Is Suspension Tuning Backwards? - A Test with RX-8
- Thread starter Maturin
- 458 comments
- 45,093 views
ok, last night I decided to put on my patience hat, start from scratch, and see what I could come up with.
Car in question: 1991 Prelude SI VTEC.
Track used: TSUKUBA
the car is lightly modified, with a N/A stage 1 tune, and the basic other stuff. 260HP, stage 1 weight reduction, stock transmission... I put the stock LSD back in, just to remove one variable. (I'm not sure if the "stock" differential is an open or limited slip unit - LSD was an option in Japan on this car...)
I unfortunately left all my notes at home (I'm at work), but I'll go off of what I remember...
to start off: EVERYTHING at its lowest setting, except for camber.
(Front / Rear)
SPRINGS
3.4 / 3.1
RIDE HEIGHT
96 / 96
BOUND
1 / 1
REBOUND
1 / 1
CAMBER
2.0 / 1.0
TOE
0 / 0
SWAY BAR
1 / 1
Each test involved about 4 laps, with an attempt at driving each of them the same. Braking is done early, and turn in is done with no braking applied. Just enough throttle is maintained to hold a constant speed until just before the apex, where throttle is applied. This technique is used in all corners. Here, I will list each change, and the results. everything else not mentioned can be considred the same as the test before.
The results were interesting I guess... some good, some bad, and some ugly.
First the ugly: I have commented on this before, but I think it should be said again If there is anything that PD could not have POSSIBLY screwed up more, its the behavior of this FWD vehicle when wheel spin occurs. I'm not sure if all FWD vehicles in the game are like this, or the prelude is just goofed up, but its really REALLY bad.
In the real world, FWD cars exibit understeer when throttle is applied, because, well, obviously, the tire only has so much traction to provide, and when you use some of that for accelerating the car, less is available for lateral grip. This problem is directly related to power output - the more horsepower, or the lower gear used, the more understeer will occur when the throttle is applied. If enough power is sent to the wheels to cause wheel spin, the cars path normally continues in whatever direction it is going - i.e. total loss of grip until power is reduced.
Well! I have found that in a slow corner, in 2nd gear, if it appears that I am a bit too hot and am going to understeer off the outside of the track, I simply have to MASH ON THE GAS. The front tires break loose in wheel spin, and yet somehow the front tucks down toward the apex, the backend slides out... if the steering angle is maintained, the car gets completely sideways, and looses a ton of speed nearly instantly. The result is that as a balanced FWD car is exiting a slow corner, when full throttle is applied in a low gear, you must COUNTERSTEER TO PREVENT OVERSTEER. WTF.
While traction is maintained, it behaves as normal... additional throttle results in less grip up front - but when power overcomes available traction, out comes the back end. So this means as you exit a corner at full power, you will have UNDERSTEER in third gear, and OVERSTEER in 2nd gear. This is ass-backwards stupid.
anyway, I had to keep this in mind as I performed my tests.
With the intial settings, body roll and pitch were very noticable, but not more then stock. This makes sense, as the lowest spring settings are ever so slightly more then a stock prelude spring. It would also suggest that a sway bar setting of "1" isn't exactly like taking the sway bar off, nor is it any noticable amount less then stock.
The car was well balanced with these settings, and though turn in was a bit slow, and it took a second for the car to "set itself" in the turn so to speak... once in steady state cornering the car was, for the most part neutral with maybe a hint of understeer. The car could be easily steered with the throttle - lifting slightly to tighten the line, or accelerating to widen it... unless the magic wheel spin occured...
There was no evidence that the car was bottoming out. Body roll in replay seemed to show the tires were not too far up in the wheel well, but that doesn't really matter because I've seen plenty of screenshots, both in photodrive and phototravel, of front tires poking through fenders.
1st change:
SPRINGS 3.4 / 4.1
Same neutrality in corners, for the most part. perhaps a SLIGHT bit more oversteer, possibily wishful thinking.
2nd change:
SPRINGS 4.0 / 5.5
I am pretty sure I noticed more rotation, and less understeer. Body roll and pitch is reduced by QUITE a bit.
This seems to suggest that the numbers mean nothnig, and if you hang out in the lower end of the spring rates, you can make the car handle like you want, and standard tuning methods are effective.
