hit-drop counter showed a nonzero value. I had 0% of hit drops on the only server, the one with the best configuration (a 1 Gb/s port and an Intel Xeon E5-1660 v2 CPU). By the same token, the next configuration after the said one in a descending order was a 100 Mbit/s port with an Intel Core 2 Duo T7700 CPU. Maybe you indeed sorted the trouble out in your perfect world, but the reality gives an example of the contrary. The further word I hand to you.

If you tried different servers with the same patch, it makes no sense. And if you are trying to show that a poor connection and hardware can affect the gameplay, well, o'rly ?

What i want to see is a comparison between the official binaries and the "patched" server (and by patch, you have to be clear : do you just mean the anticheat was removed ?)

As roflcopter i could not detect hitdrops (and i had a bad internet connection at the time i performed the tests).
My methodology was to compare the amount of hits detected on my side to the amount of hits reported by the server.
I tried it on different servers : some servers i did not know and some servers i knew to be running with the anticheat.
The only drops i had were because of my ping when i shot someone who was already dead according to the server.


You have to explain your methodology better.
Thanks given by:
#35
(20 Dec 14, 01:10AM)vonuno Wrote: - SSD and gigabit (=fast)
- Stockholm (a nice solid server for Europe)
- Having anticheat gone is itself an asset to some, and no detriment with private servers or good admins

SSD is useless while running assaultcube servers (too little filesystem IO)

However. I do not know exactly what configuration woop servers are running on. But i can tell you that :

- It is running on a dedicated server
- The CPU is a 4 core Intel Xeon E31220 @ 3.10GHz
- Ping from the US east cost can be lower then 100 ms
- The server provider network was improved even more last year making it excellent for game servers
- The guaranteed bandwidth is 100 mb, more than enough for a few AC servers (the true limit is the network and the connection to other isps...)
Thanks given by:
#36
(20 Dec 14, 01:10AM)vonunov Wrote: - SSD and gigabit (=fast)
- Stockholm (a nice solid server for Europe)
- Having anticheat gone is itself an asset to some, and no detriment with private servers or good admins

Any half-decent VPS server should be running on SSD/Gigabit these days, so no surprises there (example: Linode uses SSD, with 40Gbps (instead of 1Gbps) + Intel E5 processors (instead of E31220's)). The SSD makes little difference for AC.

Opponents with bad pings can be just as bad as yourself having a bad ping. Additionally, there's the factor of fairness.

So, I would suggest Stockholm isn't a great place for a server (in terms of latency), due to it's distance from submarine cables (this is probably why Lucas' ping was high): http://www.submarinecablemap.com/

I would suggest that UK/France/Germany/Belgium/Netherlands would be expected to be a more likely place to find a good European server, due to a combination of its proximity to international submarine cables + nation's wealth (poorer nations aren't likely to have great options). Note, I haven't tested pings, etc (useless to do so from NZ).

As such, I would expect Stockholm might be good for eastern europe - assuming you were only playing other european opponents. So, on this, it's a good addition to the pool of servers to help our eastern europe comrades :)

SPEEDTESTS
Your server runs on the Hurricane Electric network, which I rate very highly.
From as far west of USA (Fremont, LA) while staying on the Hurricane Electric network:
randum@gibbed:~$ ping se-rc.tk -c 10
PING se-rc.tk (31.220.5.44) 56(84) bytes of data.
64 bytes from hosted-by.hosthatch.com (31.220.5.44): icmp_req=1 ttl=56 time=163 ms
64 bytes from hosted-by.hosthatch.com (31.220.5.44): icmp_req=2 ttl=56 time=165 ms
64 bytes from hosted-by.hosthatch.com (31.220.5.44): icmp_req=3 ttl=56 time=164 ms
64 bytes from hosted-by.hosthatch.com (31.220.5.44): icmp_req=4 ttl=56 time=164 ms
64 bytes from hosted-by.hosthatch.com (31.220.5.44): icmp_req=5 ttl=56 time=164 ms
64 bytes from hosted-by.hosthatch.com (31.220.5.44): icmp_req=6 ttl=56 time=163 ms
64 bytes from hosted-by.hosthatch.com (31.220.5.44): icmp_req=7 ttl=56 time=163 ms
64 bytes from hosted-by.hosthatch.com (31.220.5.44): icmp_req=8 ttl=56 time=165 ms
64 bytes from hosted-by.hosthatch.com (31.220.5.44): icmp_req=9 ttl=56 time=163 ms
64 bytes from hosted-by.hosthatch.com (31.220.5.44): icmp_req=10 ttl=56 time=163 ms

