james10

Revise the current “Mid Air Collision” methodology.

30 posts in this topic

Hello All.

I do understand the reasoning behind implementing a methodology to resolve “Mid Air Collisions”. On the whole I don’t have any issues. If there was no methodology and all aircraft involved were destroyed, that would expose flying to malicious players who just wish to cause anguish from their actions by just ramming enemy aircraft with no intention to actually engage. Although ramming is a valid method of attack used in war, BGE isn’t actually war so the strict adherence to reality can be safely ignored is some circumstances.

However, in the case of the “Head on attack”, BOTH aircraft have the opportunity to NOT participate. I feel if both pilots accept the “Head on attack” and not turn away and if a “Mid Air Collision” results from the attack, BOTH aircraft should have a 100% chance of destruction, if not catastrophic damage. The thought in the pilots mind should not be “well I have a chance of getting away with this unscathed, so I’ll try my luck". Game Pilots will try their luck. I.e. “No Fear of Death”.

I feel that any methodology that adds elements to support the Real Life “Fear of Death” in a game should be at least considered, if not implemented. Will this result in player exploitation, probably yes. Is the current methodology being exploited, probably yes. I see the following comment on the chat window reasonably often: “we collided I exploded and he just flew off undamaged”.

A “Head on attack” resulting in a “Mid Air Collision”, is a choice by both players, unlike a ram (accidental or deliberate) from any other direction. It should not be coin flip to determine the result. The result should be assured for ALL participants.

Cheers

James10

Edited by james10

Share this post


Link to post
Share on other sites

I don't think it's a coin-flip. It seems to me that it's possible to "game the game" here when it comes to ramming/colliding.

Meaning that if you, say, fly underneath an EA and clip it with your V-stab, chances are you'll blow up the EA while you get out undamaged. Which is of course BS: With a large enough impact speed, any part that your plane touches on the other one should take considerable damage in a mid-air collision.

I've also had instances where the EA was pretty obviously going for the ram. I blew up, he flew along merrily without any bigger problems.

All that said: Collisions have always been a tricky thing in the MMO-sim world. Warbirds for example was notorious for this and people would avoid getting anywhere near an EA when mid-air collisions were enabled in an arena (think it was more or less a "collision bubble" in that game).

But here's how it *can* work in a 2015/2016 game:

U20BMG-RGu0

Share this post


Link to post
Share on other sites

It is not a coin flip. Currently, if two players go straight into each other, they both blow up and die. That is because player X's client will see player Y's airplane blow up only a second or so after his own plane has. It is even likely you won't see the other person blow up since you are looking forward, but from third person perspective, head-on crashes resulting in two booms are common.

Feel free to test it with a friend flying for the opposite side. However, if one chooses to pull away at the last moment, there is a good chance that one will blow up and the other will only be damaged if his wing or propeller clips. Hitting wingtips in HO merges often spin an airplane out and hitting propeller will usually take out the propeller or the governor.

Share this post


Link to post
Share on other sites

Hello Tigger6, how are you today?

It is not a coin flip. Currently, if two players go straight into each other, they both blow up and die. That is because player X's client will see player Y's airplane blow up only a second or so after his own plane has. It is even likely you won't see the other person blow up since you are looking forward, but from third person perspective, head-on crashes resulting in two booms are common.

Feel free to test it with a friend flying for the opposite side. However, if one chooses to pull away at the last moment, there is a good chance that one will blow up and the other will only be damaged if his wing or propeller clips. Hitting wingtips in HO merges often spin an airplane out and hitting propeller will usually take out the propeller or the governor.

The actual decision process was describe to me as the software (client or server, not sure which) decides which aircraft has “Right of Way” and destroys the other.

