Arma 3 Engine - What would have been a better option and what can we learn?

[fabrizio_t 58]

Despite the age, the problems and the true need for refactoring i still have to see an engine on par with RV for overall features and possibility of modding.

Edited December 24, 2013 by fabrizio_T

[ProfTournesol 956]

Despite the age, the problems and the true need for refactoring i still i have to see an engine on par with RV for overall features and possibility of modding.

Exactly.

[calin_banc 19]

Despite the age, the problems and the true need for refactoring i still have to see an engine on par with RV for overall features and possibility of modding.

In functionality... perhaps, but that's mostly because a lot of developers are concentrating on the consoles and their weak hardware. Modds not so much, a few great ones are for a lot of games.

The problem was, that until ArmA3 came around, they were doing a lot of stuff, but doing it at a mediocre level, at least for infantry. Things have changed, they've made improvements, but still, quite a little bit far from where they could be. Snowdrop engine, FB3, Rage, U4, the one behind Watch Dogs and so on, all look like good way ahead of everything is today on market. Sure, you'll have bigger maps, but it will run poorly and be mostly empty without some changes. :)

[[evo] dan 79]

Maybe they could have learned how to prevent physx from crashing the game so often, we all remember that patch, and I still have them on a semi regular basis (and I am not the only one).

[bad benson 1733]

we need no new engine because this is the one that does the job. the others still have to prove they are even remotely capable of it.

what we need though is full physX support and not some half assed "it's in there, i'm done here" approach.

gearboxes for cars have been brought up by modders trying to make cars behave more realistic.

script access to ragdoll. more "physical" objects overall. the world is static. let physX help you achieve things.

improved destruction. simpler more effective.

proxies for magazines. more control over proxies overall.

custom animation sources for weapons.

improved "attachto" that makes it actually useful to achieve things that are impossible otherwise. it needs to follow bones in characters and not mempoints.

improved intersection commands. they are highly limited. most only react to the view lod. they need an option to which lod they should react. characters are ignored by them.

clean up of the animation system. proper implementation of anim masks to move away from the full body system to a modular one.

unbelievably unflexible attachment system (what happened there?).

i could go on for a while.

what is needed is some real work on what we got. cleaning up of several messes in the data structure.

with just the two commands i mentioned modders could do amazing things.

a lot of people don't realise that a lot of the frustration comes from modders who are the ones that get the closest look on things and are often painfully aware of things that never change eventhough they are not optimal and only cause problems. this engine is very capable. it just needs some love in the details. BI needs to stop and breath and do the right things for once. a solid frame will make the future amazing.

EDIT: event handlers. we need more eventhandlers so we can make more efficient scripts. there's none for reload. the hitpart one is pretty messed up and many more i can't think of now.

Edited December 25, 2013 by Bad Benson

[Make Love Not War 10]

Excellent rundown, Bad Benson.

This engine is very capable. it just needs some love in the details. BI needs to stop and breath and do the right things for once. a solid frame will make the future amazing.

Sums up my feelings perfectly. Considering BI's current cash flow - primarily thanks to DayZ - there's no better time than the present for a serious house cleaning.

EDIT: event handlers. we need more eventhandlers so we can make more efficient scripts. there's none for reload. the hitpart one is pretty messed up and many more i can't think of now.

More than anything, we need native stackable EHs. Functions such as BIS_fnc_addStackedEventhandler are a step in the right direction, but incomplete and should be considered a stop-gap measure at best.

[fabrizio_t 58]

Dan;2585948']Maybe they could have learned how to prevent physx from crashing the game so often' date=' we all remember that patch, and I still have them on a semi regular basis (and I am not the only one).[/quote']

physx was a relatively recent implementation, i'm sure support on it by BIS will grow.

---------- Post added at 20:28 ---------- Previous post was at 20:27 ----------

More than anything, we need native stackable EHs. Functions such as BIS_fnc_addStackedEventhandler are a step in the right direction, but incomplete and should be considered a stop-gap measure at best.

This.

[progamer 14]

we need no new engine because this is the one that does the job. the others still have to prove they are even remotely capable of it.

what we need though is full physX support and not some half assed "it's in there, i'm done here" approach.

gearboxes for cars have been brought up by modders trying to make cars behave more realistic.

script access to ragdoll. more "physical" objects overall. the world is static. let physX help you achieve things.

improved destruction. simpler more effective.

This^

Physx can give us highly realistic cars, walking on vehicles native support, better destruction than BF4 (We can destroy anything), key holes, more control over friction, actual waves and many other things.

BI said we would have great opportunities for modding with Physx, but we need full Physx integration first to make this a reality.

[zooloo75 834]

We need access to PhysX or native support for languages such as C#! Many of us have been waiting to have access to PhysX scripting commands with no luck.

Edited December 26, 2013 by zooloo75

[13islucky 10]

IIRC they were holding off adding too much PhysX while waiting for an upcoming version.

