MMO Option (Subscriber P2P Option)

@InsaneMembrane

Well via Kickstarter they said we will have Castle sieges.

Source:
https://s3.amazonaws.com/ksr/assets/000/498/318/0201ce6a8500a4e24a74de94e173f46d_large.jpg?1365345004

Read that up, if this is hosted by p2p this would mean someone in the conflict would be a host and thus would have a huge advantage... And how ridiculous would it be if a "Castle SIEGE" was a 5v5 and what if someone from a European Guild is besieging an American Guild... this would be unplayable even with 5v5.

So the solution would be to host a conflict on a separate sever they own... this server would also have to be able to talk to the main server system web otherwise guilds could just declare a Siege then join the siege and proceed to dupe items and participate in other exploits with no real way to monitor what occurred..

So if they can successfully host a Castle Siege they have all the technology to let people pay to use a server. It is just about a few changes in code and WALA! it will work.

 

@Monkus

It's pronounced, 'Voila
http://en.wiktionary.org/wiki/voil%C3%A0

http://en.wikipedia.org/wiki/The_More_You_Know

^^'

 

@Vagabond Sam

I learn something new every day haha!

 

there's a Dev Chat today with Chris, where you can ask him about the scale of siege battles. I'm pretty sure they considered, how they can manage to put up fights bigger than 5v5 with their system.

I'm against an additional Subscriber-Options, since it would divide the community. I'm quite certain, that this would lead to elitist behavior and claims: I pay for this game, so I deserve more than the others. And this leads to tedious jealousy debates (I've seen a lot of those in MMOs). In the end, I think it would have a negative impact on the community and the game.

And I'm against additional slider tiers above OPO, since I don't see how there could be anything more open than open.

 

@InsaneMembrane
Could you please post a link to the video where devs covered this issue? I can't find it. I'm most interested in what they said about it, because there obviously would be an issue with bigger scale P2P servers.

 

Not really a fan of this idea, but I think it stems from the problem that all the good information on what the devs have discussed is fragmented in videos and a few posts. Would be nice if someone at Portalarium would compile details about the game in one place, but I think that will come as plans are more solidified.

First off, I think using the terms p2p or server in reference to SOTA is not entirely accurate. This will be a persistent world which is really just a big pile of data. I imagine there will be a series of servers and services which will be the bulk of our interaction of the world, and if I remember correctly I think they reference Scenes (or instances) to by more like dynamically spawned virtual machines. There will be some client to client communication, but many "transactions" will be validated by the main SOTA "cloud" to stop cheating and such.

So by suggesting there is a "subscription" server and a "free" server I think is a little short sighted. SOTA is going to be a completely different game than UO or other MMOs, but I really think that OPO will make people in these camps happy. It is just a little hard to visualize how this game will work especially with the lack of solid information or working examples. I am a little fuzzy about what MMO/UO players "won't" get in what's been explained for SOTA. The ability to forcibly and openly kill someone anytime, anywhere is the only thing I can think of that's been discussed (and even that is not true of all MMOs). Other than that in OPO I really doubt you will notice this is not an MMO.

That said, I think there are some good ideas that can spawn for this suggestion. Portalarium could charge users to spawn special instances for user generated activities. For example, maybe Bob and Jane are getting married and they want a special event. They could pay a fee and Portalarium would allow them to upload content (images for banners or table clothes), layout out furniture or other objects, maybe add script to some NPCs, etc. Then only allow this version of the Scene (instance) to be available to a select number of people in their "invitation" list.

Maybe not the best example since I bet FireLotus would do all this for free for many of us in the community just for the fun of it. However, I can see custom, private parties, events or quests might be a good candidate for a small fee to allow the customization of an instance. A small offset to the extra computing power and pee-pee review of custom content.

 

@Umbrae

If I am allowed to quote myself this is what I actually said about the server infrastructure:

"Let us simplify this, I am going out on a limb here and assuming the development team is building this game to scale into hundred thousands or millions. Here is how it would work at a very simplified scale."