--- se-rc.tk ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9014ms
rtt min/avg/max/mdev = 163.136/163.922/165.588/0.892 ms

By the way, ping for a comparable West USA (Fremont, LA) to London is:
randum@gibbed:~$ ping speedtest.london.linode.com -c 10
PING speedtest.london.linode.com (176.58.107.39) 56(84) bytes of data.
64 bytes from speedtest.london.linode.com (176.58.107.39): icmp_req=1 ttl=55 time=137 ms
64 bytes from speedtest.london.linode.com (176.58.107.39): icmp_req=2 ttl=55 time=137 ms
64 bytes from speedtest.london.linode.com (176.58.107.39): icmp_req=3 ttl=55 time=137 ms
64 bytes from speedtest.london.linode.com (176.58.107.39): icmp_req=4 ttl=55 time=137 ms
64 bytes from speedtest.london.linode.com (176.58.107.39): icmp_req=5 ttl=55 time=137 ms
64 bytes from speedtest.london.linode.com (176.58.107.39): icmp_req=6 ttl=55 time=137 ms
64 bytes from speedtest.london.linode.com (176.58.107.39): icmp_req=7 ttl=55 time=137 ms
64 bytes from speedtest.london.linode.com (176.58.107.39): icmp_req=8 ttl=55 time=137 ms
64 bytes from speedtest.london.linode.com (176.58.107.39): icmp_req=9 ttl=55 time=137 ms
64 bytes from speedtest.london.linode.com (176.58.107.39): icmp_req=10 ttl=55 time=137 ms

--- speedtest.london.linode.com ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9014ms
rtt min/avg/max/mdev = 137.188/137.276/137.383/0.337 ms
Thanks given by:
#37
(20 Dec 14, 11:30AM)Luc@s Wrote: - It is running on a dedicated server

like every other match server in existence?
[EDIT: still half asleep sry, I thought you simply meant home hosted or sth. see my comment below]


(20 Dec 14, 11:30AM)Luc@s Wrote: - The CPU is a 4 core Intel Xeon E31220 @ 3.10GHz

Mine's using an Intel Xeon E5 @ 3.50 GHz. :-)


(20 Dec 14, 11:30AM)Luc@s Wrote: - Ping from the US east cost can be lower then 100 ms

I used to get 70 to the old new york bob servers in 1.1, I don't think it's a great achievement, more to do with the server's location and the player's own connection. I've seen french people get 60ms and above to aura, and I've never once seen any east coast american player get 100ms or below to aura (xemi + waffles get about 125ms, the same as they do to se-rc which is further away). Also Sveark, for example, is getting 128ms to se-rc from (literally) the other side of the world. I don't doubt that it's possible for someone to get 100ms to aura from the usa tho.
EDIT: I would also argue that ping is somewhat redundant in this case. I get 14ms to aura and 30ish to se-rc/se-aox and I prefer the two swedish servers by a ridiculous margin. I get about 80 to xrd and 140 to gibbed.me, and I prefer gibbed.me. I don't feel like the main factor behind choosing a server should be a 20ms difference in latency, when hit registration is a far bigger plague to competitive games.


(20 Dec 14, 11:30AM)Luc@s Wrote: - The guaranteed bandwidth is 100 mb, more than enough for a few AC servers (the true limit is the network and the connection to other isps...)

Mine's running on 10Gbps, with 1Gbps bandwidth guaranteed. :-)

I agree with you about the "true network limit".