Principally my point is: 2 aircraft each weighing over 3000kg a piece with a potential head-on closure rate of in excess of 1200km/h (e.g. Bf109 G-6: 3,400kg, 640km/h; Spitfire Mk VB: 3,039kg, 595km/h) collide mid air and it is ok that there is the possibility 1 of the 2 aircraft gets out of the “Mid Air Collision” with virtually no damage while the other is an expanding stream of smoke, fire and debris just because the surviving Pilot “chooses to pull away at the last moment”. I put it to you that had the surviving Pilot, chose to pull away earlier, not stay guns-on for the kill and pull away at the last moment, there would have probably been no “Mid Air Collision” as a result.

Both Pilots placed their aircraft in a situation that resulted in a head-on “Mid Air Collision”. I believe a more correct term is “Playing Chicken” and seeing who flinches first.

Anyhow thank you both for your response.

Cheers

James10

Share this post


Link to post
Share on other sites
Hello Tigger6, how are you today?

The actual decision process was describe to me as the software (client or server, not sure which) decides which aircraft has “Right of Way” and destroys the other.

Principally my point is: 2 aircraft each weighing over 3000kg a piece with a potential head-on closure rate of in excess of 1200km/h (e.g. Bf109 G-6: 3,400kg, 640km/h; Spitfire Mk VB: 3,039kg, 595km/h) collide mid air and it is ok that there is the possibility 1 of the 2 aircraft gets out of the “Mid Air Collision” with virtually no damage while the other is an expanding stream of smoke, fire and debris just because the surviving Pilot “chooses to pull away at the last moment”. I put it to you that had the surviving Pilot, chose to pull away earlier, not stay guns-on for the kill and pull away at the last moment, there would have probably been no “Mid Air Collision” as a result.

Both Pilots placed their aircraft in a situation that resulted in a head-on “Mid Air Collision”. I believe a more correct term is “Playing Chicken” and seeing who flinches first.

Anyhow thank you both for your response.

Cheers

James10

Well, thanks!

No, there is no such thing, whoever told you didn't understand it himself. Every airplane blows up if it suffers a "significant" collision, as evaluated on that airplane pilot's client. Look:

netlag.png

Share this post


Link to post
Share on other sites

Hello Tigger6, how has the day been so far?

Well, thanks!

No, there is no such thing, whoever told you didn't understand it himself. Every airplane blows up if it suffers a "significant" collision, as evaluated on that airplane pilot's client. Look:

...

I have seen this image somewhere before. My suspicion is this methodology was devised to overcome computer hardware and internet limitations/issues way back when, just to have an online flight sim. This is good as without it there would be none. Now 2016, computer hardware and even the internet itself has significantly improved. Maybe leverage the improvements to implement a more correct resolution.

Just a little question, in the graphic the surviving aircraft passes over the top of the other. I would presume passing under the other aircraft would have the same results?

It is still strange that although the two aircraft never actually collide, 1 aircraft is treated as if it has, with a phantom aircraft as well as it is not really there.

Have a good one if you can,

Cheers

James10

Share this post


Link to post
Share on other sites
Hello Tigger6, how has the day been so far?

I have seen this image somewhere before. My suspicion is this methodology was devised to overcome computer hardware and internet limitations/issues way back when, just to have an online flight sim. This is good as without it there would be none. Now 2016, computer hardware and even the internet itself has significantly improved. Maybe leverage the improvements to implement a more correct resolution.

Just a little question, in the graphic the surviving aircraft passes over the top of the other. I would presume passing under the other aircraft would have the same results?

It is still strange that although the two aircraft never actually collide, 1 aircraft is treated as if it has, with a phantom aircraft as well as it is not really there.

Have a good one if you can,

Cheers

James10

Unfortunately, internet lag will always be a thing, new technology does not help: in the worst case, the two pilots may be in Asia, 20000km from Texas. Therefore, the information about each plane's position has to travel 40000km, which will take 113ms. Now there will be some computing overhead since there isn't a simple single optical cable connecting the end-points and so the average time the data needs to be transmitted will be about 500ms. Now add the time the server needs to process the data and forward it to the other client (remember it needs to: gather data from all clients, figure out which clients are near each other and thus need position updates, then send out those position updates). This will take at least 500ms in addition. Probably more, but let's talk about the idealized scenario.

