"Demystifying Lag" not accurate

Discussion for GU League Players
smoooth
Private First Class
Private First Class
Posts: 81
Joined: Sun Aug 05, 2007 6:13 pm

"Demystifying Lag" not accurate

Post by smoooth » Thu Jul 22, 2010 3:01 pm

For those of you that don't know, Moonpix and others have recently released a small document titled "Demystifying Lag" which attempts to explain lag to the average bzflag player. Although this is a great effort it lacks many points that need to be made:

#1 There ARE many aspects of lag and jitter that give a player an advantage. What is overlooked here is that lag is more than just a single number. When you hit the lagstats button it's just a snapshot of lag at that moment. What lagstats does not capture is increasing and decreasing lag and increasing and decreasing jitter over time. This seems to be the #1 problem with lag and jitter. People with consistent lag can be visually adjusted for by shooting early, much like Moonpix indicates, but folks with varying lag and jitter can absolutely NOT BE.

Jitter is the absolute worst aspect of lag. There is actually no justification for a player to have variable jitter or even jitter over 3ms. 90% of all jitter is created by the operating system or a computer that is poorly operated. What else causes jitter? Wireless. You hear over and over that bzflag was created for a local network and not to be player over the internet. It was not created to be played over the typical wireless connection either. Wireless connections are even less reliable than an "internet" connection because with wireless you have sporatic connection and many more dropped packets. Once you understand how wireless works you will see its not an acceptable way to play a game that was intended for a local area network. If you are serious about playing and you want to know how good you really are, you need a wired connection. Naturally, bullets will not hit you as much if you have wireless connection or one with constant jitter, it's a fact. In addition the opponent is at a disadvantage because your tank location is inaccurate, you get pauses on movements, blinking tank movements, and unpredictable hitzones. The problem with the game is that there is no positive re-enforcement to have a good connection. The better the connection one has the easier target you are. That is why some people choose to not make an effort to have a solid connection.

#2 The other aspect of lag that is overlooked is that when people are laggy or jittery it opens up the opportunity to cheat without being detected. When a player has 100-200ms lag it becomes very easy for players to hit F5 or have other cheats that add delay to prevent a tank from dying. The reason it's easier to get away with is because it's a lot harder to guage where the tank actually is, which means it can be manipulated easier without detection. Often times I will become used to the lag of a certain player and then magically on one shot, the lag is off. Was this the result of a lag spike? or did they "do" something to manipulate their lag? It's very hard to tell.

There is much more to this conversation but this should clear up some of the misleading information in the "demystifying lag" document.

FLUCTUATING Lag and Jitter is the real problem and is completely different than just "Lag." Often times I think people mean "Fluctuating Lag" when they say "Lag"

smoooth

User avatar
Cobra_Fast
Dev Monkey
Dev Monkey
Posts: 321
Joined: Sat Oct 07, 2006 11:51 am
Location: Germany
Contact:

Re: "Demystifying Lag" not accurate

Post by Cobra_Fast » Thu Jul 22, 2010 3:11 pm

smoooth wrote:FLUCTUATING Lag and Jitter
Fluctuating Lag equals Jitter
as jitter represents the ammount of time lag changes from package to package (correct me if i'm wrong, but please provide evidence).
Wikipedia wrote:In the context of computer networks, the term jitter is often used as a measure of the variability over time of the packet latency across a network. A network with constant latency has no variation (or jitter).
(http://en.wikipedia.org/wiki/Jitter#Pac ... r_networks)
BZWiki wrote:Jitter is a term used to mean the variation in lag over time.
(http://my.bzflag.org/w/Jitter)

smoooth
Private First Class
Private First Class
Posts: 81
Joined: Sun Aug 05, 2007 6:13 pm

Re: "Demystifying Lag" not accurate

Post by smoooth » Thu Jul 22, 2010 3:20 pm

No you are correct they are essentially measuring the same thing -- but the point is your jitter can be steady or even IT can fluctuate. It's possible to have a steady jitter which is bad, and then its also possible to have a fluctuating jitter, which is an essentially fluctuating lag latency.

