Author Topic: Where to from here?  (Read 835 times)

PressureLine

  • XFM Developer
  • Administrator
  • Experienced Member
  • *****
  • Posts: 192
  • Likes: +4/-0
    • View Profile
    • Rawbots Club Forums
Where to from here?
« on: March 29, 2016, 11:49:50 am »
So... I'm really starting to hit a wall with 'unofficial' development in terms of "what is actually feasible." I'm sure there are still a few tweaks I can make here and there, but none of the truly fundamental aspects of the game are all that open to editing by me. I also have it on good authority that the multiplayer builds of Rawbots never really worked (not in an en-masse playable way), and I doubt any real work was ever done on a workable in-game economy.

So where to from here? Just play Rawbots for what it is, as it is currently? Keep trying to push XFM as far as I can? Re-build in Unity? Re-build on some other engine? Build a different game based on the same concepts with a totally different style?

I really don't know at this point :-\
"Never assume that anything you've never seen before is benign. Most particularly not something that says 'MARINES' on it."

MarvinMan

  • Automation Expert
  • Experienced Member
  • ***
  • Posts: 236
  • Likes: +4/-0
    • View Profile
Re: Where to from here?
« Reply #1 on: March 29, 2016, 02:37:59 pm »
Other than the NaN issues and deleting huge bots, the game as it is now with XFM installed is stable and works well.

Obviously, if you or other suitably qualified people had an unlimited amount of free time, re-building the game in a way that you can actually make fundamental changes to would be the best option. I don't know anywhere near enough about game development to be making useful comments on how you should proceed. Attempting to build a new version of the game seems like it would be a lot of work, considering how few active players there are these days.

cupid_the_conqueror

  • Global Moderator
  • Experienced Member
  • *****
  • Posts: 89
  • Likes: +0/-0
  • A filthy casual
    • View Profile
    • the truth of rawbots
Re: Where to from here?
« Reply #2 on: March 29, 2016, 03:37:47 pm »
honestly.. unless  Neil is willing to give you the source (under a nda ofcourse) then there isn't much let to do for Rawbots. We can't finish Neil's dream for him. Asking you to rebuild the game in different engines is overkill considering the amount of work you would face for such a small community.

I am already getting ready for hibernation mode, Rawbots content can only entertain me soo much ya'know? and I think eventually we'll all get bored. The webforum will slow down to maybe a post a week , and I am okay with that.

so long as this webforum exists, it represents the dim hope that rawbots may comeback 4-5 years from now. And if it doesn't, atleast we kept the hearth burning at home'ya know?
At the end of the day, love conquerors all.

PressureLine

  • XFM Developer
  • Administrator
  • Experienced Member
  • *****
  • Posts: 192
  • Likes: +4/-0
    • View Profile
    • Rawbots Club Forums
Re: Where to from here?
« Reply #3 on: March 30, 2016, 11:08:46 am »
...considering how few active players there are these days.

..such a small community.

I am already getting ready for hibernation mode, Rawbots content can only entertain me soo much ya'know?...

There's two separate issues here:

1: The 'Rawbots' name has shot it's bolt. I feel bad for Neil, but realistically you get one shot at doing it right,  and once a game's name becomes tainted with failure it's all over. I actually quite seriously doubt there will ever be a real resurgence of a playerbase with a game under the Rawbots name, simply because most of the people who had been interested in Rawbots in the past have moved on, and simply think of Rawbots as 'that half-finished abandonware robot game.'

2: There isn't enough to do in Rawbots to keep anyone but the most hardcore going for long. Particularly applicable for the more 'casual' style of gamer, who doesn't want to have to spend hours building their playground before they can play in it (imo, taking Stardust out of the 'base' game was a massive mistake, doubly so for the constructed 'starter' world that was in 0.0.9) The deep complexity of Rawbots is great, but it's wasted if 90% of the people who pick up the game get put off by the fact that there isn't really anything to play with when you first open up Rawbots 0.1.4, only the tutorial missions and Blueshift. Even though there are some bot blueprints lying around in the resources.asset file (which a 'regular' user would never encounter), as well as TrookLue's collection of bots in tut-practice there aren't really any simple examples that a brand-new user would be able to look at the code of and be able to say "Yeah, I can see how that works. Now I can use that to do -this-."