(20 Dec 14, 11:30AM)RandumKiwi Wrote: As such, I would expect Stockholm might be good for eastern europe - assuming you were only playing other european opponents. So, on this, it's a good addition to the pool of servers to help our eastern europe comrades :)

Thanks for your comments and speedtests, found it really interesting to read :)
That was the main motivation behind choosing Stockholm as a location, specifically to cater for players within the Northern and eastern european regions, although I'm getting about 30ms from the UK too which I deem perfectly playable. If people would rather play on servers in their own country then I'm not gonna force them to do the opposite, merely providing an alternative if people want it. :P
Thanks given by:
#38
(20 Dec 14, 11:47AM)Vanquish Wrote:
(20 Dec 14, 11:30AM)Luc@s Wrote: - It is running on a dedicated server

like every other match server in existence?

no, many - if not most - of them run on a VPS...
Thanks given by:
#39
(20 Dec 14, 11:49AM)Luc@s Wrote: no, many - if not most - of them run on a VPS...

so an SSD and 100Mbit connection (as RK said, both pretty standard nowadays for any good hosting package) is overkill for hosting AC servers, but having your own dedicated server isn't?

hm ok.
Thanks given by:
#40
(20 Dec 14, 11:52AM)Vanquish Wrote:
(20 Dec 14, 11:49AM)Luc@s Wrote: no, many - if not most - of them run on a VPS...

so an SSD and 100Mbit connection (as RK said, both pretty standard nowadays for any good hosting package) is overkill for hosting AC servers, but having your own dedicated server isn't?

hm ok.
its a matter of reliability, and you know it...
sharing resources is worse than not sharing for this reason.
SSD is not common for dedicated servers yet (well it depends on your use) but in most cases its not worth it.
we are not talking about minecraft servers !
Thanks given by:
#41
(20 Dec 14, 12:11PM)Luc@s Wrote: its a matter of reliability, and you know it...
sharing resources is worse than not sharing for this reason.

I've never played on a server which has been bad because it was hosted on a VPS. I've even played on some home-hosted servers which are pretty good (ac-fraglog is homehosted on a 100mb/s connection or something and it's perfectly playable).

I can see this going on forever, so I'm just gonna summarise everything:

I know my way around AC servers and how to host them properly, but it should probably be openly stated that I'm not a server hoster first and foremost, so I don't really care about all this technical wankery or specifications bragging (even though I did it in an earlier post). Having been someone who's played AssaultCube in as close to the definition of the word "competitively" you can get in this game for about two years now, I simply wanted to create a server which creates a good playing experience. This means two things: stability and good hit registration. I've heard no complaints about either from anybody (except your initial comment where you stated you got 90ms, although as you stated in a later post the ISP's connection is probably more influential there).
I wouldn't use a dedicated server if you gave me one for free, because it's simply not necessary for the scale of what I'm doing. And I could have easily chosen another location, but decided not to.

Also, I'm not forcing anybody to play on these servers. While I would do so by choice since I believe they are amongst the best in terms of quality in the game (the only other two servers where I hit as well are gibbed.me + se.aox), I appreciate that everyone's gonna have different server preferences, which is just completely normal. It would be simply retarded for either of us to suggest that any server is the best for everybody, I'm not claiming that se-rc is and aura's certainly not either.

I feel like further discussion on this is gonna be pointless, people have had enough time to read/reply to this topic if they want to do so and if someone wants an admin pw/whitelist/etc they know who they need to contact, so could this thread be closed before it turns into a bunch of useless posts? dankeschon mods
Thanks given by:
#42

You are making things unreasonably complex. The methodology I used consists in counting shots that are ignored by the server. As you might know, there are 2 places in the source code that may cause hit drops in theory:
1) serverevents.h, 40-th line:
1.2.0.2 source code Wrote:   if(!gs.isalive(gamemillis) ||
      e.gun<GUN_KNIFE || e.gun>=NUMGUNS ||
      wait<gs.gunwait[e.gun] ||
      gs.mag[e.gun]<=0)
       return;
2) serverevents.h, 175-th line:
1.2.0.2 source code Wrote:if(e.shot.millis<c->lastevent) { clearevent©; continue; }

One shot carries maximum of one hit (except for the shotgun), so dropping whole shot causes dropping potential hit. In my tests, I only counted the shots dropped by the first block of code. Also, I have to say I had a 6-th test machine where I had zero possible unhits. The quality of that VDS' connection was explicit shit with ping getting higher and lower in a very short period of time as well as with some packet drops on a 2-3 Mbit/s uplink load. However, on that server, I had the lowest (throughout the test) ping of 50 msecs. I don't think the ACAC is capable of affecting one's hit statistics, but I think that a relatively high ping and going beyond the gunwait deadline may somehow correlate.