Add in some rendering delay and you are well above a second of lag, even if everything is perfect. In the current implementation, it's probably more.

A plane doing 500km/h is doing 140m/s and the discrepancy will thus be about those 140 meters. There is just no way about it, any other solution will include you sometimes passing through solid objects but not blowing up, or planes jumping all over the sky constantly, as some prediction algorithm tries to reconcile predictions with reality.

Have a nice day yourself!

Share this post


Link to post
Share on other sites
Unfortunately, internet lag will always be a thing, new technology does not help: in the worst case, the two pilots may be in Asia, 20000km from Texas. Therefore, the information about each plane's position has to travel 40000km, which will take 113ms. Now there will be some computing overhead since there isn't a simple single optical cable connecting the end-points and so the average time the data needs to be transmitted will be about 500ms. Now add the time the server needs to process the data and forward it to the other client (remember it needs to: gather data from all clients, figure out which clients are near each other and thus need position updates, then send out those position updates). This will take at least 500ms in addition.

Yup. I have some people that I just can't fight/kill on the ground. It's as if they're on warp-speed while I'm on impulse or something. Erasmo is one of those and it's not (always) down to me having the reflexes of a stoned turtle. It's just that when I see him, take aim and push my finger down to fire, he usually has already pulled the trigger and killed me. Even in situations where I'm sure we've seen each other more or less at the same time.

IMO a lot of this is down to our pings and locations in the world. According to him, his ping is around 150, which is roughly what I have (140 - 150ms). He connects from Spain, I connect from Germany and the server is in Texas. So with all the traveling of packets that's involved here, it's no wonder that some odd behavior will occur.

S.

Share this post


Link to post
Share on other sites

The hardest part is generating enough joules to overcome the Spitfire's front shield generators. :D

Share this post


Link to post
Share on other sites
Yup. I have some people that I just can't fight/kill on the ground. It's as if they're on warp-speed while I'm on impulse or something. Erasmo is one of those and it's not (always) down to me having the reflexes of a stoned turtle. It's just that when I see him, take aim and push my finger down to fire, he usually has already pulled the trigger and killed me. Even in situations where I'm sure we've seen each other more or less at the same time.

IMO a lot of this is down to our pings and locations in the world. According to him, his ping is around 150, which is roughly what I have (140 - 150ms). He connects from Spain, I connect from Germany and the server is in Texas. So with all the traveling of packets that's involved here, it's no wonder that some odd behavior will occur.

S.

Ping is not always what it seems it is.

I have the unique experience of actualy being in West Virginia, USA. Now if I use the net route advised for US players and do a traceroute my actual path goes to Virginia then to Sweden and then makes it's way back to CRS servers. And with that hop to Sweden and back the old net code still showed me with a 56ms ping. Tgunr even did a traceroute on me and PMed me as to how I could be in WV and coming off a Swedish hop to the servers.

So I have to use the net route that euros use, which actualy keeps all my hops in the US.

Share this post


Link to post
Share on other sites
My suspicion is this methodology was devised to overcome computer hardware and internet limitations/issues way back when, just to have an online flight sim. This is good as without it there would be none. Now 2016, computer hardware and even the internet itself has significantly improved. Maybe leverage the improvements to implement a more correct resolution.

The illustrated mechanics are fundamental to an internet connected strong-client/weak-server game, because the internet isn't instantaneous. It cannot be, as long as the game is globally marketed and various clients are connected by ordinary net links with many repeaters, routers and other electronic connections that each have finite processing times, plus the game's server array.

OldZeke also mentions the speed-of-light light limitation, but until we have players not on Earth, that won't be the dominant contributor to the total.

The need to consider gameplay-significant "lag" could be eliminated by requiring the game to be played only by individuals in the same room, connected by a direct network. Or, the significance of lag could be decreased by preventing connections by customers with "pings" over X in duration. Either however would be commercially non-viable.

Edited by jwilly

Share this post


Link to post
Share on other sites

Hello All again, thank you for your inputs, however...

