ARMA3 - Brand new engine or backward compatibility?

[Engioc 10]

It is totally useless to discuss a new engine for ArmA3 (if it is ever going to be created at all).

It just won´t ever happen and the reasons for this are simple: BIS is a type of family company, which could only survive until now, cause they found a gaming-niche, to which they created their own system.

They keep updating this engine for over 10 years now and it would be an impossible task to create a new system, which would be sufficient to meet modern requirements, like the acutal one does.

First it would need a huge amount of money let alone for the development, as they can´t just take an engine from another company and modify it to their pleasure. There is no comparable engine out there, which can perform the same way, with the huge landscape and the load of AI procedures, like the current one does, because as I already said, ArmA is a niche-product.

Second the whole BIS team would have to be trained to use the new technologies, what would also need

a huge amount of time, money and effort.

BIS can not afford to spend all this effort to make a game which then would also have to meet the expectations

of every ArmA fan out there and the gaming-scene.

To come to an end: an objective cost-benefit calculation would point out a high cost

with a low benefit (as this supernew game would also cost max. 50 euros).

The only thing we can hope for is that a business competitor picks up the idea and priorities of BIS

and makes their own new game from scratch (as seen in OFP: DR - the biggest fail in recent years)

Going by this your suggesting they stick with the current engine forever and never bother creating a new one. Can they really survive by keeping the same engine and never starting again? I dont think so.

As one other poster said, sooner or later they need to create a new engine. I think they should stick to releasing DLC and expansions to A2 for a few years and hopefully get themselves enough time and money to rewrite the engine. I doubt the current engine is going to still be good enough in another 10 years, the days of game engines being re-used over and over again are long gone, I cant see them getting more than 10 out of one game engine and surviving. I'd rather have them at least try creating a new engine rather than just die a slow death because they gave up.

Also you talk of the benefits as in how much they can sell the game for, well your right except that if they create a good enough engine maybe they can sell that to other developers so instead of being the little guys who need to buy in all the fancy tech, they can be the ones supplying it. A good game engine can be worth more money than ArmA itself.

Edited August 3, 2010 by Engioc

[dm 9]

ITT: People who [mostly] have NO idea what is involved in creating a game engine :j:

[echo1 0]

Going by this your suggesting they stick with the current engine forever and never bother creating a new one. Can they really survive by keeping the same engine and never starting again? I dont think so.

As one other poster said, sooner or later they need to create a new engine.

Could you qualify why the engine must be replaced with your obviously excellent knowledge of the limitations of the engine?

[Big Dawg KS 6]

Going by this your suggesting they stick with the current engine forever and never bother creating a new one. Can they really survive by keeping the same engine and never starting again? I dont think so.

You obviously don't know very much. There's nothing wrong with the current engine, and why the hell you people thing so is beyond my understanding, but to say that BIS shall have to start over is just insane. How is throwing out years worth of progress and functional code and starting over again going to help BIS "survive?" If anything, a complete rewrite is more likely to kill BIS, as it will most certainly drain their funds and even if they miraculously managed to build an entirely new engine and build a game on top of it without going bankrupt, I highly doubt it'd even be half as good as what they already have.

I don't know how many times it must be said but some people just aren't getting it... it's not going to happen, there's absolutely no justification.

[snkman 351]

I totally agree with that Big Dawg. ;)

The A2 engine really isn't that bad like many say.

We could have some more optimized objects like tree's and buildings which really need a lot of resources and GPU power becouse they are simply overdone.

The rest is really good from my point of view.

ArmA is by far the best moddable game i have played/modded and that's becouse it is what it is.

Of course B.I.S. could jump on the "commercial" train and re-write the whole thing but who knows what will be the result..?

I just think back to Ghost Recon I from my view of today.

Ghost Recon I was a awesome game with a lot of atmosphere and a really nice gameplay.

Just look what they have made out of it! A damn boring linear shooter like 95% of all games we have today.

The slogan of most gamers today is Pc On / Brain OFF.

Thank god we have this community here. :D

So i suggest to B.I.S. optimize the engine release DLC to keep the community running and everything is just fine. :)

[Herr_kalashnikov 10]

You obviously don't know very much.

Unneccasary way to start a reply..

There's nothing wrong with the current engine, and why the hell you people thing so is beyond my understanding, but to say that BIS shall have to start over is just insane.

