Jump to content

Sign in to follow this  
maddogx

ARMA3 - Brand new engine or backward compatibility?

Engine rewrite or backward compatibility?  

201 members have voted

  1. 1. Engine rewrite or backward compatibility?

    • Complete rewrite
      125
    • Keep it compatible
      40
    • They should do both!
      36


Recommended Posts

Well, they can for example keep the graphics engine. It is pretty good by now. However the animation system and some other basic features and functions are not there, and for example switching active addons from in-game would need quite some changes to the core of the game. Maybe it can be done by hefty modofication of the current engine. Maybe not.

That said it'll indeed be some major changes, but if they don't really have to touch graphics (much) at all, then that's a huge workload not having to be redone.

Share this post


Link to post
Share on other sites
  ch_123 said:
newsflash - poll shows 56% of people have no idea what they're on about.

hurr durr i'm so smurt lol at u faceplant

Share this post


Link to post
Share on other sites

I think that even with a complete rewrite you would still end up with mostly the same limitations(Performance, physics, ..) as they just come with the scale of a game like this.

For other things (Animations, graphics, ..) i doubt a complete rewrite would be neccesary.

Though i guess it definatly wouldnt hurt the game to rewrite everything, i think the pain would be unbearable in BI's wallet.

Edited by NeMeSiS

Share this post


Link to post
Share on other sites
  BlackLord said:
hurr durr i'm so smurt lol at u faceplant

I'm glad you agree. We need more people like you on this thread.

Share this post


Link to post
Share on other sites
  afp said:
I very much agree with your point Maddog, an editor overhaul is so much required also a better scripting language closer to C#/Java and object oriented if possible.

Just to clear up this misunderstanding, I would like to point out that the list of frequently requested engine upgrades mentioned in the first post is not my personal "wish list". Far from it.

@ch_123: If you could cut out the flamebaiting, that would be great. Thanks.

Share this post


Link to post
Share on other sites

No flamebaiting here bro, just agreeing what that poster's excellent assessment of the situation.

I wonder what percentage of BIS' income comes from VBS relative to the consumer games. I'd be willing to bet that they make more money on VBS than the games, and if so, the deciding factor on whether or not the engine needs to get replaced is whether the engine is capable of handling simulations. Things like graphics and physics are only issues insofar as they affect the utility of the software, if it's only for aesthetics then it's not worth rewriting the whole thing over.

Besides, in the wonderful world of corporate/government software, customers tend to like stuff that has been in use/development for a long time - not only for backwards compatibility, but for stability as well. Rewriting the main part of your software for the lulz is the easiest way to scare serious customers off.

Share this post


Link to post
Share on other sites

Have we been discussing about possibiities of adopting some pre-made engine?

No, not source since it does not allow big enough maps.

Not sure about cry-engine, but I guess it wouldnt allow big enough maps either.

I dont know if there even is engine that would satisfy BIS' needs. Maybe euphoria? It has proper physics, car behabiour, awesome animations engine, scale is pretty big I'd say too. And if the playground was few cities and towns, in some fictive island, I think it wouldnt be any more performance taxing than ArmA2. Just my opinnion, please dont come and say "ZOMG U R SO WRUNG!111"

Share this post


Link to post
Share on other sites
  Kristian said:
Maybe euphoria? It has proper physics, car behabiour, awesome animations engine, scale is pretty big I'd say too. And if the playground was few cities and towns, in some fictive island, I think it wouldnt be any more performance taxing than ArmA2. Just my opinnion, please dont come and say "ZOMG U R SO WRUNG!111"
Euphoria is a dynamic animation middleware, something that can be integrated into game engines but is not a game engine itself. It is used in GTA4 (which you are obviously thinking of) only for changing human animations on the fly.

Share this post


Link to post
Share on other sites
  ch_123 said:
No flamebaiting here bro, just agreeing what that poster's excellent assessment of the situation.

I wonder what percentage of BIS' income comes from VBS relative to the consumer games. I'd be willing to bet that they make more money on VBS than the games, and if so, the deciding factor on whether or not the engine needs to get replaced is whether the engine is capable of handling simulations. Things like graphics and physics are only issues insofar as they affect the utility of the software, if it's only for aesthetics then it's not worth rewriting the whole thing over.

