Awesome to hear that progress is going well, breaking up the work across developers like you did is totally the way to go. It's way more productive and efficient than putting 3 cooks in the same kitchen.
So, my comments...
Definitely open source the backend interface, because then people can write awesome backends!... That's less work for you guys, and it'll also get you guys a place on slashdot and the other nerd/Linux sites -- Linux nerds LOVE seeing games that work natively on Linux, it's what got me hooked on Heroes of Newerth... Not to mention the fact that writing our own backends gives we Linux programmers massive hardons... I know I'd contribute the the openGL implementation, along with maybe a lightweight implementation so that lesser computers can still play the game... and I also would love to try and write an AAlib backend :P
While I agree with some others that it should have been written in openGL to begin with, if you're more familiar with DX then who are we to judge the ones making this awesome game ;)
I'm *pretty* sure you can put steamworks in Linux. If you put the steamworks stuff into a DLL and then link to it through wine, you can have a native Linux binary using windows stuff... That's how Google's pacasa works... I don't know about the fine details of having the linux openGL stuff having windows drawing onto it too, but that should in theory be doable... Steam runs great under wine...
And, finally, why wxWidgets over Qt? Just wondering if Qt ever came up as an option, and if so why you decided to go against it... I'm a wicked Qt fanboy, even a core developer for libQxt! xD
<!--quoteo(post=1750500:date=Feb 3 2010, 08:31 AM:name=Draco_2k)--><div class='quotetop'>QUOTE (Draco_2k @ Feb 3 2010, 08:31 AM) <a href="index.php?act=findpost&pid=1750500"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->The DX11 tessellation is different, but the gist is the same as it was when AMD called it "Truform" five or so years ago.<!--QuoteEnd--></div><!--QuoteEEnd--> The gist is not at all the same. Truform tessellation provided very little control, and subsequently very little benefit. It was fixed-function in the worst form. Programmable tessellation however, allows for everything from simple displacement (Unigine DX11 demo), to geometry images, silhouette refinement, improved hair rendering, even hardware accelerated SVG. The domain and hull stages are no more fixed-function than vertex and pixel/fragment stages.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->Also, good job not starting a flamewar by insulting the opposition...<!--QuoteEnd--></div><!--QuoteEEnd--> You make the mistake of assuming I've taken a side, I'm only opposed to misinformation.
<!--quoteo(post=1750511:date=Feb 3 2010, 05:21 PM:name=WarLover)--><div class='quotetop'>QUOTE (WarLover @ Feb 3 2010, 05:21 PM) <a href="index.php?act=findpost&pid=1750511"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Will programmable tessellation work on DX 10.1 cards?<!--QuoteEnd--></div><!--QuoteEEnd--> There's no word and little probability that NS2 will use tessellation in any form. We may be getting a more direct answer in the upcoming Q&A.
For all the people wanting more game progress and less tools progress. They really are the same thing. The tools we see is the game itself for the most part.
I would like to see more quick twitter posts on how things are going and don't be scared of using techie concepts and wording.
There still are a lot of unknown things holding me back from starting models/textures/map. Really a lot of this stuff belongs in a SDK document or mapping guidelines.
Textures and models: Will transparency textures be used in shadow casting? Ambient Occlusion in real time or should we look at baking it into the textures belonging to models. Making parts or whole models not cast shadows? Parts of our models that emit light. Reflection? Will we be using precalculated reflection environment textures like hammer editor does? Some maps might require sky maps. Different lights such as area lights? Are we going to be using decals? Will decals work on models?
How far are you into finishing physics? So far the collision geometry of models is the same as the object. I can see we might need a way of using a simplified model for physics calculations. Will we be able to make objects that can be moved though physics. Making larger objects break apart from damage. Change gravity. Other forces to move players around.
Particles for various effects. Fog or some other type of effect for dust especially for lights.
It seems you are working on the first test map refinery. Will this be like an example map? How far is that completed and is it the only one currently?
AsranielJoin Date: 2002-06-03Member: 724Members, Playtest Lead, Forum Moderators, NS2 Playtester, Squad Five Blue, Reinforced - Shadow, WC 2013 - Shadow, Subnautica Playtester, Retired Community Developer
thank you very much for the open words (i'm beat wolf, the guy with the linux question).
For open sourcing parts of the engine, count me all in. Even though i have no open gl experience, i can certainly help to create the server browser parts etc. Really, if this is the only problem, i'm sure the linux community can help out!
