connectivity & performance
mobetta
Join Date: 2010-07-23 Member: 72693Banned
Join Date: 2010-07-23 Member: 72693Banned
<div class="IPBDescription">discussion on spark engine & more</div>Hello ns2 world. Im fairly new to natural selection. I come from the cod and quake world. I have been playing the beta of ns2. I have 47 hours logged in the beta as this is what steam shows.
Now i have got some cool ideas and suggestions that i have been thinking about and wondering about.
First off this spark engine is a lua based game engine or at least that's what i have seen and heard about. Now coming from cod and quake the engine used in thos where the id tech 3 engine. I believe that this engine in the best game engine ever made still to this date. People can talk about graphics all they want but i graphics really don't make a game good. So my point here is anything on the graphics level of css is good enough in my opinion. Ns2 has great graphics and i would like to ask you to please drop all your focus on that. I am having a problem with connectivity and performance in the beta right now. My pc can hold a solid 80 fps on crysis in dx10. Ns2 which i believe is dx9 i get about 30-60 on a good day. That needs to be improved 10 fold. Also on the connectivity. In cod and quake i can infact unplug the ethernet cable from the back on my motherboard and then plug it back in after 30 seconds and i wont drop from the server. It will lag me to the point of no packet travel but it wont drop me. In Ns2 i get pings in the 200s. Now my internet isnt the best but its good enough to play on a competitive level in cod and quake. Heres a speed test. <a href="http://www.speedtest.net/result/1219915102.png" target="_blank">http://www.speedtest.net/result/1219915102.png</a>
Now ns2 is a beta i know so please dont say u know its a beta right. Im not saying that this game sucks. Im just letting u know the current state of this game isnt anywhere good enough for release. So the engine itself needs a tone of work. I please ask of you guys to not go gold and not to release a version 1.0 for at least another 4 - 5 years. Take your time on this game.
How to spend this time. Well im sure u have a lot going on behind the doors but some ideas i have had where a game mode that allowed teams to build up a base before they get attacked. Maybe the levels could be made in a way where there are locks on the doors and vents. They become unlocked after say 5 mins and then u can go out and attack. This could allow teams to get together and come up with a strat.
Levels in general are cool. I see the way the levels are made they are basic indoor levels with vents and stuff. There true place is kind of unknown. you might be under ground or in space or underwater. I think it would be cool to see more interactive stuff in the levels. Being able to flood a room with lava or open a space vacuum that sucks anyone near it into outer space. More moving parts in the levels that breath more life into the game. Also maybe if u guys got together and made some kinda weird algorithm u could make a game mode that people could play and it would create a random levels. Also i believe the rts and fps elements need to be tied together better. There needs to be more to the rts element. It needs to play a bigger role in the game. Its cool to have the ability to have more then one commander but u really dont need more then one and thats a problem.
Really ns2 the general idea of the game is really cool and something that i want to play. I think the big keys of making this game better is making a better engine with more functionality. Getting more creative levels into the game. Making new and cool game modes. And just keep fixing the bugs that will be made during the development of the game.
Idk the tech behind the spark engine. If there is a video about it or if some of you guys know more about it. Could u please tell me why my frame rate and ping are so bad. Is there a official server that is being ran by unknown worlds?
I would really like to know more about the engine and how it is being made. I hope its being made for the pc platform.
Now i have got some cool ideas and suggestions that i have been thinking about and wondering about.
First off this spark engine is a lua based game engine or at least that's what i have seen and heard about. Now coming from cod and quake the engine used in thos where the id tech 3 engine. I believe that this engine in the best game engine ever made still to this date. People can talk about graphics all they want but i graphics really don't make a game good. So my point here is anything on the graphics level of css is good enough in my opinion. Ns2 has great graphics and i would like to ask you to please drop all your focus on that. I am having a problem with connectivity and performance in the beta right now. My pc can hold a solid 80 fps on crysis in dx10. Ns2 which i believe is dx9 i get about 30-60 on a good day. That needs to be improved 10 fold. Also on the connectivity. In cod and quake i can infact unplug the ethernet cable from the back on my motherboard and then plug it back in after 30 seconds and i wont drop from the server. It will lag me to the point of no packet travel but it wont drop me. In Ns2 i get pings in the 200s. Now my internet isnt the best but its good enough to play on a competitive level in cod and quake. Heres a speed test. <a href="http://www.speedtest.net/result/1219915102.png" target="_blank">http://www.speedtest.net/result/1219915102.png</a>
Now ns2 is a beta i know so please dont say u know its a beta right. Im not saying that this game sucks. Im just letting u know the current state of this game isnt anywhere good enough for release. So the engine itself needs a tone of work. I please ask of you guys to not go gold and not to release a version 1.0 for at least another 4 - 5 years. Take your time on this game.