This reminds me of a minister from Iraq ;), but don't u think if the current engine don't have any issues this platform would be more popular. I know from experience that alot of players try this game but loose interrest very quickly after experiencing MP and the random bugs in SP. Some of these issues are hard to resolve without rewriting a big part of the code. If u call it a complete rewrite or not is a another subject.

How is throwing out years worth of progress and functional code and starting over again going to help BIS "survive?" If anything, a complete rewrite is more likely to kill BIS..

Do u think for example that if some company lets say MS make a new platform they gonna bin all the existing code? Not gonna happen every bit of code which can be recycled/refurbished will be used again. Can't see why it should be any different with BIS products. Also if all developers would have this mentallity we all would still write our documents in WordPerfect 5.1.

[wamingo 1]

I know from experience that alot of players try this game but loose interrest very quickly after experiencing MP and the random bugs in SP. Some of these issues are hard to resolve without rewriting a big part of the code.

You're asking for the wrong thing, I think.

These are not engine related really. Bugs can be of course, but it's not really what you're talking about here. You're talking about general frustrations about the game which are things like: Slow gameplay, clunky controls, too much travel time and high death penalty.

Those are at least some of the key factors I think.

But the latter ones are mission design specific for the most part. And MP maps are probably very difficult make truely fun -- how to improve on them really needs its own topic.

The clunky controls could definitely be improved however.

But again, I'm not sure it's the engine code that needs to be touched much.

Either way we need more animation-blending, allowing especially stance changes to be interruptable, and perhaps also the ability to shoot more easily while moving.

I think this area is a critical factor in peoples frustrations over the game.

[Big Dawg KS 6]

Do u think for example that if some company lets say MS make a new platform they gonna bin all the existing code?

You're going to compare BIS to Microsoft? FPDR

Anyway, wamingo brings up a good point. It seems that many of the so called "justification" for scrapping/rewriting RV that people tend to come up with don't actually justify a new engine at all, but should be relatively achievable with RV.

Honestly though, I can't believe that some people really think RV is lacking certain things...

And I have yet to see anyone present an argument that even comes close to being justifiable for BIS scrap RV.

[echo1 0]

I know from experience that alot of players try this game but loose interrest very quickly after experiencing MP and the random bugs in SP. Some of these issues are hard to resolve without rewriting a big part of the code.

Please qualify this statement. Do you know the exact cause of these issues? Have you read the source code of the game yourself?

[Dwarden 1125]

Are you aware that engine isnt using more than 1GB of ram?

Are you aware the game does not properly use all cpu cores (i mean there is benefit but not huge one, and there are people with gulftowns and x6 phenoms here, no difference for them on 4+cores), and as far as i know 8 cores is maximum.

DX 11 is future. I know it will be huge learning curve to use OpenCL, but u realy need to.

the engine maps the memory as it sees fit,

and the engine easily maps more than 1GB

that was possible even before LAA with -maxmem=2047

memory mapped for file cache You don't see that's done by operating system

read for more details http://www.bistudio.com/developers-blog/breaking-the-32-bit-barrier-2_en.html

for multicore , please read :

http://www.bistudio.com/developers-blog/real-virtuality-going-multicore_en.html

since then more threads were added aka exThreads

http://community.bistudio.com/wiki/Arma2:_Startup_Parameters#exThreads

here is nice table where you can see how game scales with CPUs

http://forums.bistudio.com/showpost.php?p=1697744&postcount=19

yet You can await further improvements in this area ...

DX11 and OpenCL as future is still onhold

DX11 is around 6% of gaming machines on STEAM

while with rest of world it's way worse

http://store.steampowered.com/hwsurvey

OpenCL ready is even fraction of these numbers ...

GPU stress well we use whole system not just card so that's why good machine counts ...

Edited August 3, 2010 by Dwarden

[AnimalMother92 10]

OpenCL as future is still onhold

:eek: Sweet! OpenCL ftw

[tacticalnuggets 24]

You're asking for the wrong thing, I think.

These are not engine related really. Bugs can be of course, but it's not really what you're talking about here. You're talking about general frustrations about the game which are things like: Slow gameplay, clunky controls, too much travel time and high death penalty.

Those are at least some of the key factors I think.

But the latter ones are mission design specific for the most part. And MP maps are probably very difficult make truely fun -- how to improve on them really needs its own topic.

The clunky controls could definitely be improved however.

But again, I'm not sure it's the engine code that needs to be touched much.

