I think its time for Bohemia to start thinking about creating a new game engine.

[343guiltyspark 10]

Right now we are still having to deal with a decrepit engine from the early 2000s.

While it has been a incredibly good engine delivering great performance for years , its showing its age and incompatibility with new directions in computer hardware and internet technologies.

We need a brand new engine designed from the ground up (IE not another modification of the operation flashpoint engine) to deliver the best arma experience possible.

It needs to take advantage of multiple core CPU's , this just cant be "added" in an update and have it work. This is because the way the game code works is still based in early 2000s game developer logic when multicore CPU's didnt even exist yet. Now we have something like 12 core cpu's and the game is only using 1.

This would be a massive task for bohemia. But if they honestly expect to push out anymore products on this engine it will only hurt them in the long term as computer hardware becomes more advanced and their engine gets bottlenecked when new features are stapled on.

Arma 3 is a perfect example.

We need an engine that makes these things its priorities

-Optimization of current and future computing technologies throughout the engines code.

-Maximum effort put into max view distances and object counts.

-AI adaptation to using buildings and structures.

What are some things you would like to see in a new engine for arma? Specifically as it relates to development and performance and less "features" like destructable buildings and such.

[Llano 11]

I want a engine that utilize my hardware as much as it can, so i do not have to run around on Altis with 20 fps.

Also, kinda worthless thread tbh. Everything is knows, nothing is done.

[vran. 13]

Well Take on Mars runs on Enforce engine:

http://en.wikipedia.org/wiki/Enforce_(game_engine)

There's also Enfusion engine which will run the Day-Z standalone. Whether either of these will replace RV for ArmA remains to be seen.

[bullet purveyor 85]

With all the game engine experts on these forums, you should knock your heads together and create the most awesome engine ever seen.

[343guiltyspark 10]

With all the game engine experts on these forums, you should knock your heads together and create the most awesome engine ever seen.

i dont have millions of dollars to pay a staff of software engineers. If i did i would

[nikiforos 450]

Well Take on Mars runs on Enforce engine:

http://en.wikipedia.org/wiki/Enforce_(game_engine)

There's also Enfusion engine which will run the Day-Z standalone. Whether either of these will replace RV for ArmA remains to be seen.

DayZ runs really bad right now and especially in cities so I see no great progress with enfusion or whatever. To be honest I have no expectations at all with Arma engine now and in the future.

[Arma3goodCPUlowFPS 12]

343guiltyspark - i agree with you.

Llano - we need more and more theards of these. i think that BI is not aware of how the majority of the community is pissed off. so its time that this forum gets bombed with such theards and that is what BI deserved MINIMUM!

noob1 - but do we know the new things of the enforce & enfusion engine ? have we informations of what these engines support ? does both engine supports 64bit ? proper multi-threading/multi-core support ? new lighting tech ? etc.

Bullet Purveyor - and you fanboy should start to accept that the majority of the community is pissed off with the old wrecked engine and that the engine has huge performance problems.

[bullet purveyor 85]

Why don't you and the "the entire pissed of community" move on and go play some of the super awesome 60fps games instead. Then us "fanboys" can stick around with this old wrecked engine.

Everyone knows about the performance limits. There is no magic fix and pointless threads like this in not gonna make BIS flush 15 years of work down the toilet and start from scratch. Enjoy the game for what it is now, or move on to something else. Sitting around waiting for pipe dreams to come through are not very constructive.

[Nexerius 10]

Oh yes. If ARMA 4 doesnt have a totally new engine, built from the start i dont think ill be getting it soon.

There are so many problems with the ARMA games that are just incredible they havent been fixed. I love the games, but i couldnt take the terrible FPS of A3 anymore and uninstalled it.

Thinking of going back to A2 though.