User avatar
Cobra_Fast
Dev Monkey
Dev Monkey
Posts: 321
Joined: Sat Oct 07, 2006 11:51 am
Location: Germany
Contact:

Re: "Demystifying Lag" not accurate

Post by Cobra_Fast » Thu Jul 22, 2010 3:28 pm

So fluctuating jitter would mean that your lag is fluctuating in exceedingly large ammounts.

User avatar
mrapple
Sergeant Major
Sergeant Major
Posts: 460
Joined: Wed May 27, 2009 11:59 am
Location: Unknown
Contact:

Re: "Demystifying Lag" not accurate

Post by mrapple » Thu Jul 22, 2010 4:20 pm

Not necessarily.

I've had times were a certain application on my computer causes a jitter increase, but my lag stays the same.

Like smooth said, jitter does depend on the OS and the software your running
Image

User avatar
slime
Private First Class
Private First Class
Posts: 188
Joined: Fri Dec 30, 2005 3:08 am
Location: Omaha , Nebraska
Contact:

Re: "Demystifying Lag" not accurate

Post by slime » Thu Jul 22, 2010 8:35 pm

Perhaps I didn't make this clear enough for some people.

If you have any questions or comments relating to the tutorial, please private message taz on the official BZFlag forums.

smoooth
Private First Class
Private First Class
Posts: 81
Joined: Sun Aug 05, 2007 6:13 pm

Re: "Demystifying Lag" not accurate

Post by smoooth » Fri Jul 23, 2010 12:20 am

I'm making comments here, I don't like all this secret stuff. That's why message boards are here,right?

Just today I was shooting with morfeuz on quols server. His connection was from Poland to canada and mine was from north Carolina, US to Canada. Morfeuz had 140ms lag with 1ms jitter, constant. I had 44ms with 1ms jitter, constant. The lag was almost unnoticeable. This is a good example that even going across an ocean it's possible to maintain a good steady connection, and I would be willing to venture that a common path is taken when others connect across the atlantic. Folks having unsteady and worse connections than those described above should try to troubleshoot them so they can have a better connection. What I believe many will find is that their own computer is generating the lag and jitter. Try stopping all background programs before you play and even though mumble and other speaking applications are legal they definitely shouldn't be. These cause lots of additional fluctuating network traffic which may lead to a jittering connection. Just a few ideas form those looking to increase the quality of their connection.

ts
Dev Monkey
Dev Monkey
Posts: 970
Joined: Fri Jan 14, 2005 6:26 pm

Re: "Demystifying Lag" not accurate

Post by ts » Fri Jul 23, 2010 10:29 am

smoooth wrote:No you are correct they are essentially measuring the same thing -- but the point is your jitter can be steady or even IT can fluctuate. It's possible to have a steady jitter which is bad, and then its also possible to have a fluctuating jitter, which is an essentially fluctuating lag latency.
No, that would be fluctuating jitter. Whether the jitter is steady or not or if you see the jitter on the field at all depends on its reason.
smoooth wrote:90% of all jitter is created by the operating system or a computer that is poorly operated.
That is an interesting thesis but it is flawed. The OS itself hardly should cause jitter. Drivers for graphics cards do rather. Also candidates are integrated graphics cards (hope for 3.0, if it ever comes out it will likely kill the integrated graphics cards to some degree) because the have slower access to the RAM and the RAM is shared with the rest of the system and also this RAM is much slower than what modern cards have onboard. In addition to that these chipsets may not do all instructions in hardware and are generally very slow. Other candidates are cable quality, router and the internet connection itself (ISP). Of course this list is not complete, either.