It should be more like playing Checkers, Chess or Go. You can pick the basic rules of how to play in under an hour, but you can spend the rest of your life mastering the subtleties of tactics and strategy. It doesn't mean 'dumbing' the game down, but it does mean setting the game up so that it can be enjoyed by players at any level of mastery (ok, any reasonable level, I remember reading through the posts Rawbots topic over on facepunch from one user who had trouble simply grasping the basics concepts behind how Rawbots VP works. There is a limit to how much the game is supposed to compensate for people who are technically adults but have such a poor grasp of grade school maths that they can't understand how to get the 1 output of an input_sampler to drive a wheel properly)

And then there's terrain crafting. Yawnfest intensifies. Simply put: It's boring. Sure, you can do some cool looking stuff, but at the end of the day, as a 'playable' surface you have two options, dirt and metal (fine, 3, there is grass too but imo it's ugly). Ice isn't that useful (I only use it for walls that visually contrast with dirt floors), the water hexes look awful because the hexes only almost line up, and the lava is more irritating (as a player) than anything else. And then Stardust disappeared (it's actually still there in the game files, but whatever) and while apart from the water planet all the other planet types are just "red-ish, brown-ish or grey dirt" in terms of physics, they at least offer something other than 'small brown hexes to the edge of the draw distance.' Ice and Lava planets would have been cool, and grey 'rock' hexes would have at least offered the opportunity for some variety. Then we could have added Sand hexes and planets (high rolling resistance, lowish friction), a grass terrain hex that looks more like a Minecraft 'grass' block (and why not a grassy planet while we are at it!) as well as an Ore block that you can plop the resource extractor on so it can do it's thing. In the game files there's also ground/ceiling features (sized to fit on the large hex_earth terrain, but scale can be adjusted) like stalactites/stalagmites, craters and crystal formations, as well as a random asteroid field generator (and asteroid art assets!). But ofc, none of that really made it into the released versions of the game.

And that's what kills me about the whole thing. There's all this stuff there, but it can't be used because none of it was converted for use with Bullet, and it's not really possible for me to shoehorn them back into the game. And the same goes with doing what should be simple, like making a clone of the small_hex_earth (and associated ramps) or a planet type with a different texture or physical properties. And to make it worse, someone comptetent adn familiar with the Rawbots code/assets and with access to the Unity project files could add everything I listed above in a few days, even someone who is barely competent could do it in a few weeks, and Rawbots would have been a much more fleshed-out game because of it (I wasn't going to even mention the bot parts that got left behind: body/armor parts, the pressure plate, two types of switches, the magnet, resource extractor and the crystal processor. But I will, and whoever made the decision not to include those parts moving forward should be shot, or at least kicked hard in the shins a few times)

I'm sure I had a point when I started all this, but I'm not sure it was a convincing one. [/rant]
"Never assume that anything you've never seen before is benign. Most particularly not something that says 'MARINES' on it."

PressureLine

  • XFM Developer
  • Administrator
  • Experienced Member
  • *****
  • Posts: 192
  • Likes: +4/-0
    • View Profile
    • Rawbots Club Forums
Re: Where to from here?
« Reply #4 on: March 30, 2016, 11:16:25 am »
I'm sure I had a point when I started all this, but I'm not sure it was a convincing one. [/rant]

Remembered my point now 8)

There is obviously a market for Rawbots-type games, even otoh I can list a few reasonably popular 'sandboxish contruction games' as well as a few programming games (including a few that when at their 'prime' were very popular) So it's not an issue that 'Rawbots was always doomed because it is too niche' but as I posted above, development simply lost traction (especially after the always-doomed kickstarter, because lets face it, 300k was never going to happen) for the want of someone spending a few hours a week prepping the existing content that hadn't been fully implimented as a way of keeping interest (and sales!) up.

*sigh*
"Never assume that anything you've never seen before is benign. Most particularly not something that says 'MARINES' on it."

