CodeGlitch0

My second idea is that there are a bunch of gerbals in a back room, powering the servers that chimpanzees are typing wildly into until a game emerges.

I'm a developer, you're a developer, she's probably a developer too! Let's face it, DU is born to attract the developer types to it. So let's start a (serious) wild speculation thread about what we think the back-end server architecture might look like! I have to imagine that since NQ is targeting a single-shard server structure the back-end will probably need to be broken up into micro-services. But since it is a real time game, it will also need to be as compact as possible with as few network boundaries to cross as possible. A server cluster, perhaps similar to Azure Service Fabric would do the job very well, with services partitioned out and distributed across nodes for density and automatic failover and scaling. First off, there will be a single point of contact for establishing and authenticating connections to the servers. The authentication service will check credentials and negotiation the session security keys and reserve a socket connection endpoint on the gateway layer for the client. The client will then connect to the specified endpoint with provided symmetric keys to prevent tampering. The gateway layer will be scaled as necessary to provide raw network throughput. At 5 KBps bandwidth per client, that means we get a theoretical maximum of roughly 200 clients per server until the gigabit network ports are fully saturated. Incoming packets (messages) are dropped into an event queue system for processing by the back end processing services. The regions are partitioned spatially using an octree algorithm (essentially a 3-dimensional binary search, where every cubic region of space is split in half on each dimension, into eight cubes, as necessary). The regions are separate services and are spread amongst server nodes for density and scalability. Each region is responsible for calculating physics on region objects and routing events between players in the region. A level of detail system is also in place for sending lower frequency important message across regions. As more players move into a region, the region is split cubically into sub-regions that redistribute amongst nodes in the cluster. As players leave the regions, they are collapsed back into the parent region to conserve resources. When the server cluster hits specified usage limits, additional nodes are added/removed from the cluster to scale up/down as needed. The service cluster framework (for example, Azure Service Fabric) is responsible for redistribution of partitions across server nodes and replication of services for failover purposes. Each of these scale units are also distributed geographically to maintain low latency, with a backplane in place to keep geographic regions in sync. That's my initial spitball idea for the architecture. Feel free to elaborate, correct, or share your own architecture ideas. Let's get a conversation going.

Skepticism does not equal hostility. Requests for transparency, disambiguation, and clarity also do not equal hostility. Like I stated in my last post, I get that the point of the KS is to generate additional funding. But we have no idea what the end goal is. I was merely explaining how $3M over the next 2 years is probably an absolute bare minimum required funding amount. I am being patient for an NQ response. I am merely responding to other responses, which may or may not be correct, in the meantime.. Thank you. I appreciate someone else with a healthy skepticism. I have also noticed such... enthusiasm. I believe it is largely due to the precedent that being active on the forums may lead to free alpha access. That is why I always take the word of non-NQ community members with a grain of salt. I also am pledged under the assumption that the claims alpha/beta will be free are correct, for now. But I'd really like to know with absolute certainty BEFORE they take my hard earned, non-refundable money.

I'm not prone to blindly drink Kool-aid offered by people I don't know personally. I am simply asking for an official (read: from NovaQuark) statement on whether or not alpha/beta will indeed be subscription-free or not. That is not an unreasonable expectation, considering they are asking (and I have pledged) 100 Euro for said alpha access to the game. Just because the alpha/beta standard in other games is free does not mean it is the same for DU, which will have *significant* server-side cost overhead compared to most games. Please forgive me if blind chants of "It is known. It is known." by community fans isn't enough for me, I mean no offense. $150k of $500k from kickstarter blown on always-on servers over two years only leaves a healthy $7,000 annual salary for each of the 25 employees, not including any of the other expenses. Yes, the point of the kickstarter is to generate klout for loans/investments from other sources. But even 2.5 million additional funding only barely covers a modest salary (if they average at 57k) for each of the current NQ team. Again, that doesn't even cover other expenses, such as office leasing. Without an additional source of revenue, the business model doesn't hold up very well over the course of 2+ years of additional development. That might get them to launch, but not if there are delays as there always are in development. Even developers have to feed their families. I would be ok with a partial monthly fee during alpha/beta, but only if they are up front about it. Again, this is not an unreasonable request from someone pledging financial support.

