My thoughts on E:D and always-on DRM for single player
- Thread starter yamalight
- Start date
Wonder if Minecraft users wanted refunds when they realised they needed a network connection to play (even in single player)?
The argument used in the video is specious, silly even. It all just sounds like someone drumming up traffic by pandering to paranoia and nerdrage, and even plumbing the depths of character assassination (at least by implication).
I listed to almost all of it...but I wasn't prepared to follow the narrator down any of his rabbit holes of conspiracy theorising and speculation. So I started writing this instead.
The assertions used as a basis for the argument, that always-online must mean either:
What about a single-player game where the content each player receives is based upon the combined interactions of previous players, such as Akinator? Is that DRM? Whilst it could be done in a offline way, it would be completely predictable in it's responses. It's totally a single player experience, but the worldspace you navigate (with your questions) is shared.
How about something based upon an interaction with a chatbot that learns about the world from it's interlocutors? Even the Turing test is a game, of sorts. Is that DRM?
If the real-time twitch gameplay in a game is instanced and P2P, whilst all permanent non-combat interaction between the player and the shared world is server-driven, there's nothing logical about implying that transaction management is not required to play the game.
There is significant amounts of game content stored on and manipulated by the server(s), and an internet connection is required to access that.
The whole bit about offline mode has been done to death, and I fail to see what repeating those arguments in a Youtube video achieves other than giving the poster page hits.
If this game is DRM, then my email software is too, because I have to have an internet connection to receive and send new emails. And I think everyone can see that argument is laughable.
I listed to almost all of it...but I wasn't prepared to follow the narrator down any of his rabbit holes of conspiracy theorising and speculation. So I started writing this instead.
The assertions used as a basis for the argument, that always-online must mean either:
- having everyone in the same instance, or
- team-based e-sport gameplay
What about a single-player game where the content each player receives is based upon the combined interactions of previous players, such as Akinator? Is that DRM? Whilst it could be done in a offline way, it would be completely predictable in it's responses. It's totally a single player experience, but the worldspace you navigate (with your questions) is shared.
How about something based upon an interaction with a chatbot that learns about the world from it's interlocutors? Even the Turing test is a game, of sorts. Is that DRM?
If the real-time twitch gameplay in a game is instanced and P2P, whilst all permanent non-combat interaction between the player and the shared world is server-driven, there's nothing logical about implying that transaction management is not required to play the game.
There is significant amounts of game content stored on and manipulated by the server(s), and an internet connection is required to access that.
The whole bit about offline mode has been done to death, and I fail to see what repeating those arguments in a Youtube video achieves other than giving the poster page hits.
If this game is DRM, then my email software is too, because I have to have an internet connection to receive and send new emails. And I think everyone can see that argument is laughable.
I'll go with the arguments part, and keep this dry. Am happy to discuss points, as long as we're okay keeping this civil.
The argument about email is a strawman, because your email software can be used to connect to a corporate intranet, or your own personal server on a lan, or even a virtual server on your own PC. Strictly speaking, Akinator could be done in exactly the same way by providing a server API key and then downloading a database with a normalised and compressed database with it's "knowledge sphere" that handles all the local questions, though considering even compressed you'd still be staring at the better end of multiple terrabytes, it's an exercise in theory, not practicality.
The argument about significant sums of data being moved about by the server is the fallacy of false equivalency, as a singleplayer game would not require the same level of abstraction as required by a multiplayer database server, you'd use a much higher (lossier) level of abstraction for much greater levels of the database, which would mean more of it would occupy far less space, and therefore could be condensed into a far smaller area. The information pertaining to player actions and history would be stored in the save file, obviating the need to store it remotely.
Chatbots that run the turing test by nature are learning machines that have to be run locally and stateless, they're not designed to run with world sphere knowledgebases cooked in, so that one can be discarded right away. Spambots that roam the internet are pre-scripted and don't actually count anyways.
There are cases where being always connected doesn't equate to DRM, but there are not instances where a "Singleplayer" game which is "always online" is ALSO "DRM-free". There's a contradiction in the three terms.
Note specifically the three terms I use and in the phrase I use them. That's the problem right now. Online Solo is ostensibly Singleplayer, and I refute any claim that suggests otherwise. It has no active multiplayer component. It has *passive* components, but nothing that could not be simulated.
