why is it taking so long to develop this game?

Now, from the perspective of someone who has played both games,

They are two completely different animals.

NCG/WF is more traditional. It's a racing game.
BeamNG is not traditional. It's a car sandbox.

I can spend up to half an hour playing NCG/WF. I have often exceeded 4 hours in playing time on BeamNG.

Which is better? I prefer BeamNG.

 

STOP COMPARING NCG WITH BEAMNG, they are going for 2 completley different things

NCG is an Arcade racer with nice damage

BeamNG.Drive is a soft-body physics game that just happens to have amazing crash physics as a side effect

please, just stop comparing them, NCG devs have even said they don't want to be compared with beamng, SO STOP COMPARING THEM

[/rant]

sorry, it's just really annoying seeing people compare them so often

 

Side effect??....since when were "amazing crash physics" in Beamng a "side effect? That's perilously close to sedition dude!

Also, I personally do not equate Beam to the traditional description of a game, so when people bitch and whine about it being a shitty game
they are automatically disqualified from the debate, if there ever was one.

 

Hey, im just stating what the devs said (im 99.99% sure they said something like that somewhere)

Sent from the 3rd galaxy via the talks of tapping

 

I wonder... This sounds like if you want to make a proper game out of it you need GPU physics computation... even then, maybe I'd have to get a second card and dedicate it specifically to that. I don't see regular processors getting to where we need them to be within a reasonable time frame. What would the damands be to emulate NextCarGame for instance? 24 vehicles on a track with hundreds of things to crash into. Even if you go the rigid body physics route with the track objects, imagine 24 hirochi sunbursts? Thats the sort of performance hit of the average car in this game 24 times over. How can we run that at 60fps?

and can the engine be made more efficient? How unoptimized is it? How much more performance can the developers eek out of it? Its doing an incredible 2KHz simulation already. How much would things be improved by the hyperthreading in GPU's?

 

Well on the main beamng.com page it says they have opencl support planned, so we shall find out how much gpu doing physics helps then

Sent from the 3rd galaxy via the talks of tapping

 

So, if the physics was done by the GPU we'd have much higher framerates? Am I wrong?

 

Going out on a branch here but I think the languages a CPU and a GPU speak are TOTALLY different. Having GPU's read and implement the beam code would be impossible
without architectural chip die modifications?,not to mention on board caching memory etc. In essence having a GPU read "code" and there by increase sampling rates
on the physics side is probably a pipe dream. Certainly interesting but quite impossible without insane and expensive hardware changes or an emulation scheme that
would destroy any FPS gains made.


I'd like to see 16 or 24 core CPU's handle all the cars I want. Beamng uses DX9 so the graphics department is well handled by even older GTX and AMD stuff and it
would take tens of vehicles to start maxing out a graphics card.

 

that's what opencl does though, is MAKE it possible without doing all that crazy stuff (it was made by AMD, Nvidia, Intel, and a few other companys i don't remember)

 

Beam ng team is like uh what 5 people + With friend and good people that help


Next car game is.... 40 + people PAID .....


Nothing more to say...

- - - Updated - - -

how go your laptop i7 with the latest beta map im getting 40fps on my fx 6300 at 4.4 and now im like do i get a 4690-5820k 600$vs 1300$ or i jump on a laptop msi gt70 or clevo 1800$-2500$ Send a pm or post here please!

 

Can you not type an essay next time

Sent from my Galaxy S4 using Tapatalk

 

NCG and Beam are pretty much completely different games aimed at different audiences.. Just because NCG crashes aren't completely realistic doesn't make it a crappy game

Sent from my Galaxy S4 using Tapatalk

 

Considering I was running Beam on an i3 4GB Intel 4000 onboard Toshiba laptop my MSI is light years better. I run DRI and ECUSA at max settings with 30-40FPS
when on the old Toshiba I could only run GridMap with two cars max at 15-20FPS. If you ask me, your FX6300 is apparently just fine for the graphics side of things
but a better CPU is always a good bet for running Beam better. In short I would upgrade your CPU first versus spending over a grand more on a new laptop.
I'm not a tech genius but as time goes on it's fairly clear the CPU is the bigtime thing this program needs. Hope this helps. I love the setup I've got, it benches
nearly even with a R9 290 and the i7-4810MQ handles up to 10 vehicles decently by the Bananabench program. The i7 installed has turbo boost and regularly
goes from 2.8GHZ to 4.0GHZ running Beam.

As I progress with this goodness a major league desktop is going to get built because the expansion will be silly
to run and keep organized. Besides...heh heh...a 17.3" display won't cut it anymore and HDMI out is a pain in the ass.

 

Thank you

 

OpenCL is used for GPGPU applications. Aka general purpose data processing on a graphics card pipeline. AMD and NVidia both provide OpenCL drivers in their GPU driver packages alongside DirectX and OpenGL etc.
There is also CUDA which does the same but is limited to Nvidia cards. Nvidias physx physics engine uses CUDA acceleration and quite a few games do take advantage of it for processing extra effects (particles or maybe tiny bits of debris (Mafia and batman used it for broken glass bits left on ground etc) which look cool but aren't essential for systems without CUDA).

GPU for physics is a well established thing.
A CPU can calculate the result of 15*15 faster than a GPU of same clock speed, but a CPU can only do that 8 times at once if its an octa core (4 for quad, 2 for dual and a mere once for single), a gtx760 I think does it 32 times at once and some of the nvidia workstation cards do even more (they often sacrifice GPU performance for boosted CUDA performance, the nvidia OpenCL driver is actually a CUDA wrapper). Not only that but a CPU takes varied amounts of time to complete different arithmetic operations, in most GPUs the time differences are in a much narrower band, much narrower. Plus each GPU compute pipeline tends to support floating point and we all know how badly AMD CPUs in particular fail at that.

A GPU cannot crunch through 1 sum as quickly as a CPU. But it can crunch through many at a far faster rate than the CPU can. OpenCL and CUDA let us do this.

 

Beam isn't a crashing game. Beam is a driving game with nice crashing.

 

I'll buy that. It's all semantics really. You could just as well say anything in the universe is a game and it could stick.

 

Your mom is a game

Sent from my Galaxy S4 using Tapatalk

 

Was that comment really needed? Oh, right you are just here to boost your post count :|.

The game is probably taking long to develop because BeamNG is a small company/dev team. Give the devs time .

 

Who cares about post count lol? And he did say anything was a game so

Sent from my Galaxy S4 using Tapatalk