Besides, in the wonderful world of corporate/government software, customers tend to like stuff that has been in use/development for a long time - not only for backwards compatibility, but for stability as well. Rewriting the main part of your software for the lulz is the easiest way to scare serious customers off.

Okay, I get that. Perhaps my own experience in this area is skewed by the fact that I mostly develop software for internal use, not for "customers". So when new requirements pop up and it turns out that it would take an inordinate amount of effort to implement them in the current version, creating a new one is the obvious choice and often beneficial in other ways.

My main thought process while creating this poll was whether or not BIS would benefit from redesigning their game engine to accomodate features that so far seem to be too costly to implement. Not all of these merely superficial eye candy like physics or graphics, but things that would actually enhance the gaming experience for everyone.

I think we have established already that a complete engine rewrite would be bad from a business perspective, but that's not the only point of view. ;)

Edited by MadDogX

Share this post


Link to post
Share on other sites

I think the problem with these discussions is that no one outside of BIS really knows or appreciates what the engine is capable of, or not as the case may be. This leads to lots of "HURR DURR CRYSIS ENGINE" posts which are entirely pointless.

Similarly, while the main underlying engine mechanics have remained the same since OFP came out, it seems to me that lots of things have been changed or tacked on successfully... I really doubt that much of the graphics code from the OFP days is still kicking around in OA.

Edited by echo1

Share this post


Link to post
Share on other sites

I'm amazed that no one noticed that BIS is already developping a new engine...

  Quote
Poll shows 56% of people have no idea what they're on about.

You should look at the Carrier Command thread from time to time people...

At worse, it's a nice testbed for new technologies to pimp up RV and at best it's the foundation for the new "Arma 3" engine... In any case it's a complete rewrite (though it was stated it shares some algorithms and linda vegetation from RV.

Share this post


Link to post
Share on other sites
  EricM said:
at best it's the foundation for the new "Arma 3" engine...

Do you have a source for that, or is that speculation? I really haven't seen enough of (and nobody outside of the developers have) the CC engine to make any such prediction. Yes, it should make a good testbed for new technology, but the little info we have is not enough to say whether it would make a suitable replacement for RV, or that there's even a need to replace RV (which IMO there isn't). What I am hoping is that revenue from both CC and ArmA 2 DLC will be enough to drive a lot of additional development for RV4.

  Quote
Poll shows 56% of people have no idea what they're on about.

Yes, I do agree here. FPDR

Share this post


Link to post
Share on other sites

That or 56% of the people when voting for a rewrite dont believe it should be completely rewrited since that would be retarded to throw away good code.

obviously you would keep the code that works and wouldnt conflict.

Share this post


Link to post
Share on other sites
  Flash Thunder said:
That or 56% of the people when voting for a rewrite dont believe it should be completely rewrited since that would be retarded to throw away good code.

obviously you would keep the code that works and wouldnt conflict.

Yeah i voted for the 2nd options even though i dont give a poop about it, i just thought the 'complete' was so insane that i couldnt vote for it. :p

Share this post


Link to post
Share on other sites

The poll wasn't about whether to keep "good code" or not. It was about keeping compatability with older content.

I'm sure any re-write will keep useful "good code".

So, yes pls re-write if that's what it takes to move on for another decade of ArmA.

Share this post


Link to post
Share on other sites
  =WFL= Sgt Bilko said:
The poll wasn't about whether to keep "good code" or not. It was about keeping compatability with older content.

IMO the poll has no point. In the extremely unlikely event that BIS does rewrite the engine, and 15 years from now when the next game is finally released, I doubt there would be any incentive to keep the so called "backwards compatibility."

Like I said, this poll is based off ideas that are completely misguided...

Share this post


Link to post
Share on other sites

Well i dont realy care.

But all i want is to use all resorces we have.

Make it x64 only, so it can use huge ammounts of RAM, not like 1GB it uses now.

-develop it to use more than 8GB of ram.

Make use of multicore, it is obvious that multicore is Future. Design the engine to use 16 core cpus and more.

-When developing, use opterons, they have 12 core, i think even 16core is coming out soon to test it.