Last edited:
UBISOFT have been using online DRM for years (UPLAY). I have physical software (such as the Assassins Creed games) but they cannot be played until I sign into the UPLAY account. There is no problem. Why shouldn't a software manufacturer protect it's product from piracy by having some form of DRM. What would you prefer that the software records you CPU ID and then if your PC blows up or you upgrade then you will no longer be able to play the game.
MOJANG also have user verification with Minecraft. The beauty of it is that you can multiple copies on different machines but you have to verify your user details before playing. The same way that I can have multiple copies of ED but only one machine playing it.
MOJANG also have user verification with Minecraft. The beauty of it is that you can multiple copies on different machines but you have to verify your user details before playing. The same way that I can have multiple copies of ED but only one machine playing it.
But....You don't need a connection to play minecraft, just log in once and that's it.
That's not my impression. Some people seem not to care about the money they spent (or 'lost') in the kickstarter as much as they care for whining.
And really, do we need yet another thread about this?
Erm. You need to log in to Mojang servers to play so that they can verify your ID. Therefore you need to be online to play. No internet = no minecraft. Hence it is DRM by login details.
To be honest...All these users that Need OFFLINE have proved to me this last week they have a better internet connection that I do !!
Its such a shame.....They really don't know what they are missing out on...Surely a true Elite fan wouldn't care even if they could only play once a week.
In 5 years time chances are offline gaming will be down to sites like GOG to keep them alive..
Its such a shame.....They really don't know what they are missing out on...Surely a true Elite fan wouldn't care even if they could only play once a week.
In 5 years time chances are offline gaming will be down to sites like GOG to keep them alive..
Not true at all.
I've only ever "logged in" once, at that was at a friends house.
If you back up the \User\USERNAME\AppData\Roaming\.Minecraft folder, the users credentials are located in this folder, this also works for users using the FTB Launcher, which is located in \User\USERNAME\AppData\Roaming\ftblauncher
The verification is only ever used once, after that a credential file is created in the above folders depending on what launcher you use, this is also true for other third party launchers. The game is almost DRM-Free, except for the fact you need to login once, which would make sense because you need to download the game.
I've had my PC disconnected for several weeks once, and I wiped my PC and could still play due to the creation of the credential files which I had backed up along with the game files, I simply dragged and dropped them back into their original locations, and it works without the need for the internet.
Edit: Actually, now that I think about it, Minecraft is the same as what GOG.COM offers, it just takes more steps to ensure that you don't have to redownload to client.
Last edited:
However, there's not much use for a single-user email server. Unless you were to set up a chatbot that sends emails to you, in lieu of friends. Not sure how rewarding that would be though.
Practical solutions are the only ones worth discussion, surely? Otherwise it's just mental masturbation (talking of single-player offline...)
That's all well and good, but you just know a fully-offline game is getting hacked within minutes. There Will Be Spoilers. Not wanting to facilitate that happening is not DRM. Besides, how do we know what is going on server-side? The tiny amount of bandwidth used by the server-client interactions might be a result of huge processing and memory requirements under the hood. My PC might struggle as it is right now - simulating the machinations of the factions and NPCs across a galaxy might be beyond it, I fear. And I'm sure you can fake it, but that's a different game entirely, and I don't think we gave FD enough money to make 2 separate games. Anyway, this is starting to look like yet another offline mode whinefest, so let's move on...
Chatbots may have many simultaneous 1-1 interactions, or multiple 1-many interactions, or could even interact with other chatbots. But the database of all such interactions can be shared and learned from - that's the point of having all those interactions, no? You might find them lurking on IRC channels sometimes, learning and occasionally chiming in. Maybe back in the days of Eliza, you would have had a point, but chatbots have evolved a bit since then. The "official" Turing test rules may be more restrictive than what I describe, but who was talking about that really? The general concept of such a test is not to be confused with scientific conditions.
Not really. When I think of DRM, I think of a game that has no requirement for online connectivity but does it purely to authenticate the copy is valid. That's different from a game that you experience primarily as a lone player vs NPCs, but has major elements that require server-side content that would impair the experience if attempted as a pure offline thing.
That's where we agree, somewhat, but that's only right now. I have no idea what the background simulation will behave like, and nor does anyone outside of Frontier staffers. And maybe not even them, if it is coded to react to everyone's actions. (I would guess that's where the "living" part of the "living galaxy" comes in.)