smoooth wrote:I'm making comments here, I don't like all this secret stuff. That's why message boards are here,right?
I think so.
smoooth wrote:Just today I was shooting with morfeuz on quols server. His connection was from Poland to canada and mine was from north Carolina, US to Canada. Morfeuz had 140ms lag with 1ms jitter, constant. I had 44ms with 1ms jitter, constant.
This info is likely to be flawed. BZFlag is not a network quality detection tool.
smoooth wrote:The lag was almost unnoticeable.
The opposite is likely to be true, once one do not realise the jitter the lag becomes more obvious.
smoooth wrote:This is a good example that even going across an ocean it's possible to maintain a good steady connection, and I would be willing to venture that a common path is taken when others connect across the atlantic.
The pond is not the issue. The regarding ISP are more the first kilometres to the ISP fiber optical network and the routing.
smoooth wrote:Try stopping all background programs before you play and even though mumble and other speaking applications are legal they definitely shouldn't be.
On my box there are at least 59 processes running all the time. If I stop them my OS would stop working properly. What would help is reducing the CPU usage. As a CPU is connected to the rest by FSB/HT/QPI the entire CPU load for the CPU BZFlag runs on should be low, so if stuff is running, it should run on a different CPU if possible. However the question is what the difference would be. On a 8 core machine, 7 already heavily used cores would very likely make no measurable difference.

Mumble itself uses few bandwidth, so it would at max make 1ms more jitter and certainly not increase jitter from 1-2ms to 6-8ms and the jitter is visual (which several players have done according to my logs).
GU league: http://www.guleague.org/
An introduction to TCP: http://www.lafkon.net/tc/

User avatar
Cobra_Fast
Dev Monkey
Dev Monkey
Posts: 321
Joined: Sat Oct 07, 2006 11:51 am
Location: Germany
Contact:

Re: "Demystifying Lag" not accurate

Post by Cobra_Fast » Fri Jul 23, 2010 2:07 pm

ts wrote:
smooth wrote:Try stopping all background programs before you play and even though mumble and other speaking applications are legal they definitely shouldn't be.
On my box there are at least 59 processes running all the time. If I stop them my OS would stop working properly. What would help is reducing the CPU usage. As a CPU is connected to the rest by FSB/HT/QPI the entire CPU load for the CPU BZFlag runs on should be low, so if stuff is running, it should run on a different CPU if possible. However the question is what the difference would be. On a 8 core machine, 7 already heavily used cores would very likely make no measurable difference.
We (WTWRP) are playing mostly while talking in our skype conference what doesnt have a noticeable influence on our lag/jitter at all. You would only have to close other internet programs if you either use a rotten old machine or communicate to the internet via dialup or some sort of drumming...

http://my.bzflag.org/w/Server_Bandwidth - this might give you an idea of what data volumes are being sent during gameplay.
ts wrote:
smoooth wrote:No you are correct they are essentially measuring the same thing -- but the point is your jitter can be steady or even IT can fluctuate. It's possible to have a steady jitter which is bad, and then its also possible to have a fluctuating jitter, which is an essentially fluctuating lag latency.
No, that would be fluctuating jitter. Whether the jitter is steady or not or if you see the jitter on the field at all depends on its reason.
Wut?
ts wrote:
smoooth wrote:90% of all jitter is created by the operating system or a computer that is poorly operated.
That is an interesting thesis but it is flawed. The OS itself hardly should cause jitter. Drivers for graphics cards do rather. Also candidates are integrated graphics cards (hope for 3.0, if it ever comes out it will likely kill the integrated graphics cards to some degree) because the have slower access to the RAM and the RAM is shared with the rest of the system and also this RAM is much slower than what modern cards have onboard. In addition to that these chipsets may not do all instructions in hardware and are generally very slow. Other candidates are cable quality, router and the internet connection itself (ISP). Of course this list is not complete, either.
As BZFlag 2.0.x is singlethreaded id assume that only high FPS fluctuating in lower levels (like between 1 and 20) would cause noticeable jitter. This could easily be triggered by changeing the fpslimit variable by shortcuts...