I know this will work more like a large data cloud - but for explanation purposes of a concept if I added in all that meaning the emphasis on how it would work would be completely lost, here is how it works:

"...think of the MMO server as just another large peer to peer network! It will talk to the same server [infrastructure] as all the other small peer to peer networks will!"

Now if you would not mind let me quote you:

"...SOTA is going to be a completely different game than UO or other MMOs..."

Actually on the server side of things it is shockingly close to the original Guild Wars in the way it will work and behave.

The best example of a shortcoming with the Guild Wars-esque server build is the fact they said there was a Castle Siege mode, as well as a large performance stage where people will pay for seats etc....

This will take much more to do then a peer to peer network like the one seen in the original Guild Wars.

This will need a system more similar to WoW or any other current MMO to be able to pull off.

And since they have multiple stretch goals in this area I would assume they are going to be building an infrastructure that can accommodate these Stretch Goals...

If they are building an infrastructure to accommodate those features they can easily add an MMO option...

A big reason I wanted to start this thread is to show people that 2 of the Stretch Goals cannot be met in a fun and functional way without actually having a sever for everyone to hop on in these instances, and this server would be able to talk to the "verification cloud" (server that deals with all OPO, SPO, and OPO).

And capital cities... using peer to peer you would never get the feeling of a bustling city like you do in true MMO's.

 

I guess I still haven't figured out the difference between Open play online and what you get from an MMO.. other than the people you see might be relevant to you, might not be.. If it is basing it off my friends list I usually only keep about 5-10 people on that list. And in MMO's they usually aren't near me, i do my own thing.. So that means most of the people I see are going to be strangers i've never talked to.

So what is the difference?

I don't think we'll have an actual peer to peer experience like some people are discussing... Like you've said, I don't think it will be possible to do the events with enough people using p2p connections, there might be times it can offload that somehow in some areas, otherwise people, depending on their connection, are going to be experiencing lag with only a few players on their screen.

This will be stuff we'll probably figure out and know more about during alpha... One issue with discussing this is people's technical knowledge...

 

@ Sir Mishri

Your Question:
"So what is the difference."

My answer:
Okay OPO is a cool system! Let me first state that I am a fan of most peer to peer experiences but I will give you the weaknesses of OPO or peer to peer systems in general:

Take a game like WoW for example servers are geographically based, a player connects directly through a top of the line server and interacts with other players that are directly connected to the server this makes interactions occur very fast an usually at a very low latency if you are playing within the geographical regions.

When you take a peer to peer system it works perfect in some scenarios when all the players are very close to one another, the best example is LAN tournaments or events - this is a peer to peer network that performs flawlessly with virtually as many people as you want. Also if everyone had top of the line Fiber Optics internet it would work quite well!

However, lets say I have lots of friends I usually play with that live in New York, Florida, Texas, Maine, and I live in California. And most of these friends have basic DSL.

In this case someone needs to be the host, lets say I am the host in California so all my friends hop in my group 4 of them, this is pretty reasonable however some of the guys from the east coast will get lag spikes and have a higher latency... Although peer to peer is still functional at this point.

Now lets say my guild wants to have a get together at the "World is a Stage - Theater" there are 25 of us and one of us has to be the host... in this situation 1/3 of the players will have a nice connection, and the rest will have an unplayable experience...

Basically peer to peer systems scale terribly!

EXPERIMENT TIME!
If you have Team Fortress 2 or Minecraft you can test this yourself!

1. Open a server on your computer
2. Get some friends to join needs to be 20 or more!
3. Get lagged out and crash your computer server over and over.
4. Say "Man peer to peer networks on large scale do not work over significant distances!"
THEN
1. Connect to a dedicated Minecraft/TF2 server
2. Get all your friends to log in.
3. Enjoy fun gameplay with everyone! Without lag.
4. Say "Wow dedicated servers rock!"

 

@Monkus

I think you put too much into p2p. There will be p2p type communication, but I think this will not be a majority of how the game will communicate. As such you are framing this on the limitation of p2p, but that is short sighted to how SOTA as a whole will work. The problem with WOW is most game content and game state streams to you: Even Minecraft is streaming content to your machine. This is what causes a lot of that lag as it is all raw communication needs to be pushed to you and all the others. The Devs have said they want to reduce that and put much of that content on the client (locally), so communication between client/server and peers is greatly reduced thus allowing for less lag. Most world states will be synced you, and more of less will be running off a cache.