None of that bit has much to do with DRM, though. It all still smells like a zombie argument...
Last edited:
![[SI] Bobdelion](https://forums-cdn.zaonce.net/data/avatars/m/35/35943.jpg?1553527109){kind=avatar}
^ This, and rep for making me laugh.
Fair does. Stand corrected (since internet always on, never needed try that).
But as you say it is almost DRM free but not entirely. And since this thread is about DRM I'll leave it at that and get some sleep.
That's okay, just wanted to make sure you had the relevant information
My internet is constantly going up and down, so I have to make sure that the games I buy have an offline mode, or at least if I am going to be playing a lot of it.
To my knowledge there's no software level protection, comments have been made to the effect that they don't mind you allowing friends or relatives to use spare character slots, and generally it's hard to classify it as DRM. By contrast the floppy versions of previous elite games absolutely did have anti-piracy measures. There is, in any case nothing particularly wrong with DRM, provided it's implemented in a manner which is neither disruptive nor obnoxious in character. A good, largely invisible implementation that does not interfere with the ordinary enjoyment of the product nor intrude in any other way is always preferred, of course.
In effect the statement regarding this is that being online is core to the game's function, although solo online is intended to require what amounts to 1996 levels of connectivity during normal gameplay. This is slightly harder to gauge because we're still in testing, but presumably is largely true.
It would be ridiculous to conflate requiring an internet connection with DRM, and likewise to castigate all DRM as an evil simply for belonging to a loosely defined category (in this case, I suspect, simply incorrectly defined). There have always been efforts to protect intellectual property, at one stage in history this involved simply keeping trade secrets until they died with you, in the 80s it tended to involve taking the first word of the eighth paragraph of the 46th page of the manual and typing it into the game at a randomly selected interval. Sometimes it's obnoxious, sometimes it isn't, but it's an understandable and broadly appropriate intention.
In effect the statement regarding this is that being online is core to the game's function, although solo online is intended to require what amounts to 1996 levels of connectivity during normal gameplay. This is slightly harder to gauge because we're still in testing, but presumably is largely true.
It would be ridiculous to conflate requiring an internet connection with DRM, and likewise to castigate all DRM as an evil simply for belonging to a loosely defined category (in this case, I suspect, simply incorrectly defined). There have always been efforts to protect intellectual property, at one stage in history this involved simply keeping trade secrets until they died with you, in the 80s it tended to involve taking the first word of the eighth paragraph of the 46th page of the manual and typing it into the game at a randomly selected interval. Sometimes it's obnoxious, sometimes it isn't, but it's an understandable and broadly appropriate intention.
Last edited:
Actually I can provide two very good reasons for a single user email server. One I used to run at the LINX (predecessor and my successor did as well) - namely, we had a private email server setup which was specifically assigned to be the receiver for log files when something broke, the error logs would immediately route to the server along with any pertinent info and then be delivered via email to our client, which we could then collect anywhere from within the building. The advantages of that would be that the server could run on a toaster, so it took nothing out of the information systems procurement budget along with the fact that as literally nothing else was run on said toaster oven, we could reliably guarantee the email would ping within 60 sec of it landing on the server, we could check the logfiles on our phones and do diagnostics at the coffee machine, then have a solution in our head by the time we got back to the NOC.
The other primary reason for a local email server is laboratory testing, you set up a load of dummy email accounts and have them scripted to hammer the server with scripted junk, and test the server for bugs, faults and whatever. This is particularly important if you're doing things like kernel patches and security upgrades, under no circumstances whatsoever do you -ever- do that on a live server without thoroughly testing it beforehand. (We had a junior admin who suggested this once, the CIO told him if he performed maintenance unbidden, he'd be collecting his things the next day, the blood drained from his face kinda rapidly).
Any client gets hacked within minutes, hell, World of Warcraft got hacked within minutes, there's private servers for that, and that's not exactly suffering for it. Now, on the topic of not knowing what's going on serverside, on that you are entirely correct, we don't know. However, we -can- infer some basic things from the network traffic, which currently as we both agree deals in passive components save for the P2P component for multiplayer. If I was to make some educated guesses based on the traffic and what comes through and goes back, the following would be my summary:
* Server is needed to validate all players transactions, to ensure consistency and to prevent cheating and exploits. Again, no arguments there. I think we're both settled on this point.
