Darth Fez wrote:

How prototype is this video? We can't even see it yet, that's how prototype it is!

Maybe we're supposed to take the Red Pill first before we can see it... ;)

Tannim222 wrote:

That happens sometimes with updates where the front page doesn't get updated properly. Here is the link from our Kickstarter page until the forum's front page is updated.

I have to admit, my funny bone is often more bone than funny.

Why?

Because it hurts more!

It's so good to see everything coming together like this! Yes, Virginia, there is a City of Titans. Bare bones at the moment, but there's an actual game there!

Two things that really impressed me:
1) The promise of branching and adaptable missions.
2) The brilliant idea to make mission structure spreadsheet-based.

Keep up the good work, MWM. Every time you show us what you've been up to, it just cements my faith in you guys all the more.

This was very cool.

I hope my fellow viewers realize and appreciate how powerful it is to get this functionality built into the game at the foundational level. This is big.

Ohhh, maybe even have support for [url=https://www.google.com/sheets/about/]Google Sheets[/url] someday. ;)

Great update.
Lot of possibilities, i'm so excited again and again.

Wow.... Very impressive.

Always gonna have faith in you Missing World folks. You got heart.

Yess, I am so proud of you guys, cannot wait for this :))

Oh, the sweet, sweet agony of these updates.

Every time I think I understand how awesome the game is going to be, ya'll show that it will be even more amazingly awesome than I thought--but it makes me want it NAOWUH!!!

That said, spectacular work and more than well worth the wait. Take your time and get it right (BUT HURRY!!! :P)

warcabbit wrote:

It handles it pretty well, Sheep - we've got it working - the undercover guy is a conditional branch, you get an ID card from him so you can sneak through an area.

I was thinking on a scale a whole lot bigger. But not implying you should be there yet. And there *are* answers, they just have different tradeoffs (and provide different opportunities to work around some of the downsides, in some cases). To put it another way: it isn't the implementation of branching that concerned me at all. It is the ability to write stories that make heavy use of it without going insane. Or being a sheep! (which, of course, makes insanity redundant) You'll deal with it when you get there.

warcabbit wrote:

The UI is... related to the Bioware, but basically, it packs that classic picker into something that shows one response at a time, so we can show a little more text and/or significant icons like 'lawful up' and can be operated by keyboard instead of needing a mouse. (W/S to go up and down, Enter to choose by default, I think.) We're using their UI lessons, but with our goals on it. It's experimental, but so far it's shown promise. Looks like Bioware, does the _reverse_ of what Bioware does.

Much more promising, then. Although depending on what exactly you end up with, this might be an excellent opportunity to start looking at how to implement different UI options that the user can pick. Which UMG should actually make relatively sane, from what I've seen. Not saying implement all the fancy parts, just... building a second thing, when you already have a tolerable-if-not-great first thing, tends to be a pretty good way to expose design-level bugs. Which are the ones that you can't fix later for any sane cost.

warcabbit wrote:

And the CSV is not exposed at runtime, but yes, we do plan on respecting little Bobby Tables and if/when we allow handwriting missions, they'll be protected _darn_ well.
http://gafferongames.com/2016/04/25/never-trust-the-client/
For those curious about what Sheep and I are talking about, here's an example of someone messing it up. An AAA title, too.

Indeed. To reassure folks, however: all of the built-in stuff for UE4 is heavily biased towards "doing it the easy way tends to result in doing it the right way." Epic has a pretty good track record with such stuff, and provides a solid framework on which to build exactly the sort of untrusted-client model he discusses as being "the right way to do it."

My point was more pointing out that to load CSVs (or Google sheets, etc) pretty much requires being able to process them, and it would be very tempting to just grab the existing code and use it. And normally that would be exactly the right thing to do. It might still be, for this case, even -- but not without thoroughly vetting it first. The good news is that since the engine supports Unicode as a core thing, you're at least somewhat more likely to get garbage strings out than a buffer overflow if you try to feed it garbage.

Arguably more concerning would be stuff like triggering spawns, where you need to allow the permission so that the code can do it, but you don't want to allow it to spawn just any random old thing. Enforcing semantics tends to be a lot harder than enforcing syntax. :/

I much prefer the look of the UI sample than the generic text presentation in the video. Though I hope the final version won't use Times New Roman. That seems more bookish than comicbookish, if you know what I mean.

Also, if the UI is meant to show alignment effects, will the display of those indicators be optional?

Question about the phasing: if 2 characters go to the mission at the same time and Hammerdown decides to fight the NPC while Chatterbox talks his way through, does that mean Chatterbox will see Hammerdown fighting a copy of the NPC in the same spot? Or are we hoping this will rarely occur because mission doors will be semi-randomly scattered across the map as in the old game?

The ability to turn off those indicators would be nice for those people who want to make their choices without being influenced by the knowledge of how the choice affects their character's alignment/axis. Rather like how it's optional with the Light Side and Dark Side choices, IIRC, in SWTOR.

Much like in SWTOR (although that is based on a roll), I presume the team leader's choice will direct the story but individual choices will dictate any alignment/axis changes.

Fun! Being able to do things like making a character and just blindly making choices based on "theme" will ad even more entertaining possibilities for gameplay and RP.

Sweet.

Likewise

Mendicant wrote:

DeathSheepFromHell wrote:
Other than that — what on earth was that "chat UI version 2"? Other than the easily-changed opacity, the UI that showed in the window should be pretty familiar to anyone who played CoT.
You've played CoT already? Admit it, you're from the future! When's the first expansion pack come out? Does CoT eventually blow away WoW in subscribers? Does MWM put out a VR version?
DeathSheepFromHell wrote:
Not saying it shouldn't be done, mind you. Just to keep in mind the zeroth rule of server-side security: assume the client may be compromised (and actively hostile) at any time. Because someday it will be.
Indeed. It pays to be paranoid, because one day the client will be out to get you. (If I had a dime for every time someone passed a text input directly into an SQL statement without sanity checking and sanitizing it, Bobby Tables would have his college fund all paid for.)

lol. Sometimes Weird Name calling hurts?