[rekrul 7]

GTA4-5 engine is mostly optimized for consoles, as on PC it was rubbish (worse than A2 or anything). If everything is well within the engine, why rewrite it?

The thing is, no engine is "well". GTA V has horrible physics, retarded AI, terrible spawning functions and there's almost no interactions with the world you're in. And they can't do much about it.

I'm not directing this at you but at everyone here who wants a "new" enigine - absolutely all games (at least any game with a number after it) are made on "old" engines. Do you think FIFA2014 has a new engine because they added some features to it? Do you think BF4 or COD8 or whatever it is now, has a new engine because it has more shaders? Hell, even a lot of new games are made on old engines. Stop calling it engines and refer to it as a frame work, because that's what it really is. If you want to throw out the frame work and start everything over because you want e.g. better AI, you might realize how silly that claim is.

And even if that happened, do you think they would not copy/paste in the majority of the code they already have and refined over the last 15 years into this new framework? What you really want is a huge developer making an ARMA-clone with a team skilled enough to cure cancer, large enough to release it within 6 months, the end product to be perfect, cost next to nothing and since it doesn't need any bug-fixes, you should get content and features for free. Because potato.

[progamer 14]

IIRC they were holding off adding too much PhysX while waiting for an upcoming version.

Source? Last I heard was the final version of Physx was added or will be shortly.

[13islucky 10]

After 15 minutes of searching I'm unable to find the source, so disregard what I wrote unless someone else can find a source.

[roshnak 41]

Do you think BF4 or COD8 or whatever it is now, has a new engine because it has more shaders?

Um, yes? Frostbite debuted in 2008. Call of Duty is still using the IW engine, but every Battlefield game since Bad Company has been on a completely different engine than everything leading up to it. Unless you want to make the stretch of an argument that IW engine is still id Tech 3, then RV is one of the oldest engines in the business.

Also, I believe we have already established that the last 13 years has not been so much about refining as it has been tacking new stuff on, as evidenced by the large amount of cleanup work BIS did for Arma 3.

Furthermore, the kind of hyperbole in the last part of your post isn't going to do anything to further the discussion. The only thing anyone wants is for Arma to be brought up to modern standards, especially if we are going to be paying AAA prices. I doubt anyone cares how it's done, and no one is pushing for 6 month development times. Especially since Arma 3 was in development for at least 3 years and is arguably still in development.

[rekrul 7]

Um, yes? Frostbite debuted in 2008. Call of Duty is still using the IW engine, but every Battlefield game since Bad Company has been on a completely different engine than everything leading up to it. Unless you want to make the stretch of an argument that IW engine is still id Tech 3, then RV is one of the oldest engines in the business.

Aren't all BF-games (since Bad Company) using the frostbite engine (various generations)? I'm not updated on those games to be honest but it would surprise me if they started from scratch on every release.

Furthermore, the kind of hyperbole in the last part of your post isn't going to do anything to further the discussion. The only thing anyone wants is for Arma to be brought up to modern standards, especially if we are going to be paying AAA prices. I doubt anyone cares how it's done, and no one is pushing for 6 month development times. Especially since Arma 3 was in development for at least 3 years and is arguably still in development.

Agreed on the first one, that was just my take on many of the claims in this thread. But how is it not modern? The engine is state of the art in its arena. All the other engines suggested certainly doesn't hold a candle to what A3 is offering.

I'm not saying that A3 is perfect, but every 6 months or so this discussion comes up and it's silly. You don't change the engine, you change the part(s) that has the problem. And we all agree that AI needs a serious revamp, but why would you scratch all the other parts that are working adequately?

[kklownboy 43]

RV is the only engine in town that can do what we want. And its better looking than any new gen right now.

If all you go by is FPS then your a sad camper that needs more power, less bad mission making. Its funny this argument, from posters who have no clue how game engines work or even about renders. And to top it off, all the rude abuse of the devs. Yeah its a PC game, and all PC games have issues. ALL of them. When I look at the trouble-shooting threads, and see the issues players have that I never have... Have to wonder. Most are about FPS. Most are about poor mission performance. But hey even the Missions with the game can be problematic.

You could never ever make the type of missions we all play,on any of the other engines out there. None of them. Let alone be a sandbox with all the assets from previous games. You think Crytec would, or UBi(lol), EA? hell no. Activision.... hahaha. Bohemia is the the ONLY game in town. So its better to stop thinking in 60fps is the end all, and start to think about all the different stuff you can do or play at 'playable" frames.

[roshnak 41]

RV is the only engine in town that can do what we want. And its better looking than any new gen right now.