I went through a few more spring rate changes, always keeping the springs under 6kg/mm. Car seemed to behave somewhat as expected, as long as you are not expecting the numbers to be accurate.
for my next 3 changes, I played with the sway bars.
As far as balance is concerned, they don't do crap. At all.
I then played with shocks.
upping the rear shocks to 2/5 resulted in more rotation in transitions, but it SEEMED like when I got back on the throttle, there was MORE understeer. When I went back to 1/1, it still felt the same, so maybe I imagined it.
The higher bound & rebound definately resulted in more rotation in turn in, and when lifting off the throttle, so I guess thats a good sign.
I had much more extensive notes at home, but perhaps someone can build off of these results.
Car in question: 1991 Prelude SI VTEC.
Track used: TSUKUBA
the car is lightly modified, with a N/A stage 1 tune, and the basic other stuff. 260HP, stage 1 weight reduction, stock transmission... I put the stock LSD back in, just to remove one variable. (I'm not sure if the "stock" differential is an open or limited slip unit - LSD was an option in Japan on this car...)
I unfortunately left all my notes at home (I'm at work), but I'll go off of what I remember...
to start off: EVERYTHING at its lowest setting, except for camber.
(Front / Rear)
SPRINGS
3.4 / 3.1
RIDE HEIGHT
96 / 96
BOUND
1 / 1
REBOUND
1 / 1
CAMBER
2.0 / 1.0
TOE
0 / 0
SWAY BAR
1 / 1
Each test involved about 4 laps, with an attempt at driving each of them the same. Braking is done early, and turn in is done with no braking applied. Just enough throttle is maintained to hold a constant speed until just before the apex, where throttle is applied. This technique is used in all corners. Here, I will list each change, and the results. everything else not mentioned can be considred the same as the test before.
The results were interesting I guess... some good, some bad, and some ugly.
First the ugly: I have commented on this before, but I think it should be said again If there is anything that PD could not have POSSIBLY screwed up more, its the behavior of this FWD vehicle when wheel spin occurs. I'm not sure if all FWD vehicles in the game are like this, or the prelude is just goofed up, but its really REALLY bad.
In the real world, FWD cars exibit understeer when throttle is applied, because, well, obviously, the tire only has so much traction to provide, and when you use some of that for accelerating the car, less is available for lateral grip. This problem is directly related to power output - the more horsepower, or the lower gear used, the more understeer will occur when the throttle is applied. If enough power is sent to the wheels to cause wheel spin, the cars path normally continues in whatever direction it is going - i.e. total loss of grip until power is reduced.
Well! I have found that in a slow corner, in 2nd gear, if it appears that I am a bit too hot and am going to understeer off the outside of the track, I simply have to MASH ON THE GAS. The front tires break loose in wheel spin, and yet somehow the front tucks down toward the apex, the backend slides out... if the steering angle is maintained, the car gets completely sideways, and looses a ton of speed nearly instantly. The result is that as a balanced FWD car is exiting a slow corner, when full throttle is applied in a low gear, you must COUNTERSTEER TO PREVENT OVERSTEER. WTF.
While traction is maintained, it behaves as normal... additional throttle results in less grip up front - but when power overcomes available traction, out comes the back end. So this means as you exit a corner at full power, you will have UNDERSTEER in third gear, and OVERSTEER in 2nd gear. This is ass-backwards stupid.
anyway, I had to keep this in mind as I performed my tests.
With the intial settings, body roll and pitch were very noticable, but not more then stock. This makes sense, as the lowest spring settings are ever so slightly more then a stock prelude spring. It would also suggest that a sway bar setting of "1" isn't exactly like taking the sway bar off, nor is it any noticable amount less then stock.
The car was well balanced with these settings, and though turn in was a bit slow, and it took a second for the car to "set itself" in the turn so to speak... once in steady state cornering the car was, for the most part neutral with maybe a hint of understeer. The car could be easily steered with the throttle - lifting slightly to tighten the line, or accelerating to widen it... unless the magic wheel spin occured...