[Wiki 1558]

check this out:

http://www.rt-immersive.com/media

[Jona33 51]

You want a new game engine you can kiss goodbye to updates for, two, three years maybe. Look at it this way, there has been a new Image Generator (graphics engine) for DCS in the works for around three years now, they were estimating a release in 2012, and only now are we seeing teaser trailers and fairly solid confirmation it's this quarter. That's not even a brand new engine, that's only part of DCS. As has been pointed out the engine has been around for around 10 years, scrapping ten years worth of work is frankly moronic.

@ARMA3GoodCPULowFPS: You really need to grow up. Do you think you're Russell Brand? You're going to start a revolution where the players will rise up and BI will kneel down and build a new engine (going broke in the process)? All I've ever seen you do is bitch, call people fanboys, throw around words like "optimization" and make midly offensive memes.

[Arma3goodCPUlowFPS 12]

Why don't you and the "the entire pissed of community" move on and go play some of the super awesome 60fps games instead. Then us "fanboys" can stick around with this old wrecked engine.

Everyone knows about the performance limits. There is no magic fix and pointless threads like this in not gonna make BIS flush 15 years of work down the toilet and start from scratch. Enjoy the game for what it is now, or move on to something else. Sitting around waiting for pipe dreams to come through are not very constructive.

because we paid for this game so we as customers:

- has the right to say what it wrong with the game

- dont expect a 100% bugfree game, but we expect a performance friendly game which arma 3 isnt, not in MP and not in SP.

when you are pissed off about these types of theards, why dont you move on and go playing this game with its old wrecked engine and ignoring these theards instead of whining here because customers has enough good reason to be pissed off ?

[devilslayersbane 28]

because we paid for this game so we as customers:

- has the right to say what it wrong with the game

- dont expect a 100% bugfree game, but we expect a performance friendly game which arma 3 isnt, not in MP and not in SP.

when you are pissed off about these types of theards, why dont you move on and go playing this game with its old wrecked engine and ignoring these theards instead of whining here because customers has enough good reason to be pissed off ?

Here's another thing you paid for: A video game. It's a product to be sold by a company. I don't really know if you've truely considered every part of actually building a new engine from scratch. It's a process that can take years, especially when the team that works on the engine also works on the the franchise associated with it. On a purely financial level, it probably does BI a.s. no favors to actually build a new engine from the ground up when the one they have can be refined and built upon. That being said, there are many examples of really old engines being re-vamped and improved over the years, such as the ID-tech engine, the Source Engine, Crytek, and many others. While Arma 3's engine in general may not be performance friendly, there are very few open-world games that are. Examples include Skyrim, GTA IV&V, and even Minecraft all have performance issues related to CPU and memory usage. And these games are either a graphical downgrade (minecraft) or have much smaller worlds compared to ArmA 3.

Let's look at it another way, from a coding point of view. The ArmA 3 team is used to coding for the RV engine. No, it's not all they know, but they are used to it. Making such a huge shift means writing an entire new library of commands. Not only that but they'll need to develop new tools and extensions. It's taken 10 years to get the RV engine to handle a lot of this crap, and I still get 5-10 more FPS in both SP and MP on my dinosaur computer than I do on Arma 2. You have to realize that this doesn't happen overnight and it's always better to try and build on what you have.

Now, if BI doesn't want to build a new engine, frankly I'm ok with that. This engine feels right for this game. I'm not going to say that they need a new engine, but I will say that they do need work heavily on fixing performance issues once the expansion's been released.

[meshcarver 12]

Undergrounds- properly implemented... oh, and better performance.

I love the RV4 engine the way it is visually- there's something so sharp and defined about it and it fits the Stalker worlds so well too so I'm probably biased..! :)

We are not going to be getting a brand spanking new engine anytime soon so have to make do with this one.