Tajin

  • Member
  • **
  • Posts: 6
  • Likes: +0/-0
    • View Profile
Re: Where to from here?
« Reply #5 on: March 30, 2016, 02:23:26 pm »
Kinda agree with you on that. The RawBots name is most likely tainted beyond repair.
A reboot under a different could be a viable solution. It should stay true to its roots but it would have to be very clear that it is meant as a fresh start and that the mistakes of the past are aknowledged and that it aims to avoid making the same mistakes again.



All in all I think the main reason it failed was its presentation and the order in which it was developed where added.

Number 1 priority should've been to provide some basic gameplay. Something that allows people to get started, play around and have something to do. Doesn't need much complexity at that point, just a basic idea what it's aiming to become. A simple but solid foundation that you can keep adding things to.
They decided to do it pretty much the other way around and added complexity before actually implementing things to do.

I can understand that with a game like RB it was hard to draw a line where to stop adding complexity and start coding actual gameplay. Complexity is the nature of the game and coding those things surely was more interesting to the devs than working on the bread and butter of gameplay.




Anyway, the "where to" largely depends on what Neil has planned for it.
I think it would be wise to reboot it (as mentioned above) and put a somewhat different focus on it.
For example: Salvage the usable parts and rebuild it as an RTS with a basic built-in AI.
Once the RTS part is playable, start adding complexity and bot building to it. At some point allow people to replace the default (dumb) AI with their own.

More importantly... multiplayer has to be included from the ground up. No matter how simple a sandbox game is, it always becomes a lot more fun in coop or pvp. It also makes building a community so much easier and that goes a long way.
« Last Edit: March 30, 2016, 02:33:43 pm by Tajin »

cupid_the_conqueror

  • Global Moderator
  • Experienced Member
  • *****
  • Posts: 89
  • Likes: +0/-0
  • A filthy casual
    • View Profile
    • the truth of rawbots
Re: Where to from here?
« Reply #6 on: March 30, 2016, 05:33:09 pm »
The kick starter was a massive fuckup that pushed investors away. They should have asked for only 30K like TerraTech did (TT's kickstarter was there only to prove they could sell the game to players) And like TerraTech, the parts of the game should have spawned in pre-coded with basic behaviors (so when you drop for motors in a box formation they know to act like car wheels) which could have been altered when the player was competent enough to do so.


But I want to speak my mind on why the development made so many mistakes, and completely failed to capture their market. they were from SALVADOR 



Look at this small speck of a "nation", the national language is Spanish! Who on earth would think that somebody from SALVADOR could properly develop a game for North Americans and Europeans?

Go on vimeo and watch the couple videos these guys posted. Even when they are talking about the "vision" they have for their product, there is no lust behind it! its halfassed as hell!  They were a pack of bloody idiots. AND WHEN THE GOING GOT TOUGH TO THEY PACKED  UP AND LEFT! LOL! 

The fact that Neil didn't choose to invest his 250k in a local North American software team still blows my mind.
At the end of the day, love conquerors all.

z26

  • Experienced Member
  • ***
  • Posts: 108
  • Likes: +0/-0
    • View Profile
Re: Where to from here?
« Reply #7 on: March 30, 2016, 07:56:38 pm »
well, the the dev team behind kerbal space program is not from a game company, but a marketing firm from Mexico, so I'm not sure the nationality is the actual problem.
« Last Edit: March 30, 2016, 08:16:58 pm by z26 »

PressureLine

  • XFM Developer
  • Administrator
  • Experienced Member
  • *****
  • Posts: 192
  • Likes: +4/-0
    • View Profile
    • Rawbots Club Forums
Re: Where to from here?
« Reply #8 on: March 30, 2016, 09:50:13 pm »
Lower costs (both wages and floorspace) outside of the US are also a positive factor in deciding not to develop in the US. Plus countries like San Salvador also often offer grants and other financial/tax incentives for tech companies to set up shop in their country.
"Never assume that anything you've never seen before is benign. Most particularly not something that says 'MARINES' on it."

PressureLine

  • XFM Developer
  • Administrator
  • Experienced Member
  • *****
  • Posts: 192
  • Likes: +4/-0
    • View Profile
    • Rawbots Club Forums
Re: Where to from here?
« Reply #9 on: April 03, 2016, 12:23:56 pm »
Great post Tajin, I can't believe I totally missed it before (serves me right for only checking on my phone!) And you raised a few good points about complexity and when it got added in the development cycle.