There was no evidence that the car was bottoming out. Body roll in replay seemed to show the tires were not too far up in the wheel well, but that doesn't really matter because I've seen plenty of screenshots, both in photodrive and phototravel, of front tires poking through fenders.
1st change:
SPRINGS 3.4 / 4.1
Same neutrality in corners, for the most part. perhaps a SLIGHT bit more oversteer, possibily wishful thinking.
2nd change:
SPRINGS 4.0 / 5.5
I am pretty sure I noticed more rotation, and less understeer. Body roll and pitch is reduced by QUITE a bit.
This seems to suggest that the numbers mean nothnig, and if you hang out in the lower end of the spring rates, you can make the car handle like you want, and standard tuning methods are effective.
I went through a few more spring rate changes, always keeping the springs under 6kg/mm. Car seemed to behave somewhat as expected, as long as you are not expecting the numbers to be accurate.
for my next 3 changes, I played with the sway bars.
As far as balance is concerned, they don't do crap. At all.
I then played with shocks.
upping the rear shocks to 2/5 resulted in more rotation in transitions, but it SEEMED like when I got back on the throttle, there was MORE understeer. When I went back to 1/1, it still felt the same, so maybe I imagined it.
The higher bound & rebound definately resulted in more rotation in turn in, and when lifting off the throttle, so I guess thats a good sign.
I had much more extensive notes at home, but perhaps someone can build off of these results.
- 24,344
- Midlantic Area
- GTP_Duke
I have read through this post, and I have two comments/questions:
1) In a light, well-powered FWD car, turning hard into the turn and mashing the throttle does work as a last-ditch effort to drag the nose of the car around, particularly on sticky tires (I assume the stock S2s) in a low speed corner. But you've got to have decent power, otherwise you are totally correct.
Watching the SpeedWorld Challenge cars at Sebring last Sunday, I saw this happen on more than a few occasions.
But driving the MINI One around Tsukuba in the license test, there is nothing you can do, no way, no how, to make that car do anything but understeer.
2) I'm no JDM expert... but didn't the Prelude you're using have four wheel steering? That's going to fundamentally affect its perceived behaviour as a FWD car.
1) In a light, well-powered FWD car, turning hard into the turn and mashing the throttle does work as a last-ditch effort to drag the nose of the car around, particularly on sticky tires (I assume the stock S2s) in a low speed corner. But you've got to have decent power, otherwise you are totally correct.
Watching the SpeedWorld Challenge cars at Sebring last Sunday, I saw this happen on more than a few occasions.
But driving the MINI One around Tsukuba in the license test, there is nothing you can do, no way, no how, to make that car do anything but understeer.
2) I'm no JDM expert... but didn't the Prelude you're using have four wheel steering? That's going to fundamentally affect its perceived behaviour as a FWD car.
///M-Spec
Staff Emeritus
- 4,928
Maturin
News flash:
-The last 'conversation' we had, I rebuffed many of your claims in a long and detailed post. Your only response was to say you were leaving GTP. I guess we're not so lucky you actually followed through on your 'threat'.
-I am testing with a group of OLR'ers in a different forum. We are a thorough group and don't jump to conclusions after 15 minutes of testing, so it takes time to learn and exchange ideas.
-Lastly, I've been on vacation for week and then caring for a sick child upon return.
Unlike yourself, I'm not so shockingly insecure that I have to come in here and attack everyone who disagrees with my findings. There are some good posts in this thread and there are some good posters in this thread. If I have any reason to disagree, I will take my time and do my homework, not come in here half-cocked looking for a wee-wee comparing contest.
I'd like to offer quick thanks tuff240, Bad760, aarque and Greyout for their posts, as well as the always helpful sukerkin. This would be far less interesting a thread without their input.
M
- 24,344
- Midlantic Area
- GTP_Duke
See, I disagree. Real, decent FWD cars tend to be more stable through repeated transitions (such as a slalom) and usually can be driven with trail braking / throttle understeer to get a good four-wheel drift going. In SCCA club racing it comes down to whether it rains or not as to whether the Miatas beat the Sis and Neons or vice versa. I think that if anything the understeer is exaggerated in GT4; at least with the low-power cars used on the license tests. Maybe I was just traumatized by those first few license tests, but as a person who whales a FWD car around every day, the understeer is almost hyperbolic in some cases.