I am only to aware of the physical limitations of the world wide network known as the “Internet”. Yes distance is the killer and distance will always be a killer until Electricity, Electromagnetic Radiation and/or light photons are no longer used as the signal carrier. Here are some observations. This game was first released in 2001. In 2001 I would be reasonably confident to state the most common connection method to the “Internet” was via a 56kb modem. The most common computer would have been likely have been based on an Intel Pentium 4. Single core/processor running at about 166 mHz. My “Internet” connection today is currently an ADSL2+, soon moving to Fiber Optic. This computer, the one I am using to write this is an Intel i3 Quad core 2.66GHz. I am reasonably confident that my statement:

“Now 2016, computer hardware and even the internet itself has significantly improved.”

is valid as no matter how you look at it as both the Internet and in Computer Hardware, significant improvements in performance have occoured. In the case of the Internet, capacity has increased significantly as well. I am suggesting the improvements in the Internet and in Computer Hardware should be leveraged to help.

Ping/Lag:

I would suggest that “Propagation Delay” is probably a more appropriate term to describe the issue, rather that lag. Ping is a bidirectional measurement, both send and receive while “Propagation Delay” is a unidirectional measurement. Only a single direction, send or receive.

Yes alas, due to the way the data is transmitted, as described above, the “Tyranny of Distance” will plague the Internet. This is commonalty measured as “Ping” i.e. the amount of time it takes a standard packet of information to be sent to a remote (next room, next city, next country...) and be retransmitted back to the sending computer. It is measured in Milliseconds (ms) where 1 second is 1000 Milliseconds. This is a “given”.

However, an internet content provider, in this case the Cornered Rat Software Game BGE, has elected to have “One Server - One War” as one of its core selling hooks and not go down the route of more region based servers as is common with other internet based real-time games. To this end it would be at best remiss of Cornered Rat Software not to address the issues caused as a result of having far flung customers all addressing a single server cluster. When BGE or more correctly WWIIOnline was released, just getting it to work was a significant step. More steps are required now to address the easily described as demonstrated above to address the by design decision to have “One Server - One War”. The “Tyranny of Distance” was known when the decision was originally made. It is not a recently discovered problem. If anything now it is less (but NOT nonexistent) of an issue than when the decision was originally made.

Tigger6 how are you doing?

Just an observation relating to the image, Point #2. The caption says:

“2: Bf109 starts the pullup, in this moment both players can occupy the same airspace...”

Isn’t that the very definition of a “Mid Air Collision”, two (or more) aircraft occupying the same airspace at the same time? I have seen in the forums on many occasions the following, “what you see isn’t necessarily exactly what is happening”. Why does this instance appear to be different?

Jwilly, how has your day been?

No the internet is not instantaneous. I have no issue with that as it is a “given”. It’s how the “given” is addressed that is important.

I believe the game utilises a modified authoritative server model as an authoritative client model although potentially gives more accurate results, is more exposed to hacking. In a “pure authoritative server model” the game runs entirely on the server. The clients are shown the results of the simulation that has taken place on the server. Once the client is shown the results from the server the client then generates the display for the user. BGE is not a pure authoritative server model though.

From my read of what OldZeke wrote, I am unable to find any reference he made to the speed of light limitation. Yes it is a limitation inherent in the current methods of signal distribution used by the internet. Those methods being Electricity, Electromagnetic Radiation and/or light photons.

At the moment the game is not commercially viable. It is user funded. I have no issues with this game being user funded.

Thank you all for your comments. There are a lot of causes for many things that happen on an internet playable game. I can accept this.

The core point of the original post of mine is; two aircraft collide head-on. One aircraft is completely destroyed while the other can be completely undamaged. This result regardless of the how’s, whys etc, should not happen. Either two aircraft collide or two aircraft don’t.

Cheers

James10

Edited by james10

Share this post


Link to post
Share on other sites
In 2001 I would be reasonably confident to state the most common connection method to the “Internet” was via a 56kb modem. The most common computer would have been likely have been based on an Intel Pentium 4. Single core/processor running at about 166 mHz.

