Free Race Car Vehicle Simulator

Hello Sim Racers, I'm proud to present the first public release of, - RawPower, The Free Race Car Vehicle Simulator - For more information and download, please go to http://spock.mech.kth.se/rawpower Jesper Adolfsson email:[email hidden]

Simon Brown wrote: > > This is genuinely *very* impressive. Everyone on r.a.s. should download it > and take a look. The response of the car feels spot on to me. thanks > Have you thought about switching to Immediate Mode btw? > Yes, I'm currently writing an importer for SCGT tracks and that one is using DrawPrimitive. /Jesper

Hey, nice work. What about a flexible frame rate, instead of the fixed frame rate you have used? Stephen Simon Brown <[email hidden]> wrote in message news:phiF5.772$[email hidden]... > This is genuinely *very* impressive. Everyone on r.a.s. should downloadit

Quoted text (22 lines)

> and take a look. The response of the car feels spot on to me. > > Have you thought about switching to Immediate Mode btw? > > > "Jesper Adolfsson" <[email hidden]> wrote in message > news:[email hidden]... > > > > Hello Sim Racers, > > > > I'm proud to present the first public release of, > > > > - RawPower, The Free Race Car Vehicle Simulator - > > > > For more information and download, please go to > > http://spock.mech.kth.se/rawpower > > > > Jesper Adolfsson > > > > email:[email hidden] > >

Stephen Ferguson wrote: > > Hey, nice work. What about a flexible frame rate, instead of the fixed > frame rate you have used? > Hmm, flexible frame rate. I'd like that, but I don't wan't to use separate threads to implement it. The physics engine needs to be updated at approximately 500Hz or more. To have a flexible frame rate I would need to spread the drawing of graphics into these updates. To get the frames evenly spread during the physics updates, I could use the framerates 2 4 10 20 25 50 100 250. I'm not sure if it will feel jerky if they are not spread evenly. Of course I could also play with the physics update frequency to get more possible frame rates. Or I could interpolate... Maybe there is a better way? Suggestions are welcome? /Jesper

Thanks... Use the function key F2 to switch to gear and velocity info and use F1 to switch camera view. I will fix the crash when you fall off the track this weekend. I will also add some trees placed around the track. This would add some sensation of speed. /Jesper

Quoted text (11 lines)

> > Excellent! I find it a bit tricky here at the office with my digital pad, > but that's my problem. The dynamics seem to be great from what I've tried, > but knowing what gear you're in is tricky (I have no sound card). The only > fault I can find, is if you fall off the track, you just fall until the > program crashes! Also, a few drawing errors, but nothing that detracts from > the enjoyability. I take it the track is Monza, I like your choice :-) > Keep up the good work. > > Andy > --

> I'm proud to present the first public release of, > > - RawPower, The Free Race Car Vehicle Simulator - > > For more information and download, please go to > http://spock.mech.kth.se/rawpower Excellent! I find it a bit tricky here at the office with my digital pad, but that's my problem. The dynamics seem to be great from what I've tried, but knowing what gear you're in is tricky (I have no sound card). The only fault I can find, is if you fall off the track, you just fall until the program crashes! Also, a few drawing errors, but nothing that detracts from the enjoyability. I take it the track is Monza, I like your choice :-) Keep up the good work. Andy -- Visit my site... http://mail.wm-europe.com/~andrew/ Read about my Mini... http://mail.wm-europe.com/~andrew/mini/ Use console pads on a PC ... http://www.consolepads.co.uk Music videos and DC games ... http://mail.wm-europe.com/~andrew/ftp.html

This is genuinely *very* impressive. Everyone on r.a.s. should download it and take a look. The response of the car feels spot on to me. Have you thought about switching to Immediate Mode btw? "Jesper Adolfsson" <[email hidden]> wrote in message news:[email hidden]...

Quoted text (13 lines)

> > Hello Sim Racers, > > I'm proud to present the first public release of, > > - RawPower, The Free Race Car Vehicle Simulator - > > For more information and download, please go to > http://spock.mech.kth.se/rawpower > > Jesper Adolfsson > > email:[email hidden]

On Thu, 12 Oct 2000 16:25:04 +0200, "Stephen Ferguson" <[email hidden]> wrote: >Hey, nice work. What about a flexible frame rate, instead of the fixed >frame rate you have used? > >Stephen Hey! The truely Great Legend of F1 sims does just the same.. ;-) JoH

