@Hugh, well, if you make a good testing infrastructure and own update server infrastructure, you'll be able to patch like Valve: http://dota2.gamepedia.com/Patches
Well, it all feels like a pointless discussion really. What still matters most is where this game wants to go next, not how. Depending on the former, one can decide on the latter.
Sometimes you have to change the 'how' in order to change the 'what' or the 'where to go next'.
That's just the sort of pseudo-philosophical nonsensical thinking that gets you nowhere in reality. Sounds very deep and insightful though. One could probably write a very lengthy yet incredibly vague newpost about it.
@FoxyUK Let's not confuse early-access-games, like Rust, with fully released games, like NS2. People have bought this game with the expectation, that the game is finished. People bought Rust, with the expectation that it would have frequent updates and the occasional gamebreaking bug.
It is NOT fair, to force the remaining NS2 population to be beta testers after they already committed money to the product, on the premise that they are paying for a full product.
I agree completely, that's why I suggested a copy of the model. Keep everything on the dev build and work little and often, keep everything away from main branch until its guaranteed (as much as is possible) to be stable. Maybe I didn't word my post very well, but I don't think there's anything particularly good about this news and I definitely don't like how fickle UWE appear right now by seemingly abandoning NS2 for the underwater thing then changing tact and resuming NS2 work.
That's just the sort of pseudo-philosophical nonsensical thinking that gets you nowhere in reality. Sounds very deep and insightful though. One could probably write a very lengthy yet incredibly vague newpost about it.
Nope, missed.
It's just that given the recent events, if the game doesn't go forward anymore, maybe it's time for doing things differently. Reverting back to b277 isn't much of a "forward move" is it ? Good thing this game is "confidential", it would have killed any other game (going nowhere...).
Bimonthly releases has always been perfectly sustainable pacing, for server admins&modders to keep things updated, and players to adapt and embrace gameplay changes. Community members interested in beta testing are already active within the CDT, playtesting, and other initiatives. Pushing beta builds publicly is a good idea, but it's very unlikely the community can handle high frequency mainstream updates; NS2 simply lacks enough actors to keep the operational side healthy.
Why not give the development team sufficient freedom to experiment with gameplay on the old schedule, and see where that leads? We've had a complete lack of gameplay changes, certainly not caused by a lack of technical skill of the community devs. They've done some impressive engine work so far. A mandate to work on gameplay would be all they need.
Seems like everything that needed to be said has been said, but I'll chime in with echoing:
What's the point? Are the periods between updates what are truly thought to be the "problem" with ns2? Or even a problem at all?
Seems to me like again being generally out of touch with the game and its players.... when Hugh posted that first blog back with an image from an airplane, I thought to myself, what a perfect analogy for the relationship between uwe and it's poor, bastardized game...
If anything "saves" the game, I'd be very surprised if it was this.
Wow - good posts to all! As a regular public server player I am encouraged to see all of the energy spent on this subject. Long read but interesting counterpoints.
My thoughts:
* Stick with the opt in beta model I see as an option on steam games.
* Get the servers to auto update. If not then maybe a schedule.
* Put out quick little patches once a week and big patches when they are ready.
Sounds like a bastardized form of Agile Development without considering any of its downfalls. TLDR version: you introduce new game breaking bugs that take forever to fix (or nobody wants to fix) while whatever players are left leave. Sounds like a good way to finally kill this game.
First of all, I've yet to hear a really compelling advantage to a continuous live build. What's the purpose? What, specifically, is being accomplished that couldn't be accomplished with, say, weekly builds? That's still pretty darn frequent by anybody's standards. It sounds like you're suggesting that this process is going to bring more people back to the game than the alternative, and I just don't see any good reason why.
I think what's being overlooked here is the drawbacks to live code updates beyond just stability and bugs. Let's ignore that and suppose that every change is bug-free. What about public awareness? How do you communicate what's happening with the game in a format that people will notice? If you just have a continuous commit log or something, people are going to tune that out immediately. People read changelog posts because they are infrequent and reasonably impactful. So when you get to the point that people stop reading the changelogs (if in fact there are any), then you'll have people who are caught off guard by balance and game design changes, or never even find out about them except for a vague feeling that something's different.
I'm not familiar with the inner workings of you guys' build system, but however it works behind the scenes, I strongly recommend that you bundle any and all gameplay changes into a weekly update with a changelog in the blog post. And tone down the blog post spam a bit while you're at it so that changelog doesn't get lost - one post per day tops. Small hot fixes going out live are fine if you're 100% certain it's safe.