From a business perspective I think they will work on a XBOX release before Linux and OS X. A good reason for having the community taking care of those. And a reason for going for DX might be the XBOX. There isn't much work porting OGL to DX, but now they have that covered already.
Thanks for taking time answering some questions Max. I would love to see some more from Matt or Bill by the way!
Very enjoyable read. It's like a warm cozy avalanche of awesome facts and details. The way UWE is handling development of NS2 is truly nothing short of genius. You Sirs are Gentlemen and a Scholars!
path finding... WOOT! Dedicated Linux Server on release... WOOT! Steamworks included... (does this mean all the NS2 pre-orders will be integrated to steam just like the NS2 orders bought in steam?) If that the case then... WOOT! and how did you manage that deal?
<!--quoteo(post=1750408:date=Feb 2 2010, 06:38 PM:name=WatchMaker)--><div class='quotetop'>QUOTE (WatchMaker @ Feb 2 2010, 06:38 PM) <a href="index.php?act=findpost&pid=1750408"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->"...we wanted to make sure the technology was flexible enough that we wouldn't have to start again from scratch for the next game."
<!--quoteo(post=1750519:date=Feb 3 2010, 10:29 AM:name=NeoSniper)--><div class='quotetop'>QUOTE (NeoSniper @ Feb 3 2010, 10:29 AM) <a href="index.php?act=findpost&pid=1750519"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->...does this mean all the NS2 pre-orders will be integrated to steam just like the NS2 orders bought in steam?<!--QuoteEnd--></div><!--QuoteEEnd-->
No.. well not yet. Steam will allow it when they have a strong beta as they previously said. Steamworks it only some multiplayer elements that they are relying to. Voice chat, etc. Like Max said.
Any chance that we'll be able to run the finished game in a setup like the one you described? Not being dependent on Steam is a huge convenience in my book, even if it means forgoing a server browser :)
<!--quoteo(post=1750404:date=Feb 2 2010, 06:03 PM:name=Max)--><div class='quotetop'>QUOTE (Max @ Feb 2 2010, 06:03 PM) <a href="index.php?act=findpost&pid=1750404"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->Please post comments on the topic Development Blog Update - Spark Engine Questions and Answers #1 here<!--QuoteEnd--></div><!--QuoteEEnd--> It is interesting that you consider opening your backend interfaces, but do not even bother to answer your emails/PMs when people offer you guys to do the Linx/Mac/OpenGL-port for basically free... This would enable you to release the game for these platforms maybe together with the Windows-game and get you more publicity...
@Enrico: If they were to fix the ports for release then they'd had to rely on people outside the team which can be a liability. They rather do it thoroughly themselves, after the game is done (I can imagine). Having to greet, learn and work with some people who they don't pay and can't set any expectations with (hey, they're working for free right?) may just delay the release than help it.
Well, releasing the graphics interface classes <i>before launch</i> as open source would release virtually no real code, as releasing the DX implementation wouldn't be needed, and would allow people to get an openGL implementation working by launch time. Just releasing the headers to the interface classes would be enough, no real implementation code needed!... Obviously we'd also need a mechanism for changing backends so that we can test the new openGL based backend... That or just replace the existing DX .dll file, assuming the backend is entirely contained within a library and not compiled into the binary (which is obviously preferred)
Let's say: It'll get retail in germany, and it works on my "AMD64 3500+" with 4GB RAM and an "PNY 6800GTX UC", under Windows XP, then I'd be totally happy.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->"...we wanted to make sure the technology was flexible enough that we wouldn't have to start again from scratch for the next game."
unknown world is not a company doing natural-selection only. right now yes, because they are still too short on staff but if they do the same good job than they did for ns1 they gonna be in serious business. (of course not like EA or ubisoft but ...)
+ if they got a good base its gonna be easier for "mod" to be created (if they accept ppl creating mod of course) that would be nice. fps/rts game are rare and its good for a company reputation and visibility.
talking about visibility : i hope you guys at unknown world will have a minimum of budget for the "right placed pub" by "right placed pub" i mean pub at the place the "real gamer" are going like on xfire. a game with 0 pub got always their server empty and the community never grow up
question : are you guys able to give an approximately date for the close beta ? something "general" like under 3 year or under 2 year or 1 or 6 month ?
so to clear this point: you'll make a linuxoid version of this game? it would be just too good to be true...(for example, valve some many years back said, that they'll port HalfLife engine to linux...and nothing happened).
Great post, Max! I give you guys credit for not simply ignoring the ongoing linux discussion. I think releasing enough API and letting the open-source community have a go at it would be a good idea...I imagine you guys will be busy enough with the windows portion.