Yeah, it desn't look to exist on that Alpha Team Members thread. So, I'd still like a public acknowledgement/answer from NQ directly.

Ok, I'll look into that and post a link if I find it.

Yeah. Do you have an official source on it? I've only seen it describe as "Alpha Access to the game," which can be very ambiguous. My goal is to get an official answer. I'm backing at the gold level, so I want to to know with 100% certainty what will be happening. Not just conjecture.

A question for NovaQuark.... With the alpha/beta access to Dual Universe, will there be a monthly subscription requirement during this time? Or will it be free until release for those with access? Most people will probably assume it will be free (and I hope it will be also), but as a professional web/cloud developer I also know the costs of such an endeavor. According to the Kickstarter, as of today, there are 1400 or so people that will have access to the game alpha with an additional 1500 in the beta. This will require significant server investment from that point. I am assuming the server layout will likely be organized in cluster similar to Azure Service fabric, with region nodes distributed to partitions which are distributed across various servers in the cluster. Plus a gateway layer for connection negotiation, and a backplane to organize much of the cluster synchronization and backend storage and other microservices. This will probably be in the ballpark of 10 fairly decent servers (physical or virtual) to handle the networking traffic and all the back end services. If I had to estimate, I'd peg this around $4000-$6000 USD per month for the initial unscaled server cluster in a typical cloud provider. Will NQ just eat this cost for the 2+ years it will be in Alpha/Beta, or will we be expected to pay a monthly fee (or use our DACs) during this time as well as after release? EDIT: An answer!

My point was not that it is possible in real life, my point was that there is much prior art in Sci Fi for FTL communication. I'll also point to Star Trek's subspace communicators as another example. So there is no need for the delayed communication / communications relays that people are talking about. NQ can do what they want and there is lots of lore to choose from for FTL comms. I'm not sure you actually read my message... We are agreeing on the same point.

Realtime communication af any distance without relays is absolutely possible, especially in the realm of sci fi, but also in the realm of physics. theoretical, but demonstrated, i believe. It is called an ansible. Simplistically, it is a pair of quantumly tied particles that vibrate in sync, regardless of distance. Please, feel free to look it up. Edit: for those that don't feel like looking it up: https://en.wikipedia.org/wiki/Ansible and https://en.wikipedia.org/wiki/Quantum_entanglement

I wouldn't necessarily consider it a desired feature... Construct vs construct combat is the primary selling point of DU. It does strike me as concerning that it likely won't even be in game upon release unless the Kickstarter attains a stretch goal. Without it, DU is basically a glorified space minecraft MMO with fancy construct shapes beyond just cubes and pilotable transport devices. The stretch goal doesn't seem particularly likely based on the current inertia of the Kickstarter. I will wait and see what NovaQuark has to say in the AMA this week before forming any real opinions and decisions, though. But if construct vs construct combat is not a thing at release, that pretty much negates my entire reason for backing at the high-ish level I did: The goal being a head-start on ship and base defense/combat LUA scripting prior to launch.

