What are your thoughts on end-game? The idea of multiplayer being a big part of the game, yet never interacting directly with players has had me thinking.
I have been listening to a few podcasts where this was discussed, and thinking this all through to a logical... game-design conclusion : (nothing real spoilered here, just thinking and sharing some thoughts for debate)
We will all be spending at least some of our collective time cataloguing planets, naming creatures... etc. But after naming 1-2 normal-ish trees, and a few armadillos... many of us may not bother 'naming' things unless they impress upon us some sense of 'wonder', or if there is some benefit in-game (credits, resources) for doing so. The game designers must know this... going through a clunky text interface is really immersion-ruining, so people will only take the time to put their stamp on things they are proud to have discovered.
Adding to this thought... we understand that ~90% of the planets we visit will be rather desolate... the resource gathering and space combat / trading to get from System A to System B will be spattered once in a while with encounters of gorgeously lush, life-filled worlds. But most of the time, we will be landing on barren rocks, maybe with a bit of basic life.
Wouldn't these 'Awesome" worlds / creatures be a big source of any cataloguing in the game, ie: all the 'cool' things we see and bother to stop playing + pull up the interface + type in a name for?
Wouldn't that mind-blowing planet be an oasis in the universe well worth sharing?
But we all know... >99.99% of the game will never be explored.
So... ATLAS is updated constantly with these discoveries (we don't know how exactly, but...)...
Couldn't the Endgame - read "black hole" at the center of the galaxy - give us an ability to search + navigate to the named systems, worlds, and creatures 'named' by our fellow travellers? Could that be the big 'reason' to keep going Hello games has hinted at in many of their interviews?
The idea of fast travel doesn't feel to me like it meshes with the developers' vision for this game, so I don't think that'll be it. It's an interesting idea though and you could see it really changing the dynamic of the universe as more and more people reach it.
This... isn't as much variety as it looks like at first. A bunch of these ships are palette-swaps of each other, with maybe one or two minor physical differences between the two. I get the feeling that after browsing the ship store a few times they'll all start to blend together.
It's my biggest fear with this game given the small team that's developing it. The engine, ambition, and design all seem great, but they could have used about 10x the budget for asset development. The design screams "play me for 300 hours," but the eerie sameyness might result in something more akin to a 10-15 hour play time.
I think the design elements that try to push you off of a planet after spending a few minutes with it are a clever way to help deal with this. The planets, being actually planet-sized, could provide silly numbers of hours each to canvas, especially if they have planetary-level biodiversity. But then you'd only need to explore a couple planets to see every major permutation the game has to offer. So, it either hides the fact that most planets are basically the same or that most planets only have a handful of things on them by making sure you don't spend too long on any one to peer under that hood.
I think my biggest hope is that this game is successful enough to spawn a true AAA version that drastically increases the number of asset pieces. The curve of resulting combination possibilities is very favorable as you continue to add such assets; each one gives you a greater return than the last in terms of variability. Of course there's probably a ceiling you can reach where it becomes impossible to add any more components without resulting in some really awkward-looking critters, so then you just move on to the next core skeleton on which to make even more asset pieces.
I'm imagining techniques like this applied to a AAA RPG like Skyrim where every time you start a new game and explore you encounter weird new monsters to fight, and it makes me really excited for this potential future of video games.
Fleur de Alys on
Triptycho: A card-and-dice tabletop indie RPG currently in development and playtesting
This... isn't as much variety as it looks like at first. A bunch of these ships are palette-swaps of each other, with maybe one or two minor physical differences between the two. I get the feeling that after browsing the ship store a few times they'll all start to blend together.
It's my biggest fear with this game given the small team that's developing it. The engine, ambition, and design all seem great, but they could have used about 10x the budget for asset development. The design screams "play me for 300 hours," but the eerie sameyness might result in something more akin to a 10-15 hour play time.
I think the design elements that try to push you off of a planet after spending a few minutes with it are a clever way to help deal with this. The planets, being actually planet-sized, could provide silly numbers of hours each to canvas, especially if they have planetary-level biodiversity. But then you'd only need to explore a couple planets to see every major permutation the game has to offer. So, it either hides the fact that most planets are basically the same or that most planets only have a handful of things on them by making sure you don't spend too long on any one to peer under that hood.
I think my biggest hope is that this game is successful enough to spawn a true AAA version that drastically increases the number of asset pieces. The curve of resulting combination possibilities is very favorable as you continue to add such assets; each one gives you a greater return than the last in terms of variability. Of course there's probably a ceiling you can reach where it becomes impossible to add any more components without resulting in some really awkward-looking critters, so then you just move on to the next core skeleton on which to make even more asset pieces.