Either way we need more animation-blending, allowing especially stance changes to be interruptable, and perhaps also the ability to shoot more easily while moving.

I think this area is a critical factor in peoples frustrations over the game.

MP suffers drastic engine limitations, and so does the addon management. Thankfully BIS is trying to fix some of these problems. Lets hope it is done well. It is important to realize that BIS is neither god nor evil developers trying to steal your money. They are a small game company producing a product.

---------- Post added at 03:45 PM ---------- Previous post was at 03:43 PM ----------

You're going to compare BIS to Microsoft? FPDR

Anyway, wamingo brings up a good point. It seems that many of the so called "justification" for scrapping/rewriting RV that people tend to come up with don't actually justify a new engine at all, but should be relatively achievable with RV.

Honestly though, I can't believe that some people really think RV is lacking certain things...

And I have yet to see anyone present an argument that even comes close to being justifiable for BIS scrap RV.

I agree that engine is thrown around too loosely. It should be stated that components of the game need to be rewritten to achieve a better product. Unless your talking about the graphics engine or things of that matter, an engine is too generalized a term.

[Big Dawg KS 6]

MP suffers drastic engine limitations, and so does the addon management.

What does that even mean? I'd think that addon management is something of a difficulty to design, and I can not think of any game that has good integrated addon management. Besides that's not something that's particularly related to anything but the data format, and data format can probably be changed if there is a better one (not that there is), so switching to a new engine probably isn't going to make addon management any easier...

As for what else you mean, I'm not sure.

---------- Post added at 06:52 PM ---------- Previous post was at 06:49 PM ----------

I agree that engine is thrown around too loosely. It should be stated that components of the game need to be rewritten to achieve a better product. Unless your talking about the graphics engine or things of that matter, an engine is too generalized a term.

Well, all of the core components of the engine are already pretty good. I won't disagree that they could use improvement or tweaking, but certainly nothing is fundamentally flawed to the point it needs to be rewritten (optimized/refactored maybe, but that's not the same as rewritten).

[echo1 0]

MP suffers drastic engine limitations, and so does the addon management. Thankfully BIS is trying to fix some of these problems.

Wait, if it's an 'engine limitation', then it cannot be fixed without rewriting the engine surely?

And what problems with MP/addon management are caused by actual engine limitations?

There's lots of people who throw this phrase around, but don't have the facts to back it up with.

[tacticalnuggets 24]

What does that even mean? I'd think that addon management is something of a difficulty to design, and I can not think of any game that has good integrated addon management. Besides that's not something that's particularly related to anything but the data format, and data format can probably be changed if there is a better one (not that there is), so switching to a new engine probably isn't going to make addon management any easier...

As for what else you mean, I'm not sure.

---------- Post added at 06:52 PM ---------- Previous post was at 06:49 PM ----------

Well, all of the core components of the engine are already pretty good. I won't disagree that they could use improvement or tweaking, but certainly nothing is fundamentally flawed to the point it needs to be rewritten (optimized/refactored maybe, but that's not the same as rewritten).

Just think about this. Lets take a random game and look at its system equivalent for addon management (dont pull up the gameplay, its irrelevant). CSS does not force you to download everything off of random websites. How old is that game? And it will automaticaly put those files in the correct folders. So why cant BIS do something like that? Of course the content in arma2 is a little bit more complex but just link to a mirror in-game and have the game download the content and automaticaly dress it up in a mod folder. Its very simple, but the MP component of the game is so bare bone that nothing is done for you. NOTHING

BUT, now they are trying to implement solutions to those things, so i thank them.

---------- Post added at 04:01 PM ---------- Previous post was at 03:59 PM ----------

Wait, if it's an 'engine limitation', then it cannot be fixed without rewriting the engine surely?

And what problems with MP/addon management are caused by actual engine limitations?

There's lots of people who throw this phrase around, but don't have the facts to back it up with.

You should have continued to re-read my post but now you suffer the wrath of posting too quickly. LOL

Anyway, its never really an engine limitations unless of course your talking about graphics engine ect... I was just trying to speak in the same way as everyone else.

EDIT: And I forgot to add to the first part of my post that addons shouldn't have to be loaded at game start.

Edited August 3, 2010 by tacticalnuggets

[Engioc 10]

I think the problem is peoples different interpretation of what constitutes a new engine.