[Arma3goodCPUlowFPS 12]

Here's another thing you paid for: A video game. It's a product to be sold by a company. I don't really know if you've truely considered every part of actually building a new engine from scratch. It's a process that can take years, especially when the team that works on the engine also works on the the franchise associated with it. On a purely financial level, it probably does BI a.s. no favors to actually build a new engine from the ground up when the one they have can be refined and built upon. That being said, there are many examples of really old engines being re-vamped and improved over the years, such as the ID-tech engine, the Source Engine, Crytek, and many others. While Arma 3's engine in general may not be performance friendly, there are very few open-world games that are. Examples include Skyrim, GTA IV&V, and even Minecraft all have performance issues related to CPU and memory usage. And these games are either a graphical downgrade (minecraft) or have much smaller worlds compared to ArmA 3.

Let's look at it another way, from a coding point of view. The ArmA 3 team is used to coding for the RV engine. No, it's not all they know, but they are used to it. Making such a huge shift means writing an entire new library of commands. Not only that but they'll need to develop new tools and extensions. It's taken 10 years to get the RV engine to handle a lot of this crap, and I still get 5-10 more FPS in both SP and MP on my dinosaur computer than I do on Arma 2. You have to realize that this doesn't happen overnight and it's always better to try and build on what you have.

Now, if BI doesn't want to build a new engine, frankly I'm ok with that. This engine feels right for this game. I'm not going to say that they need a new engine, but I will say that they do need work heavily on fixing performance issues once the expansion's been released.

i agree. the engine is not THAT bad if you not counting the performance in it. creating a new engine is not necessary, they should only make the current engine compatible with modern hardware so that the engine use all resources, even if its require huge or full rewrite of the engine.

[Sterlingarcherz101 15]

With all the game engine experts on these forums, you should knock your heads together and create the most awesome engine ever seen.

Bahahahha class. I'm off to start a separate tread to name this new wonder engine. I call it. The Amazeballs Jizzdrop engine.

---------- Post added at 02:05 ---------- Previous post was at 02:01 ----------

i agree..

When the new Jizzdrop engine is implemented. you will unfortunately need to set up a new forum account. Called arma3smexyai120fps

[Darkhorse 1-6 16]

Useless post by a cherry #38562390. *pins your medal on you*

(search before posting, man.)

[foffy 58]

I'm unsure if they really should make a new engine. They face a very interesting dilemma themselves: if they make a new engine and require an entire new toolset to make things on said engine, they're going to require the entire ArmA community to relearn an entirely different system of editing, scripting, and modification making. For a company heavily touting its flag on community and open content creation, starting from square zero may kill such a community in big ways. If it's already getting harder and harder to make content as technology advances - especially on the make-it-for-free fan side - starting from scratch seems like a scorched earth approach.

I think they'd be better suited at trying to optimize their engine, to really shine and polish it so it functions well, instead of almost always feeling like there are clear uh ohs the community is having with it. I love the ambition of the team to always try new terrain, new ideas for gameplay, so perhaps they are heavily hindered by their ambitions versus the raw technical and manpower it takes to get that to shine. The first thing is to retool the engine so it clearly takes more advantage of the hardware of the present period, though this is a bugbear for many developers.

[zukov 490]

I think its time for Bohemia to start thinking about creating a new game engine.

why not? tomorrow is ready! the devs tonight will work hard HAHHAHAHAHHA A HA HHAHAHHAHHAHAHAHHAHAH ;)

[il_padrino 1530]

I'm not going to say that they need a new engine, but I will say that they do need work heavily on fixing performance issues once the expansion's been released.

You do realize that the performance issues are in 99% of the cases tied to the engine being old as hell, and that if they keep on using it, we will never see serious improvements, right?

And besides, performance is probably the main problem, but surely not the only one.

But maybe you think that, for example, having ugly mid range textures like these in 2015 is not a problem.

[dmarkwick 261]

You do realize that the performance issues are in 99% of the cases tied to the engine being old as hell, and that if they keep on using it, we will never see serious improvements, right?

And besides, performance is probably the main problem, but surely not the only one.

But maybe you think that, for example, having ugly mid range textures like these in 2015 is not a problem.

