Horizons A video for Frontier DEVS :)

[Kezha]

Said it before, The thing is lack off stuff within stuff

Ent was what made me interested in elite again and its nice to see someone whos not just blinded by moaning and ranting have valid points, and come up with valid good ideas in the same go, but wouldn't expect anything less from someone like ent ^^

+1!

 

[Garbarrage]

Mackwab could learn a lot from this guy.

 

[Jack Schitt]

It's a tough job, trying to draw attention to the need to revisit existing mechanics... the ones were added efficiently to fill a need, ostensibly do an "adequate" job of filling that need... but aren't really up to the scratch. Particularly for those customers who are frequently exposed to them... the frequent players, who quickly learn how everything actually works.

Relatively inexpensive "smoke-and mirrors" game mechanics will indeed tick off bullet points, and will be seen as pragmatic/realistic/achievable/sensible, etc. However, if left in place beyond their use-by date, they'll devalue the game world... because they feel cheap and nasty to the player. Unfortunately, an "onwards and upwards" design focus can leave things in place long term... things that weaken the foundations of subsequent game world construction phases.

I do agree with this, and the points raised by Ant in the video, but sometimes I worry that in trying to rush v1.00 to release in December 2014 FDEV may have painted themselves into a placeholder-filled programmatic corner from which it will be very difficult to escape. And that we may be seeing something of a repeat of this with Horizons.

I have no evidence for this beyond my own experience of course, and I fervently hope that time will prove me utterly wrong. But there's a nagging little voice at the back of my mind every time I play ED that whispers "If they could link all these elements together, wouldn't they have done it already?"

As others have pointed out, there are early signs with the branching missions in Horizons that FD may be trying to move things in the right direction, so it's a shame the current mission system is so broken because it's hard to tell what might be working and what isn't. Either way I look forward to this year's patches and version releases with interest. But I can't help but wonder whether they're starting from a more difficult position than would be ideal, brought on by the decision to cut so many corners when rushing to hit those prior release date(s).

 

[Pazmino]

His video can basically be summarised as "Please implement proper chained missions", and that IS something already planned by FDev. Hopefully it's high on their priority list, because it would make a massive improvement to the game (if used correctly).

Obsidian Ant was asking for a procedural generated missions tree on demand. With missions branching, choices and consequences on reputation, etc.

 

[Kiiyor]

Truly excellent points.

Stringing missions together will be a great way to make the universe feel... alive - something that is slightly lacking the game as it is. Your own internal narrative and imagination can only stretch so far to cover up the lack of believable human interaction. Pilots not thanking you for saving their lives is a perfect example of that. Sticks out like a sore thumb in RES points when you take on a wing of dangerous enemies to save a hapless type 9 that simply returns to their mining without a word after the debris clears.

Add in some randomization to the linked missions and you'd have a resounding success, IMHO.

 

[Kathryn Kando]

The video basically points at a weak spot in the FD team, we know about since a long time : good coders for game mechanics but no story tellers there to develop chained missions, good Galnet storylines, Powerplay content, etc...

 

[Deleted member 38366]

I'd translate and summarize the Video as "connecting the dots". Connect all those isolated mechanics islands that currently operate mostly separated from each other - leading to a lack of Sandbox-effects to at least create the illusion of having a Sandbox.

Said and suggested that since over a year, but so far no luck.

Currently, the single most turning-off issue I have with Horizons is : the lack of content.
To me, littered man-made junk every 4km on every Planet in the entire Galaxy isn't content. It's a pre-Alpha level placeholder.
Unfortunately this RNG-Generated spam is presently all we got.

I'm even sitting on a nice batch of Materials right now. Stuff others are stull hunting. I got it all.
Problem is : I have no use for it.
I initially had the idea of breaking my personal "high Altitude" Record above the SOL area of 1436LY and take Screenies from there - but found out I couldn't stand all those location-unaware, generic and nonsensical man-made POIs that are everywhere.
So I dumped the plan again, but can't even hand those Materials to someone who'd have good use for them.
It's also what entirely killed the prospect of Exploration for me, no point in heading out there and land on distant planets if I already know 100% that I'll see this kind of out-of-place stuff there.