Ashley McConnell <[email hidden]> wrote in message news:8s6l5d$65q$[email hidden]... > When will this game be out!!! ;) > > I heard it was just vapour-ware :) > Vapourware? Ha! The fidelity of the 3D modeling and coupled fluid-solid interactions is spot on, but unfortunately the frame rate is suffering. I was thinking of implementing a "Import Your Own Hip Joint" feature, but that would require the end-user to have his own medical-grade MRI scanner. If I can sort those out, then maybe I'll release it to public beta. :-) Stephen

Quoted text (15 lines)

> | > | Nice work. It must be fulfilling to use your research knowledge to also > | split off and do something fun for a while. Unfortunately for me, my > | research is in the area of orthopaedic biomechanics, and the last time I > | looked, there wasn't much of a groundswell of interest for a "3D Joint > | Arthritis Simulation" game. :-) > | > | Stephen > | > | http://cranium.unibe.ch/~ferguson/ > | > | > | > >

Vapour-ware? I thought it would be Vapo-Rub ;) In article <8s6l5d$65q$[email hidden]>, "Ashley McConnell" <[email hidden]> wrote: > When will this game be out!!! ;) > > I heard it was just vapour-ware :) > > | > | Nice work. It must be fulfilling to use your research knowledge toalso > | split off and do something fun for a while. Unfortunately for me,my > | research is in the area of orthopaedic biomechanics, and the lasttime I > | looked, there wasn't much of a groundswell of interest for a "3DJoint

Quoted text (10 lines)

> | Arthritis Simulation" game. :-) > | > | Stephen > | > | http://cranium.unibe.ch/~ferguson/ > | > | > | > >

-- Email: [email hidden] Sent via Deja.com http://www.deja.com/ Before you buy.

> > Six months is very impressive for a one man job, it's took me years to learn > 3D. > What I meant was that it took almost six months of full time work to do it. It was done in my spare time so I've been working on it for a longer period than 6 months ;) (much longer) /Jesper

Stephen Ferguson wrote: > Nice work. It must be fulfilling to use your research knowledge to also > split off and do something fun for a while. Unfortunately for me, my > research is in the area of orthopaedic biomechanics, and the last time I > looked, there wasn't much of a groundswell of interest for a "3D Joint > Arthritis Simulation" game. :-) > You must be doing some advanced math and mechanics in your field? My research field is also sort of biomechanics... We are looking at passive 3D Walkers walking down an incline. It is a model for human walking. Here are some mpg animations of our research, however the files are quite big. This page has not been updated in a while. http://www.mech.kth.se/mechwww/walking/ Sorry if its off topic for this newsgroup.... /Jesper

Simon Brown wrote:

Quoted text (8 lines)

> > Jesper, > > I'm just wondering, how difficult is it to construct a physics model up to > this standard? I noticed you mention the book "Race Car Vehicle Dynamics" > by Milliken in the readme. Is this book, along with a good advanced > mechanics book, and a good understanding of maths enough? (obviously > combined with a lot of hard work :) )

RCVD does not tell you much about how to simulate a race car vehicle. However, it has lots of information on how to model forces and torques acting on various parts of the race car. I've been toold that a better book is "Motor Vehicle Dynamics, Modelling and Simulation" if you wan't to get started doing your own race car vehicle simulator. This book is available at Amazon or World Scientific Publishing. I don't have this book but it was recommended to me by people who definitely know something about this subject. I'll will try to get hold of this book and review it here. http://www.worldscientific.com/books/engineering/3329.html It took me about 6 months to make the program. This includes learning the 3D graphics programming and the vehicle dynamics simulation programming. I'm a PhD student doing rigid multi body dynamics so I had a lot for free. > I'm writing a racing game myself in D3D IM, but my knowledge of race car > physics is practically zero, so I'm reduced to making it an arcade game, > when it could be much more. I'm planning to release my source code using some GNU license. However I need to make it a bit more structured before doing that. Maybe you could use my simulation code in your graphics code. /Jesper > > /Jesper

> > Nice work, Jesper! > It will work fine, if you get enough fps. The less fps, the more the > fps changes will be noticable. > Just continue updating your physics until you reach 'real time' (or > last time recorded as 'now'). I do that and it works fine. Thanks, I'll try that.. > Damn, have to get some more work done on my wheels in my own sim and > release the 2nd alpha. :) > BTW What kind of track method are you using? Quadtrees? 2D BSP? Or > just drawing the whole lot every time? (seems not the case) For drawing I'm relying on the retained mode to clip triangles (the track is one mesh). This is not a very efficient way of doing it. However, the simulator uses a tree structure where the leafs contain only a few triangles in a list, these triangles are searched linearly to find the intersection between a disc (the wheel) and the triangle. That way there won't be any restrictions on the geometry of the tracks.