By the way I did a quick spot check on the posts in this thread from people who expressed some opinion about this idea and this is my (rough) count:
For: 5
Against: 17
Neutral: 9
@Hugh please think VERY carefully about this before you push through everybody's objections and do it anyway. This just doesn't seem like an idea that many people are excited for, so I don't see much value in doing it. My thinking is that the power to do emergency hotfixes as needed is a very useful thing, but you should think of it as just that, and not abuse it by making the entire development process just a series of hotfixes. This is a small community at a very fragile time in the game's life, so alienating people at this stage poses a serious existential risk to the game itself.
I'm not familiar with the inner workings of you guys' build system, but however it works behind the scenes, I strongly recommend that you bundle any and all gameplay changes into a weekly update with a changelog in the blog post. And tone down the blog post spam a bit while you're at it so that changelog doesn't get lost - one post per day tops. Small hot fixes going out live are fine if you're 100% certain it's safe.
Perhaps its time to bring the Wiki out of hiding and use it for changelog documentation again. Actually, where did the wiki go? Wasn't it being 'redesigned' at some point?
My biggest concern is the compatibility with the major "mods". This HAS to be on your priority list. And it needs to be fairly high ranked. there are a LOT of servers that are defined by NS2+ and Shine (plus epsilon)... noone gives a shit if the snowballs in the RR don't work anymore or the gorge heal hearts break... (not for long and seriously) ... but these ankers of gameplay (NS2+, Shine, Epsilon) NEED to be in place reguardless of the Updates. Maybe make 'em vanilla?
On the proccess itself... It seems kinda pointless to me. you don't speed up development with it... you're just changing the time frame in wich things come out (and can go wrong) to a shorter one. In fact you slow it down by adding the update logistic in this smaler windows where you only had one very big one before. I also think that BIG updates are better for the game, cuz of the attention they draw. Smaller updates are for bigger games where you want them as unnoticable as possible. Maybe the "what" is indeed more important.
(off topic but awesome part)
I liked the Idea of DLC buildings, weapons and lifeforms. It needs to be fair priced (-.50 per thingie) and the others need to be able to pick stuff up i drop. hey... want an awesome tribal bladed Shotgun with golden shells?... *buys* *drops* there ya go... happy rookie ^^ .... but only stuff I drop follows my customization... so you can tell what comm built the macs and efed the armorywall up
Oh and maybe put in some acievments. so some are unlockable by gameplay. this motivates new players and vets alike, cuz "this is my new thing NOT everyone can buy for -.50" *proud* I did stuff
Why are people going on about "Agile Development"? Hugh never said anything about it, he talked about "rapid iteration".
Ghoul mentioned it but the whole buzz word is so vague, you can't really know what that means in the end.
Why are people going on about "Agile Development"? Hugh never said anything about it, he talked about "rapid iteration".
Ghoul mentioned it but the whole buzz word is so vague, you can't really know what that means in the end.
Just a bit of professional terminology leaking into this thread, in essence we're all talking about the same thing. Agile is one facet of a set of best practices to achieve rapid iterations (amongst other things). It just frames the discussion for the devs in here a bit.
DC_DarklingJoin Date: 2003-07-10Member: 18068Members, Constellation, Squad Five Blue, Squad Five Silver
Also another point.. No build numbers? Thats how I read it.
Thats an even bigger mistake.
I had the last 2 big patches, excluding the 'oops' patch and the rollback, the problem that steam did not have the latest patch yet.
Updating, hell even reinstalling the server from scratch would give me the 'latest' server download. Which was one behind the patch just released within the last lets say 3 hours.
The only 2 reasons this was apparent were the servers not showing up on my updated client & the server logs stating... yes the build number.
Imagine the pain without a build number. 'did you update?' "well yes I downloaded the latest from steam."
And for the folk who wonder, yes the problem in both cases was not on our side, and yes I bothered to ask cdt/uwe.
By the way, this post reminds me of a communist manifesto Let's make the world better by throwing out good-sounding buzzwords that all can agree upon (who doesn't want NS to be better?), but the whole idea seems to lead nowhere in reality. And the 'what' is vastly more important than the 'how'... Do you want pancakes or icecream for dessert? You have to have completely different toolsets for both...