I'm imagining techniques like this applied to a AAA RPG like Skyrim where every time you start a new game and explore you encounter weird new monsters to fight, and it makes me really excited for this potential future of video games.
Yes, agree. One thing that seems like it will be the routine is:
Touch down
Scan around
Encounter beast types A/B/C
Encounter plant types D/E/F
Shop ship types G/ H / I
Search for resources common to this solar system / planet
Search for structures for boosts
blast off for next planet
What I am hoping for is a steady progression of 'weirder, cooler, more dangerous beasts / plants / ships / resources / structures' the closer you get to the center of the galaxy. ie: you start at distance Z from core...
I'm fine with terraria-like progression through planets of similar distance from galactic core as various 'biomes' for creatures, ships, etc... ie very basic example ....
'green slime on planet A'
Blue bigger slime planet B
Red small quick moving slime planet C
Black slow moving, poison slime planet C
Once you get past a certain threshhold in the galaxy, where you are now closer than "Z distance", between X-Y distance from core, so now you encounter new entire types of aliens / ships etc...
I hope that will keep me moving forward naturally wanting to progress to see 'what else is out there'.
The developers have said they want you to move, explore, so you hope that moving from system to system really 'wows' you... there have been a number of times hello games demos have wound up surprising the devs with what they find... so we can wait + hope and see.
Either way, as you said it's looking like a huge leap forward in the application of this methodology.
Anybody see the pool? They flipped the bitch!
0
Options
RandomHajileNot actually a SnatcherThe New KremlinRegistered Userregular
The beauty of the design in this game is that they don't need a huge asset team. Hopefully the procedural generation will achieve more than any asset team could think of.
The beauty of the design in this game is that they don't need a huge asset team. Hopefully the procedural generation will achieve more than any asset team could think of.
The procedural generation is limited by the number of asset pieces that the generation has to work with.
Scroll up a bit to that big image of ships. Look at the first row, columns 4 and 6. Those two ships both share the same set of parts at the center of the ship, but they have slightly different parts on either side of it (col 6 seems to lack a back piece while col 4 has a shorter front piece). They each have a different coat of paint.
Now look at row 5, column 3. It's basically the ship you get if you shove the first two ships we just looked at together -- the front piece of 1,4 and back of 1,6, same center configuration, and another coat of paint. Now look right next to it at column 4. It looks like it has one piece in the center different, but they're otherwise identical (and even have very similar paint jobs).
There are many other ships that share similar configurations to these. 2,6 and 2,9 have lots of familiar stuff going on with the ships we just looked at. 7,8 and 7,9 are looking remarkably familiar here. And after looking at 7,8 we can jump to 7,4 to find a nearly identical ship, or to 3,1 and 3,6 which have nothing new to show us at this point. There are a number more we could look at from here, but I think you get the point.
That image only features 72 procedurally-generated ships, yet I don't think there's a single one on there that looks unique compared to the rest of them. In fact I bet you'd find pretty much all the major pieces used throughout the image by just looking at the first two rows, or 18 ships. Overall the image gives the impression that the game has 4 or 5 ships with a few minor options to tweak (or rather ways to mix-n-match those 4 or 5), which actually feels quite paltry.
Now, perhaps this image was generated using a small subset of their actual ship parts so that we'd still be surprised when we saw the game. But that seems ill-conceived; if they have enough ship parts that they could make 72 ships that all actually look different, then they'd have enough parts to generate hundreds of ships that look fairly unique (even if they do share a few aspects).
What's limiting them here is their set of assets that the procedural generation is using to create the ships.
Fleur de Alys on
Triptycho: A card-and-dice tabletop indie RPG currently in development and playtesting
I'm less concerned about the ship uniqueness, since from what I understand, you probably are never going to see another person the entire game. Your ship is going to be unique by default. I'm kinda worried about the differences in planets, since we've only seen one so far.
The beauty of the design in this game is that they don't need a huge asset team. Hopefully the procedural generation will achieve more than any asset team could think of.
The procedural generation is limited by the number of asset pieces that the generation has to work with.
Scroll up a bit to that big image of ships. Look at the first row, columns 4 and 6. Those two ships both share the same set of parts at the center of the ship, but they have slightly different parts on either side of it (col 6 seems to lack a back piece while col 4 has a shorter front piece). They each have a different coat of paint.