Quoted text (7 lines)

> > I'm still looking at suspensions myself; looking at the axis systems, > and whether to see the wheel-surface contact as collision detection or > rather use forces if a wheel gets too low (using tire rate and wheel > center rate). Don't know whether I should look at the car from the > wheel perspective (wheels are the basic axis system), or to start with > the sprung mass and force the 5 degrees of constraint on the wheels.

From the wheels perspective, the car body is my main reference system. The suspension unit moves along a line which is defined in the car body reference frame. The inertial reference frame of the wheels is the car body reference frame. I assume that the suspension forces and contact forces are much higher than the inertial forces acting on the suspension unit. This is a major simplification but I beleive it's a reasonable simplification. Any comments?? In the future I will add contrained rotation of the suspension units. > > Nice job, in any case, I wonder what the SCGT track stuff will bring! > Works fine on a Banshee/Win2K btw, in the default 1152x... res. Thanks, the SCGT stuff will probably take some time to implement, unless someone is interested in helping. I also plan to release my source code using some GNU license. /Jesper > > Ruud van Gaal, GPL Rank +53.25 > MarketGraph : http://www.marketgraph.nl > Pencil art : http://www.marketgraph.nl/gallery/ > Car simulation: http://www.marketgraph.nl/gallery/racer/

Jesper Adolfsson <[email hidden]> wrote in message news:[email hidden]... > > It took me about 6 months to make the program. This includes learning > the 3D graphics programming and the vehicle dynamics simulation > programming. I'm a PhD student doing rigid multi body dynamics so I had > a lot for free. > Nice work. It must be fulfilling to use your research knowledge to also split off and do something fun for a while. Unfortunately for me, my research is in the area of orthopaedic biomechanics, and the last time I looked, there wasn't much of a groundswell of interest for a "3D Joint Arthritis Simulation" game. :-) Stephen http://cranium.unibe.ch/~ferguson/

Jesper, I'm just wondering, how difficult is it to construct a physics model up to this standard? I noticed you mention the book "Race Car Vehicle Dynamics" by Milliken in the readme. Is this book, along with a good advanced mechanics book, and a good understanding of maths enough? (obviously combined with a lot of hard work :) ) I'm writing a racing game myself in D3D IM, but my knowledge of race car physics is practically zero, so I'm reduced to making it an arcade game, when it could be much more. Any advice or book recommendations would be much appreciated. Simon. "Jesper Adolfsson" <[email hidden]> wrote in message news:[email hidden]... > > > Simon Brown wrote: > > > > This is genuinely *very* impressive. Everyone on r.a.s. should downloadit

Quoted text (12 lines)

> > and take a look. The response of the car feels spot on to me. > > thanks > > > Have you thought about switching to Immediate Mode btw? > > > > Yes, I'm currently writing an importer for SCGT tracks and that one is > using DrawPrimitive. > > > /Jesper

On Thu, 12 Oct 2000 17:16:36 +0200, Jesper Adolfsson <[email hidden]> wrote:

Quoted text (7 lines)

>Hmm, flexible frame rate. I'd like that, but I don't wan't to use >separate threads to implement it. The physics engine needs to be updated >at approximately 500Hz or more. To have a flexible frame rate I would >need to spread the drawing of graphics into these updates. To get the >frames evenly spread during the physics updates, I could use the >framerates 2 4 10 20 25 50 100 250. I'm not sure if it will feel jerky >if they are not spread evenly.

Nice work, Jesper! It will work fine, if you get enough fps. The less fps, the more the fps changes will be noticable. Just continue updating your physics until you reach 'real time' (or last time recorded as 'now'). I do that and it works fine. Damn, have to get some more work done on my wheels in my own sim and release the 2nd alpha. :) BTW What kind of track method are you using? Quadtrees? 2D BSP? Or just drawing the whole lot every time? (seems not the case) I'm still looking at suspensions myself; looking at the axis systems, and whether to see the wheel-surface contact as collision detection or rather use forces if a wheel gets too low (using tire rate and wheel center rate). Don't know whether I should look at the car from the wheel perspective (wheels are the basic axis system), or to start with the sprung mass and force the 5 degrees of constraint on the wheels. Nice job, in any case, I wonder what the SCGT track stuff will bring! Works fine on a Banshee/Win2K btw, in the default 1152x... res. Ruud van Gaal, GPL Rank +53.25 MarketGraph : http://www.marketgraph.nl Pencil art : http://www.marketgraph.nl/gallery/ Car simulation: http://www.marketgraph.nl/gallery/racer/

"Jesper Adolfsson" <[email hidden]> wrote in message news:[email hidden]...

