"Alternate" Age Creation?

Anything that isn't directly related to Age Creation but that might be interesting to Age developers.

Re: "Alternate" Age Creation?

Postby Korovev » Mon Sep 02, 2019 4:05 am

Speaking of which, is there any plan to import some of the already released ages on Gehn? I suppose the main hurdle, besides the actual conversion, would be to get in contact with the creators.
Gehn #137074, Deep Island #194197. Gahreesenomee biv keneet pishoeet b’set!
Want to know what’s going on in the Cavern? Get tweets from the GoMe.

MacOS wrappers, PlayOnLinux/Mac scripts, DniTools, goodies.
User avatar
Korovev
 
Posts: 165
Joined: Sun Jan 27, 2013 5:18 am

Re: "Alternate" Age Creation?

Postby Tsar Hoikas » Mon Sep 02, 2019 1:40 pm

Sirius wrote:Two: C++ is NOT a scripting language, and Blueprint is NOT a language at all, which is a serious deal breaker for me. Three: the UI is still pretty ugly.
But even I have to accept that it's THE best free game engine ever, period. There is a reason Cyan and a lot of other people work with it.


I've gained a bit of experience with Unreal as of late, so I feel like I should chime in here. As you know, I implemented the logic node system in Korman, largely based off a suggestion from Deledrius. The goal there was to expose something that allows age authors to use some of the more interesting internal Plasma features without exposing all of the gritty details thereof. When I started using Unreal's blueprints, I was instantly enlightened. This was exactly what I wanted Korman's logic nodes to be and more. It also carried the same pitfall--you had to understand some of Unreal's inner workings for it to be useful. It does a appear to be a very thin layer over the C++ API. You can, in fact, compile blueprints to machine-readable C++ (they call it "nativizing"). For sure, blueprint is not a language. It is, however, a form of visual scripting, and it is fairly compelling IMO. I like it, but to be honest, I can write the same thing in C++ much faster.

Sirius wrote:In the end, either you pick Plasma and you have built-in networking BUT it's a pain to use, or you pick another engine which is much easier to work with BUT have no multiplayer... In either case, people will enjoy exploring what you create.


This was probably true several years ago. However, in Unreal 4.22, multiplayer is very much a core feature. I would argue that Unreal's multiplayer is much stronger than Plasma's. The server situation in Plasma is currently non-ideal and the overall design is rife with ways to exploit the game due to the inheritance of a decentralized model from 20 years ago. Unreal, on the other hand, is used in Fortnite, an actively developed game that has an interest in defeating exploits.

Sirius wrote:I hate C++. I like that it runs fast, yes, but writing and troubleshooting it is slooooow, and interoperability with other languages is a nightmare


It works very well with VC++2019 and its built-in CMake support. Writing C++ gets faster and easier as you practice with it.

Korovev wrote:Speaking of which, is there any plan to import some of the already released ages on Gehn? I suppose the main hurdle, besides the actual conversion, would be to get in contact with the creators.


The main hurdle, presently, is a lack of cooperation and effort from the gate keepers at OpenURU. You do raise an interesting point that, as time goes on, it will be more difficult get in contact with original age creators. When we wanted to make some small adjustments to Vothol Gallery, it was somewhat challenging to get into contact with rustee in that he is no longer active here. To his credit, once we got in touch with him, he was very accommodating.
Image
User avatar
Tsar Hoikas
Councilor of Technical Direction
 
Posts: 1900
Joined: Fri Nov 16, 2007 9:45 pm
Location: South Georgia

Re: "Alternate" Age Creation?

Postby Deledrius » Mon Sep 02, 2019 2:41 pm

Tsar Hoikas wrote:As you know, I implemented the logic node system in Korman, largely based off a suggestion from Deledrius. The goal there was to expose something that allows age authors to use some of the more interesting internal Plasma features without exposing all of the gritty details thereof. When I started using Unreal's blueprints, I was instantly enlightened. This was exactly what I wanted Korman's logic nodes to be and more. It also carried the same pitfall--you had to understand some of Unreal's inner workings for it to be useful. It does a appear to be a very thin layer over the C++ API.


That sort of thing was what I had in mind from the start, but unfortunately (as you found) Unreal's implementation is so bare metal that I had a hard time finding good examples of the kind of abstractions we want to provide. It's tough to find the right balance between flexibility (it can do whatever you want, but you have to understand how to build from scratch) and usability (common things are simple, but uncommon things are hard or impossible).
User avatar
Deledrius
Gehn Shard Admin
 
Posts: 1281
Joined: Mon Oct 01, 2007 1:21 pm

Re: "Alternate" Age Creation?

Postby Sirius » Tue Sep 03, 2019 12:16 am

Tsar Hoikas wrote:As you know, I implemented the logic node system in Korman [...]

I guess my main gripe with visual scripting is that it makes it much harder to get into complex systems or scenarios involving inheritance, templating, etc. But you're right, it's still very efficient for scripting and a good introduction for non-programmers.

Tsar Hoikas wrote:This was probably true several years ago. However, in Unreal 4.22, multiplayer is very much a core feature.

