Work Blog - JeffR

455 posts Page 43 of 46
Bloodknight
Posts: 288
Joined: Tue Feb 03, 2015 8:58 pm
by Bloodknight » Sun Dec 01, 2019 11:36 am
Might seem a little late in the day, but I wonder if we shouldn't have a specific forum for just 4.0 Preview?

I'm doing a lot of testing and as a consequence of my anti-Midas touch gift... a lot of breaking, my general dislike of GitHub black hole system for tracking 'issues' hasn't changed, so a more central and usable location for all discussions specific to preview 4.0 would be useful I believe.

It would also clear the cruft out of this particular thread, which should be just an announcement thread imo.
marauder2k9
Posts: 372
Joined: Wed Feb 18, 2015 7:36 am
by marauder2k9 » Sun Dec 01, 2019 2:53 pm
there is a specific forum for 4.0 preview viewtopic.php?f=40&t=1594
Bloodknight
Posts: 288
Joined: Tue Feb 03, 2015 8:58 pm
by Bloodknight » Sun Dec 01, 2019 3:37 pm
I did actually mean a forum so that individual issues could be posted, from my end mostly testing and bugs, but also requests for documentation on features, not an individual thread.

I could continue to use discord, but most of it is unread/ignored/forgotten; discord has its uses but it is pretty useless for most things that need to kept or located or referenced.
suncaller
Posts: 22
Joined: Mon Apr 09, 2018 1:03 am
by suncaller » Sun Dec 01, 2019 6:27 pm
I think there are a few critical realization here, that might be missing, and are important to actually speaking about Torque -- or any software development tool -- in a meaningful way, that isn't just, "Well, I like the engine, but add this, that, and the other thing, and I'll consider using it." Torque3D fulfills a different function, for a different set of people, than many other engines (i.e. Unity, Unreal, etc.). If there's any triple-A engine it could be somewhat -- but not very -- reasonably compared to, setting aside budgets, staffing, adoption, and other similar concerns, then it would be Unreal Engine. Even then, only in the sense of why someone would adopt T3D instead of Unreal Engine, and it probably applied more before the current iteration. And, yes, there are very good reasons. They just probably aren't the reasons most of the people making demands or offering vague suggestions have ever thought about in a clear and meaningful way.

The reality is, most novice developers don't even know where to start asking the right questions, and the problem is that they start offering vague, and offensively worded solutions in the face of that. Do those making such claims, in similar situations, perhaps see why that makes people angry, and may not be entirely appropriate?

More realistically, T3D might be compared to other open source engines such as Godot, Urho, Quake, etc, and then only along the vectors that matter. These engines also don't solve the same problems for the same sets of people. That's a good thing. They are options that exist for different groups with different needs.

What are appropriate questions to ask?

For starters, let's try: given options x, y, and z, all tools that seem to me like they will contribute to the completion of my goals, which will help me reach those goals while possibly also, as a bonus, meeting my secondary and tertiary requirements? As opposed to: tool x has good features, but it's missing other good features from tool y -- why aren't the people behind tool x implementing the features from tool y, so that I don't have to think about this problem?

Other important questions, that I'm not sure enough people ask themselves, might be: what am I doing; why am I doing it; what do I see as the best and worst case outcomes; am I trying to maximize one side, or minimize the other? It might be useful to start there, and iterate until the right path shows itself.

Another, possibly unrecognized realization, is that unlike other parties offering "free" tools, the reason for this one are very different. Az has done a good job of pointing out what some of those are. But is it clear why other companies are offering free solutions, for those that seem to think "consumers" are powerful, driving entities in this domain? This engine has no need to harvest vast amounts of personal data from users; or foster adoption from a truly mind-boggling number of people to recoup losses or drive sales; or to establish dominance in terms of mind and market share with the hope of creating a monopoly in the industry.

Anyway, that's all just a brief attempt at offering perspective, from one hobbyist to others that seem to need it, and genuinely not intended to be offensive. If it did offend anyone, I'm glad, and I hope you take that pain and doing something positive with it.
Razer
Posts: 82
Joined: Tue Jan 10, 2017 11:29 am
by Razer » Mon Dec 02, 2019 9:06 pm
Finally, what is the roadmap priority about :
- New character physic driven system with capsule and ragdoll collisions

not a high priority
I think beginners will ask for a good and versatile character system with physics, this is perhaps one of the most important features, because most beginners wil try to mimik games like Uncharted , Dark Souls or Halo, and many times they will try to make some open world game around characters.
The old T3D character and strange cubes collision system should be replaced with new characters physics and new separate control and camera scripts.