<!--quoteo(post=1750537:date=Feb 3 2010, 06:59 PM:name=Triggerman)--><div class='quotetop'>QUOTE (Triggerman @ Feb 3 2010, 06:59 PM) <a href="index.php?act=findpost&pid=1750537"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->@Enrico: If they were to fix the ports for release then they'd had to rely on people outside the team which can be a liability. They rather do it thoroughly themselves, after the game is done (I can imagine). Having to greet, learn and work with some people who they don't pay and can't set any expectations with (hey, they're working for free right?) may just delay the release than help it.<!--QuoteEnd--></div><!--QuoteEEnd--> This is a lame excuse for not sending a simple "no". What is the difference for you between opening the render backend interfaces so people can work on an alternative renderer and opening the render backend interfaces <i>earlier</i> so people can work on an alternative renderer?
<!--quoteo(post=1750517:date=Feb 3 2010, 09:15 AM:name=perfectheat)--><div class='quotetop'>QUOTE (perfectheat @ Feb 3 2010, 09:15 AM) <a href="index.php?act=findpost&pid=1750517"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->From a business perspective I think they will work on a XBOX release before Linux and OS X. A good reason for having the community taking care of those. And a reason for going for DX might be the XBOX. There isn't much work porting OGL to DX, but now they have that covered already.
Thanks for taking time answering some questions Max. I would love to see some more from Matt or Bill by the way!<!--QuoteEnd--></div><!--QuoteEEnd-->
That's an interesting point. However, it is safe to say that it's a fairly big leap from forming an independent game company and releasing something through online distribution to publishing for the XBOX. Even if that was the end result of the venture, I think that an OpenGL port would happen before something on that level of publishing support fell into place.
I am definitely interested in Linux support and even thought I don't use OSX, adding support for it will increase the community. I know how to program in C++ and OpenGL and I'd be willing to contribute some code to help make Linux support a reality.
The most important thing to me though is to remove any dependency on Steam. A lot of people love Steam and I used to but a lot of people have also been burned by Steam. I bought GTA4 from Steam and it didn't run so I asked for a refund and they refused. I got my refund through the credit card company and lost my entire account. I've learned my lesson and will not use Steam again. Please make sure we can still have the full game experience without Steam.
There plenty of third party applications outside of Steam that allow people to find their friends and voice chat in any game not just NS2 so I think having those pieces missing is not so critical. The critical piece that I saw mentioned was the server list. Please make sure everyone can access the server list somehow even if it ends up just being a webpage with ip addresses on it.
I think anyone who is a hardcore PC gamer runs Windows. You just miss out on a lot of games if you don't. That said I would much rather prefer to run a game using Linux when I can.
<!--quoteo(post=1750770:date=Feb 4 2010, 10:45 PM:name=chrsghoyt)--><div class='quotetop'>QUOTE (chrsghoyt @ Feb 4 2010, 10:45 PM) <a href="index.php?act=findpost&pid=1750770"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->The most important thing to me though is to remove any dependency on Steam. A lot of people love Steam and I used to but a lot of people have also been burned by Steam. I bought GTA4 from Steam and it didn't run so I asked for a refund and they refused. I got my refund through the credit card company and lost my entire account.<!--QuoteEnd--></div><!--QuoteEEnd--> That's... That's pretty horrible. Although you did violate quite a bit of common sense there.
<!--quoteo(post=1750770:date=Feb 4 2010, 10:45 PM:name=chrsghoyt)--><div class='quotetop'>QUOTE (chrsghoyt @ Feb 4 2010, 10:45 PM) <a href="index.php?act=findpost&pid=1750770"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I've learned my lesson and will not use Steam again. Please make sure we can still have the full game experience without Steam.<!--QuoteEnd--></div><!--QuoteEEnd--> You don't need an account to play, you just need the client to be running (well, I think). On Linux, you might not even need that.
<!--quoteo(post=1750770:date=Feb 4 2010, 10:45 PM:name=chrsghoyt)--><div class='quotetop'>QUOTE (chrsghoyt @ Feb 4 2010, 10:45 PM) <a href="index.php?act=findpost&pid=1750770"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->There plenty of third party applications outside of Steam that allow people to find their friends and voice chat in any game not just NS2 so I think having those pieces missing is not so critical. The critical piece that I saw mentioned was the server list. Please make sure everyone can access the server list somehow even if it ends up just being a webpage with ip addresses on it.<!--QuoteEnd--></div><!--QuoteEEnd--> Server browser is pretty much the main reason they use Steamworks.