I don't particularly care about RL money for game work. It just struck me as odd that the DAC system would promote builder work to earn RL money for NQ solely, with nothing for the people adding content to the game, besides free play. Perhaps it is because I am viewing in-game built items more like player-driven DLC. Using Steam Workshop as an example, there would at least be profit sharing. Using Second Life as an example (because it is by far the closest existing game model to Dual Universe), there are definitely well-off players that spend real-life money to contract developers to make things for them. I've seen it first hand, and DU seems like an excellent candidate for the same. There are those players who spend $1200 for an early copy of No Man's Sky and people who buy the $139.99 coin pack for free to play games. Somehow, people have money to blow on ridiculous things. I'm not necessarily sure I agree with this statement. The basement nerd who lives online might have distinct advantages over those (like me) with families and super busy lives. Time spent in game means more time to build and script bigger, better, and badder objects. I fear that DU might become a prime example of pay to win. If the builder decides not to sell blueprints regularly on open markets, but rather sell single copies to the highest bidder. If those items have distinct advantages over what is currently available, then simply buying DACs and selling to earn in-game money might be all you need to become the next tyrant leader until the next person buys something better and takes you down. Or perhaps the builder decides to become the tyrant for a time and later decides to offer their Death Star to the highest bidder on ebay. Just a couple of examples. Perhaps I'm just a cynic at heart, but I really hope that doesn't happen to DU. This topic was started mainly in hopes of promoting discussion around the topic in hopes NQ sees it and figures out a way to prevent it.

Yeah, it's not so much about DACs as a form of payment, so much as what seems a general pointlessness to the whole DAC system. Obviously, payment in-game would always be the in-game currency, and it would be up to the currency-holder (in this case, the builder) to decide if they want to purchase a DAC or not. It applies to resource gatherers / sellers just as much as builders. When I think about it a bit deeper, I suppose it does make some sense to have them in-game as both a way for NQ to make additional money (as they are more expensive than regular sub), but also as a way to dump currency to keep the economy flowing.

As a future developer in Dual Universe, the idea of DACs as payment seems like an interesting prospect. But the more I think about it, the more it seems like a bit of a d-bag move on the part of the developers. I don't want this to be an overtly negative post, but bear with me.. I intend to be creating a whole lot of stuff in game, ships, defenses, tools, etc. with full Lua scripting to back them all up. These things will take a TON of time to develop if you hope to have a decent result from them. However, with the Dual Access Coupons being the only *real* form of payment I can hope to receive in game, it kind of feels like a bit of a face slap. Think about it. I spend 25 hours working on a super awesome engine control module with advanced auto-maneuvering for someone who requested it custom. Joe Space Captain goes to the Dual website and buys himself two DACs to give to me as payment for the work I did. Great, I got this month and next in game for free as payment. = 30 Euro savings or whatever for me. Now, instead of a custom job, I make a super awesome base turret blueprint that I sell on open market. Let's say 250 people buy it at 500 credits each, and I trade those 125,000 credits for 5 DACs that people are selling on the market. What happened here? I now have 5 free months in game, which I may or may not ever use, but NovaQuark just made 75 euro in real money cash for all of that work that I did. Once I start collecting DACs like candy because now 10,000 users have realized how good it is and have purchased my super-awesome turret, NQ is rolling in the money for work I did, and those DACs I accrued have become virtually useless to me. Obviously, I made assumptions the monthly fee is 12 euro and DAC are 15 euro. Actual cost doesn't really matter. The more I make and sell on market, the more trivial the DACs will become. Meanwhile, NQ makes tons of constant value cash on my work. My alternative would be to just hoard the in-game currency, which has equally nil real-world value. I have to imagine that with this system of payments, serious developers and builders will eventually just start dealing outside of the game and make contracts to do in-game work with payments in real cash. That's the only way it would actually be worth it to build things over the long term. All of the really good stuff in game will only be accessible through various user-run black markets outside of game, with in-game market only used for trivial purchases just to get items transferred and to earn the in-game money needed for resource accumulation. I could be wrong, but that's what I foresee happening. What are your thoughts?

that is why they invented the internet.