I experimented with crafting Ammo etc. - but usually fighting just around the corner of a Station or Outpost, the benefit of that isn't worth the time needed to gather these Materials by a far margin.
I'd rather dock and come back in a few Minutes. Getting those Materials was a little bit more than just a few Minutes

That leaves Horizons with : including Planetary Bases for landing and just buzzing a Planet's surface just for kicks once in a while.
That's what Horizons means to me in its current incarnation.

PS.
I've never been a friend of scripted pseudo-content that doesn't even vary or always make sense in specific locations. But I'm afraid that's all we'll ever have.

 

[SentineI]

His video can basically be summarised as "Please implement proper chained missions", and that IS something already planned by FDev. Hopefully it's high on their priority list, because it would make a massive improvement to the game (if used correctly).

Obsidian Ant was asking for a procedural generated missions tree on demand. With missions branching, choices and consequences on reputation, etc.

Hmmm, OK. I guess he was asking for slightly more complexity than the most basic implementation of chained missions one could imagine. And knowing FDev, that's what the first version will probably look like :-/ .

P.S. I think the "really basic missions" problem that ED has suffered from since release is down to the fact that they do everything using database tables (so it can run on top of Amazon's distributed cloud-computing infrastructure). It's got to be quite hard to do fancy missions when you have to use fixed table structures for a wide range of different mission types, particularly when any bugs risk messing-up lots of player data.

 

[wstephenson]

P.S. I think the "really basic missions" problem that ED has suffered from since release is down to the fact that they do everything using database tables (so it can run on top of Amazon's distributed cloud-computing infrastructure). It's got to be quite hard to do fancy missions when you have to use fixed table structures for a wide range of different mission types, particularly when any bugs risk messing-up lots of player data.

If I may point out, some of the databases they use are NoSQL (MongoDB, about 22 minutes into that presentation), NoSQL databases strength and weakness being not being bound to the static schema that the databases we grew up with enforce. The missions are probably JSON documents that can have arbitrary structure (so this is why sometimes a class of missions in the BBS lacks an expected field, or has it twice due to copy and paste ). Now, did I get in before Pecisk?

EDIT: There's a good example of this at 20 minutes in where some of the JSON describing a FdL is shown:

Spoiler

[video=youtube_share;EvJPyjmfdz0]https://youtu.be/EvJPyjmfdz0?list=LLopBqe_8-v-4Ka70j41sCWA&t=1212[/video]

The factor that I think makes their missions a bit buggy is the vast range of possible input data coming from Stellar Forge, so they are adding post hoc sanity checks to make sure missions are doable.

 

[eqinox]

If I may point out, some of the databases they use are NoSQL (MongoDB, about 22 minutes into that presentation), NoSQL databases strength and weakness being not being bound to the static schema that the databases we grew up with enforce. The missions are probably JSON documents that can have arbitrary structure (so this is why sometimes a class of missions in the BBS lacks an expected field, or has it twice due to copy and paste ). Now, did I get in before Pecisk?

Yeah, they are using quite a number of persistence mechanisms, MySQL, MongoDB, DynamoDB, S3... I doubt the tech is really their problem though, unless you can attribute lack of persistence or randomness with trying to avoid too large databases leading to costly AWS infrastructure or something.

 

[Marc_Hicks]

Sadly, I am seeing a few "glass half empty" types on this thread...which is unfortunate. I've watched this game evolve for the last year, so am not remotely fearful that we'll be "stuck with place-holders". Also, having listened to the plans of the developers for the coming season, I don't think they lack the "story-telling nous" to link together all these disparate elements. They just wanted to make sure they had these elements in place-& working properly-before playing around with linking them all together.

 

[Marc_Hicks]

Based on what is already in the game, &what is planned for this coming season, this is how I believe we willsee an increase in “content” & “game-play” over season 2.Then I might indulge in a little bit of speculation in the finalpart.


Procedurally generated, fully-chainedmissions: This will increase content for exactly the reasons putforward by ObsidianAnt. This will be even more the case if thesefully-chained missions-& even one-off missions-can be obtainedmore often via Signal Sources, Points of Interest & Super-cruisecontacts. Lastly, if what happens when we go to complete thesemissions is more randomized, then you'll almost never find yourselfdoing the same group of missions.