Comments
So, my comments...
Definitely open source the backend interface, because then people can write awesome backends!... That's less work for you guys, and it'll also get you guys a place on slashdot and the other nerd/Linux sites -- Linux nerds LOVE seeing games that work natively on Linux, it's what got me hooked on Heroes of Newerth... Not to mention the fact that writing our own backends gives we Linux programmers massive hardons... I know I'd contribute the the openGL implementation, along with maybe a lightweight implementation so that lesser computers can still play the game... and I also would love to try and write an AAlib backend :P
While I agree with some others that it should have been written in openGL to begin with, if you're more familiar with DX then who are we to judge the ones making this awesome game ;)
I'm *pretty* sure you can put steamworks in Linux. If you put the steamworks stuff into a DLL and then link to it through wine, you can have a native Linux binary using windows stuff... That's how Google's pacasa works... I don't know about the fine details of having the linux openGL stuff having windows drawing onto it too, but that should in theory be doable... Steam runs great under wine...
And, finally, why wxWidgets over Qt? Just wondering if Qt ever came up as an option, and if so why you decided to go against it... I'm a wicked Qt fanboy, even a core developer for libQxt! xD
The gist is not at all the same. Truform tessellation provided very little control, and subsequently very little benefit. It was fixed-function in the worst form. Programmable tessellation however, allows for everything from simple displacement (Unigine DX11 demo), to geometry images, silhouette refinement, improved hair rendering, even hardware accelerated SVG. The domain and hull stages are no more fixed-function than vertex and pixel/fragment stages.
<!--quoteo--><div class='quotetop'>QUOTE </div><div class='quotemain'><!--quotec-->Also, good job not starting a flamewar by insulting the opposition...<!--QuoteEnd--></div><!--QuoteEEnd-->
You make the mistake of assuming I've taken a side, I'm only opposed to misinformation.
My card supports it but it isn't DX11.
There's no word and little probability that NS2 will use tessellation in any form. We may be getting a more direct answer in the upcoming Q&A.
I would like to see more quick twitter posts on how things are going and don't be scared of using techie concepts and wording.
There still are a lot of unknown things holding me back from starting models/textures/map. Really a lot of this stuff belongs in a SDK document or mapping guidelines.
Textures and models:
Will transparency textures be used in shadow casting? Ambient Occlusion in real time or should we look at baking it into the textures belonging to models. Making parts or whole models not cast shadows? Parts of our models that emit light. Reflection? Will we be using precalculated reflection environment textures like hammer editor does? Some maps might require sky maps. Different lights such as area lights? Are we going to be using decals? Will decals work on models?
How far are you into finishing physics? So far the collision geometry of models is the same as the object. I can see we might need a way of using a simplified model for physics calculations. Will we be able to make objects that can be moved though physics. Making larger objects break apart from damage. Change gravity. Other forces to move players around.
Particles for various effects. Fog or some other type of effect for dust especially for lights.
It seems you are working on the first test map refinery. Will this be like an example map? How far is that completed and is it the only one currently?
Can skulks walk on all surfaces yet?
How far is the dynamic infestation finished?
Sorry about so many questions :)
For open sourcing parts of the engine, count me all in. Even though i have no open gl experience, i can certainly help to create the server browser parts etc. Really, if this is the only problem, i'm sure the linux community can help out!
i'm really looking forward to this
Thanks for taking time answering some questions Max. I would love to see some more from Matt or Bill by the way!
path finding... WOOT!
Dedicated Linux Server on release... WOOT!
Steamworks included... (does this mean all the NS2 pre-orders will be integrated to steam just like the NS2 orders bought in steam?)
If that the case then... WOOT! and how did you manage that deal?
<!--quoteo(post=1750408:date=Feb 2 2010, 06:38 PM:name=WatchMaker)--><div class='quotetop'>QUOTE (WatchMaker @ Feb 2 2010, 06:38 PM) <a href="index.php?act=findpost&pid=1750408"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->"...we wanted to make sure the technology was flexible enough that we wouldn't have to start again from scratch for the next game."
NS3 confirmed?<!--QuoteEnd--></div><!--QuoteEEnd-->
Maybe NS2:Episode 1 :P
No.. well not yet. Steam will allow it when they have a strong beta as they previously said.
Steamworks it only some multiplayer elements that they are relying to. Voice chat, etc. Like Max said.