This is it. Just about everything in game (that you'd actually care to protect) is the code for how things operate. If the devs have a creator protection on scripts, which they'd better or the game probably won't be worth playing, then it shouldn't be a problem. Copying the construct without the code would leave the copier with a nice ornamental paperweight. That's about it, unless they put in a ton of work to recode everything. That said, do we really need a legal system in game? Maybe the players could set something up if they can convince everyone to follow the rules. But I think that will be unlikely. I will be a builder/coder in game. And I intend to protect my creations. If people decide to copy my work and sell it to others, I will be willing to pay large sums to any mercenary who can prove they have terminated said copycat and destroyed everything they've built. Ruthless? Yes. But it should be a sufficient deterrent.

I read this suggestion and can't help but think of Second Life and the thousands of lame dance clubs with zero to three people in there because everyone wanted their own club and the mass number made them all nearly useless and boring. Don't want to be a downer, but I really hope this doesn't actually become possible. I can only see it taking away from the purpose of the actual game.

When you die, your will spawn back at wherever the nearest accessible resurrection pod is. That will likely be back at the arkship for quite a while in game. After time, you might make your own pod, maybe on your ship where you died, but that will require huge quantities of resources and power to maintain, plus a potentially significant recharge time for the pod. You will also lose some of your resource inventory likely, due to the differences in the other multi dimensional self that was resurrected. But my main point: Once you are dead, it is very likely that whoever killed you will likely scavenge everything you had at that moment, and the only thing left there of your base and ship will be a few piles of dirt.

The reason Dual gets meet with more skepticism is because the general population does not truly understand how proper utilization of a cloud backend would work. Most games and game reviewers and even game developers are stuck in the rut of shard instances, game lobbies, single servers. If you really want to understand, look into the current development of a game called Age of Ascent. They are making a space game capable of tends of thousands of people in a single space battle. I've been in their occasional play tests and the tech is amazing. They also have a LOT of documentation describing exactly how it all works to support that. Basically, it comes down to distributed clusters micro services with very specific purposes and an ultra high speed communication layer between them. Servers in the cluster can be spun up and down to meet demand and take over partial zone control and processing. Simply put, Dual Universe is the second game I've seen lately that aims to meet this target. The age of a single server game instance and player "lobbies" is dying or already dead. They just don't realize it yet.

There's a few good cloud providers I can think of, namely Azure, that have extremely good mobile support. If they develop the backend well and all of the Lua / DPUs run on the server, then it should be a simple matter of pushing an event name and a small bit of metadata to a queue. That would have minimal game impact (a couple network packets to a local service) and another service would costume and push the mobile notifications. It might work for define-your-own events, but I imagine there are only so many types of game events you would care about outside of the game: build complete, intruder detected, item sold/purchased, etc.

The closest thing to Dual Universe is actually Second Life. It isn't a space based game, but you can build things based on simple shape tools and script them with Lua to do anything you want. But objects have no resource costs, so if you own it you can build it. Mind you, there's no real "game" to SL. You can't die and it is a free for all that quickly became a cesspool of human corruption. I am hoping that Dual doesn't get to that point. It had hope because of the space aspect and the building limitations being imposed by the devs.

Since this game will be cloud based for the single shard system, I would love to see integration with a mobile app and push notifications that can be set up to respond to in game events, such as a perimeter radar that notifies you immediately if someone comes in range or breaches your compound or ship. That would provide the ultimate hook to lock me to the game. If I know someone is attacking my base, I would make every effort to be back on game as soon as possible to defend.

I personally love the idea of a common starting area. It would make it so much easier for a trader to set up a ship targeted to new players. I intend to establish early on where all noobs can find me easily for their first products and to buy resources in insane quantities.

The Lua scripting topic covered the flight stabilizer DPU as an example. I think that was meant as an actual example of a minor construct we would be required to build as players. It includes controlling individual thrusters, and taking input from the system DPU (such as a bank right command keystroke) to tell the stabilizer to lean appropriately to make the turn. So, I am extrapolating from that we will be able to make any type of flight control we desire. And the most optimally made will likely dominate in battle.

I would also add in angles and surfaces of design as a part of stealth. No hard angles or too much curvature would throw of radar signals at least, similar to the stealth bomber design.