My personal read on it is they got a bit too carried away with the emergent gameplay concept, where they playerbase would provide most of the 'fun' things for players to do while Rawbots itself functions as little more than a venue, so the focus became more on 'increasing' (used with a sense of bitter irony, since as time went on at least as many things got taken away from the game as were added, especially in the case of bot parts which is particularly inexplicable) the capabilities of Rawbots rather then providing a 'base' gameplay to emerge from.

The multiplayer point is a valid one, and at the very least if the game doesn't actually include MP in it's alpha/early beta versions it needs to be taken into account, something which doesn't seem to have been done all that much with Rawbots. The whole way Rawbots works, with it's multiple hierarchies and the way things are linked between the VP code 'world' and the physical worlds to me doesn't seem to be all that well set up for multiplayer (this is borne out by the sources I have claiming that while there are/were MP enabled builds of Rawbots, they never really worked) The alternative is a really good, engaging singleplayer game experience which Rawbots also lacks. I tend to see Rawbots (as it is now) as being more along the lines of those bridge-building games, or even like the old Incredible Machine games, but missing the essential 'game' aspect of it by only including the limitless sandbox mode (this ties in with my point above) which really sucks away the longevity of what gameplay there is.

Quote from: ZeroPunctuation
https://www.youtube.com/watch?v=4wgQvij3rVE

Give yourself a project. You have to find your own entertainment, whether that be skipping through the wilderness picking flowers or building an eighty-foot golden ####...
 
...But then again, part of the appeal of building is the effort required to gather materials. That eighty-foot golden #### wouldn't be so satisfyingly turgid if it weren't for the entire continent you had to turn upside-down to find all the ore. And that's what sets it apart from other creativity toys like Garry's Mod. If you can just clap your hands and summon fifty explosive barrels to pile ragdolls around, then the spectacle is as fulfilling as eating your own snot.

Thinking in terms of a 'reboot' or more a 'do a similar thing with a different name' is more or less the only real way to go imo. As I said at the start of the topic, the Rawbots name is tainted with fail:
  • Failed kickstarter
  • 'game ded' posts all over Steam, social media (Facebook, Reddit mainly) and forums
  • Loads of old 'reviews' (that were apparantly based solely off the trailer and/or other marketing material) but nothing new or relevant
  • The overwhelming response of user reviews/reactions (where available) is that the game is too hard to get into even though once you do get the hang of it the game really does open up

With regards to further development of Rawbots in-situ (continuing development the existing game essentially as-is in Unity) I'm really not sure about two main things (bearing in mind that I am not a professional programmer, and have strictly limited experience with working in Unity):
  • How much more is realistically possible
  • How much more is realistically worth doing
"Never assume that anything you've never seen before is benign. Most particularly not something that says 'MARINES' on it."

MarvinMan

  • Automation Expert
  • Experienced Member
  • ***
  • Posts: 236
  • Likes: +4/-0
    • View Profile
Re: Where to from here?
« Reply #10 on: April 03, 2016, 01:04:59 pm »
I suppose in the old days the devs did run official challenges with real prizes, which you could consider to be a more game like element.

Given how open ended the potential solutions are, I suspect it would be quite difficult to implement actual objective based gameplay in a way that works reliably.

While I wouldn't want to lose the current unlimited sandbox mode, adding some sort of part cost/limited power/fuel requirement type systems would add a whole now layer to gameplay. Making the terrain more interesting and varied (and much harder to navigate) in combination with some collectibles/achievements as incentive to explore could add significantly to gameplay with relatively little change to the base game.

Regarding multiplayer, it would be nice to have, but I don't see the physics engine handling it well, seeing the problems it has with running a couple of larger bots or running on slower computers.

PressureLine

  • XFM Developer
  • Administrator
  • Experienced Member
  • *****
  • Posts: 192
  • Likes: +4/-0
    • View Profile
    • Rawbots Club Forums
Re: Where to from here?
« Reply #11 on: April 03, 2016, 03:00:25 pm »
I suppose in the old days the devs did run official challenges with real prizes, which you could consider to be a more game like element.