It is interesting that you consider opening your backend interfaces, but do not even bother to answer your emails/PMs when people offer you guys to do the Linx/Mac/OpenGL-port for basically free... This would enable you to release the game for these platforms maybe together with the Windows-game and get you more publicity...
If they were to fix the ports for release then they'd had to rely on people outside the team which can be a liability. They rather do it thoroughly themselves, after the game is done (I can imagine). Having to greet, learn and work with some people who they don't pay and can't set any expectations with (hey, they're working for free right?) may just delay the release than help it.
NS3 confirmed?<!--QuoteEnd--></div><!--QuoteEEnd-->
nah not ns3 lol.
unknown world is not a company doing natural-selection only. right now yes, because they are still too short on staff but if they do the same good job than they did for ns1 they gonna be in serious business. (of course not like EA or ubisoft but ...)
+ if they got a good base its gonna be easier for "mod" to be created (if they accept ppl creating mod of course) that would be nice. fps/rts game are rare and its good for a company reputation and visibility.
talking about visibility : i hope you guys at unknown world will have a minimum of budget for the "right placed pub" by "right placed pub" i mean pub at the place the "real gamer" are going like on xfire. a game with 0 pub got always their server empty and the community never grow up
question : are you guys able to give an approximately date for the close beta ? something "general" like under 3 year or under 2 year or 1 or 6 month ?
me going and spreading the word...
I love you Boys :)
Even if us Mac and Linux users aren't vocal about it (we're not all like the dude above me), the effort is extremely appreciated.
If they were to fix the ports for release then they'd had to rely on people outside the team which can be a liability. They rather do it thoroughly themselves, after the game is done (I can imagine). Having to greet, learn and work with some people who they don't pay and can't set any expectations with (hey, they're working for free right?) may just delay the release than help it.<!--QuoteEnd--></div><!--QuoteEEnd-->
This is a lame excuse for not sending a simple "no".
What is the difference for you between opening the render backend interfaces so people can work on an alternative renderer and opening the render backend interfaces <i>earlier</i> so people can work on an alternative renderer?
Thanks for taking time answering some questions Max. I would love to see some more from Matt or Bill by the way!<!--QuoteEnd--></div><!--QuoteEEnd-->
That's an interesting point. However, it is safe to say that it's a fairly big leap from forming an independent game company and releasing something through online distribution to publishing for the XBOX. Even if that was the end result of the venture, I think that an OpenGL port would happen before something on that level of publishing support fell into place.
The most important thing to me though is to remove any dependency on Steam. A lot of people love Steam and I used to but a lot of people have also been burned by Steam. I bought GTA4 from Steam and it didn't run so I asked for a refund and they refused. I got my refund through the credit card company and lost my entire account. I've learned my lesson and will not use Steam again. Please make sure we can still have the full game experience without Steam.
There plenty of third party applications outside of Steam that allow people to find their friends and voice chat in any game not just NS2 so I think having those pieces missing is not so critical. The critical piece that I saw mentioned was the server list. Please make sure everyone can access the server list somehow even if it ends up just being a webpage with ip addresses on it.
I think anyone who is a hardcore PC gamer runs Windows. You just miss out on a lot of games if you don't. That said I would much rather prefer to run a game using Linux when I can.
That's... That's pretty horrible. Although you did violate quite a bit of common sense there.
<!--quoteo(post=1750770:date=Feb 4 2010, 10:45 PM:name=chrsghoyt)--><div class='quotetop'>QUOTE (chrsghoyt @ Feb 4 2010, 10:45 PM) <a href="index.php?act=findpost&pid=1750770"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->I've learned my lesson and will not use Steam again. Please make sure we can still have the full game experience without Steam.<!--QuoteEnd--></div><!--QuoteEEnd-->
You don't need an account to play, you just need the client to be running (well, I think). On Linux, you might not even need that.
<!--quoteo(post=1750770:date=Feb 4 2010, 10:45 PM:name=chrsghoyt)--><div class='quotetop'>QUOTE (chrsghoyt @ Feb 4 2010, 10:45 PM) <a href="index.php?act=findpost&pid=1750770"><{POST_SNAPBACK}></a></div><div class='quotemain'><!--quotec-->There plenty of third party applications outside of Steam that allow people to find their friends and voice chat in any game not just NS2 so I think having those pieces missing is not so critical. The critical piece that I saw mentioned was the server list. Please make sure everyone can access the server list somehow even if it ends up just being a webpage with ip addresses on it.<!--QuoteEnd--></div><!--QuoteEEnd-->
Server browser is pretty much the main reason they use Steamworks.