Now look at row 5, column 3. It's basically the ship you get if you shove the first two ships we just looked at together -- the front piece of 1,4 and back of 1,6, same center configuration, and another coat of paint. Now look right next to it at column 4. It looks like it has one piece in the center different, but they're otherwise identical (and even have very similar paint jobs).
There are many other ships that share similar configurations to these. 2,6 and 2,9 have lots of familiar stuff going on with the ships we just looked at. 7,8 and 7,9 are looking remarkably familiar here. And after looking at 7,8 we can jump to 7,4 to find a nearly identical ship, or to 3,1 and 3,6 which have nothing new to show us at this point. There are a number more we could look at from here, but I think you get the point.
That image only features 72 procedurally-generated ships, yet I don't think there's a single one on there that looks unique compared to the rest of them. In fact I bet you'd find pretty much all the major pieces used throughout the image by just looking at the first two rows, or 18 ships. Overall the image gives the impression that the game has 4 or 5 ships with a few minor options to tweak (or rather ways to mix-n-match those 4 or 5), which actually feels quite paltry.
Now, perhaps this image was generated using a small subset of their actual ship parts so that we'd still be surprised when we saw the game. But that seems ill-conceived; if they have enough ship parts that they could make 72 ships that all actually look different, then they'd have enough parts to generate hundreds of ships that look fairly unique (even if they do share a few aspects).
What's limiting them here is their set of assets that the procedural generation is using to create the ships.
That image was taken from a demonstration showing how the procedural generation system modifies parts of a "genus" to generate new "species." They showed similar examples with a few different kinds of animals. I expect that those were chosen sort of "sequentially" in terms of the generation to show the variation mechanics.
You will probably start recognizing the underlying genus in various animals, but as long as that doesn't happen to me until at least hour ~50, I'll be pleased with the investment. I plan to focus more on intergalactic trading than exploring, anyway.
I can't wait to be Commander Skullnumbers: Space Explorer(and sometimes Space Asshole probably)! Gonna stock up on adult diapers and cases of Ensure so I never have to leave my chair.
Everyone has a price. Throw enough gold around and someone will risk disintegration.
Ir reckon this is going to disappoint many people.
Yeah, I imagine it will capture a lot of people but also turn away quite a lot.
Also, have they talked about patching and adding content much? I recall that being a thing, but not sure where I heard it.
The beauty of the design in this game is that they don't need a huge asset team. Hopefully the procedural generation will achieve more than any asset team could think of.
The procedural generation is limited by the number of asset pieces that the generation has to work with.
Scroll up a bit to that big image of ships. Look at the first row, columns 4 and 6. Those two ships both share the same set of parts at the center of the ship, but they have slightly different parts on either side of it (col 6 seems to lack a back piece while col 4 has a shorter front piece). They each have a different coat of paint.
Now look at row 5, column 3. It's basically the ship you get if you shove the first two ships we just looked at together -- the front piece of 1,4 and back of 1,6, same center configuration, and another coat of paint. Now look right next to it at column 4. It looks like it has one piece in the center different, but they're otherwise identical (and even have very similar paint jobs).
There are many other ships that share similar configurations to these. 2,6 and 2,9 have lots of familiar stuff going on with the ships we just looked at. 7,8 and 7,9 are looking remarkably familiar here. And after looking at 7,8 we can jump to 7,4 to find a nearly identical ship, or to 3,1 and 3,6 which have nothing new to show us at this point. There are a number more we could look at from here, but I think you get the point.
That image only features 72 procedurally-generated ships, yet I don't think there's a single one on there that looks unique compared to the rest of them. In fact I bet you'd find pretty much all the major pieces used throughout the image by just looking at the first two rows, or 18 ships. Overall the image gives the impression that the game has 4 or 5 ships with a few minor options to tweak (or rather ways to mix-n-match those 4 or 5), which actually feels quite paltry.
Now, perhaps this image was generated using a small subset of their actual ship parts so that we'd still be surprised when we saw the game. But that seems ill-conceived; if they have enough ship parts that they could make 72 ships that all actually look different, then they'd have enough parts to generate hundreds of ships that look fairly unique (even if they do share a few aspects).
What's limiting them here is their set of assets that the procedural generation is using to create the ships.
That image was taken from a demonstration showing how the procedural generation system modifies parts of a "genus" to generate new "species." They showed similar examples with a few different kinds of animals. I expect that those were chosen sort of "sequentially" in terms of the generation to show the variation mechanics.
You will probably start recognizing the underlying genus in various animals, but as long as that doesn't happen to me until at least hour ~50, I'll be pleased with the investment. I plan to focus more on intergalactic trading than exploring, anyway.