Throwing arbitrary large percentage numbers around from your imagination coupled with some sort of unwarranted accusation on an unconnected topic will not get you much in the way of respect around here. If your belief is that ONLY a complete rewrite is the answer, then simply say so.

[CaptainAzimuth 714]

The Engine is good for Arma. Free look alt holding alt is best feature.

Haha, ok, but on a serious note now, it's completely reasonable for BIS not to completely throw out RV4. Perhaps what BI should do is build a new one, based on the RV Engine, and leave out irrelevant features and functions, and coding that is keeping it back. This would mean a new Engine, but able to do what it does now, but better. But again, the reality of this is the same as Solar Roadways. It's a great idea, it could potentially fix mufuggin everything, but to put this in motion would take lonesome amounts of time. Years. The only thing they can really do is optimize, and build on/improve on the Engine where possible.

[tonic-_- 53]

In a technical aspect the RV Engine is pretty rock solid, the part that holds it back and cripples performance is the way textures are rendered, as stated the engine is old and they use a pretty old technique for rendering the game. As obvious as it is the rendering part does not fully utilize a persons GPU. The engine was reviewed about two years ago by Nvidia and they gave their assessment on the engine which was stated above. If they actually spent the time to go through and get rid of some of the old ways and bring it up to current times I think with my opinion it could be one of the best engines out there from a technical standpoint. As far as being user friendly well we all can't be like CryEngine or Unreal Engine which is made in mind to be user friendly. You don't just design and program something to be user friendly unless you are looking to outsource it, if you know the ways who cares about the other people.

I see a lot of people complaining about the engine being old and all these graphical problems but lets be honest, what game have you seen that builds open terrain on the scale that Bohemia does and keep some-what high-fidelity? I would love to know because I don't know of any. Games built on this scale look like utter crap if you ask me and mainly because they are not meant to look great, hell ARMA 2 has some really sexy angles to it and that game was built in/around 2008 and released in 2009. If you have a problem with the engine then by all means go make your own, have fun dumping 5-6 years into that while also spending millions. No one re-invents the wheel anymore because it costs to much and it's not like they can switch because Bohemia is not the type of company to use another engine that they don't own rights to nor will any studio/publisher out there. They all use their own proprietary engine and software so stop barking up that tree.

I think they have been on the right track for 5 years now slowly getting up with the times and chucking out old parts that are obsolete but they are really just focusing their time on the wrong parts which is one of the problems with ARMA 3 as a game is currently facing, development being spent on the wrong problems, more focus towards the easily fixed problems but ehh.. That's game development/business for you.

That's just my two cents on the RV-Engine, I think it is a Splendidâ„¢ engine overall and have no doubt that it will see further improvement once they shift development of it in the right path and no I am not a fan-boy of Bohemia, actually the opposite.

[IndeedPete 1038]

I find it interesting that there are so many people claiming low hardware utilisation and poor optimisation.

I can't find any evidence with my system supporting the multi-core / CPU usage, I actually did some tests a while ago: http://forums.bistudio.com/showthread.php?183647-Good-Hardware-Monitoring-Gadgets&p=2782966&viewfull=1#post2782966

As you can see by usage and numbers only my CPU would be capable of running another Arma instance but my GPU runs at 99% with reasonable framerates considering the test cases. Of course, these are just numbers of my very own system and can't be seen as representatives for the whole community. But every time I see a thread about low FPS in SP or MP I'm thinking the OPs would help more with providing actual system specs and benchmarked numbers.

Anyway, what makes this engine so different from other game's engines is that one never knows what the users will do to it. Great modding possibilities like we have in Arma come with a certain unpredictability. Also, user generated content is often very pooly optimised or with many AI acting in small space - not necessarily the engine's fault. Plus the engine seems to be designed to be very flexible, it can be used for a wide range of tasks, like different addons, a vast amount of mission types - you can do everything with it. That's really great and one of the things I love about Arma but this generalisation comes with a price I guess. Other really great looking shooters, like CoD, Crysis or Battlefield can be optimised to death (for example with fully scripted AI), the devs of mentioned games know the possible scenarios and can optimise towards these.