The test clearly showed unhit existance, mainly for mid-ping (100 msecs) cases. Having this in mind, I come to believe Larry's hitfix patch is still viable, even for the 1.2.0.2 version.
Thanks given by:
#43
(20 Dec 14, 12:27PM)Vanquish Wrote:
(20 Dec 14, 12:11PM)Luc@s Wrote: its a matter of reliability, and you know it...
sharing resources is worse than not sharing for this reason.

I've never played on a server which has been bad because it was hosted on a VPS. I've even played on some home-hosted servers which are pretty good (ac-fraglog is homehosted on a 100mb/s connection or something and it's perfectly playable).

I can see this going on forever, so I'm just gonna summarise everything:

I know my way around AC servers and how to host them properly, but it should probably be openly stated that I'm not a server hoster first and foremost, so I don't really care about all this technical wankery or specifications bragging (even though I did it in an earlier post). Having been someone who's played AssaultCube in as close to the definition of the word "competitively" you can get in this game for about two years now, I simply wanted to create a server which creates a good playing experience. This means two things: stability and good hit registration. I've heard no complaints about either from anybody (except your initial comment where you stated you got 90ms, although as you stated in a later post the ISP's connection is probably more influential there).
I wouldn't use a dedicated server if you gave me one for free, because it's simply not necessary for the scale of what I'm doing. And I could have easily chosen another location, but decided not to.

Also, I'm not forcing anybody to play on these servers. While I would do so by choice since I believe they are amongst the best in terms of quality in the game (the only other two servers where I hit as well are gibbed.me + se.aox), I appreciate that everyone's gonna have different server preferences, which is just completely normal. It would be simply retarded for either of us to suggest that any server is the best for everybody, I'm not claiming that se-rc is and aura's certainly not either.

I feel like further discussion on this is gonna be pointless, people have had enough time to read/reply to this topic if they want to do so and if someone wants an admin pw/whitelist/etc they know who they need to contact, so could this thread be closed before it turns into a bunch of useless posts? dankeschon mods

everybody is grateful for the servers but you are suggesting there is a problem with hit registration and that your server is flawless regarding this supposed matter while you are not giving any kind of evidence. And that is a problem for me.

By the way, i moved to paris and i have a new, much better ISP. I have a ping of 3 ms on woop servers and 90 on yours with this new connection. 90 is not bad but deceiving for a server you first advertised as the best available at the moment.

More servers is always great... Just.. dont make up reasons to claim they are better.
Thanks given by:
#44
he never said they are flawless?

he only said 'i believe they amongst the best' -> opinion, not a fact. you make it seem like he stated a fact. and that is a problem for me.
Thanks given by:
#45
I'm claiming that they are better than match servers which have the anticheat for the reasons stated above, and I stick by that. I'm not claiming, have never claimed, and will never claim that everyone agrees or should agree with me (although so far the significant majority of reactions to the servers have been positive).

And I've never said they're flawless, I don't believe such a thing is possible in this game (the closest thing was probably uk bob which gave me about 6 ping and that's hax), although I do believe the hit registration is flawless.
Thanks given by:
#46
luc@s invented "hitloss" after he was mad from losing cm, he secretly added it to the sauce bottle with other shady devs namely ones of Semitic backround.... if you ever lost an aim duel it was probably because of this guy......I recently found this tucked away
if I.P= not baguettebro
then cry= yes
if 2nd tier french baddie= true
then luc@s= you
I studyied camel case 1 year at the university of homeopathy TX so naturally I would be the first to pick up on such things
Thanks given by:
#47
(20 Dec 14, 02:05PM)Marti Wrote: he never said they are flawless?

he only said 'i believe they  amongst the best' -> opinion, not a fact. you make it seem like he stated a fact. and that is a problem for me.