User avatar
blast
General
General
Posts: 4744
Joined: Fri Mar 21, 2003 3:49 pm
Location: playing.cxx
Contact:

Re: "Demystifying Lag" not accurate

Post by blast » Fri Jul 23, 2010 2:18 pm

ts, you forget that many people are on old hardware, so many don't even have a dual-core system... I myself have a single core Pentium M laptop that I play on from time to time, as well as a P4 system with hyperthreading. (As well, I have a Core 2 Duo E8400 system with a nV 9800GTX+ [and yes, it can run Crysis] that I sometimes use)
"In addition to knowing the secrets of the Universe, I can assure you that I am also quite potty trained." -Koenma (Yu Yu Hakusho)

Image

ts
Dev Monkey
Dev Monkey
Posts: 970
Joined: Fri Jan 14, 2005 6:26 pm

Re: "Demystifying Lag" not accurate

Post by ts » Fri Jul 23, 2010 5:23 pm

blast you forget that newer hardware is not equal to better. More cores are often counter-productive.
GU league: http://www.guleague.org/
An introduction to TCP: http://www.lafkon.net/tc/

User avatar
blast
General
General
Posts: 4744
Joined: Fri Mar 21, 2003 3:49 pm
Location: playing.cxx
Contact:

Re: "Demystifying Lag" not accurate

Post by blast » Fri Jul 23, 2010 5:53 pm

ts wrote:blast you forget that newer hardware is not equal to better. More cores are often counter-productive.
[citation needed]
"In addition to knowing the secrets of the Universe, I can assure you that I am also quite potty trained." -Koenma (Yu Yu Hakusho)

Image

ts
Dev Monkey
Dev Monkey
Posts: 970
Joined: Fri Jan 14, 2005 6:26 pm

Re: "Demystifying Lag" not accurate

Post by ts » Fri Jul 23, 2010 6:16 pm

Well, would you say a 1 year old workstation is worse than a brand new netbook? Or from laptop with a "real graphics card" to netbook? Or from a laptop with "real graphics card" to a laptop with a chipset graphic (most likely Intel)? There is definitely a switch in progress.

Regarding CPUs the multi-core architectures are worse than multiple CPU architectures regarding performance. You only have 1 uncore part on each CPU. The more cores you have on a CPU the lower are the clock speeds of each core thus single-threaded applications will benefit a lot more from less cores because of the higher clock speed.
GU league: http://www.guleague.org/
An introduction to TCP: http://www.lafkon.net/tc/

User avatar
blast
General
General
Posts: 4744
Joined: Fri Mar 21, 2003 3:49 pm
Location: playing.cxx
Contact:

Re: "Demystifying Lag" not accurate

Post by blast » Fri Jul 23, 2010 6:29 pm

ts wrote:Well, would you say a 1 year old workstation is worse than a brand new netbook? Or from laptop with a "real graphics card" to netbook? Or from a laptop with "real graphics card" to a laptop with a chipset graphic (most likely Intel)? There is definitely a switch in progress.
So you switch from talking about an 8-core CPU to a crappy netbook CPU with 2 cores to suit your argument? Nice. The atom is only one example of mult-core CPUs, so definitely not "most".
ts wrote:Regarding CPUs the multi-core architectures are worse than multiple CPU architectures regarding performance. You only have 1 uncore part on each CPU. The more cores you have on a CPU the lower are the clock speeds of each core thus single-threaded applications will benefit a lot more from less cores because of the higher clock speed.
Clock speed isn't the only thing that matters. I'm confident that my 3GHz Core 2 Duo with one core behind it's back would outperform a 3GHz (or faster) Pentium 4. Heck, my 1.6Ghz Pentium M can outperform a 2.4GHz or so Pentium 4.
"In addition to knowing the secrets of the Universe, I can assure you that I am also quite potty trained." -Koenma (Yu Yu Hakusho)

