New Technology Support

[srdsprinter]

How about, support for multi-threaded dual core optimization (for those so well endowed with a X2 or dual core), and support for emerging technology like the new advanced physics support by the physX processor from ageia.

The physics processor is quite a stretch for LFS in some regards, as it is not fully proven technology, and it certainly hasn't been utilized with regards to a racing sim. Already incorperated into the new ghost recon game, bet on soldier and rise of nations, the physX is planned to be incorporated into many new titles. The technology is amazingly promising, and would be incredible if unleashed into a sim like lfs. Imagine if you will:

A close race between 20 FOX's racing down the front stretch of blackwood, white flag, last lap. 5th place going down the back stretch as all drivers jockey for position in the draf. The second place car goes for the lead, but locks the brakes up. The smoke pours off the front tires as he slides through the gravel. As the cars in front of you pass, the smoke vortexes and spirals off the aerodynamic surfaces of their cars. Into the right-left chicane bits off ruber fling towards your visor as the new found second place driver battles off the racing line. Going wide out of the turn, down hill the leader puts his right rear off into the grass. The blades of grass and dirt are sprayed in perfect detail at 100mph in all directions as you concentrate on making it through. In an instant, the leader helplessly crosses the track, clipping the left rear of the next driver sending his car cartwheeling. Before the leader crashes into the armco, the engine cover of his car flutters through the air, while the suspension bits of the 2nd place car are ripped off and thrown in every direction. The former leader's tire and suspension bounce off the armco barrier and fly towards the helpless car in front off you. He dodges right but to no avail, as the red-hot, smoking headers of the cart-wheeling FOX fall downward demolishing his rear wing meters from your nose. Through the smoke, grass, flying tires and carnage, you see the drivers hopes go disappear as you blow by to take the checkerd flag.

Now I KNOW this isn't what LFS is about, but the technology the dedicated physics processor (or using part of the huge processing power of SLI for physics processing) can do all of this without affecting really affecting the frame rate, or causing LFS to loose what makes it so great, the vehicle dynamics. In 5 years, it will probably be mainstream to expect this level of interactive environment, but it is possible now (alienware, falcon, and the dell renegade are shipping with this processor, ATI and Nvidia are experimenting with the idea, and Agia plans on shipping the physX cards this summer). LFS can use all the cpu processing power on the physics engine of the vehicle dynamics, while the physics processor can make sure that every piece of rubber, blade of grass, bit of armco, piece of suspension and engine can be destructed and will interact with you the driver. If not, the technology is still very exciting and I imagine will at some point be integrated by someone into a driving sim.

Info on physX:
http://gear.ign.com/articles/697/697450p1.html

http://media.gear.ign.com/articles/697/697450/vids_1.html
(check out "hanger of doom" as it shows the how it can release the main CPU from bogging down. also the "cell factor" movie is fairly amazing, looking more like 3d rendering rather than fast paced fps)

http://physx.ageia.com/footage.html
The real time ghost recon comparison is amazing.

http://www.ageia.com/physxinaction/demos.html
These demo videos (the boulder one) are neat

http://physx.ageia.com/titles.html

The games it will optimize, maybe LFS some day... :shrug:

[Bob Smith]

There was a big thread about this on RSC... don't you have to use their own propriety physics engine to use the card?

[filur]

From what i've gathered, ageia is:

Very expensive
Not very accurate

Pretty poor combination that, and muxing two different physic models doesn't sound good to me, and replacing the current one seems like a really bad idea and probably an incredibly long process.

.. don't you have to use their own propriety physics engine to use the card?
..it's also a closed piece of hardware. So unless you want to "sell" your physics SDK to those guys, you won't be able to expose this hardware with your own SDK.
http://saschawillems.blogspot.com/2005/09/ageia-putting-end-to-realistic-in-game.html

[srdsprinter]

There was a big thread about this on RSC... don't you have to use their own propriety physics engine to use the card?

I'm not sure, the articles I read were pretty vague. (i dont go to RSC much :( )

From what i've gathered, ageia is:

Very expensive
Not very accuratePretty poor combination that, and muxing two different physic models doesn't sound good to me, and replacing the current one seems like a really bad idea and probably an incredibly long process.



From what I've heard the add-in cards available this summer will be produced by manufactures of video card like BFG and Sapphire(not sure exactly which brands). They will have a MSRP of about $299 USD. This is about the price of a mid-high end graphics card, which is high, but not a huge amount when you consider a SLI or top of the line Crossfire can cost you well over $1000 USD.


EDIT* I intended this as a what-if, not just to be bashed for technicalities. IF they can use there own physics, and IF its not impossible to implement into current programming, Would it be Cool?

Once they work things like this out, it should be neat IMO. Game Programming seems to be about optimization now more than ever. Dual-core optimization (cell processor in the not-too-distant future with ps3 and pc's), multithreading, sli/crossfire, and advanced physics modeling all now challenge game developers and programmers.

[filur]

.. They will have a MSRP of about $299 USD.

I meant the cost of licensing ageia, but you're right, the addon card is also pretty expensive :)

[srdsprinter]

I meant the cost of licensing ageia, but you're right, the addon card is also pretty expensive :)

Ah, I hadn't crossed that path. Mainly I just wanted to express that this technology, is here in somewhat infancy, but it is pretty mindblowing. It will become more mainstream just as sli and dual core have become because it can take gaming to new levels. It will come sooner or later, it seems these things sometimes come later to racing sims than other games, and it could be cool if LFS was a leader (as it already is in terms of online gameplay, driving physics, etc...)

[filur]

It will become more mainstream just as sli and dual core have become

I really don't think you can call SLI and dual core cpu's mainstream just yet, they're all still in the very bottom of an ascending price-sorted list of items in any online computer store. :)

[srdsprinter]

IMO PC gaming hardware is already too expensive. If additional garbage like physics processors become commonplace I'll just switch to consoles like everybody else seems to be doing.

I agree, if you use a PC solely for gaming and gaming only in place of a console, then it is highly expensive. I (and most people) dont, it is a multimedia tool I use for gaming, studying and writing reports, web browsing, watching dvds, burning dvds, watching tv, tivoing listening to music, mp3 management, file storage, networking, messaging, getting the weather, checking news, stock markets, buying equipment, buying gifts, and countless other things.

If I didn't have a computer to do ALL these, I would need a seperate tv, dvd player, file storage hard drive, a tivo, some way to get mp3's to an ipod, a radio, cd player, typewriter, internet web brower, digital cable, more phone service, have to DRIVE to stores and deal with people, watch the news, read newspapers, buy newspaper subscriptions and magazine subscriptions, get a broker, and still not be able to game! I'd need a good console (there really aren't any anymore) and then buy another steering wheel, and buy games I couldn't try before I buy, which I have to go out and actually buy in Person! Now, I'm not a math major (i'd need like 6 more credits) but it seems to me that the savings in space alone would be worth a good multi-purpose pc, but the cost of the individual replacements would be greater than the sum of all the parts of a good pc.

Sorry, but when you think what a pc REALLY does for you, the price is not that high...

[srdsprinter]

I really don't think you can call SLI and dual core cpu's mainstream just yet, they're all still in the very bottom of an ascending price-sorted list of items in any online computer store. :)

sorry, you are correct. Gamers know SLI, multi-taskers know dual-core. The general public might not know yet, but they are learning. With both amd and intel making relatively inexpensive dual core chips now, mainstream computer users are being introduced to what dual core offers.

Also, sli is really only known with gamers. I'm not even sold on the idea. But it is something you that is being devoloped heavily, and hasn't up-and-dissappeared-like-a-fart-in-the-wind after a year of existence.

physics processors haven't been around that long. But they are being embraced (by mainstream pc maker dell) as part of the ongoing evolution of the home computer. If it is embraced and used in games like ghost recon, some hardcore gamers will go out and grab it without thinking. Most gamers will wait (as with sli and dual core) and get a feel for the impact that it REALLY makes on games. After mainstream gamers support it and buy it, and tell dell and hp to use it, it will become mainstream IMO.

[mr_x]

i think after time as this technology is improved then it will be spot on. but as with all technology.. nothing is perfect

[filur]

gaming, studying and writing reports, web browsing, watching dvds, burning dvds, watching tv, tivoing listening to music, mp3 management, file storage, networking, messaging, getting the weather, checking news, stock markets, buying equipment, buying gifts.

Task(s)(-s) requiring a cpu over 500MHz marked in bold type above. :)

[srdsprinter]

Me neither. I need a PC at home for work, and I use it for all the myriad other reasons you mentioned, but the fact is that a pretty slow box could do all that stuff - the only real reason I've got performance hardware (or it was when I paid for it anyway ;)) in there is for gaming.

Anyway, you're american, your cost of living is practically nothing compared to the rest of the world (and I've lived in Raleigh and Myrtle Beach, so that appraisal is based on experience!), so your definition of "expensive" is probably quite different.

Agreed, people all over the world have different ideas of expensive, no doubt about it. I feel, for me anyway, that once you have a basic computer, the additional cost of making it a good gaming pc will also make it a much better all round pc, and if it can replace my need for an 800 dollar ps3 (which pretty much nessecitates a REALLY expensive HDTV to utilize the effect), all the better. I think the sooner a young new technology like this embraced, the sooner it becomes reasonable in price, and the sooner it makes my pc gaming That much better. With a console, you get shafted because there are no upgrades, and IF there are, it comes in the form of a new model, which you DONT have! My psp for example, is about to have 4 gigs flash memory and a camera on it, I had to buy the aftermarket 180 dollar 4 gig hard drive on top of the price of the psp... bummer.

[srdsprinter]

Task(s)(-s) requiring a cpu over 500MHz marked in bold type above. :)

Not if your getting them all done at the same time in the 10 minutes before work/class. :D

[Vain]

1. On the games-and-hardware-situation:
I will think about buying new, modern and fancy hardware when people start to make good games again. When I go to the shop to look at new games I always end up thinking "and where... are the good games?". No need to buy hardware for 800€ just to play games I don't like.
2. On the PPU (let's abbreviate physics processing unit):
Let's look at how the GPUs came to the world. First graphics were done on the CPU. Then someone said "hey, look at this GPU, with it you get fancy textures!" and a hand full of developers began optimizing their code for this product, and later products. And it was a hassle, because every device did it's thing differently. Then people came up with libraries to make things more simple. OpenGL, DirectX, you know them. After the hardware became compliant to these standards the GPUs became much more popular, because about every program could support these devices very easily.
Let's go back to the PPU. We should wait until somone releases a OpenPL or DirectX 273*2,71/3,14 or the like that offers a common library for developers to utilize all current and future PPUs on the market.

Vain

[tristancliffe]

I think I'm just getting a bit jaded with the PC gaming scene in general. To be honest, that's how I ended up at LFS - because the big publishers are releasing the same tired old crap over and over again with an extra <x> polygons each time and a few re-hashed effects.

The returns have diminished to the point that investment in new hardware seems futile. Do I really want to spend £500 to get the next-detail-up textures in Battlefield 2? Even if I do, it won't make all the other players on the server suddenly become less dickhead-ish, and honestly I'd rather upgrade the gaming public than my gaming PC.

Lol at your avatar :D

[filur]

We should wait until somone releases a OpenPL

Very good point. :thumb:

[srdsprinter]

I'm familiar with ventrilo, and teamspeak, but has it been suggested that built in communication be written into the software? (i searched but couldnt really find anything) If it was a clearly marked well defined option would it be good to be able to do this without an additional 3rd party program running just for your microphone?

(im sorry if this has already been discussed, hence why no new topic:shrug: )

[SamH]

I watched the 1stPS video with the PhysX thing doing its thing in one clip, and a high-end card doing the same thing in another. It looked pretty well identical, except for a rather impressive explosion with bits flying all over the place. Very gory, I'm sure.

What I did notice was that in the PhysX clip, a guy (not the 1st person) appeared to get hit by flying debris and finished up on the ground, either dead or seriously wounded. In the other clip, he didn't get hit by anything because there wasn't the impressive collection of flying debris, and didn't finish up on the ground.

So I came to wondering.. if PhysX WERE included in LFS, where would the information about the flying bits come from? And how quick would that flow of info be, between different racers? I don't wanna be wiped out by an invisible piece of flying debris from one guy's car, just because the PhysX processor of some guy, behind us both, said it happened that way.

I may be wrong, but I suspect the PhysX engine doesn't actually contribute to the multiplayer physics data. It merely makes "big" what it sees locally. But if this is the case, then wouldn't PhysX-users be swerving around before our eyes, to avoid trailing bumpers and bouncing wheels that the rest of us can't see, and resulting in erratic and unpredictable driving for seemingly no reason? :shrug:

[filur]

I may be wrong, but I suspect the PhysX engine doesn't actually contribute to the multiplayer physics data.

I'd guess you could tie in the solved physics to whatever you'd like, question of design, and a very huge amount of network traffic if you'd like it all to be interactive and not just fancy effects.

[srdsprinter]

I may be wrong, but I suspect the PhysX engine doesn't actually contribute to the multiplayer physics data. It merely makes "big" what it sees locally. But if this is the case, then wouldn't PhysX-users be swerving around before our eyes, to avoid trailing bumpers and bouncing wheels that the rest of us can't see, and resulting in erratic and unpredictable driving for seemingly no reason? :shrug:

Yeah, who knows... :shrug:

the technology is in its enfancy, but the cell factor video looked pretty good, with those gravity mine things... it seemed like things would fly "through" you in a lot of those videos. Don't know what they're planning on doing about that.... but it certainly seems as if it should effect the user. Interesting conundrum. :scratchch

[hideaki-san]

The problem is that both PhsyX, ATI, and Nvidia are working on their own solutions regarding physics rendering, so it's better to wait until the dominant player emerges.

[spankmeyer]

The problem is that both PhsyX, ATI, and Nvidia are working on their own solutions regarding physics rendering, so it's better to wait until the dominant player emerges.
I Agree. I'm imagining most probable thing is that either ATI or nVidia integrates a physics processor on their gaming GFX cards. But once again it's the catch 22 situation (you need a killer app with amazing physics only available with the new hardware, but no company wants to spend boatload of cash and release new hardware without software support on consumer lever).

Hopefully I got that description right, confusing stuff for non-native English speaker. :)

[Vain]

I believe, if NVidia and ATI get involved in the PPU-market, that there will propably be a library like I talked about above that has a software-engine built in. That means all game developers can use this library to have objects automatically behave physically correct (of course, within a given frame) and the library decides wether the CPU has to take the burden or wether there is a PPU to do all that stuff. That'd mean that you can either have no PPU and a high-end-CPU to do the physics at expense of otherwise-cpu-related performance, or a dedicated PPU and an idle CPU that can make calculations on AI or calulate pi or whatever.
That'd be my approach and it's obviously the best. :D

Vain

[filur]

I believe, if NVidia and ATI get involved in the PPU-market, that there will propably be a library like I talked about above that has a software-engine built in. That means all game developers can use this library to have objects automatically behave physically correct

Probably, but only as with all other standards after there have been 10 "standards" competing for some time.

[Woz]

This will be a case of until a solution that people accept is ready it will not take off. That coupled with a killer app. It took years for Voodoo to establish the 3D market and the physics card is the same.

Trouble is they do not have the time to establish the market. NVidia and ATI are looking at SLI to do physics processing. But worse than that we now have dual core CPUs and quad are in the pipeline. I would expect quad to be later this year or early next

So with 4*CPUs running at 3GHz why would you need a physics card.

[sgt.flippy]

First of all, it's spelled fizzics.

Secondly, I barely understand what great things this does, I think the truck blowing up looks overdone, but after all, I don't know how a real explosion looks like. About those libraries, I would love to see a library for materials. Wouldn't it be grand to just model something, slap on the color you like, give it a flag or whatever, and it shines or doesn't shine like it should do. That's what keeps me away from modeling too much... Texturing :D

Oh yea, I'm not really pro-updating LFS to support all this fancy stuff. It'll just make it a tool for hardware developers to get you to buy better stuff every time something new comes up. Buy a computer today, dump it tomorrow because there is a new game.

[gezmoor]

Sorry to drag up an old thread but it would appear that NVidia have aquried Ageia and therefore the PhysX engine.


http://www.nvidia.com/object/io_1202161567170.html

[Töki (HUN)]

/me wants too (as I said this in about 100332 threads)

[Jakg]

nVidia will implement PhysX using CUDA - imo all those PPU cards are about to become useless.

WE NEED MULTITHREADING!

[Woz]

WE NEED MULTITHREADING!

+1

Dual core is common and 4 core cpu's starting to hit the market, more than enough power to do what is required. Also if you make your app multi threaded it will scale to use the cores that are available, not the case for the PPU.

Its just pointless for any developer to support PPU's IMHO.

In a years time people will all but forgotten about them :)

[Dajmin]

As long as it's just optimal multithreading and those of us with single cores don't suffer because of it.

We already have a community divided amongst the "I like LFS because my 3 year old PC runs it" who become the "OMG you killed my frame rates with your extra 4 polygons in South City" and the people like yourselves with high-end new kit, we don't want that gap to get any wider.

[Jakg]

Multi-threading means slightly improved FPS for HT users and GREAT FPS improvements for us with high-end kit (and those with average dual cores). For single core users it yeilds no improvement or disadvantage.

At the end of the day the more Scawen tries to make LFS realistic with enhancements for stuff like brake wear/fade, and improved Engine Modelling, CPU load WILL go up. It makes sense that in a game that can bring a Quad-core to 40 FPS in a full grid that theres 300% more CPU power going to waste, that this should be tapped into.

[DarkTimes]

I think it's too early for Scawen to consider things like PPUs before that market has developed, but the game does need to start taking advantage of multicore CPUs soon really. Especially so as the complexity of the physics simulation increases. I'd be very happy for LFS to use both my cores, instead of just sitting at 50% most of the time.

[wien]

For single core users it yeilds no improvement or disadvantage.That's not necessarily true. Depending on how multi threading is implemented design decisions made to accommodate multiple threads in parts of the pipeline may affect its efficiency on single core systems, even if the engine is still running in one thread on these systems.

[GobLox]

I meant the cost of licensing ageia, but you're right, the addon card is also pretty expensive :)

Ageia had some part of their operation here in St.Louis and they came out to demo the hardware for the LAN Community. My understanding is that there were no licensing costs for developers; Ageia desperately wanted to garner support for their hardware and they gave away, more-or-less, plug and play physics code for development use.

Physics accelerators aren't really applicable to online gaming. There were some pretty excellent mods that took advantage of HL2's revolutionary physics engine but it did point out that synchronizing more than a few 100 moving objects for more than a handful of clients via the internet presented a bandwidth issue. The PhysX card is capable of working with hundreds of thousands or even millions of objects/collisions at once but there's no way to harness that power for elements that effect gameplay over the internet. Synchronizing Position/Orientation as well as vector information for 300,000 objects 10 times each second doesn't work for online gaming and those are the kind of numbers that make it worthwhile to utilize a Physics Accelerator. Objects/Gases/Liquids that are superfluous and don't effect gameplay (Tire smoke which does not have to appear exactly the same to each client and does not need to be synchronized accurately) could take advantage of an accelerator but I doubt you'll see this implemented in LFS.

[Woz]

Multi-threading means slightly improved FPS for HT users and GREAT FPS improvements for us with high-end kit (and those with average dual cores). For single core users it yeilds no improvement or disadvantage.

Good multi threading requires an overhead to keep in sync and to protect data overwrite etc. So a move to multithread does have an impact on single core users.

Also HT actually has a negative impact sometimes because there is a single shared core so the extra context switching can slow things down.

[SgtH3nry3]

PhysX isn't interesting anyway. CUDA is a proprietary language and will be superseded by standard API's like DirectX compute shaders and OpenCL which will work on all GPU's.

@GobLox: You don't need continuous synchronizing between clients and servers or huge data transfers.
There is a lot possible with partial prediction and low-level synchronizing.


Multithreading is a must though, Valve Software has proven it doesn't have to strain singlethreaded systems and can increase performance in huge factors.

[DragonCommando]

I have a P4 prescott 3.2ghz HT, when I play any source games my performance is nearly doubled when I have HT enabled.

My framerate goes up from 100fps to 170fps, this is with the same high settings and the same (complex) scene in single player.

[arco]

*shameless bump*

Nvidia has released beta drivers (http://www.nvidia.com/object/win7_winvista_32bit_195.39_beta.html) with OpenCL (http://www.nvidia.com/object/cuda_opencl.html) support, as has ATI too I believe. Would LFS benefit from using OpenCL for its physics calculations, and how much work would it be to implement it in LFS?

[wien]

and how much work would it be to implement it in LFS?Rewriting the entire physics engine? I'm sure that can be done in a jiffy. :) Could be useful for new developments like proper aerodynamics simulation (http://www.youtube.com/watch?v=dYumgkhn8lQ) though, but I'm not certain if even modern GPUs can do that at a high enough detail level to be useful (compared to faking it I mean).

[arco]

Yeah, I wasn't sure if this was something fairly easy to implement or would require a lot of work. But if it means a total rewrite of the physics engine, I suppose it's out of the question at this time.

[wien]

OpenCL (and The DirectX equivalent - Compute shaders), are basically completely new languages. They're both similar to C in syntax, but that's where the similarities end. The current C++ physics code from LFS would have to be completely rewritten to take advantage of either. Scawen would probably also need to re-jig the algorithms he has spent years developing to allow for the massively parallel execution model of these new languages. That's where the real tricky stuff comes in.

In the near future I think multi-threading would be a much more realistic goal in terms of speeding up the physics.

[Bluebird B B]

"physX processor from ageia"

Not many successful attempts to use it and as mentioned earlier, also i never read about anybody being enthusiastic about it. Just as mentioned earlier, its a very, very, very bad idea to supporting technology's specific for one manufacturer. So physX, No i disagree.

There is an other possibility to get a lot of extra physics calculations:
Todays pc's come with at least dual core cpu's and quad-core cpu's are now available below $100 dollars or less then 85 euro's. So multi-threading would give huge possibilitys to do many, many physics calculations But somewhere on this forum i read a post from scawen rewriting the engine of lfs is a lot of work meaning it will take a long time to do it. But i do think it is becoming more important everyday since performance for a individual core is hardly improving over the last two years but the number of cores is still growing. Soon we can plug in hexa-cores from amd and a few monthes later intel will be producing with 6 to 8-core cpu's. In fact, amd is already supplying hexa-core cpu's for x86 servers.

So if the crew wants to add realism, for example model complex aerodynamics into lfs, i think there will be no other possibility than to rewrite the lfs engine to really take advantage of at least two cpu cores. It will increase available processing power on the average system by about 80%. With quad-cores up to 200% extra cpu-power (using more cores rarely gives linear improvements in performance).

So i do think lfs should use multiple cores, it will open up great possibility in terms of added realism. Waiting to get faster individual cores.. is going to take a very, very long time..

[SgtH3nry3]

OpenCL is very interesting to run physics engine on. It is very scalable and runs on CPUs, GPUs and even DSPs when OpenCL-capable.

Aerodynamics simulation is possible, to have proper CFD you need dual precision floating point operations and many of them simultaneously.
An AMD HD 5800 series GPU should be fairly capable of doing CFD, not to mentioned nVidia's Fermi.

The problem is that you will need brand new hardware to run this. Which doesn't really fit Live for Speed since that is a very light optimized software package.
Maybe it is best to add OpenCL support now, but to add more demanding physics in the future.
Multithreading should be integrated in the game as soon as possible.