as Vanquish himself said, flawless regarding that specific "issue" (which is still not proven)
read http://forum.cubers.net/thread-7959-post...#pid158851 and also read my post again.
(suggest = insinuate)
he is encouraging a belief based on 0 fact, biasing the "debate"
Thanks given by:
#48
I'm not gonna attempt to argue about it, people were wrong about the hitdrop issue in 1.1 and they are wrong again now. I don't have proof and you can use this to try and sweep the issue under the rug, which is fine because if I were in your shoes I'd do the same.
But if I shoot at a stationary playermodel with 100hp and it takes 8 bullets (1 bullet misses) to kill him, that's all the proof I need. We could open a whole new thread about hitreg if you want, I'd be interested to hear what other people think.
Thanks given by:
#49
(20 Dec 14, 03:05PM)Vanquish Wrote: We could open a whole new thread about hitreg if you want, I'd be interested to hear what other people think.
Thanks given by:
#50
(20 Dec 14, 03:05PM)Vanquish Wrote: I'm not gonna attempt to argue about it, people were wrong about the hitdrop issue in 1.1 and they are wrong again now. I don't have proof and you can use this to try and sweep the issue under the rug, which is fine because if I were in your shoes I'd do the same.
But if I shoot at a stationary playermodel with 100hp and it takes 8 bullets (1 bullet misses) to kill him, that's all the proof I need. We could open a whole new thread about hitreg if you want, I'd be interested to hear what other people think.

makes no sense. Remind me who made the first stats about the 1.1 hit bug and actually proved its existence ? Oh yeah it was me.

Also, i made stats myself on this version and found no issue.
So i feel offended by pour post because its just a lie.

What happened to you (8bullets to kill) was probably that :
because your ping is > 0, you hit the player while its already dead from the server point of view. And it does not affect the gameplay AT ALL.

I don't really care about what other people think. I want facts. Simple as that.
Thanks given by:
#51
So my weak hitreg on pussy irl is because of the anticheat?
fuaaaaaaaark
luc@s for head dev
Thanks given by:
#52
@my previous point: the hitdrop occurs in the middle of the rest of the bullets, which all connect as hits perfectly. It's something I've seen many times on many other servers.

(20 Dec 14, 04:26PM)Luc@s Wrote: I don't really care about what other people think. I want facts. Simple as that.

As I said, I can't prove it so trying to do so would be useless on my end [EDIT: read sveark's posts, he may be on to something though]. If we do assume now that anticheat does in fact drop no hits and the whole thing is purely a placebo, then refer back to my previous post about placebos and how they affect your performance when you play games. I could list more examples (maxfps, sensitivityscale, smoothmove, smoothdist, crosshair, crosshair colour, playermodel colour, hitsound, resolution, aspect ratio, etc) where somebody may feel like one setting is perfect and helps them to uhm let's say kill people better, when in reality it's just a placebo and they find it easier to adjust their playstyle to certain settings.
The only difference between all of the aforementioned variables and the hit registration you get on certain servers is that you are in control of the former and not the latter. As I said, let's assume anticheat is a placebo: if myself and many others believe that they hit better on servers that do not have anticheat enabled, that will affect their performance in game. At a high level in any game (yes even AC), placebos and mindsets when you play are incredibly important. I can tell you this as someone who used to rage a lot when playing ingame (I'm human, I still think I rage sometimes although much less than for example, six months to a year ago, and I can clearly tell the difference in my performance between when I'm focused and when I'm not).

I've known people who play with no hitsound simply because they feel like they're having to adjust their aim according to how many hitsounds they're hearing when they fire over a playermodel and would rather just focus their crosshair on to an enemy without having the auditory distraction. This is similar to the way some people feel like they hit better with /hitsound 1 and others with /hitsound 2 - some people (most that i've asked) feel like they hit better on servers with no anticheat.

You said you don't care about what other people think and that's great, but when people are playing a game what's going through their head is very important. If they're focused on what they believe to be bad hitreg (or any external factor that they're not able to control, but in this case we're explicitly talking about hitreg) they will be playing worse since they're not focusing on the game itself and playing the game as well as they can in each situation. That leads back to my point about wanting to create servers that provide a good quality environment in which people can play games. I literally couldn't give a rat's arse about statistics, if people like the servers and agree that the hitreg is good then that's all I ever intended to accomplish by setting them up.
Thanks given by:
#53
Well, next time, just run basic servers and lie about the anti-cheat, same result.
Thanks given by:
#54
(20 Dec 14, 04:59PM)ExodusS Wrote: Well, next time, just run basic servers and lie about the anti-cheat, same result.