Image

ts
Dev Monkey
Dev Monkey
Posts: 970
Joined: Fri Jan 14, 2005 6:26 pm

Re: "Demystifying Lag" not accurate

Post by ts » Fri Jul 23, 2010 7:47 pm

blast wrote:So you switch from talking about an 8-core CPU to a crappy netbook CPU with 2 cores to suit your argument? Nice. The atom is only one example of mult-core CPUs, so definitely not "most".
The extrem switch from a workstation to a netbook was the most obvious example I was able to think of, to demonstrate it depends on what people buy, not on the date they buy something.
blast wrote:Clock speed isn't the only thing that matters. I'm confident that my 3GHz Core 2 Duo with one core behind it's back would outperform a 3GHz (or faster) Pentium 4. Heck, my 1.6Ghz Pentium M can outperform a 2.4GHz or so Pentium 4.
I nowhere wrote clock speed would be the only thing that matters. However if you stay within one architecture (e.g. Nehalem) the clock speed gives a strong hint.
If one buys more, slower cores than fewer or equal, faster cores in comparison to a previous architecture then BZFlag would run slower (provided there is not much background load while playing).
Cobra_Fast wrote:We (WTWRP) are playing mostly while talking in our skype conference what doesnt have a noticeable influence on our lag/jitter at all. You would only have to close other internet programs if you either use a rotten old machine or communicate to the internet via dialup or some sort of drumming...
Skype is not in our whitelist, please don't use it. :p
GU league: http://www.guleague.org/
An introduction to TCP: http://www.lafkon.net/tc/

User avatar
Cobra_Fast
Dev Monkey
Dev Monkey
Posts: 321
Joined: Sat Oct 07, 2006 11:51 am
Location: Germany
Contact:

Re: "Demystifying Lag" not accurate

Post by Cobra_Fast » Fri Jul 23, 2010 8:26 pm

ts wrote:
Cobra_Fast wrote:We (WTWRP) are playing mostly while talking in our skype conference what doesnt have a noticeable influence on our lag/jitter at all. You would only have to close other internet programs if you either use a rotten old machine or communicate to the internet via dialup or some sort of drumming...
Skype is not in our whitelist, please don't use it. :p
In our experience teamspeak has definetely more effect on your lag/jitter than Skype and Skype got much better audio quality while using very few bandwidth. So we cannot understand the whitelist at that point and therefore we're not following it in this particular point.

smoooth
Private First Class
Private First Class
Posts: 81
Joined: Sun Aug 05, 2007 6:13 pm