I don't agree with the fact an entire rewrite is needed. I also don't believe all this needs to be done for A3. However going by many statements here its as if nobody wants any changes made because the current engine is fine forever, and that is just as crazy as making huge changes before A3.

As far as I understand it in order to achieve all the items on the long wishlist it will require major changes to the current engine, sure they will re-use code from the current engine and they wont be making all the changes at once but at some point there will be enough significant changes that in a way its no longer the same engine. So call it what you like, new engine, huge update to current engine, it all equals the same thing.

I would never want to see them throw away all the code they have now and start over, that is crazy I agree and pointless because you could end up with a new engine that simply has different limitation, you'll never have the perfect engine if you keep starting from scratch.

My reason for agreeing with a new engine is because the OP stated new engine doesn't mean you simply throw away all current code and start from scratch. My belief is sooner or later BI will need to make major changes to the current engine, effectively creating a new one. All game engines go through this and so will BI.

So to those who are complaining we never need a new engine, get it straight that new engine doesn't mean you need to throw away the last 10 year worth of work.

You can be sure BI will make some significant changes to the current engine before we see A3, you wont see all the items from the wishlist implemented, but your sure to see some of them, even some that do require major changes. By the time we see A4/A5 we're likely to see what is essentially a new engine because of the number of changes made.

Also if all developers would have this mentallity we all would still write our documents in WordPerfect 5.1.

Don't you mean Word * :).

Edited August 4, 2010 by Engioc

[gossamersolid 155]

As much as people want a full re-write and because of the fact that if I remember correctly, BIS said after OA they wouldn't have a new mil sim for a long time, a Full re-write will not happen unless they score a huge publishing deal or run into a bundle of cash. ArmA 2 already improved a lot of ArmA 1 and whenever/if they make an ArmA 3, I'm sure it'll impress as well.

[PuFu 4600]

A lot of ppl are thinking that the current engine doesn't support stuff la physics (which i am sorry to break your bubble but it physics engine runs parallel with the game and animation angines), advanced lighting calculations, render to texture technology or what not.

That is not entirely true. You can definatly see the difference between how lighting is calculated and rendered in A2 and the one in OA. And i am sure dynamic lights is possible with BIS engine as well, but what would be the cost of it?

Most people are comparing different games and game engines with what BI has developed. Well, no other game out there has the same scope A2/OA has, nor its size, depth, its calculations, its unscripted AI etc etc. On top of all that, it is a big SDK that anyone can play with..Basically it has very little limitation to what one can do with it.

Obviously, i would like to see some changes here and there (especially in the modding department and the tools that are provided - especially the availability of plug-ins for the established modelling/animation software out there). But even here there was improvement over what was happening back in the OFP days (the use of proxies springs to my mind here).

So in the end, it would obviously be a waste to throw all this code away. Sure, rewriting PARTs of the code would only be beneficial to all of us, but pointing (yet again) that BIS needs to rewrite code so it can bring euphoria, ro havok (with its ragdolls - which i am sorry to say but its getting old now) it's nothing but a sign that you have no idea what a game engine is, and the difference between it and a physics engine.

Another request i see way too often, especially coming from new forum members, is stuff such as DX11 support (with its tessellation), opposed to OpenGL and OpenCL. Sorry to say but i would always go for an open dev environment vs a m$ exclusive one...

[tacticalnuggets 24]

I agree with pufu on this, you really cant describe what an engine is and isn't if you have no idea what an engine is. I like to divide a game into 3 engines. Graphics, Physics, and MP, but there are more elements than that, so engine is way to general a term for so many parts of a game. Getting away from UDP, as far away BIS can get, is a good start though on a major engine change.

[zaira 10]

the engine maps the memory as it sees fit,

and the engine easily maps more than 1GB

that was possible even before LAA with -maxmem=2047

memory mapped for file cache You don't see that's done by operating system

read for more details http://www.bistudio.com/developers-blog/breaking-the-32-bit-barrier-2_en.html

for multicore , please read :

http://www.bistudio.com/developers-blog/real-virtuality-going-multicore_en.html

since then more threads were added aka exThreads

http://community.bistudio.com/wiki/Arma2:_Startup_Parameters#exThreads

here is nice table where you can see how game scales with CPUs

http://forums.bistudio.com/showpost.php?p=1697744&postcount=19

yet You can await further improvements in this area ...

