Comparing SotA to Ultima VII

Yes, why not release the MMO part first as a sandbox, fine tune everyting, optimize it so that it is a great sandbox MMO. Then, take the time to create an epic single player game with significant differences to the MMO part. A game that does not feel tacked on to an MMO but can stand on its own.
Right now I simply can't help myself but think that neither the MMO fans nor the SP fans will be satisfied in the end as too much time has been taken away fom building a great MMO experience and too much time has been taken away from building a great SP experience.

I am wondering who from the devs is a real SP RPG fan and would love to create a game similar to Ultima 7? I only hear that they are big MMO fans and love UO. Is there no love for the SP part of SotA in the team? Somehow it feels as if the SP part will be in the game just because it was announced during the Kickstarter and because they have to fulfill this promise...

 

I can imagine that Lum is real SP RPG fan

 

Sadly they officially quashed this concept. As there isn't any way to stop hacking, the offline and online worlds will be completely independent with no hope of offline ever bouncing over to online.

 

According your calculation Port needs 212,500 $ / month to pay the developers. Now calculate how long the money goes

The Kickstarter fee. 5% of total funds raised plus payments processor applies an additional 3–5% fee. This makes around $100,000 fee and maybe another $100,000 for payment transactions.

The pay roll numbers are of course pure speculation and may be a lot less. Anyhow - below the line the money isn't infinite and yes, I think there is a kind of financial pressure.

Again, it is not my intention to spread bad temper. I'm a huge fan of Richard Garriott and I bow to his work and his many ground breaking ideas. And I also appreciate the hard work of his developer team. I just want that SotA will be successful - with enough single player story content

 

I would point out Portalarium's frugal ways.. and while no numbers were ever actually given.. they have stated that at least some of the devs are taking below industry standard pay.

 

I can testify that programing good chat AI can get pretty complicated pretty quickly. Simply catching the key word "where" just isn't going to get the job done.. but that seems to be where that dialogue is at the moment. Where is|am doesn't quite work either. Rather.. where is + ???? Well that could require a different response for every ????. I don't even Lum's job.

 

Problem is I think that new players won't understand it. They will type in every kind of questions and probably won't get satisfying answers most of the time. Now imagine what they will do? Will they understand that the system is limited and that they have to find the right keywords? Or will they make jokes of it and give up, frustrated?
I think that most people here wear rose-tinted glasses because they want this game to succeed and because they are on board for a very long time and know how everything works.
But after Early Access, when new players pour in they will see this as a released game - and I have to say: rightfully so. Without the "protection" of Early Access people are basically paying money for a finished game. This is also when the game will be reviewed and rated and also: rightfully so.
But if you play the game like you would any other released game, SotA offers some very queer features. Like the dialogue system. I predict that many players won't understand it and will get frustrated and they won't care that the game will be developed further - and again they are right. Why? Because they paid money for a game that is out of Early Access. And by paying money they have a right to play a fully functional and fun game and not take part in an Alpha test that is no longer labeled as Alpha. And this is a risk Portalarium has to acknowledge because they charge money for the game.
SotA has to deliver in so many ways when the Early Access ends... I am sceptical if they can manage to do this...

 

I just want to get this off my chest: I HOPE that at some point, probably not in Episode 1, but at SOME point in the future, in one of the future Episodes, the NPCs' conversation AI will be as good as the Alicebot (http://alice.pandorabots.com/). Shroud's conversation interface is literally PERFECT for this kind of AI.

Here is an example of a recent conversation I had with Alicebot:

Me: What's your name?
ALICE: I am called ALICE.
Me: what's your job?
ALICE: chat bot.
Me: Where am I?
ALICE: You haven't told me where you are. Where are you?
Me: I'm in California
ALICE: I am in California too!
Me: Where are you?
ALICE: Oakland, California.
Me: Do you like adventuring?
ALICE: No, I don't care for it.
Me: What do you like?
ALICE: I like movies about robots.
Me: Have you seen I Robot?
ALICE: No I have not seen it. What's it like?
Me: It's a movie about robots
ALICE: Maybe I will try to rent the movie on video.
Me: Do you like playing games?
ALICE: We are playing a game right now.
Me: True
ALICE: Take it easy.

There will be many, many challenges to make NPCs this intelligent. The bright side is that it can be done, and the community members of Shroud can actually be a great resource for making this happen, as they will be throwing a lot of conversations at the NPCs, which will get recorded, and the NPCs responses can then be improved to respond to them. One of the challenges is that each NPC should respond with his or her own personality to make the game fun. This is a lot of work to program each NPC like that. Perhaps a compromise is that only a few NPCs will be able to respond as well as Alicebot can, maybe only the companions of the Avatar.

Again this is just a dream of mine.

 

To be fair, I think some of this will be a learning curve for new players, regardless of it being Early Access or at Release, since the days of parser-based RPGs has not been around for awhile. People are used to point and click style conversations (or even the Bioware dialogue circle menu) where they can just pick choices. Since this goes back more to the early Ultima style, it will take some getting used to for people to learn and react to it. I don't think that's completely a bad thing though; it brings back a lot of the investigative process that RPGs used to include.

That said, of course there's still polishing that needs to be done, and I'm sure the glaring issues will continue to get patched and updated. I'm more concerned with them fixing the style of the chat window itself, since it's kinda fugly, imo.

 

The big problem I'm having with your whole argument with the dialogue system here is you're ignoring the whole fact that keywords are underlined by default and they're clickable. Most players are just going to click on keywords most of the time. You're focusing on the worst and assuming the worst. Yet no player is going to be forced to type out dialogue to get a response. The option is there yes, and it's not perfect or finished, true. It's one of those things that hopefully will never stop improving.. but it remains optional. Not required.

 

Exult rocks and let you play U7 with the U7-Serpent-Isle "dolls", amongst some other modes. Highly recommended.

Did work like a charm on my linux notebook.

// Armar

 

I would play it if I had the original Ultima VII files.

 

You can buy it for a few bucks on GOG.

https://www.gog.com/game/ultima_7_complete

 

Sweet! I'm going to compile Exult for my phone.

 

*Cough* if "someone" in software development tells you it takes "X whatever" you _at_least_ multiply X by 3 to get a halfway realistic timeline.

// Armar

 

Its a bit different but Baldurs Gate I+II are pretty close.

// Armar

 

@Truthveyor Chat bots like those are done with Markov Chains. The concept is pretty simple and the AI is actually more stupid than you think. It's a neat little trick that does a pretty good job at selling the illusion.

I've done a ton of research on the subject.

The idea is to grab multiple words per line of text; not trying to identify a series of words in succession.

So if "where" is found in the string, look for another keyword that has to do with a location (like "Ardoris"). If there's no keyword that matches a location, look for a NPC name. If neither was found, the user was probably talking about something else.
Repeat the process for multiple keywords until you found a plausible solution.
Output a response based on that.

You could essentially write "Hey, looking at you Ardoris buddy. Where's my cat?"
and (assuming that there's no keyword for cat) the system would register [Where][Ardoris]

The problem with this method is that you have to account for multiple synonyms and spellings of particular words. I'd go as far as to say that a "markov chain" Chatbot is practically easier to pull off; it's just reliant on your "text source"... most people use the novel "Art of War" in text format since it's has an immense catalog of words.

Portalarium could use the entire "Sword of the Avatar" novel as a starting point.

There's more AI involved with the keyword system.