Quoted text (14 lines)

> > Hello Sim Racers, > > I'm proud to present the first public release of, > > - RawPower, The Free Race Car Vehicle Simulator - > > For more information and download, please go to > http://spock.mech.kth.se/rawpower > > Jesper Adolfsson > > email:[email hidden] >

Well done Jesper, just playing around with low speed donuts you can see like GPL the the cars motion is based on real world physics. Its my opinion that the next great simulator will start off just like this with the community adding the tracks and graphical gloss. It already feels better than a few racing games Ive played this year. Todd.

When will this game be out!!! ;) I heard it was just vapour-ware :)

Quoted text (13 lines)

| | Nice work. It must be fulfilling to use your research knowledge to also | split off and do something fun for a while. Unfortunately for me, my | research is in the area of orthopaedic biomechanics, and the last time I | looked, there wasn't much of a groundswell of interest for a "3D Joint | Arthritis Simulation" game. :-) | | Stephen | | http://cranium.unibe.ch/~ferguson/ | | |

Woh - this will spawn a whole new genre and newsgroup :) rec.medical.simulation :) Cant wait to get my hands on the beta

Quoted text (10 lines)

| Ashley McConnell <[email hidden]> wrote in message | news:8s6l5d$65q$[email hidden]... | > When will this game be out!!! ;) | > | > I heard it was just vapour-ware :) | > | | Vapourware? Ha! The fidelity of the 3D modeling and coupled fluid-solid | interactions is spot on, but unfortunately the frame rate is suffering. I | was thinking of implementing a "Import Your Own Hip Joint" feature, but

that | would require the end-user to have his own medical-grade MRI scanner. IfI

Quoted text (7 lines)

| can sort those out, then maybe I'll release it to public beta. :-) | | Stephen | | | > | | > | Nice work. It must be fulfilling to use your research knowledge to

also | > | split off and do something fun for a while. Unfortunately for me, my | > | research is in the area of orthopaedic biomechanics, and the last timeI

Quoted text (13 lines)

| > | looked, there wasn't much of a groundswell of interest for a "3D Joint | > | Arthritis Simulation" game. :-) | > | | > | Stephen | > | | > | http://cranium.unibe.ch/~ferguson/ | > | | > | | > | | > | > | |

Thanks for the info Jesper. I looked both books up on amazon. It seems mechanics books cost even more than 3d programming ones :) I'll order Motor Vehicle Dynamics when I have the funds and see how difficult it all looks. It shouldn't be beyond me though. Six months is very impressive for a one man job, it's took me years to learn 3D. Also thanks for the offer of your code at some later point. I may be interested, I don't know really. Obviously I'd like to try and learn it for myself first. Most of the games I'm planning to write, although only a few are car-oriented, will use a lot of physics, so I really need to take the time to learn how to code physics models. Best of luck in the future, Simon.

Quoted text (22 lines)

> RCVD does not tell you much about how to simulate a race car vehicle. > However, it has lots of information on how to model forces and torques > acting on various parts of the race car. I've been toold that a better > book is "Motor Vehicle Dynamics, Modelling and Simulation" if you wan't > to get started doing your own race car vehicle simulator. This book is > available at Amazon or World Scientific Publishing. I don't have this > book but it was recommended to me by people who definitely know > something about this subject. I'll will try to get hold of this book and > review it here. > > http://www.worldscientific.com/books/engineering/3329.html > > It took me about 6 months to make the program. This includes learning > the 3D graphics programming and the vehicle dynamics simulation > programming. I'm a PhD student doing rigid multi body dynamics so I had > a lot for free. > > I'm planning to release my source code using some GNU license. However I > need to make it a bit more structured before doing that. Maybe you could > use my simulation code in your graphics code. > > /Jesper

Ok, understood :) "Jesper Adolfsson" <[email hidden]> wrote in message news:[email hidden]... > > What I meant was that it took almost six months of full time work to do > it. It was done in my spare time so I've been working on it for a longer > period than 6 months ;) (much longer) > > /Jesper

On Fri, 13 Oct 2000 10:27:41 +0200, Jesper Adolfsson <[email hidden]> wrote:

Quoted text (7 lines)

>> Damn, have to get some more work done on my wheels in my own sim and >> release the 2nd alpha. :) >> BTW What kind of track method are you using? Quadtrees? 2D BSP? Or >> just drawing the whole lot every time? (seems not the case) > >For drawing I'm relying on the retained mode to clip triangles (the >track is one mesh). This is not a very efficient way of doing it.