- 2,063
Enlightening post Greyout; I'm glad to see you're still on-side and investigating hard. Some of your results are heartening too 👍
Just to back up Duke tho', flooring the throttle is a normal racing technique in FF cars to get the nose to bite and the car to stabalise in a 'sideways' situation (the BTCC drivers do it all the time if their car gets out of shape).
Generally tho', I don't think anyone'll disagree with your take on the traction budget for FF's. In the driving I've done so far in the game, I too have noticed that they've 'cured' the problems that FF and 4WD cars used to have with chronic understeer (once you get above a certain minmal power level (i.e. not in a Mini on glass tyres
)). Whether they've gone too far I don't know as yet - all I can say is that my S4 is a joy to drive round El Capitan right now :grins from ear to ear:.
For FR/Mr cars, they've robbed us of some power oversteer on the whole, from what I can see. There's still enough for my TVR's tho' so that's fine. The braking plough understeer is monumental however
. I find that I have to progress the brakes on to hard and suddenly release so I can get some steering back (the rebounding chassis helps with weight transfer oversteer when the brakes come off .. thank God).
Sorry for the rambling - bordering on off-topic there :blush:. The braking issue and its 'realism' could perhaps be grist for our Enquiry Mill at a later date.
I'm still hoping that the past five years I've spent on learning how to tune and generating suspension model equations (generally failing due to lack of data
) haven't been wasted ones :everything firnly crossed:.
One thing to add that may help the study. Try calculating out a 1Hz front suspension frequency for the test vehicle and see how it behaves as you vary the rear frequency up from 1Hz. I don't know how the math will work out in GT4 (GT3 often wouldn't let you get the rear springs stiff enough for the desired 1.3Hz) but if it turns out that 1Hz each end gives reasonable handling then we have a starting point for further investigation.
Just to back up Duke tho', flooring the throttle is a normal racing technique in FF cars to get the nose to bite and the car to stabalise in a 'sideways' situation (the BTCC drivers do it all the time if their car gets out of shape).
Generally tho', I don't think anyone'll disagree with your take on the traction budget for FF's. In the driving I've done so far in the game, I too have noticed that they've 'cured' the problems that FF and 4WD cars used to have with chronic understeer (once you get above a certain minmal power level (i.e. not in a Mini on glass tyres
)). Whether they've gone too far I don't know as yet - all I can say is that my S4 is a joy to drive round El Capitan right now :grins from ear to ear:.For FR/Mr cars, they've robbed us of some power oversteer on the whole, from what I can see. There's still enough for my TVR's tho' so that's fine. The braking plough understeer is monumental however
. I find that I have to progress the brakes on to hard and suddenly release so I can get some steering back (the rebounding chassis helps with weight transfer oversteer when the brakes come off .. thank God).Sorry for the rambling - bordering on off-topic there :blush:. The braking issue and its 'realism' could perhaps be grist for our Enquiry Mill at a later date.
I'm still hoping that the past five years I've spent on learning how to tune and generating suspension model equations (generally failing due to lack of data
) haven't been wasted ones :everything firnly crossed:.One thing to add that may help the study. Try calculating out a 1Hz front suspension frequency for the test vehicle and see how it behaves as you vary the rear frequency up from 1Hz. I don't know how the math will work out in GT4 (GT3 often wouldn't let you get the rear springs stiff enough for the desired 1.3Hz) but if it turns out that 1Hz each end gives reasonable handling then we have a starting point for further investigation.
On the FWD topic, I was doing the Italian series last night using 2 different Alfas, these cars being totally stock as per the rules except for using S3 tyres, both with 250hp and no driver aids.
If I could keep the front wheels stuck then I could definitely get the car to turn in better by winding on more lock and nailing the gas.
Coming out of relatively slow 2nd gear corners and hairpins as soon as the car lit one or both of the front tyres up I lost a lot of lateral grip and the car would just understeer off the track unless I backed off the gas.
Now that is different behavior to my worked Focus, it has been a while since I have driven it, but I am sure I experienced what Greyout did, I could just turn the front wheels and hit the gas and it would pull the front around almost all of the time no matter if the front wheels were gripping or wheel spinning.