I will admit this is an ambitious goal which is difficult to visualize, but it is completely different than the technology that is used today. There will be a limit to how many people you can see, but your communication to other will be limited by the space you share thus the scope of your "Scene" will be smaller then open world MMOs. MMOs today are like amusement parks, and you will see more people in a location than is reasonable. In SOTA, the amount of people will be smaller and more organic/realistic.

If communication is optimized properly than the latency you normally see in p2p and server based MMOs should be greatly reduced and in most cased be seamless. Although until we get to play with some of this tech, it all speculation. I just believe some of your concerns still steam from an older model which SOTA is not trying to replicate.

 

@Umbrae

Can you please try to create a concept where it would be extremely minimal upkeep costs per month, and you could have around 30-50 people in an instance (developers stated this was the rough number.) this cannot sustain on peer to peer networks at all... it would have to be dedicated servers if this is the case.

You also state that Minecraft, TF2, and WoW all stream information to your PC that causes lag...

So SotA will not show other players on your screen, no PvP in SotA, no Marketplace in SotA, no chat... come on man it will be WAY WAY more data streaming to your computer than Minecraft...

You can have as much data on the client as possible you can have every graphic, full engines...etc.

But 1 thing you cannot do is add player interactions, market trades, PvP damage calculations, loot tables, currency tables...

This is why all games that are played online have a robust server side code, as well as client side code.

The way you describe it is "ambitious" what it actually is, is impossible with the current technology.

You got to think if there are 40 people in an instance with peer to peer or any other system you are hinting at your computer and internet connection will need to constantly update all those player locations, if they are dealing damage that has to go through the connection, trades, chat, status changes, let alone the fact that the server has to trigger animations on the other characters like triggering all 40 people to do running animation from point a to point B then do attack animation 1-2-3 and skill animation 1-2-3 on creature all this data has to come down your pipeline in a system you are describing. And all of this has to happen between all the involved players vs just through a server.

That system is silly and really clumpy... It is Guild Wars (1) but with way too many players per instance GW1 limited it to 2 groups or 8-10 people. GW1 was an expansion based game just like this one with peer to peer technology with a LAN integrated client for competitive play and they still sold cosmetics to become profitable.

Now if they do a Guild Wars 2 approach they will go bankrupt unless they sell cosmetics is huge quantities since GW2 is a completely different server architecture they also have to sell loads of cosmetics to not lose money.

Seems like the majority of backers have unrealistic expectations for a game if this game was only Single Player, and Single Player Online + Co OP this server structure would work flawlessly; however, we hit the Stretch Goals.

Something will have to be compromised I think it will be OPO...

 

Basically you cannot have your cake, and eat it too.

I have reviewed all the dev chats so far that covered this information, it seems like the system is built for Single Player Online, and Friends Play Online (Tram). And the idea is to scale it to the point that Ultima Online fans will love the game also...

It just is not scale-able past 20 players in a true online environment.. where people do not live next door to one another.

 

@Monkus - Numbers of onscreen people has not been set, but "dozens" have been used. Not 40-50. Most MMOs send more than player locations. Most monsters, world data, etc. live on the servers and all that information is passed to you. They way they have explained it as most data that MMOs pass from the server will be handled on the client. There will be some p2p communication with the server doing validation checks and handling certain transactions.

I have been working on Internet related technologies since there was an Internet, so I understand what you are trying to say, but you are still applying old methods to something they have said is new. I believe RG said the Ultimate Collector game Portalarium made uses the same network tech that will be used in SOTA, so most of the technology for this already works. I personally am not going to nay-say anything until there are more details, but having built tons of internet based applications for Fortune 500 companies I would hardly say their goals are impossible.

 

@Umbrae

As I stated previously I worked with Maximus Healthcare an extremely large insurance provider as a Senior Database Administrator.

I am not fresh off the boat either. Time can only tell.