Ah, true. I guess Unreal comes with a lot of features that make it easy to setup good multiplayer (unlike Unity which recently deprecated its whole networking module for the second time).
I guess what I meant by "lack of multiplayer" is that Plasma already comes with fan-run Shards, a community, all sorts of social features (KI, avatar customization, emotes), and that Age builders can usually rely on PyPRP/Korman default settings to let the game synchronize clickables and animations.

Tsar Hoikas wrote:It works very well with VC++2019 and its built-in CMake support. Writing C++ gets faster and easier as you practice with it.

I see what you mean... But as you said, it takes practice, and a lot of it due to the number of pitfalls with its syntax, the lack of safety net, how platform-dependant it is, etc.
In most situations, C# and Java still provide you with plenty of performance, better autocompletion, and an easier way to deal with libraries, without spending too much time to "git gud".
Obviously, in the end it all depends on whichever language the programmer is more comfortable with and what are the requirements of the software he's writing... No language is intrinsically bad, but some are tougher than others.

Thank you for your insight, it's always good to hear different opinion on these things ! ;)
User avatar
Sirius
 
Posts: 1184
Joined: Mon Jul 26, 2010 4:46 am
Location: France

Re: "Alternate" Age Creation?

Postby Aloys » Tue Sep 03, 2019 5:28 am

The problem I have with Unreal's Blueprints or most similar tools is that they are too low level; with too much clutter.

For over a year and half at work I've lead the development effort of a new visual scripting plug in for Unity. And prior to doing that I had tried most big similar tools in Unity (Playmaker, Uscript, Bolt) and Unreal's Blueprints. So I'm starting to have a decent experience in this area. And the problem I had with all of those was the same. If you are trying to make a specific kind of game with limited features (we are working on a 3d point and click) then it often feels overkill and like reinventing the wheel. And as Hoikas pointed out there are just things that are just simpler to code anyway so you end up with two different development paradigms. You can do a lot with these tools, and often that's just too much. It is like using Blender to make Ages. It's great , as long as you have no other options. ("this tool is great, just ignore 95% of the stuff that's on the screen") The reason we went with our own solution (which is very much like reinventing the wheel) is a long term business decision where we want to make several games with the same gameplay, and having our custom tools is a great productivity boost. They are limited, but they suit our needs and they are very accessible (whenever we have someone new on the team they are fully operationnal withinh a couple days)... You couldn't do a first person shooter game with those, but that's not the point either.

Finding a good balance with specialized tools that works great but are limited, and general purpose tools that do much more bur are also more confusing to use is difficult. But from my own experience I end up liking better the specialized tools... And most of the tools I've tried aren't.

Korman's current node system may not be perfect, but I still thinks it's a better option than Unreal's Blueprints or Unity's Playmaker. (that tool is terrible I don't understand why it's so popular)

For the record: In Unity I really like Bolt, it's a super polished prototyping tool, and I could even see myself producing a full game with it. But if you are trying to make specific gameplay features there's 90%s of the stuff in there's you'll never use.
User avatar
Aloys
 
Posts: 1869
Joined: Sun Oct 21, 2007 7:57 pm
Location: France (GMT +1)

Re: "Alternate" Age Creation?

Postby Sirius » Wed Sep 04, 2019 5:47 am

Aloys wrote:Finding a good balance with specialized tools that works great but are limited, and general purpose tools that do much more bur are also more confusing to use is difficult.

This is a very good point IMHO.

Game engines (and lots of softwares in general) are all about trading some flexibility in exchange for a more specialized framework. A good software properly balances the loss of flexibility with the gain of specialization, depending on the target user.

I like visual scripting, but my main gripe with it is how most people think it's a way to do a programmer's job without hiring a programmer :? As you and Hoikas said, it's still a very thin wrapper around machine code. While it does remove the syntax ("language") barrier, it still means you're dealing with actual low-level coding. It's not a panacea that will completely remove the requirement of programming knowledge, but it will help people acquire that knowledge.

However in my opinion, visual scripting can become a panacea once a programmer enriches it with higher level concepts, that the level designer can later reuse. In Uru's case, those higher-level concepts could be "Door", "Lever", "Journey", etc; which the level designer can chain together, as in "Lever + Button = Door opening". But usually those concepts are too specific to each gameplay to be included in an actual game engine (which is one of the reasons Plasma started moving to global xSomething Python scripts).

Hence why switching to a new engine is often hard... You have to recode part of your framework from scratch, on top of having to learn how to use the new engine.
User avatar
Sirius
 
Posts: 1184
Joined: Mon Jul 26, 2010 4:46 am
Location: France

Re: "Alternate" Age Creation?

Postby Lehnah » Fri Sep 06, 2019 1:36 am

Chuckles58 wrote:And yet there may be more folk like me that would like to learn Age Building, but have the same Busy Life situation. I would hope this community can hang in there. During Mysterium, Rand mentioned that he really loves Uru.


This! I'd LOVE to create Ages, but unfortunately my experience with any kind of digital 3D creation is zilch. If there was a way to learn that I'd have a go, but I've no idea where to even start. :roll:
User avatar
Lehnah
 
Posts: 12
Joined: Sun Jun 17, 2018 1:37 pm

Re: "Alternate" Age Creation?