Oh!
Well. Um.
Hooray!
Triptycho: A card-and-dice tabletop indie RPG currently in development and playtesting
To be clear, I am expecting Star Wars style worlds, with a single biome each, and about a dozen or so unique animals on each. If you have a thousand different base animals, and you recognize a genus (my word, not theirs) after about the third time seeing it, then that's still likely to be several hundred worlds visited before you you start seeing widespread duplicate genus. Likewise, if they have a hundred different ship genus and you only have a couple of options for new ships each system, you can expect to go pretty far before you recognize a duplicate. Considering the amount of time put into this game by even the small development team, neither of those two numbers seems unreasonably large to me.
Honestly, the part I'm most looking forward to is the seamless transition from surface to space. If that continues to be as smooth in the game as it looked in the demo, then I'm sold.
Apparently the disc size is only 6 GB (with most of that being audio), according to Murray. I don't know if there's a download in addition to that, but that's kind of nuts.
IS the game an always on the internet game? Most of the beef might be on site for them, and we just get a client. Or the download is going to be huge if just 6 gigs is on the disk.
IS the game an always on the internet game? Most of the beef might be on site for them, and we just get a client. Or the download is going to be huge if just 6 gigs is on the disk.
Or it's all just math, and that type of procedural generation doesn't take up much space.
IS the game an always on the internet game? Most of the beef might be on site for them, and we just get a client. Or the download is going to be huge if just 6 gigs is on the disk.
Or it's all just math, and that type of procedural generation doesn't take up much space.
There was this famous comment from a little while back...
What's the best preview to read right now on this? I need to decide if I want it or not and truth be told I just know soundbite descriptions of the systems.
Ah definitely, I'm just wary on if it will hold up. I fell for the Simcity 2013 fiasco, where everyone was given a short demo of the game, and it was insanely fun. But once you played for about 6-8 hours, you realized how bad it really was.
I am not sure if I am sold on the game yet or not but I cannot help but compare it to Spore. I find it odd that so few other people are doing the same because they seem so similar in the surface.
Posts
Well I guess if I spin the procedural wheel enough, that's the next best thing
Well I got served
I NEEDS IT NOW!
I have been listening to a few podcasts where this was discussed, and thinking this all through to a logical... game-design conclusion : (nothing real spoilered here, just thinking and sharing some thoughts for debate)
Adding to this thought... we understand that ~90% of the planets we visit will be rather desolate... the resource gathering and space combat / trading to get from System A to System B will be spattered once in a while with encounters of gorgeously lush, life-filled worlds. But most of the time, we will be landing on barren rocks, maybe with a bit of basic life.
Wouldn't these 'Awesome" worlds / creatures be a big source of any cataloguing in the game, ie: all the 'cool' things we see and bother to stop playing + pull up the interface + type in a name for?
Wouldn't that mind-blowing planet be an oasis in the universe well worth sharing?
But we all know... >99.99% of the game will never be explored.
So... ATLAS is updated constantly with these discoveries (we don't know how exactly, but...)...
Couldn't the Endgame - read "black hole" at the center of the galaxy - give us an ability to search + navigate to the named systems, worlds, and creatures 'named' by our fellow travellers? Could that be the big 'reason' to keep going Hello games has hinted at in many of their interviews?
Your thoughts?
It's my biggest fear with this game given the small team that's developing it. The engine, ambition, and design all seem great, but they could have used about 10x the budget for asset development. The design screams "play me for 300 hours," but the eerie sameyness might result in something more akin to a 10-15 hour play time.
I think the design elements that try to push you off of a planet after spending a few minutes with it are a clever way to help deal with this. The planets, being actually planet-sized, could provide silly numbers of hours each to canvas, especially if they have planetary-level biodiversity. But then you'd only need to explore a couple planets to see every major permutation the game has to offer. So, it either hides the fact that most planets are basically the same or that most planets only have a handful of things on them by making sure you don't spend too long on any one to peer under that hood.
I think my biggest hope is that this game is successful enough to spawn a true AAA version that drastically increases the number of asset pieces. The curve of resulting combination possibilities is very favorable as you continue to add such assets; each one gives you a greater return than the last in terms of variability. Of course there's probably a ceiling you can reach where it becomes impossible to add any more components without resulting in some really awkward-looking critters, so then you just move on to the next core skeleton on which to make even more asset pieces.
I'm imagining techniques like this applied to a AAA RPG like Skyrim where every time you start a new game and explore you encounter weird new monsters to fight, and it makes me really excited for this potential future of video games.