Probably more like the Pentium III, IIRC. I sold my aging P II 233 around the year 2000 to get a P III. 166 MHz is way too low a number for that period. Even some of the first-generation Pentiums in the mid/late 90s ran at higher clock-speeds than that.

Connection-wise, I wouldn't be too sure. I had my first internet-connection at home in the mid 90s and that already was an ISDN-connection which was at least 128 kbit/s (minimum speed of ISDN). At the office I had T1-speeds as early as the mid/late 90s but I would agree that this sort of connection was out of reach for private households back then.

By the time we got to WW2ONL, DSL wasn't that far off. Here's an old traceroute I did back in 2002 - I'm not 100% certain, but I'm very sure that this was done on my first DSL-connection.

Trace%20ient-com_zpswdfe7aaw.jpg

My current ping (on a much faster DSL-connection and literally living next door to the DSL-"hub") isn't much better than that (140 - 150ms, pre 1.35).

But you have to remember how different the game was back then. Graphics were nothing like what they are today. FPS were pretty horrible - I'd say the very first versions of the game were nearly unplayable on then-current PCs. I remember that the first version I ever played (black loading screen with a PzIII and some other stuff on it) would take an eternity to load when you first started the game.

Plus: The version we're currently playing is "frozen in time", so to speak. I don't recall exactly when the big overhaul of the engine happened but it happened quite a long time ago and how much "radical" change have we seen since then?

All that doesn't change the fact that distance from the server will always be a factor in this type of game. Especially when we're talking about air-combat where speeds and closure-rates are much higher than on the ground.

The core point of the original post of mine is; two aircraft collide head-on. One aircraft is completely destroyed while the other can be completely undamaged. This result regardless of the how’s, whys etc, should not happen. Either two aircraft collide or two aircraft don’t.

Oh, they do collide. The problem seems to be that in a lot of instances, only one AC will take critical damage/blow up. I've run into EA myself and only suffered a mild flop/spin as a result while they blew up. And I've had EA run into me and fly on as if nothing happened while I exploded.

I do agree that this behavior is annoying and doesn't seem to make much sense. As I've stated earlier: If you run into an EA with enough smack to make it blow you, your own plane should be seriously damaged as well.

Edited by sascha

Share this post


Link to post
Share on other sites

The core point of the original post of mine is; two aircraft collide head-on. One aircraft is completely destroyed while the other can be completely undamaged. This result regardless of the how’s, whys etc, should not happen. Either two aircraft collide or two aircraft don’t.

Cheers

James10

No offence intended here but you should understand that you are the 10000th person (no exaggeration here) to run into this very issue, hundreds of them being on this forum alone, yet this never changed in any game.

Your problem is that you do not understand what is going on. The issue is that there are at least 2 distinct realities and there always will be 2 distinct realities. Those are not identical due to internet lag (this is how we called it for 20+ years, we won't change terminology) , and they cannot be reconciled. The current behavior results in your airplane only blowing up if you crash into an object in your reality and not otherwise. Any other implementations, such as A: If one client crashes, both clients blow up or B: client position and timestamps are compared by the server and if a collision is detected, both blow up, will result in many instances of your plane spontaneously blowing up. This is absolutely not something I want to have.

Share this post


Link to post
Share on other sites

Hello again.

Probably more like the Pentium III...

That may be the case.

No offence intended here but you should understand that you are the 10000th person (no exaggeration here) to run into this very issue, hundreds of them being on this forum alone, yet this never changed in any game.

Your problem is that you do not understand what is going on. The issue is that there are at least 2 distinct realities...

If I am the only person to post on this issue with then that’s OK. However if I am, as you have posted, the “10000th person” to indicate this then it could reasonably be taken to indicate there is at least a potential problem that should be addressed.

In truth and for all practical purposes the number of realities in existence at any one given point in time for a typical real-time internet game, is equal to the number of players currently on line, plus 1. The plus 1 is the actual reality for the game running on the games servers. This is the only reality that matters as all of the rest are only reflections of that reality presented to the user via their computer connected to that server. The clients reality (current player on line) does have some predictive ability but it still salves itself to the server reality (the plus 1) as soon as it is able, to correct any errors that may have crept in.