I do run the Focus with TC on 2 so this could possible be effecting it (330hp and FWD are not good bedfellows).
I will test the Alfa and the Focus back to back tonight to confirm what I have just said.
If I could keep the front wheels stuck then I could definitely get the car to turn in better by winding on more lock and nailing the gas.
Coming out of relatively slow 2nd gear corners and hairpins as soon as the car lit one or both of the front tyres up I lost a lot of lateral grip and the car would just understeer off the track unless I backed off the gas.
Now that is different behavior to my worked Focus, it has been a while since I have driven it, but I am sure I experienced what Greyout did, I could just turn the front wheels and hit the gas and it would pull the front around almost all of the time no matter if the front wheels were gripping or wheel spinning.
I do run the Focus with TC on 2 so this could possible be effecting it (330hp and FWD are not good bedfellows).
I will test the Alfa and the Focus back to back tonight to confirm what I have just said.
- 2,063
"Hertz me noggin" indeed. Most humerous, Aarque ... I believe that you'll find under the GTP charter that puns are my balliwick ... hands off!
.What I'm on about in my above post is a fairly simplified method I devised for being able to calculate the suspension frequency for a vehicle in GT3.
You need to know the weight distribution of the car to make it work but it does allow you to either pick a frequency and it'll give you the necessary spring rates or you can pick the spring rates and it'll tell you what frequency that gives.
I'll post it up tonight if anyones interested? Bear in mind that, given the things we don't have values for in-game, it's only an approximation but it seems to work quite well.
Oh ... and it's "Einsteinian" or "Einsteinean" by the way
j/k
- 13,827
- Down under
I havent read through this thread, but have the gist of what the debate is. He is my bit of experience posted elsewhere that may be of interest to certain sections:
I havent meddled too much with regulsr FF or FR cars, but a RR car (as the alpine is) is pretty extreme to test spring rates. Adjustments appear to follow logic in this scenario anyway.
I havent meddled too much with regulsr FF or FR cars, but a RR car (as the alpine is) is pretty extreme to test spring rates. Adjustments appear to follow logic in this scenario anyway.
- 466
I am very interested in any formulas or algorithms people have devised or discovered. It is my belief that polyfunny carefully mapped only one car in stock form and then the various stages to fully modified. I'm guessing it would be a Supra because of all the blueprinty sketches you can see in almost every GT tuning screen. They then quantified these values (into polygrams?) and developed a formula where they could get a little loose data; skid pad, dyno, a bit of a radar tickle, and then BAM- as that culinary virtuoso Emiliosukerkin
would say- they'd have another car in their virtual pocket.If you think about it for just 5 seconds, there is no way they went to Lotus and bolted on a spoiler; or convinced somebody to let them supercharge his Chevelle. Of course they could check their models against known performance and adjust accordingly. They even got so confident, they extrapolated it back to 1886, who alive would know the difference anyway; and into Nike's future, "building" literal dream cars. So....
If we discover the formulas by which the laws of GT4 physics are governed, we can more accurately tune our cars for the characteristics we desire.
Just to chime in here after aarque's last post (with which i agree 100%) with my opinion on how the game was put together... keep in mind this is all speculation, since i don't know exactly how they did it; however i know how its generally done, if that makes sense. So you have your original car physics model, based on real world specs etc. from which all other cars are born. This is very likely what they did, with the possible exception of creating a few basic physics models for different classes of cars; say LMP's, Prototypes, and F1 in the first class, GT cars and other race models in class 2, and production models in the last. Or possibly models based on drivetrain. From here they extrapolate all the cars in the game tweaking here and there. Now most of this so far is based on code, we havent gotten into much 3D yet. I'm pretty sure that about 90% of the graphics engine of GT4 is based on GT3 code, and its very likely that a good portion of the physics were based on GT3 as well so they didn't start from scratch by any means.
hmmm i kind of typed all that and forgot what my point was... well i guess i'll just get back to racing the nsx-r lm... i swear i was getting somewhere with that, if i remember i'll post again hahaha =D
-Shmak
hmmm i kind of typed all that and forgot what my point was... well i guess i'll just get back to racing the nsx-r lm... i swear i was getting somewhere with that, if i remember i'll post again hahaha =D
-Shmak
im more than a bit befuddled...
for all the effort that has gone into this thread...id assume you technophiles might invest in a PC and a *real* auto sim.