Ok, will probably be a problem once you get complete SCGT files in, but I understand this is something to be worked on when it becomes a problem. >However, the simulator uses a tree structure where the leafs contain >only a few triangles in a list, these triangles are searched linearly to >find the intersection between a disc (the wheel) and the triangle. That >way there won't be any restrictions on the geometry of the tracks. Hm, I intend to use a separate track description from the actual geometrical representation (the triangles). This way I can decrease the level of detail of the track without making curves to handle like digitized ones. I mean, getting a banking turn to smoothly change angle instead of relying on the actual polygons (which is like making a turn out of wooden boards, sort of). This is something that GPL uses; sectors with start altitudes and ending altitudes, so you can interpolate, no matter what the resolution of your triangles is. However, this means that I'm doing my own track creation script mini-language. (the first totally boring straight with no textures is already born).

Quoted text (15 lines)

>> I'm still looking at suspensions myself; looking at the axis systems, >> and whether to see the wheel-surface contact as collision detection or >> rather use forces if a wheel gets too low (using tire rate and wheel >> center rate). Don't know whether I should look at the car from the >> wheel perspective (wheels are the basic axis system), or to start with >> the sprung mass and force the 5 degrees of constraint on the wheels. > >From the wheels perspective, the car body is my main reference system. >The suspension unit moves along a line which is defined in the car body >reference frame. The inertial reference frame of the wheels is the car >body reference frame. I assume that the suspension forces and contact >forces are much higher than the inertial forces acting on the suspension >unit. This is a major simplification but I beleive it's a reasonable >simplification. Any comments?? In the future I will add contrained >rotation of the suspension units.

RCVD contains a chapter on axis systems. I have little time currently (also just received 'Fundamentals of Vehicle Dynamics' today, but gosh, these books are expensive as the dollar is expensive compared to the Dutch guilder, so have to slow down on buying interesting stuff like this), but that's high on the To-Read list. I believe it proposes one set of axis systems. I believe your simplification about suspension intertial forces is quite correct. The body weight and road upward forces are much more important. And sometime you have to stop doing secondary effects, as this will only bring your fps down. >> Nice job, in any case, I wonder what the SCGT track stuff will bring! >> Works fine on a Banshee/Win2K btw, in the default 1152x... res. > >Thanks, the SCGT stuff will probably take some time to implement, unless >someone is interested in helping. I also plan to release my source code >using some GNU license. Great to hear. I'm interested in the SCGT stuff myself too, although ouch, I don't have SCGT! Or uhm, I have 'Grand Touring' here from Empire, but that's different, is it not? And have to get suspension working before I'm doing track stuff. Ruud van Gaal, GPL Rank +53.25 MarketGraph : http://www.marketgraph.nl Pencil art : http://www.marketgraph.nl/gallery/ Car simulation: http://www.marketgraph.nl/gallery/racer/

On Fri, 13 Oct 2000, Simon Brown wrote: > Thanks for the info Jesper. I looked both books up on amazon. It seems > mechanics books cost even more than 3d programming ones :) I'll order Motor > Vehicle Dynamics when I have the funds and see how difficult it all looks. Here is the <shameless plug>! For our book, "Race Car Vehicle Dynamics", it's much faster (and maybe cheaper outside the USA) to order direct from the publisher. Search for R-146 in the bookstore at <http://www.sae.org>. They also sell a number of other books on racing. -- Doug Douglas Milliken <[email hidden]> Milliken Research Associates Inc.

On Fri, 13 Oct 2000, Jesper Adolfsson wrote: <snip of nice comments about our book>

Quoted text (9 lines)

> acting on various parts of the race car. I've been toold that a better > book is "Motor Vehicle Dynamics, Modelling and Simulation" if you wan't > to get started doing your own race car vehicle simulator. This book is > available at Amazon or World Scientific Publishing. I don't have this > book but it was recommended to me by people who definitely know > something about this subject. I'll will try to get hold of this book and > review it here. > > http://www.worldscientific.com/books/engineering/3329.html

Or if in the USA, order through your local Barnes & Noble, for a lot less money -- they quoted me US$59, and 1 week delivery to the store. -- Doug Douglas Milliken <[email hidden]> Milliken Research Associates Inc.