As for the latter (poor optimisation) I don't really understand how so many people seem to know so much about that. Do you guys have access to the source code and are qualified to judge where it could be optimised? I don't. I've been studying and working in IT / Software Engineering for around six years now and I think I'm nowhere near the competence and inside knowledge needed to judge if the RV engine is a poorly optimised piece of rust. As most of you, I can only notice and report symptoms. The engine is certainly not perfect and needs a lot of attention to stay up to date. But re-creating an entirely new engine from scratch seems a bit too ambitious (if we want a new Arma game before 2020). Most software iterations rely on older code. I'm no expert but I bet one coudl still find code from the 90's in Windows 8.^^

[Arma3goodCPUlowFPS 12]

if BI want to stay with the current engine - ok

if BI will not add 64bit & proper multi-threading / multi-core (i hope all here knows that the game runs mostly on one core, uses only a small % of the other cores) support - ok

but they should write then on the boxes that this game is a very limited sandbox game where you cant have huge battles/wars with good FPS, where you can only have small spec ops/SWAT/Counter strike like battles.

about big maps: ok, you can have big maps, but there are other engines out where you can have big maps. cryengine allows you huge terrains in arma level, but not everyone has the access for this. you must contact crytek and provide them your work and inform them about your company (if you have one) and inform them about your project etc to get access to the terrain code. but not everyone knows this because of lazyness to research, this is why i read bullshit from some people here saying that cryengine doesnt allow huge terrains. there are some games WIP with cryengine which has terrain streaming (and which works better than the arma terrain system). anyway, arma terrains can be big, but also cause problems, begining with performance, one word: kavala (and no i dont talk about place yourself alone at kavala and you have good FPS). the engine is in many aspects more limited than we can imagine, from performance/CPU/GPU usage, to terrain, to modeling (in modeling there are many limitations too) and and and...

you can do alot in arma, but the few things of the alot runs smooth, so this makes not only the game limited, but the playing too.

the only game i know that can be A BIT compared is: garrys mod, why ? because its a sandbox too where you can spawn objects, units etc like in arma. it can also run custom mods, scripts, maps etc like in arma. heck you can even have helicopters and jets there and tanks and military stuff like in arma. but maps are different sizes, there are small maps and some big (not big as arma but still bigger than the usual source maps) yet the source engine is in many aspects more advanced:

- 64bit support

- proper multi core and multi threading

- better ragdolls/physics system

- proper bulletholes, sprites, effects

- and many more things.

there are even life mods running on servers and zombie mods like in arma 3.

there isnt a game out which has the same scale of arma 3, but the closest which comes to arma series is garrys mod.

i personally prefer garrys mod over arma. in garrys mod, i can do more than in arma, because i can play in garrys mod proper wars/battles with thanks, jets etc which arma 3 cant handle. in garrys mod i can have maps with skyscrapers where arma 3 has problems in terms of performance aswell in terms of model limits. in garrys mod, i can play all the life, zombie and many more mods too. in garrys mod, i can play awesome addons too and in garrys mod, i can play everything with proper multi core, multi threading and 64bit support, with good ragdolls and physics which i cant in arma 3.

there is a motto: more is less, less is more.

for arma 3 i say: more is less, why ? because arma 3 gives you so much but its very limited which makes you less usage of it. less AI, less FPS, less huge battles etc etc

for garrys mod, i say: less is more, why ? because you cant have some things from arma in garrys mod like for example big maps, but you can have proper battles/wars, be it with military, with aliens, or zombies, you have proper 64bit, multi core and multi threading support, good physics, good FPS, proper city maps which can be huge.

so guys, i say more is less and less is more :)

btw, you can have campaigns and missions in garrys mod, but the creating is far different than in arma series.