Re: "Demystifying Lag" not accurate

Post by smoooth » Fri Jul 23, 2010 10:58 pm

smoooth wrote:90% of all jitter is created by the operating system or a computer that is poorly operated.
That is an interesting thesis but it is flawed. The OS itself hardly should cause jitter. Drivers for graphics cards do rather. Also candidates are integrated graphics cards (hope for 3.0, if it ever comes out it will likely kill the integrated graphics cards to some degree) because the have slower access to the RAM and the RAM is shared with the rest of the system and also this RAM is much slower than what modern cards have onboard. In addition to that these chipsets may not do all instructions in hardware and are generally very slow. Other candidates are cable quality, router and the internet connection itself (ISP). Of course this list is not complete, either.
I don't disagree with what you are saying but you're just being more specific than i am. I am generically saying "OS" which typically includes your graphic drivers and i said "poorly operated computer" which means a variety of things, bad old technology, not enough memory, too many taxing programs, etc etc.
smoooth wrote:I'm making comments here, I don't like all this secret stuff. That's why message boards are here,right?
I think so.
smoooth wrote:Just today I was shooting with morfeuz on quols server. His connection was from Poland to canada and mine was from north Carolina, US to Canada. Morfeuz had 140ms lag with 1ms jitter, constant. I had 44ms with 1ms jitter, constant.
This info is likely to be flawed. BZFlag is not a network quality detection tool.
This is not flawed. Either BZFlags lagstats are a representative of real connections or they are not. I will say that lagstats seems to be a good indicator of the actual connection; Is it a super-accurate network quality tookit?of course not, it's a game.
smoooth wrote:The lag was almost unnoticeable.
The opposite is likely to be true, once one do not realise the jitter the lag becomes more obvious.
I think we're having a language miscommunication here.
smoooth wrote:This is a good example that even going across an ocean it's possible to maintain a good steady connection, and I would be willing to venture that a common path is taken when others connect across the atlantic.
The pond is not the issue. The regarding ISP are more the first kilometres to the ISP fiber optical network and the routing.
I agree with you, ISP, router, and also wireless is way more of the issue however all of that is included in "across the atlantic". Clearly being across the atlantic (a long distance) does increase lag with absolute certainty.
smoooth wrote:Try stopping all background programs before you play and even though mumble and other speaking applications are legal they definitely shouldn't be.
On my box there are at least 59 processes running all the time. If I stop them my OS would stop working properly. What would help is reducing the CPU usage. As a CPU is connected to the rest by FSB/HT/QPI the entire CPU load for the CPU BZFlag runs on should be low, so if stuff is running, it should run on a different CPU if possible. However the question is what the difference would be. On a 8 core machine, 7 already heavily used cores would very likely make no measurable difference.

Mumble itself uses few bandwidth, so it would at max make 1ms more jitter and certainly not increase jitter from 1-2ms to 6-8ms and the jitter is visual (which several players have done according to my logs).
[/quote]

Absolutely some people have high end machines and are able to run operating systems with processes and experience very little jitter or system degredation. HOWEVER, some people have slower/older computers, poorly maintained operating systems, rouge processes, spyware, torrents, connections with poor upload speeds, and other things that interfere with the network, the processing of video, and other system information. Absolutely, without a doubt, an older computer can get slowed down by other running processes, and to say otherwise has absolutely no merit. Also how much bandwidth is used depends on how much bandwidth you actually have. It is possible to overload a connection with small upload capacity with just a little bit of extra data.

User avatar
slime
Private First Class
Private First Class
Posts: 188
Joined: Fri Dec 30, 2005 3:08 am
Location: Omaha , Nebraska
Contact:

Re: "Demystifying Lag" not accurate

Post by slime » Sat Jul 24, 2010 5:34 am

Smoooth, it is clear you either did not read the tutorial very closely, or not at all.

What I will ask of you is to read my post thoroughly before you quote or comment on it, so we can minimize misunderstandings.

Your entire #2 point:
smoooth wrote: #2 The other aspect of lag that is overlooked is that when people are laggy or jittery it opens up the opportunity to cheat without being detected. When a player has 100-200ms lag it becomes very easy for players to hit F5 or have other cheats that add delay to prevent a tank from dying. The reason it's easier to get away with is because it's a lot harder to guage where the tank actually is, which means it can be manipulated easier without detection. Often times I will become used to the lag of a certain player and then magically on one shot, the lag is off. Was this the result of a lag spike? or did they "do" something to manipulate their lag? It's very hard to tell.
Is nearly exactly what section 5.4 is in the tutorial:
5.4) Solid signs of cheating
Now that you already have an usable method to "see through lag", you will be able to discard the most common fake signs of cheating caused often by lag. Although, what are the possible solid signs that will still remain noticeable even when compensation methods are used?

# Tanks going through the compensated (lagged) bullet line often.
# Tanks getting high jitter conveniently in important moments, as in flag carring, etc (F5 cheat).
# Tanks fitting into tight doubles where a tank should not be able to fit.
# Tanks going through bullets without any delayed dodges even after the expected hit point.