- Support modern character import like GLFT or get Blender 2.8 support

assimp does that. may or may not need a few additional tweaks. will have to check with one of the folks that already added it to the engine instead of demanding others do all the work for them.
Beeing unable to import your characters and animations you made or you bought or having issues and you're stuck on your project.
I got some characters from some stores, i can do tests when T3D will get import fully working.
Also Blender 2.8 is the open source leading software, getting some direct import for Blender files can be something to consider as T3D is also open source.


Like @ Bishop said, Godot can be a great inspiration for T3D in some areas like project management that seems to be better.

Image

Finally, what is the roadmap priority about :
- New character physic driven system with capsule and ragdoll collisions
not a high priority
- Support modern character import like GLFT or get Blender 2.8 support
assimp does that. may or may not need a few additional tweaks. will have to check with one of the folks that already added it to the engine instead of demanding others do all the work for them.
- Add a general game input system no more relying on old T3D code and system, able to map any joystick values or keys to actions
no. we've already got joystick et al support. it's pointless to redo it from scratch.
- Adopt a popular scripting languages like C# instead of Torsion nobody knows
you're getting also-c# from lukas. drop the instead of garbage. it's not going to happen.
- Entity system, and the ability to add scripts to objects
frontend is already protoyped. and mentioned in this thread.
- Better projects organization with scenes management per project
better is a useless datapoint. scenes are in, and they trip game modes. be specific
- Easy system to store, read game data and organize data structures, similar to databases.
asset system button is
- XML integrated save/load game system
that's a 'you' problem. scripts already got hooks to write/read text-based files. I've even done html output.
that being said, if you actually tested the system you might note that .taml files *are* xml.
- Terrain,foliage and water enhancement with modern shaders and lighting
being worked on
- Cinematic quality post effects : DOF, Aces tone mapping, cinematic bloom
being worked on
I'm pretty happy with T3D direction. But when all those will be ready :mrgreen: ?
Last edited by Razer on Thu Dec 05, 2019 8:52 pm, edited 1 time in total.
Bloodknight
Posts: 288
Joined: Tue Feb 03, 2015 8:58 pm
by Bloodknight » Tue Dec 03, 2019 12:49 pm
- Support modern character import like GLFT or get Blender 2.8 support

assimp does that. may or may not need a few additional tweaks. will have to check with one of the folks that already added it to the engine instead of demanding others do all the work for them.
Beeing unable to import your characters and animations you made or you bought or having issues and you're stuck on your project.
I got some characters from some stores, i can do tests when T3D will get import fully working.
Also Blender 2.8 is the open source leading software, getting some direct import for Blender files can be something to consider as T3D is also open source.
I don't understand the 'this is open source we should worship the same false idols' mantra, do you go to the same Stalman Zealot group as Duion?

you are harassing the wrong people try here https://github.com/assimp/assimp :p

also, if you are asset flipping, unity is #1
Razer
Posts: 82
Joined: Tue Jan 10, 2017 11:29 am
by Razer » Thu Dec 05, 2019 9:02 pm
- Support modern character import like GLFT or get Blender 2.8 support

assimp does that. may or may not need a few additional tweaks. will have to check with one of the folks that already added it to the engine instead of demanding others do all the work for them.
Beeing unable to import your characters and animations you made or you bought or having issues and you're stuck on your project.
I got some characters from some stores, i can do tests when T3D will get import fully working.
Also Blender 2.8 is the open source leading software, getting some direct import for Blender files can be something to consider as T3D is also open source.
I don't understand the 'this is open source we should worship the same false idols' mantra, do you go to the same Stalman Zealot group as Duion?

you are harassing the wrong people try here https://github.com/assimp/assimp :p

also, if you are asset flipping, unity is #1


For example UE4 has a very strong character system as a starting template, does that mean that all published games are called assetflip ?
Using assimp is one more step to convert files, while T3D should use a modern import, like showing up the import dialogbox when it detects some new files on the project folder.

And i don't know about Duion group and what it is about :lol:
Bloodknight
Posts: 288
Joined: Tue Feb 03, 2015 8:58 pm
by Bloodknight » Thu Dec 05, 2019 11:17 pm
Assimp is an inline process with torque, so arguably it does detect all the model types it can convert, and if you want to volunteer to rewrite the torque blender importer every time blender destroys backward compatibility, and you know somebody who can do the same for all the other asset types out there or even the common ones there's only a dozen of those.

I also actually disagree about the asset import method, it should be on-user-demand not driving the demand. As far as existing models are concerned, torque wholly lets you mess up by auto importing a 'changed' model, which btw is semi hilarious when you misname your export...

