Bloodknight

A lot of the assertions being sent from the server won't update much beyond a set necessity, the only major things will be player positions and their impacts therein on the world ... of course all being managed on an authoritative basis.

Translation, all the server will be doing is everything.

The torque server will interface pretty securely with all of your 3 different servers pretty easily. There is clear separation of client and server so thats secure, and if you believe you have a network developer thats smart enough to create a better networking system than the torque one, then they should be smart enough to know how to spend 10% of that time adding extra security to that system instead of re-inventing the wheel.

You seem to be fixated on an area of security which tbh is probably unachievable in real terms, regardless of which engine you use, even if you write your own. There are ways of course, but then you are heading somewhat down the same path as starforce and the same sort of destructive issues. Simply put, security and inconvenience are partners in crime, the more secure you want something the more you inconvenience your users, this will remain a truism until the day night cycle of the planet ends.

I guess this could be merged with a showoff forum or merging of more than one, but there seems to be more and more games getting to completion, and getting on greenlight, so i thought it might be an idea so that we can track various games, by we i mean people like me who like infinite level categorisation, and by track i mean 'oh i saw such and such do that with a game a while back, let me go spend 30 minutes not being able to find the link again'

you get the picture :p

My two pet missing features (haven't checked lately).

Properly and correctly working snapping (worked like a dream in TGE/A)

press and hold hotkeys for switching between move scale and rotate

These two things together made banging out a level very quick, making prototyping incredibly efficient.

Sure, some of the tools have become more advanced, this out of necessity often makes them slower/less efficient, we had a short discussion last week about a level editing video from GDC, which makes some interesting points about various level editors. The end result being a pseudo agreement that some review might be worth doing at some point as a solid single focus project improvement path for T3D. There is also some work done on a new editor layout which i think is now discontinued, which looked very promising (Provides a kick start for this project or at least a prototype).

I think tbh the best thing to do would be to start with an asset store, modeled somewhat after opengameart but with a requirement that packs come 'torque ready' so that they become easily usable for novices. I would also recommend that in the beginning at least while the repository gains 'stock' that the requirement starts of non-commercial. Expansion to commercial and alternate categories could be added relatively easily.

Also, it has to be noted that my base pessimist nature has to make comments about the insane levels of lawyerised bullshit around software licences and image copyrights emanating from the USA which makes even a free 'store' something of a logistical and legal nightmare in some respects

In the past getting through greenlight was very hard, after some time it became easier, but maybe it has changed again and it has become a little harder again, I don't know.

Steam is fast becoming one of the worst and most expensive platforms, despite the never ending sales.

Just because steam has millions of users does not mean that sales are higher on steam.Finding something worth playing on steam is borderline impossible, its like trying to find decent software for a mobile device, pointless time waste. There is so much to wade through, and steam picks and chooses what sells

Given the amount of trash on steam, along with games that still cannot get greenlit, it appears that steam greenlight is 100% arbitrary.

Simple proof of this, ubergame a minor mod of stock torque that just happened to get greenlight vs Airship dragoon that had to 'fight' for thousands of votes.

While i understand that no mother admits their baby is ugly in public....

If you have a real serious game and not some obvious ripoff or trash than you should be able to get through greenlight very easily.

Did you seriously keep a straight face while typing this?

... since Torsion does not work or not work well on LInux...

These terms should be mutually exclusive, it either works or not, if it works then its quality is variable.

Given that torsion is now MIT, and given that it uses wx as its UI it should operate on a cross platform basis. I know its old and outdated but it functions, and its minimalist feature set already improves over and above just simple syntax highlighting that other editors offer.

If all you need is an editor with highlighting, then probably geany is a good choice, it has project options and highlighting which covers most of torquescript most people ever need *.

* Also has plugin capability so it is possible to fully adapt to torque

the matrix functions are all inside source/math/mMatrixF.h-cpp, but a quick search for matrixF should find you some examples, as of yet i have not looked into how either GL or DX implement these, youd have to ask one of the smarter guys.