Just keep in mind, though, that isolated "positives" may occur caused by isolated lag issues, spikes, packet loss, etc. In example (despite rare), a bullet may go through if the exact packet where the bullet was got lost in the way, etc. So, these methods will only point to a solid sign if:

# Interpreted by an experienced player that fully understand the methods.
# Along with some margin of mistake for exceptions (the ammount of incidents must be considerable).
# The lag is confirmed as constant and proportional to the compensation after every positive incident.
I am wondering how the above information is in any way misleading?

Now, I will admit that your point #1 is a good point that should be included into the tutorial. The reason it is not in there now is because the tutorial deals primarily with distance lag. Unsteady lag/jitter is only brushed upon. If you wish to explain it more, yet again, I ask that you send moonpix a private message with the information, not to keep secrets, but because there is a much higher chance of him seeing that than every post in this thread, and he can decide whether or not to add it to the tutorial. Keeping this thread is fine, but use it to help fix the tutorial, rather than to discredit it by generalizing that the whole tutorial is "not accurate".

smoooth
Private First Class
Private First Class
Posts: 81
Joined: Sun Aug 05, 2007 6:13 pm

Re: "Demystifying Lag" not accurate

Post by smoooth » Sat Jul 24, 2010 12:04 pm

Slime, excellent points. My thread title was poorly chosen and wasn't intended to discredit the whole document. I'm glad someone wrote it and wish admins would further organize the league with documents such as these to include more specifics, like this document does. I'd like to see one with the rules written better and make new rules about game play, such as the use of pauses. Lot's of work left to be done to make this league even better. The overall document is great -- my only concern was the few comments I made which you pointed out meshes up completely with what the document already says. One general underlying bad thing is in some places the tone suggests it's OK to have to lag and jitter and that downplay the effects it has on game play. I think lag and jitter is a real problem in bzflag and we should try to work and figure out how to reduce it, as much as possible, for as many players as possible. I think that is what you guys are trying to do with the tutorial but in some sections it seems a little vague and maybe wants to justify lag and jitter as an "OK" part of game play. Naturally some folks have bad connections and are far away and will never be able to "correct" them so to speak, and that's no problem. We should try to encourage folks not to use wireless and emphasize the effects of wireless on game play. Understand that wireless is popular way to connect these days but has real issues when playing BZFlag -- we should heavily discourage this but also be understanding that some folks will have to use wireless because of social-economic conditions.

Great job writing the document -- I wanted to create discussion, which is good, and always gives everyone an opportunity to provide input and perhaps make small suggestions here and there.

ts
Dev Monkey
Dev Monkey
Posts: 970
Joined: Fri Jan 14, 2005 6:26 pm

Re: "Demystifying Lag" not accurate

Post by ts » Sat Jul 24, 2010 12:07 pm

smoooth wrote:I don't disagree with what you are saying but you're just being more specific than i am. I am generically saying "OS" which typically includes your graphic drivers and i said "poorly operated computer" which means a variety of things, bad old technology, not enough memory, too many taxing programs, etc etc.
Drivers aren't exactly part of an OS and the latter is rather a pooly operating computer.
smoooth wrote:
smoooth wrote:Just today I was shooting with morfeuz on quols server. His connection was from Poland to canada and mine was from north Carolina, US to Canada. Morfeuz had 140ms lag with 1ms jitter, constant. I had 44ms with 1ms jitter, constant.
This info is likely to be flawed. BZFlag is not a network quality detection tool.
This is not flawed. Either BZFlags lagstats are a representative of real connections or they are not. I will say that lagstats seems to be a good indicator of the actual connection; Is it a super-accurate network quality tookit?of course not, it's a game. [/quote]Here is a new word for you: Heuristic. Every player from PL I watched had a worse connection that lagstats did indicate, reason unknown.
smoooth wrote:Also how much bandwidth is used depends on how much bandwidth you actually have. It is possible to overload a connection with small upload capacity with just a little bit of extra data.
Well, not overload (that would mean transfering more data than possible) but heavily degrading connection quality because bandwidth is saturated.
slime wrote:# Tanks going through the compensated (lagged) bullet line often.
Wrong, that is not necessarily a sign of cheating. You forgot the entire DR strategy.