* Low level abstracted data such as markets in places the players occupy and the "bubbles" around each player. As the players get more spread out, there's more bubbles to process, and therefore more data is processed at the lowest level of abstraction, more load on the servers. AWS cloud solutions work well in this sense because they scale up and down based on what's required, but they don't always do it -quickly-. Then again, there's not likely a need for that unless there's a mass player migration and a huge spawning of new bubbles.
* Persistent NPC location data. So far this part feels very debatable, most of the NPC's genuinely feel proc-gen.
* Stellar cartographic data - this is where the biggest "data sink" is potentially (because you can't abstract it, even when the player bubbles vanish, the data has to remain persistent), because there's got to be some kind of database which tracks the discoveries of stellar objects. Thankfully those discoveries don't provide residual revenue to those people who find them otherwise you'd have to tie it to names and that would result in a -huuuuuuuuge- database set.
Now, as to simulating or faking it, that's actually not the hard part. Procedural generation these days can do a very, very good job of "hiding the magic" behind the curtain if it's done right. If you've seen the nemesis system in Shadows of Mordor you'll know what I mean by this. That's a good example of Procedural Generation being mixed with dynamic and emergent events to create enemies with distinct personality and looks based on your own actions and influence. It's certainly not out of the realms of impossibility and most definitely not a computationally impossible problem, games are advancing rapidly and what looks impossible to solve now is easy six months away. Limit Theory is rapidly redefining what Proc-gen is capable of as well.
The best ones are either written to be "Stateless" AI, which can effectively come to a conversation without prior knowledge and make intelligent decisions based on the information that's provided, or "Stateful" where they accrue information over the course of the conversation and refer back to it. You're discussing the latter, and there's plenty of good examples out there, but a good chunk of them critically can be downloaded and run locally if you so desire, there's no need to run them on some huge supercomputer, as long as you're willing to spend time teaching it. They're open source in many cases, which allows people to study the code, and if they can, improve upon them.
At least at the point of newsletter 49, I did not see anything that was present that couldn't be simulated or that really screamed at me "This needs a server". There's dynamic markets, but in all fairness you can simulate -those- too. The words "Living Dynamic Galaxy" still cut far too close to "Glassbox Engine", but perhaps FDEV will surprise everyone.
I won't reply to much of that, because there's a danger of leading the thread down too many blind alleys.
That last part may be true, or it may not - there's not really a good way to determine that though, since the server is a black box.
To be honest, I don't really care either way. I want all the expansions to be made, and for this game to fulfil it's potential. I can't see any of that happening if the game gets cracked and torrented within a week, so if they need to hoard a pile of data to ensure that the game has enough of a lifetime to reach maturity, I'm all for it. I have enough faith in the developers, given how the game has blossomed thus far, to think that they will demonstrate the value of their server-driven architecture in due course, without it needing to go full P2P (where P2P in this case = "pay to play", and not "peer to peer").
Although if someone has any idea of how to do a fully P2P (where P2P in this case = "peer to peer", and not "pay to play"), massively multiplayer, DHT-based shared galaxy with both low latency and reliable transaction management and verification, coupled with a suitably jawdropping graphics and game engine, and if they can explain their vision in terms simple enough for me to understand, I'm ready with my Kickstarter funds (I evidently didn't learn my lesson)...
That last part may be true, or it may not - there's not really a good way to determine that though, since the server is a black box.
To be honest, I don't really care either way. I want all the expansions to be made, and for this game to fulfil it's potential. I can't see any of that happening if the game gets cracked and torrented within a week, so if they need to hoard a pile of data to ensure that the game has enough of a lifetime to reach maturity, I'm all for it. I have enough faith in the developers, given how the game has blossomed thus far, to think that they will demonstrate the value of their server-driven architecture in due course, without it needing to go full P2P (where P2P in this case = "pay to play", and not "peer to peer").
Although if someone has any idea of how to do a fully P2P (where P2P in this case = "peer to peer", and not "pay to play"), massively multiplayer, DHT-based shared galaxy with both low latency and reliable transaction management and verification, coupled with a suitably jawdropping graphics and game engine, and if they can explain their vision in terms simple enough for me to understand, I'm ready with my Kickstarter funds (I evidently didn't learn my lesson)...