One reality that does appear to have escaped notice is to have a Collision, “Mid Air” or otherwise, requires two objects, whether the two objects are an aircraft, the ground, a tree, a building or in this case another aircraft.

No offence taken.

Cheers

James10

Share this post


Link to post
Share on other sites
Hello again.

That may be the case.

If I am the only person to post on this issue with then that’s OK. However if I am, as you have posted, the “10000th person” to indicate this then it could reasonably be taken to indicate there is at least a potential problem that should be addressed.

In truth and for all practical purposes the number of realities in existence at any one given point in time for a typical real-time internet game, is equal to the number of players currently on line, plus 1. The plus 1 is the actual reality for the game running on the games servers. This is the only reality that matters as all of the rest are only reflections of that reality presented to the user via their computer connected to that server. The clients reality (current player on line) does have some predictive ability but it still salves itself to the server reality (the plus 1) as soon as it is able, to correct any errors that may have crept in.

One reality that does appear to have escaped notice is to have a Collision, “Mid Air” or otherwise, requires two objects, whether the two objects are an aircraft, the ground, a tree, a building or in this case another aircraft.

No offence taken.

Cheers

James10

Usually people ask and then say, aah! I get it at some point in the thread.

It seems you understand what the issue is here, I suppose you propose that the invisible reality be the one taken into account. I do not understand why I would want a game mechanics that makes my airplane randomly blow up without any apparent reason, just because some computer somewhere calculated there was a collision that was invisible to anyone. That makes no sense whatsoever. In practice you would be able to safely fly through enemy airplanes (promoting the shoot until you ram attitude), on the other hand you would blow up without touching an airplane. Worst of all worlds.

Share this post


Link to post
Share on other sites

Hello again,

Usually people ask and then say, aah! I get it at some point in the thread.

It seems you understand what the issue is here, I suppose you propose that the invisible reality be the one taken into account. I do not understand why I would want a game mechanics that makes my airplane randomly blow up without any apparent reason, just because some computer somewhere calculated there was a collision that was invisible to anyone. That makes no sense whatsoever. In practice you would be able to safely fly through enemy airplanes (promoting the shoot until you ram attitude), on the other hand you would blow up without touching an airplane. Worst of all worlds.

There is no invisible reality. When we spawn in and play we both see the "invisible reality". So its not all that invisible. It is the games true reality, all of the others are representations only.

Yes as it is the only true reality for the game. It is not some computer somewhere it is the computer that calculates the game for everyone else to see.

If you are OK with the way things are, that's fine by me. I am not fine with the way things are and have suggested altering them a little to conform better with what I and apparently a lot of other people would also be fine with. I am not suggesting changing ALL of the collision mechanics, just the one where two pilots choose a head-on attack that results in a collision. Two pilots, two planes and one collision.

Cheers

James10

Share this post


Link to post
Share on other sites
Hello All again, thank you for your inputs, however...

I am only to aware of the physical limitations of the world wide network known as the “Internet”. Yes distance is the killer and distance will always be a killer until Electricity, Electromagnetic Radiation and/or light photons are no longer used as the signal carrier. Here are some observations. This game was first released in 2001. In 2001 I would be reasonably confident to state the most common connection method to the “Internet” was via a 56kb modem. The most common computer would have been likely have been based on an Intel Pentium 4. Single core/processor running at about 166 mHz. My “Internet” connection today is currently an ADSL2+, soon moving to Fiber Optic. This computer, the one I am using to write this is an Intel i3 Quad core 2.66GHz. I am reasonably confident that my statement:

“Now 2016, computer hardware and even the internet itself has significantly improved.”

is valid as no matter how you look at it as both the Internet and in Computer Hardware, significant improvements in performance have occoured. In the case of the Internet, capacity has increased significantly as well. I am suggesting the improvements in the Internet and in Computer Hardware should be leveraged to help.