Postby Tweek » Fri Sep 06, 2019 2:19 am

There is, my 3D skills were zilch until I just started doing it.

The GoW Wiki has a whole bunch of guides and info on how to build Ages.

Just grab a copy of Blender, a copy of Korman and just start tinkering away. That's how most of us have learned.
Beneath - IC Blog.
Beneath: Ages of Tweek - FB Age Dev Page.
User avatar
Tweek
 
Posts: 685
Joined: Sat Sep 29, 2007 6:37 am

Re: "Alternate" Age Creation?

Postby Sirius » Fri Sep 06, 2019 9:51 am

Tweek wrote:There is, my 3D skills were zilch until I just started doing it.

True, we all start from zero. But it's more a question of available time and dedication, IMHO. It's extremely hard to learn something completely new after a full workday.
But I agree whoever wants to build Ages should just get started whenever they feel like it, and focus on the fun parts of Age building. An Age doesn't need perfect visuals or complex puzzles to be good. Unless you're looking to be a professional 3D artist, but that's another story.

I'd love an MMOG based on the Uru universe, where the gameplay loop revolves around the idea of exploring other player's Ages to gather "knowledge", then build your own Age using a simple editor, let other people explore it, and so on. I think this would suit a lot of Myst fans, and provide a nearly unlimited amount of fan Ages.
Of course, creating such a game would definitely not be trivial, but I've given it a lot of thought over time.
User avatar
Sirius
 
Posts: 1184
Joined: Mon Jul 26, 2010 4:46 am
Location: France

Re: "Alternate" Age Creation?

Postby Deledrius » Fri Sep 06, 2019 11:15 am

Lehnah wrote:
Chuckles58 wrote:And yet there may be more folk like me that would like to learn Age Building, but have the same Busy Life situation. I would hope this community can hang in there. During Mysterium, Rand mentioned that he really loves Uru.


This! I'd LOVE to create Ages, but unfortunately my experience with any kind of digital 3D creation is zilch. If there was a way to learn that I'd have a go, but I've no idea where to even start. :roll:


The really wonderful thing about Blender (well, I personally think there are a lot of wonderful things about it, and I've tried some of the super-pricey tools, too) is that there is no shortage of video tutorials on YouTube that will teach you how to get started and do some really amazing things. The only other tool I tend to see as many videos for is probably ZBrush, but that's not particularly relevant directly to the sort of thing one would be making for Uru anyway.

Like any other skill, learning to make art requires a lot of time, devotion and determination.


Sirius wrote:
Tweek wrote:There is, my 3D skills were zilch until I just started doing it.

True, we all start from zero. But it's more a question of available time and dedication, IMHO. It's extremely hard to learn something completely new after a full workday.
But I agree whoever wants to build Ages should just get started whenever they feel like it, and focus on the fun parts of Age building. An Age doesn't need perfect visuals or complex puzzles to be good. Unless you're looking to be a professional 3D artist, but that's another story.

In fact, I would say that a good way to get started which doesn't require too much forward investment is to block things out with a minimum of detail. Get a sense of the size, layout, and gameplay of the Age you want to make. Sadly, a lot of Fan Ages seem to stop here, throwing at best some basic tiled textures onto things with no lighting at all and releasing it, but it's a good place to start, and you can add to it as you learn. Importantly, it also means it costs less (in time and effort) to change things, and you will need to change things a lot from your initial design when you're inexperienced.

Of course, this doesn't work for everyone. If you're doing something a lot more artistically-focused and just making an artistic Garden with little interactivity, you may focus mostly on the composition, detail, lighting, right from the start and do something tight and small.

Honestly, I wish I were better at following this advice myself! I have a tendency to get stuck on details before I've gotten far in planning.

Sirius wrote:I'd love an MMOG based on the Uru universe, where the gameplay loop revolves around the idea of exploring other player's Ages to gather "knowledge", then build your own Age using a simple editor, let other people explore it, and so on. I think this would suit a lot of Myst fans, and provide a nearly unlimited amount of fan Ages.
Of course, creating such a game would definitely not be trivial, but I've given it a lot of thought over time.

People ask for this all the time. I know that you know how much work it would be, but generally it seems people don't. Not if the goal is something that doesn't just produce cookie-cutter Ages, or Minecraft-Alpha-level structures. Just like was discussed in the other thread regarding tooling, finding the sweet spot between easy-to-use and flexibility is not simple, and in the case of something like this it's also a bell curve in terms of quality. Make it too simple and you can't make anything good. Make it too flexible and the difficulty in making anything good is immense. Right in the middle is great, but that's the only place where novices can still get anything done worth making, and it takes the most effort to create a tool like that (and the assets to go with it).
User avatar
Deledrius
Gehn Shard Admin
 
Posts: 1281
Joined: Mon Oct 01, 2007 1:21 pm

PreviousNext

Return to Off-Topic Discussion

Who is online

Users browsing this forum: No registered users and 1 guest

cron