Torque has already been tested and working with blender 2.7 files (screw blender org and its backward screwing design changes), FBX (the single most common asset used in games today) GLTF 1 & 2 (looking forward to trying this myself soon) Lightwave (still surprisingly common) aside from a few proprietary formats for which there are weird issues anyway, I think it does pretty well for asset handling.

At this point, I wonder if you are even trying to look at torque or just stamping your foot because your list of demands have not been met yet :p
Razer
Posts: 82
Joined: Tue Jan 10, 2017 11:29 am
by Razer » Sat Dec 07, 2019 12:29 am
At this point, I wonder if you are even trying to look at torque or just stamping your foot because your list of demands have not been met yet :p
Already downloaded and tried T3D 4.
This is not requests but observations about things that matters to me in a game engine, while i understand some people can find those irrelevant or useless.
I already have a game engine and i don't need a new one, but i keep an eye on T3D 4, it's open source and it could become very interesting when it will reach Godot actual state about features and usability.
I also actually disagree about the asset import method, it should be on-user-demand not driving the demand. As far as existing models are concerned, torque wholly lets you mess up by auto importing a 'changed' model, which btw is semi hilarious when you misname your export...
I like to drop some 3D models to a project folder directly, and automatically have the game engine show me the import options right away ( like Godot for example ). While this could be some option people could set "on" or "off", i can understand some people does not like that "modern" way. But forget this idea.
Torque has already been tested and working with blender 2.7 files (screw blender org and its backward screwing design changes), FBX (the single most common asset used in games today) GLTF 1 & 2 (looking forward to trying this myself soon) Lightwave (still surprisingly common) aside from a few proprietary formats for which there are weird issues anyway, I think it does pretty well for asset handling.
I never tested T3D import, that's good news.
Shape editor proposes "dae" import mostly, there is no gltf or fbx ? Is there some import tutorial ?

Be reassured I won't stamp more ideas for T3D.
I never felt like writing those as "requests" someone would felt forced to work on right away.
Did someone really started working on all those ideas ? :lol:

While i agree 100% with @ JeffR to focus on indie games, the majority of indie games are not aiming at AAA graphics, that's not possible for three or ten person team.
"I've made it pretty clear on the discord that I angle at things from a "does this actually help people get their games made" and all the extra cool doodads are just bulletpoint fluff. Yeah, it's NEAT to have a full augmented reality, realtime cinematic blending of prerendered movies and game render and all that jazz. But 99.99% of people aren't ever going to use it for their games. We're not in the business of making movies. I'll leave that to hollywood and 3ds max, maya, blender, and all that"
Razer
Posts: 82
Joined: Tue Jan 10, 2017 11:29 am
by Razer » Sat Dec 07, 2019 10:56 am
At this point, I wonder if you are even trying to look at torque or just stamping your foot because your list of demands have not been met yet :p
Already downloaded and tried T3D 4.
Those i wrote are observations about things that matters to me in a game engine, while i understand those can be irrelevant and useless for other people. I already have a game engine and don't need a new one, but i keep an eye on T3D 4, it's open source and it could become very interesting when it will reach for for Godot for example about usablility and interface maturity.
I also actually disagree about the asset import method, it should be on-user-demand not driving the demand.
I like to drop 3D models into a project folder and have the engine automatically show the import options right away (like Unity or Godot do).
While this could be some option people could set "on" or "off", i understand some people does not like that "modern" way. But forget this idea.
Torque has already been tested and working with blender 2.7 files (screw blender org and its backward screwing design changes), FBX (the single most common asset used in games today) GLTF 1 & 2 (looking forward to trying this myself soon) Lightwave (still surprisingly common) aside from a few proprietary formats for which there are weird issues anyway, I think it does pretty well for asset handling.
Shape editor proposes "dae" import mostly, there is no gltf or fbx ? Is there some import tutorials ?

Be reassured I won't stamp more ideas for T3D and i never felt like writing those as "requests" someone would felt forced to work on right away. Did someone really started working on all those ideas ? :lol:

While i agree 100% with @ JeffR to focus on indie games, the majority of indie games are not aiming at AAA graphics, that's not possible for three or ten person team, even most indie games are not about realistic and complex characters graphics. T3D should focus on user experience about usability.
"I've made it pretty clear on the discord that I angle at things from a "does this actually help people get their games made" and all the extra cool doodads are just bulletpoint fluff. Yeah, it's NEAT to have a full augmented reality, realtime cinematic blending of prerendered movies and game render and all that jazz. But 99.99% of people aren't ever going to use it for their games. We're not in the business of making movies. I'll leave that to hollywood and 3ds max, maya, blender, and all that"
455 posts Page 43 of 46

Who is online

Users browsing this forum: No registered users and 1 guest