You had me nervous for a little bit. Seeing you guys were a week late. But definitely worth it. I was tempted to post a reply to a thread about the lateness along the lines of "that just means it's going to be a super good update" but I didn't want to over hype an update. In hindsight I probably should have.

As for more videos, they are not really necessary and counter to some of the previous goal and direction statements. Certainly we enjoy these updates with moving pictures, but we understand that video's take a lot of time which could be better spent on further development and testing. Especially considering the limited time each of you has to volunteer. We believe in this little dev studio.

I would be happy with an update along the lines of, "Art is working on some new building pieces for X district. And some costume pieces, something that looks like a space age Victorian wig. Networking is ...uh...networking? Business is discussing the 394 posts about the business model and store stuff. World builders are world building. And story and lore are writing something involving a flying R.A.T." It's deep, it's teasing and whets the appetite. And takes less time to write if you aren't making it up off the top of your head like I am.

That's good enough for me, but I understand there are quite a bit of people following you so do what you feel is best.

To me this sounds less like a complication on the tech end and more of a complication on the human end. Those kinds of branching missions have more moving parts, so it will require considerably more care on the developer side to make certain everything fits together neatly and works well together. Missions that have "If A, then C; If B, then D" approaches are trickier than missions that have go "Start - Step 1 - Step 2 - End".

Especially if they have [url=http://www.gamasutra.com/blogs/JoshuaWeinberg/20160502/270103/Puzzle_Dependency_Graph_Primer.php]more out-degrees than they have in-degrees[/url].

Darth Fez wrote:

To me this sounds less like a complication on the tech end and more of a complication on the human end. Those kinds of branching missions have more moving parts, so it will require considerably more care on the developer side to make certain everything fits together neatly and works well together. Missions that have "If A, then C; If B, then D" approaches are trickier than missions that have go "Start - Step 1 - Step 2 - End".
Especially if they have more out-degrees than they have in-degrees.

Remember that doing a check for unbound logic targets is trivial, and doing a test that there is at least one way to reach each of the final exit conditions is only marginally less so. The sane way to deal with those sorts of dependencies is to only put them in where they make sense, and let them be driven by state that is contained in the mission object. Or, more formally, keep your decision graph state-independent except for a single state that is passed in (and passed along at exit).

DeathSheepFromHell wrote:

Shadow Elusive wrote:
Ah, you're talking team dynamics. I believe those will go by the leader. It's been discussed. Basically if you're in the same mission together (instead of two private instances of it) then you're on one course together.
Disclaimer: all of the following is a discussion of "when it was proposed, up until the time I stopped being involved with it". Things may have changed, and may change further, as always. But this is roughly how the original design vision was laid out:
The idea of phasing is fairly simple: at all times, you will see the world "through the eyes" of… well, I would say "the team leader," but some of the corner cases folks brought up might mean it would be "through the eyes of a designated team member" (probably defaulting to the leader).
But whatever the case, consider the situation to be playing "crossover" — while you will probably get the option to complete *your own* mission with the same consequences-for-choices (if you have it outstanding), you wouldn't be forced to. Choosing not to means to complete your own copy of it you'll have to run it again and make *your* choices, but presumably that's why you're choosing not to in the first place.

A few players will want to be the stars of their own UGC created missions, with them doing scripted Gesturing's/Sayings in the cut scene. Some of the time they wont be the Teams leader, but, granted.. Most of the time.. they probably Will Be! ;)

desviper wrote:

If the possibility of less-humanoid characters (e.g., four arms, no legs floating, etc) is still on the table, that'd make emotes and such problematic for a cutscene, so it should probably default to a humanoid character, if there is one.

I would imagine that it would be limited to either emotes, or "animations you have access to" (emotes plus any other standard animation, like combat animations — always dream big). For other reasons related to the aesthetic decoupling feature, they're pretty much going to need to have a predefined list of things that a character has a bound animation for anyway (even if it is just a default, very generic sort).

But in order to have something there I would expect you to need to define it in the mission, anyway, so.

Very nice - step by step this game is coming together.

Baalumbral wrote:

This was very cool.
I hope my fellow viewers realize and appreciate how powerful it is to get this functionality built into the game at the foundational level. This is big.

Good to hear for us novices.

When I found out about the dynamic maps, I immediately thought of all the procedurally-generated maps that some games like to do. Monstrum, SCP Containment Breach, Minecraft, Evolve, No Man's Sky... If CoT winds up like any of those games, then hats off to you guys! Obvious dynamic environments aside, I understand that replacing map and texture files with a single algorithm can [i]greatly[/i] reduce file size, so not only would this radically increase immersion, but it would also be a lot more hard drive-friendly. Heck, .kkrieger was able to pack all of its procedurally-generated textures into 96 kilobytes, as opposed to the 2-gigabyte texture packs that other modern video games typically use.

Well, the rule of thumb is to send all inputs to the server AND to yourself. The client acts in response to your inputs but if the server disagrees its correction overrides you actions. That's the cause of "rubber banding". It makes the client more responsive to your actions too. Things to improve overall throughput is to reduce client-server traffic and to keep the workload on the server to a minimum. In short you want to do as much as possible on the client rather than on the server. If it doesn't affect the other players' games then you should only do it locally.