As a side note, even opengl removed these functions from their implemntations, there is currently some hapless attempt to add the glm library to the engine, which adds these functions independently https://github.com/Bloodknight/Torque3D/tree/add_GLM/Engine/lib/glm if this is more useful

Surely the player just needs to know the gravity vector, as far as i can tell this is the only difference between 'regular' terrain and spherical terrain

There was somebody else working on speherical terrains a couple of years ago, tho in a much smaller scale, more like the british gas ads or that one little world in dragonball z, unless that was you, have you changed usernames in the past half decade or so?

anyway, upshot is, if you aregoing to have multiple planets of multiple sizes then you are likely to need a variable gravity vector anyway, how difficult this is i dont know, since at least some of the gravity system is hard coded magic numbers

i have a folder full of all sorts of things..

a complete dump of abighole unsure of date..

a collection of single page saved web pages from the resources section, granted most of this stuff is still live and almost certainly very out of date.

i have a collection somewhere of TDN pages, but i am also acutely aware that not everyone believes as i do about the sharing of knowledge so generally i dont share, however, since 1) pretty much most of if not all previous paid resources are now MIT 2) this junk is a decade out of date and of debatable use and 3) i'm not sure who can even access TDN anymore.

as a side note, short of the doc master (@GameDevMich) i dont know who has a full or complete copy of TDN, so if anybody has knowledge of how to rip a mediawiki site that needs a login let me know.

anyway, like a bunch of people, i have a lot of accumulated junk/wisdom taking up HDD space, not sure if theres somewhere i can dump it, or if i should, while i consider this if there is something specific you wanted i could always look for it

So, 3rd time in a week I've crashed the engine in pngwrite.c

callstack of most recent: http://pastebin.com/SiBZQSbX

I will endeavour to be more diligent, since the crashes while in the same file has been at 2 different points, so they are not consistent, nor to they trigger consistently since I've repeated the work and not crashed. I haven't posted an issue on GitHub as I believe this need more exploring, the next week or so will decide if it needs to be promoted i guess.

For the record:

NVidia GPU

AMD CPU

64-bit build

Windows 10

DX9 (yeah yeah, I know)

The answer is always to change the C++ code if you cannot learn how to be flexible.

You need to check on the callbacks, some objects have onCollision callbacks some do not, as I recall TSStatics are pretty much geometry only, they aren't real objects, if you want to know hat your running over then you need to change the objects to a scripted object type, but bear in mind this will obviously have some overhead.

Coding alternatives are to create a new lightweight class based on tsstatic and add a few choice options or alter the tsstatic class if you want all objects to be smart objects in some way.

Also, I feel I should add, that for a decade or so now I have been the community leader as far as grumpy sarcastic is concerned, and feel you are cruising a little close to my turf, also I should warn you that this is a poor way of getting answers, trust me, I'm an expert.

You will get more fidelity at a cost, the terrain LOD operates on a division type algorithm last I looked at the parts I understood, so while it does work, its still going to be heavier on memory in some respects. That being said 1 2k height map is a 2k heightmap probably not much difference, but as mentioned if you start getting bigger then more memory will be an issue, I do some experimenting pre x64 times with large terrains, I should probably revisit my 8k terrain at some point.

the height is pseudo based on the squaresize if you experiment with square size in the game, and then re-import a height map at the same square size you will see what I mean.

Well, getting stuck more is probably a result of having the higher horizontal resolution while keeping the same vertical size - so more polygons that exceed the "max walkable" slope. So one would have to ensure that the source height map was relatively free of noise.

No, you also get stuck because of more edged terrain.

both saying the same thing since you can only get 'edges' with angled polys.

disclaimer: only read the last post, i'm lazy, sue me.

If you are moving files and folders around why would you move them into core rather than out of core? seems to me core is more about the engine initialization and less game related core functions maybe.

Why not separate files in a way that makes separating client and server easier for those that want to make multiplayer games

I think the best solution might be to identify the material as one that doesn't use decals at all, while I don't know if it is in the engine or not (I don't use weapons) there is a resource somewhere that implements different decals on different materials, this should include a method to prohibit decals.

I'm wondering if your base approach is even the best option here, seems to me that much like a PC/NPC that has a variety of armour sets and your vehicle seems to be an 'upgradeable' in a similar way to the ship in darkstar one. At which point it might be better to create the extras as actual parts of the vehicle model and use mesh hiding techniques, this allows you to create better/custom collision meshes. Just a thought :)

and to hijack the thread a little from an earlier comment;

"'these things seem to have been solved by other but haven't been added to the main repo" paraphrasing sorry for that

Partly as az says there are limited numbers of active devs around at the moment, but if you have found a collection of 'solutions' be it useful mods or actual solutions to real problems that haven't been added to the main engine itself, then it would be a great help if you could make a thread and linking these. Even if you don't want to list them all, then out of self-interest post a list of ones that you think would help you and your project, and perhaps others will follow suit.