I'm curious as to what makes you say that RV is the only engine capable of doing what it does. Don't mistake the fact that certain engines have only been used for certain games as evidence that they are only capable of making those types of games. Star Citizen is a pretty hugely ambitious project that is being developed for CryEngine. I'm certainly not advocating for licensing a different engine, but let's not make false statements about other engines' capabilities. It's also worth noting that game engines are not just a delivery vehicle for graphics, so the graphical fidelity of Arma 3 is pretty irrelevant.

So its better to stop thinking in 60fps is the end all, and start to think about all the different stuff you can do or play at 'playable" frames.

I'm not sure I understand this. You are saying that we should not be striving for a solid 60 FPS? I feel like performance is a pretty huge aspect of game, and since 60 FPS is pretty much the standard for smooth gameplay, then it should probably be our benchmark. Especially since so many things in Arma are framerate dependent. Ever notice how your machine gun starts shooting slower when you have less than ~40 FPS?

[MANTIA 55]

I think most of the replies in this thread come from people who think they know what a game engine is but really have no idea. Which is the case in almost all "this engine is better than this engine" debates.

[bad benson 1733]

What you really want is a huge developer making an ARMA-clone with a team skilled enough to cure cancer, large enough to release it within 6 months, the end product to be perfect, cost next to nothing and since it doesn't need any bug-fixes, you should get content and features for free. Because potato.

sounds good to me :p

seriously though. there is surely something in the middle of that and what arma is. one of the big problems with the arma engine is the lack of documentation. just look at how the devs themselves are tackling the AI issues. it's like they are modders.

code only becomes cryptic if you don't document it. by not doing that properly they created kind of a monster. i read that somewhere in an article about ofp's developement. not sure how deep that problem goes though.

the way things are done is just sloppy a lot of times. also on higher levels that are accessible to modders.

many things are just so unclever and unflexible. i mean i can only bring up the attachment system again. it's a mess if you consider that BI advertise "mod support" for their game themselves.

and actually a lot of these things that are engine level but not at the very core are like that. so they ARE changeable. the few new things in arma 3 show that. the inventory and clothes system is pretty good in my opinion.

dayZ shows that BI are capable of drastic changes, if they want to. it just needs more people to be watching them. too bad arma doesn't have such a hype to make them work harder...

i just hope that if they drop arma that they will make something as big and moddable as a replacement. after all that's what made something like dayZ even possible. pretty much everything that makes dayZ special comes from arma. that's why all the clones can never recreate that atmposphere.

Edited December 27, 2013 by Bad Benson

[artisanal 22]

This is what I read from a developer, posted on the forum of a moderatly known graphic engine/game, concerning the Outerra engine :

"But Arma3 is too soon for the technology ... and given how happy their players are now, and how mediocre their competition is, I feel nothing really bad would happen even if Arma4,5 were on the same RV engine."

Edited December 27, 2013 by Artisanal

[skreaper 10]

ive been obsessed with this game for a while now, thinking dayz will fill my craving, once i got that for christmas, i was wrong, i really needed the actual ARMA

could someone please gift me a copy of arma 2 or arma 3?

im a 20 year old beyond-full-time college student who doesnt have any money and i have to wait months and months to get my financial aid

all of my friends are on arma 3 because dayz is all cheaters and stuff, and everyone is telling me that if i love the combat in dayz, its really all about arma

im tired of playing stupid stuff like "armericas army" while im obsessed with arma, i wanna actually play arma

ive been wanting arma 2 for years and im obsessed with arma 3

[zooloo75 834]

ive been obsessed with this game for a while now, thinking dayz will fill my craving, once i got that for christmas, i was wrong, i really needed the actual ARMA

could someone please gift me a copy of arma 2 or arma 3?

im a 20 year old beyond-full-time college student who doesnt have any money and i have to wait months and months to get my financial aid

all of my friends are on arma 3 because dayz is all cheaters and stuff, and everyone is telling me that if i love the combat in dayz, its really all about arma

im tired of playing stupid stuff like "armericas army" while im obsessed with arma, i wanna actually play arma

ive been wanting arma 2 for years and im obsessed with arma 3

ArmA2 has a free version which is the full game minus high res textures, no campaign and no mod support.

[skreaper 10]

free version was taken down, apparently it can also be a bannable offense to hack ur way into downloading it again

[roshnak 41]

This is what I read from a developer, posted on the forum of a moderatly known graphic engine/game, concerning the Outerra engine :

"But Arma3 is too soon for the technology ... and given how happy their players are now, and how mediocre their competition is, I feel nothing really bad would happen even if Arma4,5 were on the same RV engine."

These could certainly be factors limiting the advancement of Arma. The competition isn't mediocre, it's nonexistent. There is also a group of people that react with a great deal of hostility when people criticize this series (which is to be expected with any niche product). So there really isn't anyone challenging BIS to be better but themselves. I don't know, that might be enough.

[TSAndrey 1]

free version was taken down, apparently it can also be a bannable offense to hack ur way into downloading it again

There are ways to download the A2 free version. And I don't see how anyone can ban you for doing that