Ping/Lag:

I would suggest that “Propagation Delay” is probably a more appropriate term to describe the issue, rather that lag. Ping is a bidirectional measurement, both send and receive while “Propagation Delay” is a unidirectional measurement. Only a single direction, send or receive.

Yes alas, due to the way the data is transmitted, as described above, the “Tyranny of Distance” will plague the Internet. This is commonalty measured as “Ping” i.e. the amount of time it takes a standard packet of information to be sent to a remote (next room, next city, next country...) and be retransmitted back to the sending computer. It is measured in Milliseconds (ms) where 1 second is 1000 Milliseconds. This is a “given”.

However, an internet content provider, in this case the Cornered Rat Software Game BGE, has elected to have “One Server - One War” as one of its core selling hooks and not go down the route of more region based servers as is common with other internet based real-time games. To this end it would be at best remiss of Cornered Rat Software not to address the issues caused as a result of having far flung customers all addressing a single server cluster. When BGE or more correctly WWIIOnline was released, just getting it to work was a significant step. More steps are required now to address the easily described as demonstrated above to address the by design decision to have “One Server - One War”. The “Tyranny of Distance” was known when the decision was originally made. It is not a recently discovered problem. If anything now it is less (but NOT nonexistent) of an issue than when the decision was originally made.

Tigger6 how are you doing?

Just an observation relating to the image, Point #2. The caption says:

“2: Bf109 starts the pullup, in this moment both players can occupy the same airspace...”

Isn’t that the very definition of a “Mid Air Collision”, two (or more) aircraft occupying the same airspace at the same time? I have seen in the forums on many occasions the following, “what you see isn’t necessarily exactly what is happening”. Why does this instance appear to be different?

Jwilly, how has your day been?

No the internet is not instantaneous. I have no issue with that as it is a “given”. It’s how the “given” is addressed that is important.

I believe the game utilises a modified authoritative server model as an authoritative client model although potentially gives more accurate results, is more exposed to hacking. In a “pure authoritative server model” the game runs entirely on the server. The clients are shown the results of the simulation that has taken place on the server. Once the client is shown the results from the server the client then generates the display for the user. BGE is not a pure authoritative server model though.

From my read of what OldZeke wrote, I am unable to find any reference he made to the speed of light limitation. Yes it is a limitation inherent in the current methods of signal distribution used by the internet. Those methods being Electricity, Electromagnetic Radiation and/or light photons.

At the moment the game is not commercially viable. It is user funded. I have no issues with this game being user funded.

Thank you all for your comments. There are a lot of causes for many things that happen on an internet playable game. I can accept this.

The core point of the original post of mine is; two aircraft collide head-on. One aircraft is completely destroyed while the other can be completely undamaged. This result regardless of the how’s, whys etc, should not happen. Either two aircraft collide or two aircraft don’t.

Cheers

James10

I started playing this game on 500mhz K7 amd rig..barely. September 2001 I gave myself a 1.7ghz P4 rig for my birthday.

Share this post


Link to post
Share on other sites
it is the computer that calculates the game for everyone else to see.

An authoritative server game (i.e. what you describe) with fast moving player objects--aircraft, or even running infantrymen--would be unplayable with global or even continent-wide players, due to lag.

That's the fundamental physics faced by every MMOLG design team.

Your suggestion isn't novel. How to design WAN-based games like this one to best deal with the physics facts has been an ongoing topic among game developers for over a generation how. This game's design approach will be the best choice until either the internet has faster-than-light connections, or time is directly controllable. Neither of those is expected in the near future.

Share this post


Link to post
Share on other sites
until either the internet has faster-than-light connections, or time is directly controllable. Neither of those is expected in the near future.

We need a subspace-connection upgrade for the internet!! :D

The funny part is that if the netcode is done right, you'll hardly notice the fact that you're playing over distances of 10s of thousands of kilometers.

I remember playing Warbirds over a LAN at the office back in 2000 or 2001 (that game came with a limited LAN-mode) and it pretty much felt the same flying against a colleague sitting in the office nextdoor as did flying against people from all over world on the live-server.