Part and parcel of small numbers of active devs are their ability to move outside of their own zones, they check out the forums here the issues on GitHub and maybe IRC channels, and they pursue their own games interests, if their game genre is different to that of yours or somebody else then the chances are they may not have even seen the problem or the solution. A list allows them to both see that a solution has been created at one point so that they don't have to re-invent the wheel while implementing a solution.

you are assuming 4.0 is the next branch, which is potentially flawed. it might be 3.10, 3.11 and so on.

Unless there are plans to maintain more than one release then the master/develop branch system is sufficient.

All major features should be on independent branches anyway. I'm assuming its the SC's job to fix all the conflicts caused by integrating PRs in whatever order they do them in (would seem a little unfair to send somebodies work back because it no longer merges cleanly)

I guess some of those might be more useful on the main repo instead of having to look through a dozen different repos for a specific feature.

you broke it here btw,

http://i.imgur.com/6ukPosP.png

cross branch communication should only be one way

I have no idea who moderates or works the forums, but it might be useful if common internet methodology is used and threads like this that have an actual solution posted could be locked off and a [solved] tag prepended to the title. I know we have a small community here, but i still think there's a larger quieter community who use the engine and run into issues finding a [solved] link always makes my day a little brighter when searching solutions for problems. :)

Most/many file extensions are used more than once anyway, this is not uncommon. The chances of finding an 'unused' extension are slim to none, the only real concern is that the extension clashes with another current and more commonly used extension.

Somewhere and somehow doesn't matter what you pick short of breaking the whole 8.3 or at least the .3 part (which I personally find very annoying when I find this) there is going to be a clash.

Personally, I change almost all of my 'visual studio' file associations so that I don't mistakenly launch the beast anyway, so I set codeblocks or notepad++ to open single code files.

The vagrant tutorial on the wiki fails due to out of date software, I'm going to review a document I made a long time ago if for no other reason I'm pretty sure somebody somewhere linked it to the public lol

that looks a massive improvement over the last screenies i saw for macs

I guess the first test should be opening the collada export back in the artists package to review what the exporter has done to the ngon.

next test would be what the game engine does to the collada file.

I am not artist enough to know anything about ngons other than its trickery at best, horrible organisation at worst, there is a reason triangles are used over and above all other geometric shapes, quads at least have a logical reason for existing, i can't see any logical reasoning for ngons personally, not even topology.

somebody who actually writes pure graphics routines would have to confirm my theory that even opengl/directdraw only use triangles, even if they do order them into strips, fans, etc.

Bonus points for not adding an ugly soundtrack

Nice to see a visualisation of the cmake process, would have been useful for me last year when i first got introduced to cmake.

I have couple of points i think could be raised/reviewed.

-- CMAke will create the folders as appropriate for its own build files if the actual folder doesnt exist, so creating the cmake folder is not needed.

-- Small amount of confusion may be cause over the needless copying of the template folder a) to a new template b) to the my projects folder; As long as the torque app name is the same as the folder name you chose under my projects you dont need to copy any files manually at all, the build install project does this all for you

-- More small confusion may be caused by you using two different projects in the videos, full and newfull

*side note I'd recommend making a new CMake build folder rather than using buildfiles and then delete the entire buildfiles folder unless you need it for some reason.

my generic complaint many of these video types is that 'shortcutting' cause confusion, eg not actually building the project (you can always edit out the time it trakes to compile and copy files), usually not an issue if the video is voiced as sometimes its explained that shortcuts are taken.

please take comments as helpful tips rather than nitpicking :)

I'm not forcing to use any distro, but as recommendation don't use Ubuntu or Ubuntu family :lol: and a professional recommendation, if you want at some point work in Linux, stay away of Ubuntu, if you used, you will learn Ubuntu not Linux, and that is the problem.

http://i.imgur.com/UgKrxPl.png

also

http://i.imgur.com/Xa1nuwu.png

meh

I was under the impression that triangulation occurs in the end regardless of quads/ngons which really as far as io know are just 3d editor things hiding triangles.

The simplest thing would be to just try, and then inspect the model in the engine