Cobra_Fast wrote:In our experience teamspeak has definetely more effect on your lag/jitter than Skype and Skype got much better audio quality while using very few bandwidth
As skype is a peer to peer technology your mileage may vary, depending on network traffic conditions.

Cobra_Fast wrote:So we cannot understand the whitelist at that point and therefore we're not following it in this particular point.
I'd ban you for doing that and even doing without asking first. However as you hardly play in GU league that's likely not being worth the effort.
GU league: http://www.guleague.org/
An introduction to TCP: http://www.lafkon.net/tc/

smoooth
Private First Class
Private First Class
Posts: 81
Joined: Sun Aug 05, 2007 6:13 pm

Re: "Demystifying Lag" not accurate

Post by smoooth » Sat Jul 24, 2010 1:28 pm

TS, you are nitpicking semantics.

1. Hardware Drivers ARE part of every operating system. Some are included in the kernel and others are added in later. Love to see any operating system work without hardware drivers.
2. I don't disagree that every player may have a worse connection than is indicated in lagstats -- but that doesnt change the fact that lagstats is "representative" of the actual connection. Either lagstats is somewhat accurate or it is not. Chose which it is.
3. You're trying to provide your own definition of "overload" -- degrading quality because the bandwidth is saturated IS OVERLOADING. Once again this is semantics or just language barrier.

Clown.

User avatar
Cobra_Fast
Dev Monkey
Dev Monkey
Posts: 321
Joined: Sat Oct 07, 2006 11:51 am
Location: Germany
Contact:

Re: "Demystifying Lag" not accurate

Post by Cobra_Fast » Sat Jul 24, 2010 4:02 pm

ts wrote:
Cobra_Fast wrote:So we cannot understand the whitelist at that point and therefore we're not following it in this particular point.
I'd ban you for doing that and even doing without asking first. However as you hardly play in GU league that's likely not being worth the effort.
Go ahead, I don't want to play in a league which's dictating players on what software to use. Additionaly, mumble and teamspeak are only approved to be used, not the only to be granted to be used. Please read and understand your whitelist before accusing players.

So either the whitelist is lacking off clearness or you're gonna ban me for an ambiguous reason or you simply didn't quite understand the whitelist rules. Either way shows incompetence but in different ways.

ts
Dev Monkey
Dev Monkey
Posts: 970
Joined: Fri Jan 14, 2005 6:26 pm

Re: "Demystifying Lag" not accurate

Post by ts » Sat Jul 24, 2010 5:14 pm

Cobra_Fast wrote:Additionaly, mumble and teamspeak are only approved to be used, not the only to be granted to be used. Please read and understand your whitelist before accusing players.
This means that any other item might be considered a cheat and we will ban for it if it is used on GU League related public or match servers.
GU league: http://www.guleague.org/
An introduction to TCP: http://www.lafkon.net/tc/

User avatar
Cobra_Fast
Dev Monkey
Dev Monkey
Posts: 321
Joined: Sat Oct 07, 2006 11:51 am
Location: Germany
Contact:

Re: "Demystifying Lag" not accurate

Post by Cobra_Fast » Sat Jul 24, 2010 6:56 pm

You still haven't read it. It says:
It is explicitly allowed to use voice communication for playing.

ts
Dev Monkey
Dev Monkey
Posts: 970
Joined: Fri Jan 14, 2005 6:26 pm

Re: "Demystifying Lag" not accurate

Post by ts » Sun Jul 25, 2010 9:22 am

Cobra_Fast wrote:You still haven't read it. It says:
It is explicitly allowed to use voice communication for playing.
The voice communication is allowed with the mentioned tools and other tools are treated regaring my quote.

I've read it before it was even published.
GU league: http://www.guleague.org/
An introduction to TCP: http://www.lafkon.net/tc/

Post Reply