Try to use dx11 for calculating physics, and AI. It is allot cheaper even to buy most expensive gpu, than to upgreade all. And this game doesnt use gpus to their full potential.

I just hate the fact that even if i get the most expensive components (cpu, gpu ram), overclock them and still i will have low fps at some parts of game, not becouse limitations of components, but becouse game engine doesnt us it to their full potential.

Share this post


Link to post
Share on other sites
  zaira said:
Well i dont realy care.

But all i want is to use all resorces we have.

Make it x64 only, so it can use huge ammounts of RAM, not like 1GB it uses now.

-develop it to use more than 8GB of ram.

Make use of multicore, it is obvious that multicore is Future. Design the engine to use 16 core cpus and more.

-When developing, use opterons, they have 12 core, i think even 16core is coming out soon to test it.

Try to use dx11 for calculating physics, and AI. It is allot cheaper even to buy most expensive gpu, than to upgreade all. And this game doesnt use gpus to their full potential.

I just hate the fact that even if i get the most expensive components (cpu, gpu ram), overclock them and still i will have low fps at some parts of game, not becouse limitations of components, but becouse game engine doesnt us it to their full potential.

erm ...

are You aware the engine is already able use 4GB directly (LAA on 64bit OS) and N GBs via file cache (see the blog)?

are You aware You can use -cpuCount to support N cores ?

DX11 is way too limited spread and OpenCL ready drivers same ...

Share this post


Link to post
Share on other sites

well, i first wanted to vote "complete rewrite".

But then i voted "keep it compatible" because i think the engine is nearly perfect for what it is supposed to do! BUT there are a few things that should be improved/added in the next years:

-Physics (seriously)

-destruction model

-gfx-engine (not now, but maybe in a few years)

I know that these are things that aren't necessary for the simulation aspect, but they would add a lot of atmosphere and... fun! :-)

just my 2 cents...

Share this post


Link to post
Share on other sites
  Dwarden said:
erm ...

are You aware the engine is already able use 4GB directly (LAA on 64bit OS) and N GBs via file cache (see the blog)?

are You aware You can use -cpuCount to support N cores ?

DX11 is way too limited spread and OpenCL ready drivers same ...

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.

(i have 1.07 patch, and i read the blog)

I am talking about future.

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

Im shure you know your market, but if you do that, imagine people just need to buy better video card to run game better, not whole pack, like now.

And, im engineer (electronic systems and tech), on a way on geting master of science degree, and im very interesting in weapon systems (2nd generation anti aircraft).

Edited by zaira

Share this post


Link to post
Share on other sites
  zaira said:
Are you aware that engine isnt using more than 1GB of ram?

It uses 4Gb+ on my rig...

  Quote
Are you aware the game does not properly use all cpu cores (i mean there is benefit but realy big 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.

Tested on 2 machines, same clock speed, on is quad, one is dual. I have seen the game using all 4 cores on the quad (with a peek on the first 2 though), and both cores on the dual

(i have 1.07 patch)

  Quote
DX 11 is future,

No, it is not...Really

  Quote
i know something about programing

You obviously don't know enough..

  Quote
and im shure that it would be major pain in the arse to program in OpenCL, but that is the future, accept it.

Why? Because and OpenCL product can run on any OS system, not only M$?

Share this post


Link to post
Share on other sites
  pufu said:
It uses 4Gb+ on my rig...

Tested on 2 machines, same clock speed, on is quad, one is dual. I have seen the game using all 4 cores on the quad (with a peek on the first 2 though), and both cores on the dual

(i have 1.07 patch)

No, it is not...Really

You obviously don't know enough..

Why? Because and OpenCL product can run on any OS system, not only M$?

I dont want to argue m8, but can u pls take a SH of task manager showing proceses. Becouse on my system with 4GB of ram u it does not use more than 1GB. How much ram do you have?

There is difference 2vs4 core, but not much difference on 4+.

Share this post


Link to post
Share on other sites
  zaira said:
And, im engineer (electronic systems and tech), on a way on geting master of science degree, and im very interesting in weapon systems (2nd generation anti aircraft).

What colour is the boathouse in Hereford?

Share this post


Link to post
Share on other sites
Guest
This topic is now closed to further replies.
Sign in to follow this  

×