I don't see a point in running a (non-public) server with anticheat, not gonna explain why for the nine-hundredth time. I simply felt the need to defend some ridiculous accusations. This incident aside, I do have a great deal of respect for both Lucas personally and for what he's done for the game, and it's never been my prerogative to denounce the usefulness of his contributions to the anticheat or attempt to oust the it's inclusion from AC on the whole. I was simply trying to state the reasoning for why I chose to run match servers without anticheat (I still consider the reasons to be perfectly valid), and people started getting their titties in a twist.

But yeah, it's never been my intention to start, prolong or instantiate any type of conflict or perpetuate negativity. I wanted this thread closed about ten posts ago, and still do.
People have the information they need if they want to contact someone with control over the servers.
Thanks given by:
#55
not trying to add to the technical discussion, but i've definitely noticed more hit drops in 1.2. it also seems to get much worse as your ping gets higher. its definitely there, as i've dropped shots with the carbine and the sniper rifle.
Thanks given by:
#56
(20 Dec 14, 05:27PM)Undead Wrote: not trying to add to the technical discussion, but i've definitely noticed more hit drops in 1.2. it also seems to get much worse as your ping gets higher. its definitely there, as i've dropped shots with the carbine and the sniper rifle.

I hear ya on that one. I've also had the same thing happen to others while shooting at myself (Yes, I've literally seen a bullet hit but not register). I think I may have a demo of this somewhere.
Thanks given by:
#57
(20 Dec 14, 11:47AM)Vanquish Wrote: Although I'm getting about 30ms from the UK too which I deem perfectly playable.

Your typical user will need to negotiate his own routes to the Hurricane Electric network first, which is probably why Lucas' ping is 90ms.

The following pings are from routers on the Hurricane Electric network (the same routers that provide your backbone and thus, have no other routers to negotiate).

Paris 1: core1.par1.he.net to 31.220.5.44 is 36ms
Paris 2: core1.par2.he.net to 31.220.5.44 is 38ms
London: core1.lon1.he.net to 31.220.5.44 is 48ms

Map of the Hurricane Electric network: http://he.net/HurricaneElectricNetworkMap.pdf

So based on the above, I find it hard to believe that you would ping better to London, than Paris. Keep in mind, that "about 30ms" you stated for UK, is probably before your data leaves the Hurricane Electric network.

You can test yourself via http://lg.he.net/

Again, a good addition for our Eastern Europe comrades. If the players on your servers are central europe though, they would likely be better off with a Frankfurt server (look at how many routes go through that city from that map earlier). Paris and Amsterdam would be good too of course.
Thanks given by:
#58
(20 Dec 14, 11:30AM)Luc@s Wrote:
(20 Dec 14, 01:10AM)vonuno Wrote: - SSD and gigabit (=fast)
- Stockholm (a nice solid server for Europe)
- Having anticheat gone is itself an asset to some, and no detriment with private servers or good admins

SSD is useless while running assaultcube servers (too little filesystem IO)

However. I do not know exactly what configuration woop servers are running on. But i can tell you that :

- It is running on a dedicated server
- The CPU is a 4 core Intel Xeon E31220 @ 3.10GHz
- Ping from the US east cost can be lower then 100 ms
- The server provider network was improved even more last year making it excellent for game servers
- The guaranteed bandwidth is 100 mb, more than enough for a few AC servers (the true limit is the network and the connection to other isps...)


CPU is also useless, and network bandwidth is also useless.
You don't need a lot of ram and surely not an SSD.

The only thing you want is a decent network interface card. Game servers send a lot of small packets at high rates. This doesn't consume a lot of bandwidth, but can be very hard for low spec NICs. When hosting on a dedicated hardware, you don't have to share this NIC, so yes, if you really want to have the best gaming server, you can place it on a dedicated server. :)

I have already tested some servers on Rasberry Pi, Intel Atom (and other low profile hardware). Network I/O requests per second is what matters :)

@RandumKiwi:
Nice post about geolocation of servers. Netherlands are indeed one of the best location to host.
Thanks given by: