When we become a LFS version that ist DualCore optimzed?
We will have much better performance!

GreeZ

WIGGA

the improvments/suggestions section would have been a better place to post this... check the 1st page to see if this has already been posted.

If LFS is ever going to be multi-threaded I think the performance will stay the same but the simulation and features will be more detailed/advanced/sophisticated and of course more realistic.

Quote from Gabkicks :

the improvments/suggestions section would have been a better place to post this... check the 1st page to see if this has already been posted.

Upps.

Moved.

Quote from Bob Smith :

Moved.

THX

Quote from SlamDunk :

If LFS is ever going to be multi-threaded I think the performance will stay the same but the simulation and features will be more detailed/advanced/sophisticated and of course more realistic.

that would be nonesense, because you would have a better simulation with 2 cores than with 1.
that means a dual core cpu would give other results than a normal cpu.
that would cause a advantage/disadvatnage for one side.
i think, depending on graphicscard, a performance advantage from 30 to 60% could be realistic.

@black vampire: I disagree, I don't think the video card would affect it any more than it does now, which is to say it doesn't as long as it's even remotely recent (i.e. DX8, such as GF4). I know for a fact that minimum framerates in LFS are entirely related to CPU performance (unless you're running insane levels of AA and AF at a ridiculous resolution, then it starts to have a small effect).

ok.. i understand you, but why i think the graphicscard will limit it more ist because from cpu side (X2 3800 +) i can do 100 - 150 frames. i tested it and my graphics card (6600GT) didnt affect anything, ok. but i doubt that my GC can do 300 frames with 1280x960 and all details etc.
for my PC with dualcore 200-250 frames in average would be more realistic than >300

Quote from Forbin :

@black vampire: I disagree, I don't think the video card would affect it any more than it does now, which is to say it doesn't as long as it's even remotely recent (i.e. DX8, such as GF4). I know for a fact that minimum framerates in LFS are entirely related to CPU performance (unless you're running insane levels of AA and AF at a ridiculous resolution, then it starts to have a small effect).

no aa or af and my 6600's (s because i have SLi), top of the range 256 Mb GT's, are always the limiting factor

but then i do have an AMD 4000....

EDIT - D'oh! GT isnt highest model, Ultra is!

Not for the 6600. There's only a 6800 Ultra. The one and the only. The geforce 7 series top of the line is GTX now.

Quote from Primoz :

Not for the 6600. There's only a 6800 Ultra. The one and the only. The geforce 7 series top of the line is GTX now.

No the 7950GX2 is the top of the line

That's debateable. It's merely two 7900GT's sandwiched together. In the grand scheme of things, though, you can SLI two 7950GX2's in the same way you can SLI two 7900GTX's, and the two 7950GX2's will always be faster.

back to topic please.

The way LFS is handling physics will make it hard to do multithreaded, I think. Only advantage would be to run all AI and gfx stuff on one thread and physics on another. i THINK the physics thread need to be on a single core, because each calculation depend on the one before it. I could be wrong though. Would be cool if it could be done, though.

Actually, physics calculations depend on the exact mathematical models LFS employs, which most of us here don't know. Anyway, multithread is particaularly good for parrallel processing. If its some algorithm that requires massively fast serial processing, they're nothing special.

Actually, all new games on the market are heavily dependant on parralel processing to perform anyway. That's why we need to change those expensive graphics card all so often. If a single core was sufficient, we'll need no 2nd CPU to process graphics.

Point is, LFS will benefit grealy from proper multi-core support, since newer and faster CPUs will all be multicore as the limits of single core tech are already being reached.

what i can think about without knowing how the physics are proceeded:

cpu0: physics of all cars
cpu1: shadows and other gfx things, network, AI

or:

cpu0: car 1,3,5,7...
cpu1: car 2,4,6,8...

Have any of you chaps done any multithreaded programming?

I have (and do), and I don't get to choose which cpu/core/etc the thread runs on - you create the thread and leave it to the operating system (in Windows, at least).