GTs main problem is that it takes no account of actual suspension physics or dynamics. (thats the big friggin problem)
also, you cant just mess with spring rates or dampers singularly and call a physics engine completely crap...especially when you dun know anything about the actual linkages, and the game engine's compromises to afford so many cars.
besides. if the system was so crap - is the game no fun anymore? would you stop posting here? i dunno, but the gt series has been as much about f'n around and figuring out your own driving style, that analyzing every minute detail. if thats what youre into check Grand Prix Legends as thats the only real, real sim ive ever driven. fun as **** tho|
(case in point: people saying the Triumph is a decent car, when the prod model came with positive camber on the rears -more than a little dangerous-...but not in the game. and with is swing rear axle, has a problem with 360s on straights, not in the corners)
for all the effort that has gone into this thread...id assume you technophiles might invest in a PC and a *real* auto sim.
GTs main problem is that it takes no account of actual suspension physics or dynamics. (thats the big friggin problem)
also, you cant just mess with spring rates or dampers singularly and call a physics engine completely crap...especially when you dun know anything about the actual linkages, and the game engine's compromises to afford so many cars.
besides. if the system was so crap - is the game no fun anymore? would you stop posting here? i dunno, but the gt series has been as much about f'n around and figuring out your own driving style, that analyzing every minute detail. if thats what youre into check Grand Prix Legends as thats the only real, real sim ive ever driven. fun as **** tho|
(case in point: people saying the Triumph is a decent car, when the prod model came with positive camber on the rears -more than a little dangerous-...but not in the game. and with is swing rear axle, has a problem with 360s on straights, not in the corners)
- 2,063
Hi Biz
I'm not quite sure what you meant to achieve with your post (other than mildly ridicule our somewhat obsessive intent to pick apart the shroud of mystery around the physics engine of GT4
).
There are in fact those of us who do exactly as you propose and have PC driving simulators (or, in my particularly sad case, proper suspension modelling software
).
The whole point of this thread is to figure out how the physics engine ticks (warning: mixed metaphors alert!) and is not solely about just twiddling a spring or a damper in isolation. However, until we know how the individual components behave we can't postulate a more coherent and wholistic view.
People have their own ways of extracting fun from any game and tinkering with the hardware side of things happens to be our way of playing GT4.
In short, leave us be and we promise not too make too much noise.
I'm not quite sure what you meant to achieve with your post (other than mildly ridicule our somewhat obsessive intent to pick apart the shroud of mystery around the physics engine of GT4
).There are in fact those of us who do exactly as you propose and have PC driving simulators (or, in my particularly sad case, proper suspension modelling software
).The whole point of this thread is to figure out how the physics engine ticks (warning: mixed metaphors alert!) and is not solely about just twiddling a spring or a damper in isolation. However, until we know how the individual components behave we can't postulate a more coherent and wholistic view.
People have their own ways of extracting fun from any game and tinkering with the hardware side of things happens to be our way of playing GT4.
In short, leave us be and we promise not too make too much noise
.Heres the thing about "ground up re-writes". Companies who build software programs, games, and the like, do this stuff all the time. They say, "Hey! Our stuff is great but we are going to make it even better by starting over and rewritting everything from the ground up!" While im not going to say the PD is lying exactly, they are bending the truth. Heres how im going prove this: If you compare the GT3 credits to the GT4 credits (which i have just done) you will see that the catagories in question (graphic optimization, system optimization, data optimization, simulation algorithim, visual program, etc, etc, etc) retained the same programmers for both titles. Now heres my thinking. Would a programmer who spent years writing code that works very well simply throw all that out the window? Most likely not. Writting code is pretty much an art form, and a good programmer can look at a program and recognise who wrote it if they have a grasp on that person's style. Organization, clarity, and speed are highly valued things. So one may reorganize the code, read through all of it again and refine it, make it faster and more concise, but hardly ever start over completely, because the basic commands on the lowest level remain pretty much the same, and starting over from scratch is pretty much like rewritting an old book with the same plot, characters, and setting, but changing the prose, sentence structure, and eliminating unnecessary statements so that the climax is reached in less pages. So they say its a "whole new thing" but its largely based on the old version. Its the same way with the cars and the tracks. GT1: 11 tracks, 200 some cars. GT2: 27 tracks, 500 some cars. Since these 2 games were on the same console and were rather similar its reasonable to assume that since they had the tracks and cars from GT1 when developing GT2 and simply modified them a bit. Same goes for GT3 and GT4. They had to rework most of the game for the PS2 but even then stuff carried from the first titles. GT3: 20 tracks, 200 some cars. GT4: 52 tracks and 600 some cars. And as for simply playing GT3 and GT4 games back to back, i have to say that if by some horrible motivation the programmers did throw everything out the window and start over COMPLETELY, then there had to be alot of wasted time rewritting the same statements since both games definately look and feel quite similar no doubt about it.