Given how open ended the potential solutions are, I suspect it would be quite difficult to implement actual objective based gameplay in a way that works reliably.

The challenges are a good idea, but I see them as more of thing to extend gameplay, like achievements in some games (the garden gnome achievement in Half Life 2 Ep 2 springs to mind) and not really a replacement for gameplay. At it's heart, Rawbots is a game where once you are gud@rawbots you can do some amazing stuff, but the only way to become gud@rawbots is to play a lot of Rawbots (a certain amount of 'predisposed to be gun@rawbots' is there too) Which is an investment in time and effort that a lot of players simply aren't going to to be willing or able to put in. In most games, even if you aren't very good at them you are still able to enjoy the game (especially in SP or non-competitive MP) but if you aren't good enough (and/or are unable/unwilling/disinclined to spend the time/effort in learning the intricacies of learning to VPcode) at Rawbots coding to participate in something like the ACL, or even to create control systems for the flying or vtol challenges there isn't much you can do.

While I really love the Rawbots VP system, I just feel it doesn't have the 'difficulty' scaling that a written language that can easily have a gui to construct code (either internal to the game, or as a separate IDE editor) like the RobotC gui for the VEX robotics kits for use by 12 year olds:



Coupled with 'example' scripts to control bots with 'common' layouts like the robot rovers in Take on Mars:




Players who are at a 'user' level can enjoy the game without having to get all the way into the details of programming (especially if the stock scripts provide all the basic functionality) without resticting the ability of those who can and do want to write custom code for their 'outside the square' custom robot designs.

Ofc, from the very start there needs to be some thought put into not only providing 'sandbox without limits!' but also game aspects like resource extraction & management and bot energy management, as well as things AI mooks for SP players. Things that aren't terribly difficult to have in a game, provided that their inclusion (or possible inclusion) is integrated from the time the first line of code gets written.

Another thing to provide for right from the start is mods/modding. What good is a game based around custom player-built robots without the ability for (sufficiently proficient) users to create their own custom parts, or rebalance the 'base game' parts to their liking?
"Never assume that anything you've never seen before is benign. Most particularly not something that says 'MARINES' on it."

z26

  • Experienced Member
  • ***
  • Posts: 108
  • Likes: +0/-0
    • View Profile
Re: Where to from here?
« Reply #12 on: April 03, 2016, 03:17:38 pm »
That gui looks good, its basically a text based language you can use in graphical chunks.  Kinda blurs the line between visual and text languages and should be easily converted from one to the other.
« Last Edit: April 03, 2016, 03:21:58 pm by z26 »

MarvinMan

  • Automation Expert
  • Experienced Member
  • ***
  • Posts: 236
  • Likes: +4/-0
    • View Profile
Re: Where to from here?
« Reply #13 on: April 03, 2016, 03:42:03 pm »
I don't think the VP system is too hard to understand, the problem is that the game just drops you on an empty planet with no objective or instructions. I think the best way to improve on this would be to have a proper tutorial that shows the user how to make a basic wheeled/flying platform, and to allow blocks of pre-made code to be saved and shared in custom VP hexes (or possibly integrated into parts in the case of control systems).

The current VP system and more conventional sequential code have their own advantages and disadvantages. The VP system is good at handling lots of control loops in real time, but implementing branched or sequential behaviors can get messy very quickly. Conversely, sequential code is well suited to implementing complex behaviors, but would have problems dealing with the control loops for multiple actuators. Ideally, the game would keep aspects of both, allowing VP hexes with custom sequential code to be created, or allowing mathematical relationships to be defined at the start of the code for the engine to solve and update on each time step.

Regarding content, perhaps having an easily accessible library of bots, pre-coded parts and terrain structures would be a good idea.

cupid_the_conqueror

  • Global Moderator
  • Experienced Member
  • *****
  • Posts: 89
  • Likes: +0/-0
  • A filthy casual
    • View Profile
    • the truth of rawbots
Re: Where to from here?
« Reply #14 on: April 03, 2016, 06:17:57 pm »
I got it! All we have to do is insert Rawbots , Homebrew/scrapmechanic and minecraft into a blender and then BLAM, we have the game we want :P
At the end of the day, love conquerors all.