I'd hazard a guess that it wouldn't be practical for LFS to have it's own interface to the hardware to a) query it to find out what's there and how to use it and b) force a thread to run on a given cpu. Imagine if all programs did this - it'd be a nightmare if you ran 3 programs that badly weighted everything to cpu1.

IF the decision is made to multithread things (if they aren't already - multithreading has been around for a while) you won't know what's running where, only that the game appears to be running a bit smoother.

Quote from bLaCk VaMpIrE :

what i can think about without knowing how the physics are proceeded:

cpu0: physics of all cars
cpu1: shadows and other gfx things, network, AI

or:

cpu0: car 1,3,5,7...
cpu1: car 2,4,6,8...

AI is in both CPU's in your first taught.

+0,1 for the original idea

I think a more benifical peice of hardware will be phisics calculators, like ageia and the soon to be releassed havok on gpu type from nvidia, that would have a better effect then dual core optimization, in terms of gameplay
and before you all jump and say some ppl will have it and some wont which will make an imbalance, we are racing sim finatics, if we really want to experience the best buy a phisics card simple

@seinfeld: Quoted from the_angry_angel in this thread: http://www.lfsforum.net/showth ... 288&highlight=physics

Quote from the_angry_angel :

As a speculation, Novadex is only general. i.e. vehicle physics are basic vehicle physics, rather than each indepedent part of the vehicle, and how it affects it as a whole. Whilst it maybe good at some things - like being using in a FPS (such as UT2003/4), its technically not designed for racing sim's, and a fair chunk of stuff would need to be written from scratch, to talk and add on bits that were missing, anyway - so why use Novadex? A lot of the physics is already written, stripping it out and replacing it with Novadex and rewritten code for Novadex compatibility would be nuts - its a huge task. 1 person can only do so much, and theres not much real benefit to do so.

Dont get me wrong, I like Novadex, its good, but it has its place. You wouldnt write a kernel in QBasic, because its the wrong tool for the job. The same applies here.

I'm not sure about this, but LFS isn't really CPU dependant.

By optimization, I assume you mean something to the equivelance of reverse hyper threading, my current Core Duo runs LFS just fine, my Opteron 165 from about 3 months ago with much lower specs ran LFS exactly the same as my e6300 which is almost 2 times faster. So I don't see how LFS really needs to optimize dual core technology cuz its not CPU dependant.

I'm not sure, RockyZ, but I suspect there may be something strange going on with the Core 2 Duo (Conroe) and LFS. I say that because I hear people claiming to have better performance than me with Athlon 64's. Note that I also happen to have a relatively high end video card, a 7900GT.

Quote from Forbin :

I'm not sure, RockyZ, but I suspect there may be something strange going on with the Core 2 Duo (Conroe) and LFS. I say that because I hear people claiming to have better performance than me with Athlon 64's. Note that I also happen to have a relatively high end video card, a 7900GT.

Actually, I'm running an Allendale and not a Conroe. But I really don't see ANY difference at all. I've played LFS with a 900MHZ AMD CPU, 2.2GHZ Athlon XP, 2.6GHZ Athlon FX, 2.2GHZ Opteron, and a 2.45 Core 2 Duo, and the CPUs that were in the 2GHZ range all felt the same since I ran them with the same video card.

I recently upgraded from a 6600GT to a 6800GS and the FPS are more consistent but I have it locked at 100, as far as graphics I have every maxed out with AA and AF on, FPS is still higher than the 100FPS limit I set. So I'm thinking the requirements for LFS is really low.

Heh, I tried maxing out LFS (all in-game settings maxed, 2048x1536 with 6xAA + temporal AA + adaptive AA + 16x HQ AF), and I get ~70fps on my own on some parts of the track, but as low as 17fps in some places. Dunno why that is really, everything looks about the same, I'm assuming my minimum FPS is CPU limited.

Even reducing the resolution down to 1600x1200 doesn't get my FPS that high, peaks at 85fps, yet down to just under 30fps in some sections. So I dunno how the hell you can sit there at 100fps.

I'm running Athlon 64 3200 (2.02GHz) + Radeon X1800XT (256MB version) + 1GB 667MHz DDR2.