How to spend this time. Well im sure u have a lot going on behind the doors but some ideas i have had where a game mode that allowed teams to build up a base before they get attacked. Maybe the levels could be made in a way where there are locks on the doors and vents. They become unlocked after say 5 mins and then u can go out and attack. This could allow teams to get together and come up with a strat.
Levels in general are cool. I see the way the levels are made they are basic indoor levels with vents and stuff. There true place is kind of unknown. you might be under ground or in space or underwater. I think it would be cool to see more interactive stuff in the levels. Being able to flood a room with lava or open a space vacuum that sucks anyone near it into outer space. More moving parts in the levels that breath more life into the game. Also maybe if u guys got together and made some kinda weird algorithm u could make a game mode that people could play and it would create a random levels. Also i believe the rts and fps elements need to be tied together better. There needs to be more to the rts element. It needs to play a bigger role in the game. Its cool to have the ability to have more then one commander but u really dont need more then one and thats a problem.
Really ns2 the general idea of the game is really cool and something that i want to play. I think the big keys of making this game better is making a better engine with more functionality. Getting more creative levels into the game. Making new and cool game modes. And just keep fixing the bugs that will be made during the development of the game.
Idk the tech behind the spark engine. If there is a video about it or if some of you guys know more about it. Could u please tell me why my frame rate and ping are so bad. Is there a official server that is being ran by unknown worlds?
I would really like to know more about the engine and how it is being made. I hope its being made for the pc platform.
Comments
I understand that you're a small team, and that you're doing an amazing amount of work in a short time. That's why I feel your next focus after the new binding layers should be on the networking. As a multiplayer only game, your server to client networking will make or break you. Work on the server software, tweak the code, whatever you gotta do - get that connection working smooth.
the game will get better and more will be optimized. interactivity (flooding levels with lava) will come much later, after 1.0, i do believe they want to make elevators and door switches interactive before 1.0 however. right now max is working on optimization of the server and c++/lua code. the engine will get better, we just need to be patient.
as far as commander/player gameplay, many of us agree that there needs to be more to tie them together. search the forum for that thread, it's heated (and currently being retalked about with the latest welding news)
the tech behind the engine is powerful, just youtube spark engine. and while you are okay with css graphics, i am not. games like splintercell and farcry 2 have made me a graphics snob. lighting is one of the coolest things ever. while you say its graphic, i say its gameplay: especially in this game.
I don't take a quake like game as competitive, no offense. that game is mindless, just like halo: hmmm, lemme run the same "pro" route thru the level, making my way to the super powerup and then making my way to the largest explosive weapon and then rinse and repeat killing everything on the way. that's not tactical, do you need to coordinate with other humans to do that? no. might as well take off your headset and go play cod.
i want a game where if you don't talk to eachother and the other team does, you lose. css was like that, but one guy with a awp who knew the level and ran the "pro" route could beat your entire team (see quake reference above). this is the sole reason i'm not buying crysis 2 after playing the demo. it's halo meets cod, but its still the same multiplayer mindless firefight crap that has been thrown at me for years now. it's not innovative. it's not challenging besides overcoming the frustration of learning what the "pro" route is.
in short, just be patient, this game is going to be everything we all hope it to be, framerates will come up, performance will improve, and we'll all get gorge plushies.
What?
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->I please ask of you guys to not release a version 1.0 for at least another 4 - 5 years.<!--QuoteEnd--></div><!--QuoteEEnd-->
You think they have that much time?