DX11 and OpenCL as future is still onhold

DX11 is around 6% of gaming machines on STEAM

while with rest of world it's way worse

http://store.steampowered.com/hwsurvey

OpenCL ready is even fraction of these numbers ...

GPU stress well we use whole system not just card so that's why good machine counts ...

Thank you very much for your time. I will check everything.

[Flash Thunder 10]

No new engine required for what my top wants are

Performance issues that plague Arma 2/OA, crossfire,SLI config issues, stuttering, constant LOD thrashing, Missing textures on terrain when viewdistance is High, bushes with low LOD until you're right next to them

More realistic lighting, I guess this could be added in as a graphical option under Shader quality??

Performance and stability should be the keen thing that BIS focuses on for their next version of RV and arma 3.

things like better animation and physics dont need a new engine they're just heavy on resources which is a good thing they werent very well done in A2 or I wouldnt be able to play the game at all.

[echo1 0]

I wonder just how easy it's going to be to design an engine that would offer a worthwhile increase in efficiency and scalability over the current one.

It's very easy for people to say "Game X looks better than ArmA 2 and yet it runs better on my machine." The reality is that you have to consider the scale of what BIS' engine is generating. Things like the Source engine, championed as the wonder engine that runs great on the slowest of PCs, would shit itself over half of what the BIS engine has to do (case in point - Garry's Mod will eat a well spec'd computer on big maps with lots of objects) Hell, I'm sure there's probably an argument to be made for BIS' engine being the most efficient computer game engine out there. Who knows? Certainly not going to be simple to improve on it in a way that justifies the years of work and millions of euros that a substantial rewrite would take.

As for SLI. Last time I checked the Valve hardware surveys, only 2% of Steam users (which I would consider a good representative sample of PC gamers as you will ever get) have SLI systems. Should BIS go to all that effort to cater to them? I'm not so sure.

Edited August 4, 2010 by echo1

[hobojoe 10]

there is nothing seriously wrong with the game as it is.

its just lacking what the usual gamer expects from a GAME

Physics and a whole lot more character animations

the rest is hardware related. once our pc's can run 4-5 times faster than now all the detailed stuff can be added like completely open buildings with hundreds of objects and plants

i just hope they can sort out the traffic and road issues with the ai

(if there was ever a problem with Arma) its that vehicles and units use the same paths to move around.

[Big Dawg KS 6]

its just lacking what the usual gamer expects from a GAME

Physics and a whole lot more character animations

Again, are you trying to say that RV lacks physics? FPDR

Anyway, pufu made some good points. We've seen BIS significantly improve the animations and physics components, and what's there already works well, so there's no need to rewrite anything. It would be much more worthwhile for BIS to just build on what's there (which is a lot more significant than most of you seem to realize).

And really, physics is one of the last things that needs attention right now, they are pretty much perfectly fine for what ArmA needs. As for the animations, I'm not sure why everyone has such high expectactions. What BIS has been able to achieve with improvements in the animations since OFP is incredable. I guess some people just don't realize how many animations ArmA actually has, and how much work it takes to create & configure them all.

[CarlGustaffa 4]

Pet "engine issues":

* Shooting speed limited by framerates (and in AIs case, sometime also by distance to target).

* One bullet per frame.

* Inability to access addon resources for everything (i.e. sounds).

* Hard limits to what simulations are available from where (i.e. no handgun rockets).

* (Assumed) limitations arising from single point precision. I.e. dim #lightpoints at altitude fail, probably due to base unit value and rounding errors.

* Lack of lowlevel commands and parameters in the scripting language (i.e. the Advance command or the sceneBrightness parameter).

* Highly limited difficulty system, which makes it impossible to choose more difficult options on a low difficulty server.

Physics is for the most part fine, for those items that are affected by it. Is enough affected? No, but I don't know if the outcome is synchronized. Has always been a problem with bounced smoke grenades, they end up in different locations for each client - NOT good for marking purposes. No suggestions how to fix it either :)

Animations are pretty good, but for the most part it needs content. Opening and closing doors on vehicles as you get in and out would be highly welcome, even if it's not game breaking. The only thing that needs rewrite is hos the anims are executed; ability to break current animation would be awesome. We could have longer walk cycles with some variations thrown in, but when you get down it shouldn't wait for the previous animation to finish, but rather morph between the two.

Not sure what is meant with "leave the UDP". How many high bandwidth shooters exist that is based on TCP, and how many players do they support?