I'm also glad to see so many enthusiastic pple on both sides :] This game is far from dead, even in the current state!
Only thing that I understand here as a developer (but for other things) is that small and fast iterations means you're able to for example change just the fade movement and thoroughly test just that. If you have a huge update in front of you, you have to be damn systematic and thorough not to miss something out during testing as well as be aware that when you throw tons of stuff in a game at once, they might conflict more often than if you add them faster with smal changes and testing it every time for just that specific thing.
Which means changes can roll out faster and are more reliable despite being rolled out more often (which would suggest opposite results).
Only thing that I understand here as a developer (but for other things) is that small and fast iterations means you're able to for example change just the fade movement and thoroughly test just that. If you have a huge update in front of you, you have to be damn systematic and thorough not to miss something out during testing as well as be aware that when you throw tons of stuff in a game at once, they might conflict more often than if you add them faster with smal changes and testing it every time for just that specific thing.
Which means changes can roll out faster and are more reliable despite being rolled out more often (which would suggest opposite results).
This is a solid point that I for one haven't been considering.
Dedicating new patches to singular balance changes to see how they change things is an interesting path.
I suppose this comes down to specificity, all we know at the moment is that the thought exists to run a "rapid iteration" route, we don't know what exactly we're iterating. Balance edits could be interesting to see rapid iterations of, though I gotta say that I'd get pretty tired of having a new game change every day to figure out how to exploit/counter/break. Even weekly balance edits would get old after two weeks.
Changes like that wouldn't affect much in the way of mods (presumably), and really would give pretty valuable insight on any impacts (at the cost of the average player's confusion).
Bigger engine/ui/balance changes iterating in small steps is really what's worrying.
Trust is the problem here, the community is still pretty unsure what to think when it comes to trusting UWE again. Communication before the crazy ideas hit the servers is key, and again, ideally people should be able to opt-out of these changes on the serverside and clientside.
EDIT: And if this communication feels as fruitless as replying to one of @Hugh 's posts, people who really care will start to just give up and leave. There's supposed to be two streams in a conversation. We really are interested in UWE's perspective on this, and if we're not being told it because they think we'll just shout and complain about it, doesn't that say something about how right they are (and thus that this might all be a bad idea)?
I play this game alot and I don't think it's fun most of the time lol @Deck_
Yeah I know what you mean. Maybe a better way to describe this game is, we play in the hope that we have that fun game. The quest for it, even if you go through a lot of crappy rounds to get there. The reason we come back is because you have those amazing rounds that trump every other game. I still think game-play wise this game is the best. If I didn't think that, I would have stopped playing.
I'm in favour of a blue sky research mod, if that is the goal. Making a game more fun is a many dimensional landscape with a few lonesome peaks in a vast desert. Hill-climbing will only get you to the top of the peak you have already found; it will not get you to potentially higher peaks in the distance.
The servers that opt into this mod know what they are doing, and the players who inhabit the server know what they're getting. A rapidly mutating, unbalanced mess with a series of wacky changes. The big problem is getting feedback rapidly enough to have something to iterate on; this is potentially unsolvable. Every major change it takes up to a few weeks for a new meta to be establish.
If you find the fun, but it is a hideous unbalanced mess; then you may be onto a winner. Incorporating any of the changes from this mod into the real game would be a long slog; you have to climb the peak to find out if it was higher than the previous one.
Ditto on what Soylent said: Servers running a test mod (or just one official testing server) can be a good way of exposing the wider community to potential changes before committing to them. Another advantage with a test mod would be that it could give modders a chance to see how their mods will interact with the game (at least, in theory). The only difficulty I can think of is that sometimes users will shy away from joining test servers, but I've seen the idea work before.
I play this game alot and I don't think it's fun most of the time lol @Deck_
Yeah I know what you mean. Maybe a better way to describe this game is, we play in the hope that we have that fun game. The quest for it, even if you go through a lot of crappy rounds to get there. The reason we come back is because you have those amazing rounds that trump every other game. I still think game-play wise this game is the best. If I didn't think that, I would have stopped playing.
Well I play the game with the hope that a much higher percentage of rounds will be fun. And @rantology is finally working on making that possible
I play this game alot and I don't think it's fun most of the time lol @Deck_
Yeah I know what you mean. Maybe a better way to describe this game is, we play in the hope that we have that fun game. The quest for it, even if you go through a lot of crappy rounds to get there. The reason we come back is because you have those amazing rounds that trump every other game. I still think game-play wise this game is the best. If I didn't think that, I would have stopped playing.