Another useless whine post about performance I guess.
i want a game where if you don't talk to eachother and the other team does, you lose. css was like that, but one guy with a awp who knew the level and ran the "pro" route could beat your entire team (see quake reference above). this is the sole reason i'm not buying crysis 2 after playing the demo. it's halo meets cod, but its still the same multiplayer mindless firefight crap that has been thrown at me for years now. it's not innovative. it's not challenging besides overcoming the frustration of learning what the "pro" route is.
in short, just be patient, this game is going to be everything we all hope it to be, framerates will come up, performance will improve, and we'll all get gorge plushies.<!--QuoteEnd--></div><!--QuoteEEnd-->
Im no monkey. Just stating my view on the beta. Idk who this max is and dont really care.
Quake is a very very competitive game. Its not a team base game tho. I agree with the idea of team work playing a big role in ns2. One thing i noticed is that in the beta people tend to play together and thats really cool. Thats is a aspect that i think many console ported games lack. I very patient in fact i dont want them to release the game for awhile. Or it will just get a bad rap and people will never come back to it.
Quake is a very very competitive game. Its not a team base game tho. I agree with the idea of team work playing a big role in ns2. One thing i noticed is that in the beta people tend to play together and thats really cool. Thats is a aspect that i think many console ported games lack. I very patient in fact i dont want them to release the game for awhile. Or it will just get a bad rap and people will never come back to it.<!--QuoteEnd--></div><!--QuoteEEnd-->
you missed my sarcasm emoticon, monkey. :>
max is the technical director, he coded everything, the engine, the level editor, the cinematic editor, decoda (the code editor). charlie helped.
yes quake has competitions, but do they require much skill? imho no.
and they will release 1.0 when they need the money, it's a business after all.
my suggestion to everybody making threads like this one (and they come in waves apparently) is that if you're not happy with the state of the game, put it down for a month, and check back with us later.
Personally I think the post was a waste of time and a quick look at the past few pages of the forum or getsatisfaction will demonstrate the issues that are currently outstanding.
Yes, there are performance issues. Yes, this is BETA. Yes, this will take time.
Enough said really.
You apparently haven't heard of CryEngine 3 then...
Quake-like games non competitive? That is practically competitive FPS at its purest form, and anyone knowing anything about competitive games is aware of this. What is defined as competitive to RTS is strategy, what is competitive to FPS is player skill (and a little strategy too). Since this game is RTS/FPS, we have commanders playing from a near pure RTS standpoint planning strategy, but what they control (more like guide) are FPS players of varying skill levels. If the commander doesn't send them in the right direction they can be as good as they want but all that skill isn't going to add up to a team win necessarily. On the other hand, if they are all cooperative but horrible at killing, should also be quite likely to lose too regardless if the comm happens to be the only competent player. We could level the playing field by nerfing everything to the max and limiting skill growth potential among the FPS players, but that detracts highly from a good FPS game. I guess what I'm trying to say is individual player skill should matter or you've eliminated one of the most important features of what many players will be looking for in the FPS side of this title's game play.
I think he ment how incredibly efficient the Quake3-engine was, and how nice it looked (and looks still), as well as it's very smooth gameplay. I don't care for CryTek's work really. Just throwing a quazillion polygons\shaders in there does not a good engine make.
Anyways, I'm really not sure on the validity of comparing a tuned in game engine to a under construction game engine is.
and being small company like this, it might actually be higher even, who knows.
if they didn't go with building their own engine we would most likely have 1.0 version some years ago. I'm sure some of them might regret going this way.
I've been waiting for ns2 since ns1 ever came out, but currently I do not know where this game is headed, too much gameplay from ns1 has changed for it to be as it was though as some will most likely say, beta beta beta.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->Initial tests show my binding layer is about 2-5x faster than luabind. Will take some more work before it's in game for a real test though<!--QuoteEnd--></div><!--QuoteEEnd-->
Taking this with a grain of salt (or more like a cube), this could potentially be a big contribution to both the server-tickrate, as well as the client's FPS (I'm sure I've already said this at some point). Add to that, that a lot of the problems encountered are symptomatic of a poor performing server\client, we <u>might</u> see a significant improvement upon implementation of the new Lua binding-layer (which we may even see in the coming build).
This.
You can rest assured that everything you've mentioned has been said before, addressed by UWE, and will be fixed.
You can rest assured that everything you've mentioned has been said before, addressed by UWE, and will be fixed.<!--QuoteEnd--></div><!--QuoteEEnd-->
Good news. Tbh the game in its current state while annoying on some aspects is really fun and competitive. I can only dream of what this will turn into with the coming years.