Branched missions: The missions thatget offered to you to a faction's rivals do need a massiveoverhaul-with more variety in what they ask you to do to throw yourmission, the odds of a branched mission leading to a fully chainedmission, & the quantity & quality of the rewards they giveyou. That, coupled with what I said above, will massively improve thesense of having more “things to do” within the game. I do alsobelieve that there are still some mission types that are eithermissing, or incomplete-like Convoy & Convoy Raiding stylemissions & Interdiction/Scan missions.


Looting & Crafting: I believe thisactually does have enormous potential for increasing the “content”of the game, for several reasons. Firstly, the potential for massivecustomization of ships means that the chance of any 2 ship encountersbeing exactly the same will be much, much lower. Also, it will add awhole new set of missions to do, namely those involving the pickingup of modules & scrap metal from ship-wrecks (both land based &space based) on behalf of employers, & going out to simply findship-wrecks on behalf of an employer (for them to pick up later).Lastly, it gives an avenue for employers to pay you in somethingother than credits. The criminal organization that pays you for yoursmuggling with a module that allows you to remain in Silent Runningfor longer is better than just another 500,000Cr pay-out.


Points of Interest: I sincerely believethat if these-& the missions around them-follow a similarevolution as what we saw with signal sources, then they can become amajor strength in the 2^nd season game. Having the ship'sscanner being able to differentiate between Strong, Medium & Weaksignal strengths would help people a lot. So too would having surfacemissions spawn semi-persistant PoI's that will show up as thespecific “Mission Target”. Of course, adding more PoI's (bothnatural & artificial, persistent & transitory) would be greattoo. Natural geological formations that you can discover, &possibly even mine for rare gems & metallic ores, is a great casein point.


Avatar Creation: Again, I feel thiscould really help with game-play, as it could result in us havingpersistent & semi-persistent tier 2 NPC's, that we can interactwith-be it via Power Play or via the bulletin board....or just thatperson who gets in touch with you whenever you arrive at astation/outpost, to give you that special deal-or special job-thatyou can't get anywhere else.


Power Play: Admittedly this is a bit ofa grind right now, but if what Michael has said about making it moredynamic (i.e. the way in which powers rise & fall), & theaddition of Bulletin Board style missions as alternative ways ofgaining merit (& cash) could make Power Play one of the mostengaging parts of the game.


Background Simulator: Again, what I amhearing from Michael on this front makes me very happy. Getting moreobvious indicators of what the BGS is doing (without damagingimmersion) will help the game significantly (like economic statesmore clearly impacting the buying & selling cost of trade goods,or the security states having more impact on the likelihood of beingattacked in system-as well as these states being better represented,visually, within the system).


Military Missions: Once again, MichaelBrookes has indicated previously that he is looking to totallyoverhaul how the entire military side of the game works. I sincerelyhopes this means a return to something more like what we had in EliteII & Elite III! If so, then this will undoubtedly improve contentquite significantly.


I also disagree with the notion thatthe FD team lack the ability to make these happen. What I think we'veseen so far is them merely getting “all their ducks in arow”-coding wise-so that it will be much easier to join it alltogether later on.


Anyway, apologies for the excessivelength of the post, but I felt it was all worth saying. It will bevery interesting to see where things stand 1 year from now. Hopefullyall of my above predictions will have come to pass!

 

[Marc_Hicks]

Sorry, double post. My bad!

 

[dxm55]

The biggest factor now stopping me from actually buying ED:H (even more than the price point, in fact), is the lack of ability to land on an atmospheric world!

You could do it in Frontiers Elite II... without the need for an expansion. But I digress...
Take a look at the screenie below and imagine what it would be like in EDH's level of visuals.

Or how it's done in Infinity: Battlescape. Starts at the 2 minute mark.
https://www.youtube.com/watch?v=dp8RTHR0HWQ

The transition of the sky from space-black to blue, the re-entry effects. All this counts in the immersion factor.
Landing on barren rocks is meh ... But I'll bite when I can re-enter an M-class planet and see vegetation and a live cityscape under my ship.

 

[Marc_Hicks]

The biggest factor now stopping me from actually buying ED:H (even more than the price point, in fact), is the lack of ability to land on an atmospheric world!