Well I play the game with the hope that a much higher percentage of rounds will be fun. And @rantology is finally working on making that possible
This game will always have a level of frustration. Any good competitive game should be frustrating when you lose and when things don't go your way. Can't have the sweet without it - but yes, it would be nice to have less annoying things in the game.
Sometimes you have to change the 'how' in order to change the 'what' or the 'where to go next'.
That's just the sort of pseudo-philosophical nonsensical thinking that gets you nowhere in reality. Sounds very deep and insightful though. One could probably write a very lengthy yet incredibly vague newpost about it.
Similarly, prefixing appropriate words with "pseudo" is a symptom of nonsensical sentence construction that also gets you nowhere.
Much like the people who talk about "pre-conditions."
Comments
Still more than 25 minutes.
That's just the sort of pseudo-philosophical nonsensical thinking that gets you nowhere in reality. Sounds very deep and insightful though. One could probably write a very lengthy yet incredibly vague newpost about it.
I agree completely, that's why I suggested a copy of the model. Keep everything on the dev build and work little and often, keep everything away from main branch until its guaranteed (as much as is possible) to be stable. Maybe I didn't word my post very well, but I don't think there's anything particularly good about this news and I definitely don't like how fickle UWE appear right now by seemingly abandoning NS2 for the underwater thing then changing tact and resuming NS2 work.
It's just that given the recent events, if the game doesn't go forward anymore, maybe it's time for doing things differently. Reverting back to b277 isn't much of a "forward move" is it ? Good thing this game is "confidential", it would have killed any other game (going nowhere...).
Why not give the development team sufficient freedom to experiment with gameplay on the old schedule, and see where that leads? We've had a complete lack of gameplay changes, certainly not caused by a lack of technical skill of the community devs. They've done some impressive engine work so far. A mandate to work on gameplay would be all they need.
What's the point? Are the periods between updates what are truly thought to be the "problem" with ns2? Or even a problem at all?
Seems to me like again being generally out of touch with the game and its players.... when Hugh posted that first blog back with an image from an airplane, I thought to myself, what a perfect analogy for the relationship between uwe and it's poor, bastardized game...
If anything "saves" the game, I'd be very surprised if it was this.
Guess we'll see.
My thoughts:
* Stick with the opt in beta model I see as an option on steam games.
* Get the servers to auto update. If not then maybe a schedule.
* Put out quick little patches once a week and big patches when they are ready.
* Remove the Halloween candy (like, today).
I think what's being overlooked here is the drawbacks to live code updates beyond just stability and bugs. Let's ignore that and suppose that every change is bug-free. What about public awareness? How do you communicate what's happening with the game in a format that people will notice? If you just have a continuous commit log or something, people are going to tune that out immediately. People read changelog posts because they are infrequent and reasonably impactful. So when you get to the point that people stop reading the changelogs (if in fact there are any), then you'll have people who are caught off guard by balance and game design changes, or never even find out about them except for a vague feeling that something's different.
I'm not familiar with the inner workings of you guys' build system, but however it works behind the scenes, I strongly recommend that you bundle any and all gameplay changes into a weekly update with a changelog in the blog post. And tone down the blog post spam a bit while you're at it so that changelog doesn't get lost - one post per day tops. Small hot fixes going out live are fine if you're 100% certain it's safe.
For: 5
Against: 17
Neutral: 9
@Hugh please think VERY carefully about this before you push through everybody's objections and do it anyway. This just doesn't seem like an idea that many people are excited for, so I don't see much value in doing it. My thinking is that the power to do emergency hotfixes as needed is a very useful thing, but you should think of it as just that, and not abuse it by making the entire development process just a series of hotfixes. This is a small community at a very fragile time in the game's life, so alienating people at this stage poses a serious existential risk to the game itself.
Perhaps its time to bring the Wiki out of hiding and use it for changelog documentation again. Actually, where did the wiki go? Wasn't it being 'redesigned' at some point?
On the proccess itself... It seems kinda pointless to me. you don't speed up development with it... you're just changing the time frame in wich things come out (and can go wrong) to a shorter one. In fact you slow it down by adding the update logistic in this smaler windows where you only had one very big one before. I also think that BIG updates are better for the game, cuz of the attention they draw. Smaller updates are for bigger games where you want them as unnoticable as possible. Maybe the "what" is indeed more important.