S.

Edited by sascha

Share this post


Link to post
Share on other sites
Hello again,

There is no invisible reality. When we spawn in and play we both see the "invisible reality". So its not all that invisible. It is the games true reality, all of the others are representations only.

Yes as it is the only true reality for the game. It is not some computer somewhere it is the computer that calculates the game for everyone else to see.

If you are OK with the way things are, that's fine by me. I am not fine with the way things are and have suggested altering them a little to conform better with what I and apparently a lot of other people would also be fine with. I am not suggesting changing ALL of the collision mechanics, just the one where two pilots choose a head-on attack that results in a collision. Two pilots, two planes and one collision.

Cheers

James10

I don't think you understand what is going on. Try to look at that image again.

Share this post


Link to post
Share on other sites

Hello OldZeke, how has your day been?

I started playing this game on 500mhz K7 amd rig..barely. September 2001 I gave myself a 1.7ghz P4 rig for my birthday.

I hope the newer birthday computer showed some improvements in game play.

Cheers

James10

Share this post


Link to post
Share on other sites

Hello all.

Jwilly, how are you doing?

“...

it is the computer that calculates the game for everyone else to see.

...”

It is true I did say that in post #17, however when I first referenced the “authoritative server model” in Post #12 I also said the following:

“BGE is not a pure authoritative server model though.”

Hello Tigger6 how has your day been so far?

I don't think you understand what is going on. Try to look at that image again.

I have, have you?

What it does suggest to me is the “Collision Detection” is currently been completed on the Players Computer Client and not the Server. What do you think?

Maybe moving the collision detection to the server would alleviate the problem.

As I have indicated in a previous posting, If you are OK with the way things are, that's fine by me. I am not fine with the way things are and have suggested altering them.

Cheers

James10

Edited by james10

Share this post


Link to post
Share on other sites
Maybe moving the collision detection to the server would alleviate the problem.

If the server mediated collision detection, it also would have to control all objects' movement. Otherwise there would be no single time-reality in which collisions could be detected.

This has all been explained. Such a system would result in a game in which every player would see objects (including their own) moving irrationally, because of the physically mandated time loop between each player's control of their object and the server's implementation of that control. It'd be impossible to visually model realistic looking aerodynamics and ballistics because the time lag between control input and implementation would mean that faster-than-directing movement changes would be needed to "catch up" with what players were directing.

I am not fine with the way things are and have suggested altering them.

Understood. What we're trying to explain is that all of your suggested ideas were tried decades ago, and don't and can't result in a marketable MMOLG.

Share this post


Link to post
Share on other sites

Hello jwilly how has your day been?

If the server mediated collision detection, it also would have to control all objects' movement. Otherwise there would be no single time-reality in which collisions could be detected.

...

It appears the current BGE methodology the server mediates very little collision detection. I believe the only objects the server has anything much to do with are “Server Tracked Objects”. Currently these objects are only the light mortar rounds. The server may control the various AI installations but it is unclear where the control of those objects are resides.

“Otherwise there would be no single time-reality in which collisions could be detected.”

Isn’t this the issue here? “There is no single time-reality in which collisions are detected”. The game server would appear to do very little, probably no, collision detection. It appears to only be performed on individual player’s computer clients.

...

Understood. What we're trying to explain is that all of your suggested ideas were tried decades ago, and don't and can't result in a marketable MMOLG.

Decades ago, this was probably the case. However what was difficult if not impossible, decades ago may now be possible with the cumulative advances that HAVE occurred in the intervening decades. Below is a list of firm and by general consensus at the time they in vogue, irrefutable statements:


  • The Earth is Flat.
  • The Sun revolves around the Earth.
  • Man cannot Fly.
  • World War One was the war to end all wars.
  • With the advent of the atomic bomb armies are no longer needed.

I’m sure you can think of some as well. Maybe it might just be time to revisit the idea and see if it is now possible.

Cheers

James10

Edited by james10

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.