You could do it in Frontiers Elite II... without the need for an expansion. But I digress...
Take a look at the screenie below and imagine what it would be like in EDH's level of visuals.
https://upload.wikimedia.org/wikipedia/en/3/37/Frontier_elite2_screenshot.gif

Or how it's done in Infinity: Battlescape. Starts at the 2 minute mark.
https://www.youtube.com/watch?v=dp8RTHR0HWQ

The transition of the sky from space-black to blue, the re-entry effects. All this counts in the immersion factor.
Landing on barren rocks is meh ... But I'll bite when I can re-enter an M-class planet and see vegetation and a live cityscape under my ship.

The graphics technology required to do atmospheric landings in a believable fashion is much, much greater than it was back in the early 1990's. I have no doubt that we shall see low atmosphere (less than 1 atmosphere) landings at the end of this year, & full atmosphere landings by the end of 2017. I'm patient enough to wait.

 

[nexxo]

And add to this that Frontier is doing it the hard way. Barren planets without atmosphere are actually harder to model because you cannot hide surface rendering flaws under vegetation and atmospheric haze.

We're not in the era anymore of smooth spheres, slap some oblongs on it for buildings and a few lines for roads and call it a planet.

 

[Ozram]

His arguments are very well thought out and it would be exceptionally good if FD would implement those ideas. +1 to ObsidianAnt.

 

[wstephenson]

Yeah, they are using quite a number of persistence mechanisms, MySQL, MongoDB, DynamoDB, S3... I doubt the tech is really their problem though, unless you can attribute lack of persistence or randomness with trying to avoid too large databases leading to costly AWS infrastructure or something.

I think you can, to a degree. Since they are reading from/writing to different databases all the time (e.g. BGS player actions to MongoDB, player progression/credits/CG actions to MySQL, session local state to DynamoDB, Redis, reading from StellarForge if all else fails) the game needs to have the mother of all persistence layers and cache control systems, and extending these to handle new structures (Persistent NPCs, for example) is going to be costly in engineering terms.

You don't lightly add dynamically evolving stations to the galaxy, when this will require exporting static values returned by StellarForge to hundreds of thousands of documents in a NoSQL db, and ensuring that nothing deadlocks when a player enters a star system for the first time, for example.

In the video above, Dav Stott mentions the exploration data set as potentially #systems * #players, and is happy that this is only ever written once - now consider the relative size of a persistent NPC data set, and the amount of processing to update this as things happen to them. Which is not to say that there is anything wrong with FD's backend, just that it was always going to be complex for a game of this size and adding new uses to it requires consideration of what can happen in multiple layers.

 

[wstephenson]

Procedurally generated, fully-chained missions: This will increase content for exactly the reasons put forward by ObsidianAnt. This will be even more the case if these fully-chained missions-& even one-off missions-can be obtained more often via Signal Sources, Points of Interest & Super-cruise contacts.

The 'workflow' for obtaining new mission chains/branches needs to be streamlined. Right now, dropping out of supercruise to receive an alternate branch is punishing (even if they did offer you a branch with a worthwhile reward). I'd suggest alt mission sources should be able to give missions in supercruise, if you target them, and it should be possible to Nav Lock to appropriate NPCs as with Wings, to allow private comms (or ambushes).

Branched missions: The missions thatget offered to you to a faction's rivals do need a massive overhaul-with more variety in what they ask you to do to throw your mission, the odds of a branched mission leading to a fully chained mission, & the quantity & quality of the rewards they give you.

I'd also like to see non-material rewards such as alt branches that give massive Influence or Reputation changes, or missions that will cause NPC factions to change their behaviour - to attack other factions, or stop smuggling in a system. Internally this would simply be modelled as a BGS action, but in gameplay terms it could result in a new CZ spawning or temporarily change the NPC/PoI spawn tables.

I do also believe that there are still some mission types that are either missing, or incomplete-like Convoy & Convoy Raiding style missions & Interdiction/Scan missions.

Hear hear - having a way for players to non-lethally oppose smuggling as a sort of Sheriff's Deputy would allow great opposed CGs.

Looting & Crafting: <snip> Also, it will add a whole new set of missions to do, namely those involving the picking up of modules & scrap metal from ship-wrecks (both land based & space based) on behalf of employers, & going out to simply find ship-wrecks on behalf of an employer (for them to pick up later).

Yup, I'd like Scavenger to be a worthwhile minor profession - just like in HardWar.

 

[Ozram]

On a side note: https://forums.frontier.co.uk/showthread.php?t=221498