(off topic but awesome part)
I liked the Idea of DLC buildings, weapons and lifeforms. It needs to be fair priced (-.50 per thingie) and the others need to be able to pick stuff up i drop. hey... want an awesome tribal bladed Shotgun with golden shells?... *buys* *drops* there ya go... happy rookie ^^ .... but only stuff I drop follows my customization... so you can tell what comm built the macs and efed the armorywall up
Oh and maybe put in some acievments. so some are unlockable by gameplay. this motivates new players and vets alike, cuz "this is my new thing NOT everyone can buy for -.50" *proud* I did stuff
ya dig?
Ghoul mentioned it but the whole buzz word is so vague, you can't really know what that means in the end.
Just a bit of professional terminology leaking into this thread, in essence we're all talking about the same thing. Agile is one facet of a set of best practices to achieve rapid iterations (amongst other things). It just frames the discussion for the devs in here a bit.
Thats an even bigger mistake.
I had the last 2 big patches, excluding the 'oops' patch and the rollback, the problem that steam did not have the latest patch yet.
Updating, hell even reinstalling the server from scratch would give me the 'latest' server download. Which was one behind the patch just released within the last lets say 3 hours.
The only 2 reasons this was apparent were the servers not showing up on my updated client & the server logs stating... yes the build number.
Imagine the pain without a build number. 'did you update?' "well yes I downloaded the latest from steam."
And for the folk who wonder, yes the problem in both cases was not on our side, and yes I bothered to ask cdt/uwe.
I'm also glad to see so many enthusiastic pple on both sides :] This game is far from dead, even in the current state!
Im no gourmet by far, but I KNOW, its awesome with strawberries. :DD
But you still cant make icecream in a frying pan ;]
Really?
Which means changes can roll out faster and are more reliable despite being rolled out more often (which would suggest opposite results).
Really, FRYINGpan :P
This is a solid point that I for one haven't been considering.
Dedicating new patches to singular balance changes to see how they change things is an interesting path.
I suppose this comes down to specificity, all we know at the moment is that the thought exists to run a "rapid iteration" route, we don't know what exactly we're iterating. Balance edits could be interesting to see rapid iterations of, though I gotta say that I'd get pretty tired of having a new game change every day to figure out how to exploit/counter/break. Even weekly balance edits would get old after two weeks.
Changes like that wouldn't affect much in the way of mods (presumably), and really would give pretty valuable insight on any impacts (at the cost of the average player's confusion).
Bigger engine/ui/balance changes iterating in small steps is really what's worrying.
Trust is the problem here, the community is still pretty unsure what to think when it comes to trusting UWE again. Communication before the crazy ideas hit the servers is key, and again, ideally people should be able to opt-out of these changes on the serverside and clientside.
EDIT: And if this communication feels as fruitless as replying to one of @Hugh 's posts, people who really care will start to just give up and leave. There's supposed to be two streams in a conversation. We really are interested in UWE's perspective on this, and if we're not being told it because they think we'll just shout and complain about it, doesn't that say something about how right they are (and thus that this might all be a bad idea)?
Yeah I know what you mean. Maybe a better way to describe this game is, we play in the hope that we have that fun game. The quest for it, even if you go through a lot of crappy rounds to get there. The reason we come back is because you have those amazing rounds that trump every other game. I still think game-play wise this game is the best. If I didn't think that, I would have stopped playing.
The servers that opt into this mod know what they are doing, and the players who inhabit the server know what they're getting. A rapidly mutating, unbalanced mess with a series of wacky changes. The big problem is getting feedback rapidly enough to have something to iterate on; this is potentially unsolvable. Every major change it takes up to a few weeks for a new meta to be establish.
If you find the fun, but it is a hideous unbalanced mess; then you may be onto a winner. Incorporating any of the changes from this mod into the real game would be a long slog; you have to climb the peak to find out if it was higher than the previous one.
Well I play the game with the hope that a much higher percentage of rounds will be fun. And @rantology is finally working on making that possible
This game will always have a level of frustration. Any good competitive game should be frustrating when you lose and when things don't go your way. Can't have the sweet without it - but yes, it would be nice to have less annoying things in the game.
Similarly, prefixing appropriate words with "pseudo" is a symptom of nonsensical sentence construction that also gets you nowhere.
Much like the people who talk about "pre-conditions."
What UncleCrunch said was philosophical. Period.