Yes, agree. One thing that seems like it will be the routine is:
Touch down
Scan around
Encounter beast types A/B/C
Encounter plant types D/E/F
Shop ship types G/ H / I
Search for resources common to this solar system / planet
Search for structures for boosts
blast off for next planet
What I am hoping for is a steady progression of 'weirder, cooler, more dangerous beasts / plants / ships / resources / structures' the closer you get to the center of the galaxy. ie: you start at distance Z from core...
I'm fine with terraria-like progression through planets of similar distance from galactic core as various 'biomes' for creatures, ships, etc... ie very basic example ....
'green slime on planet A'
Blue bigger slime planet B
Red small quick moving slime planet C
Black slow moving, poison slime planet C
Once you get past a certain threshhold in the galaxy, where you are now closer than "Z distance", between X-Y distance from core, so now you encounter new entire types of aliens / ships etc...
I hope that will keep me moving forward naturally wanting to progress to see 'what else is out there'.
The developers have said they want you to move, explore, so you hope that moving from system to system really 'wows' you... there have been a number of times hello games demos have wound up surprising the devs with what they find... so we can wait + hope and see.
Either way, as you said it's looking like a huge leap forward in the application of this methodology.
This is a clickable link to my Steam Profile.
http://steamcommunity.com/id/pablocampy
Scroll up a bit to that big image of ships. Look at the first row, columns 4 and 6. Those two ships both share the same set of parts at the center of the ship, but they have slightly different parts on either side of it (col 6 seems to lack a back piece while col 4 has a shorter front piece). They each have a different coat of paint.
Now look at row 5, column 3. It's basically the ship you get if you shove the first two ships we just looked at together -- the front piece of 1,4 and back of 1,6, same center configuration, and another coat of paint. Now look right next to it at column 4. It looks like it has one piece in the center different, but they're otherwise identical (and even have very similar paint jobs).
There are many other ships that share similar configurations to these. 2,6 and 2,9 have lots of familiar stuff going on with the ships we just looked at. 7,8 and 7,9 are looking remarkably familiar here. And after looking at 7,8 we can jump to 7,4 to find a nearly identical ship, or to 3,1 and 3,6 which have nothing new to show us at this point. There are a number more we could look at from here, but I think you get the point.
That image only features 72 procedurally-generated ships, yet I don't think there's a single one on there that looks unique compared to the rest of them. In fact I bet you'd find pretty much all the major pieces used throughout the image by just looking at the first two rows, or 18 ships. Overall the image gives the impression that the game has 4 or 5 ships with a few minor options to tweak (or rather ways to mix-n-match those 4 or 5), which actually feels quite paltry.
Now, perhaps this image was generated using a small subset of their actual ship parts so that we'd still be surprised when we saw the game. But that seems ill-conceived; if they have enough ship parts that they could make 72 ships that all actually look different, then they'd have enough parts to generate hundreds of ships that look fairly unique (even if they do share a few aspects).
What's limiting them here is their set of assets that the procedural generation is using to create the ships.
That image was taken from a demonstration showing how the procedural generation system modifies parts of a "genus" to generate new "species." They showed similar examples with a few different kinds of animals. I expect that those were chosen sort of "sequentially" in terms of the generation to show the variation mechanics.
You will probably start recognizing the underlying genus in various animals, but as long as that doesn't happen to me until at least hour ~50, I'll be pleased with the investment. I plan to focus more on intergalactic trading than exploring, anyway.
http://steamcommunity.com/id/pablocampy
http://steamcommunity.com/id/pablocampy
I can't wait to be Commander Skullnumbers: Space Explorer(and sometimes Space Asshole probably)! Gonna stock up on adult diapers and cases of Ensure so I never have to leave my chair.
Yeah, I imagine it will capture a lot of people but also turn away quite a lot.
Also, have they talked about patching and adding content much? I recall that being a thing, but not sure where I heard it.
Yep, I'm cautiously optimistic, but I lived through Spore, so... you know.
Well. Um.
Hooray!
Honestly, the part I'm most looking forward to is the seamless transition from surface to space. If that continues to be as smooth in the game as it looked in the demo, then I'm sold.
Or it's all just math, and that type of procedural generation doesn't take up much space.
There was this famous comment from a little while back...
obviously it's not ready for the big time until it can take up more space.
"60 bucks for this trash?! There isn't even a million lines of code! Rip-off garbage!"
edit: n/m, found one!
http://www.polygon.com/2016/4/1/11340322/no-mans-sky-preview-ps4
meh. it gave me a good idea what to expect.
PSN:Furlion