-Shmak
*edit* i hope im not too far off topic here... i just posted and now im thinking i rambled a bit heh. oh well =p
-Shmak
*edit* i hope im not too far off topic here... i just posted and now im thinking i rambled a bit heh. oh well =p
- 2,063
But other than the lack of paragraphing, a pretty good post I reckon.
It says pretty much what I think too from the professional side of things (I have some knowledge of the software engineering trade {cough, cough, National Grid Substation Control Systems cough}).
Anyhow, on to those very rough approximation calculations I talked of earlier. These are by no means what the engineer in me would call accurate but they do seem to function in giving reasonable Spring Rates for non-race bodied cars in GT3.
____________________________________________________________________
To get the Spring Rate for a desired frequency you need the car weight in kilograms and the Weight Distribution (as a decimal fraction). The latter can usually be found by research but in a pinch take the ratio of the default springs as a guide.
With the Car Weight and Weight Distribution calculate the Quarter Weights for Front and Rear i.e. the total weight multiplied by the Weight Distribution all divided by two:
(CW * WD)/2
Find the product of (2*PI*Desired Frequency) and square that result.
Finally multiply by the Quarter Weight in Kg and divide by 1000.
To determine what Frequency a given Spring Rate will give you take one divided by two times PI and multiply by the Square Root of the Spring Rate times 1000 divided by the Quarter Weight:
(1/(2*PI)) * SQRT ((Spring Rate * 1000)/Quarter Weight)
_____________________________________________________________________
Not exactly complicated (tho' getting there was a long journey I have to say given the 'plaiting fog in the dark scarcity of really useful car data in the GT series).
If they work for people that's a bonus. If not, feel free to improve, spindle, fold or otherwise mutilate the equations. But above all, I'd appreciate intelligent input into how to make them qualatively better (no flaming if you don't like them tho', I'm sensitive and may cry).
It says pretty much what I think too from the professional side of things (I have some knowledge of the software engineering trade {cough, cough, National Grid Substation Control Systems cough
}).Anyhow, on to those very rough approximation calculations I talked of earlier. These are by no means what the engineer in me would call accurate but they do seem to function in giving reasonable Spring Rates for non-race bodied cars in GT3.
____________________________________________________________________
To get the Spring Rate for a desired frequency you need the car weight in kilograms and the Weight Distribution (as a decimal fraction). The latter can usually be found by research but in a pinch take the ratio of the default springs as a guide.
With the Car Weight and Weight Distribution calculate the Quarter Weights for Front and Rear i.e. the total weight multiplied by the Weight Distribution all divided by two:
(CW * WD)/2
Find the product of (2*PI*Desired Frequency) and square that result.
Finally multiply by the Quarter Weight in Kg and divide by 1000.
To determine what Frequency a given Spring Rate will give you take one divided by two times PI and multiply by the Square Root of the Spring Rate times 1000 divided by the Quarter Weight:
(1/(2*PI)) * SQRT ((Spring Rate * 1000)/Quarter Weight)
_____________________________________________________________________
Not exactly complicated (tho' getting there was a long journey I have to say given the 'plaiting fog in the dark scarcity of really useful car data in the GT series).
If they work for people that's a bonus. If not, feel free to improve, spindle, fold or otherwise mutilate the equations. But above all, I'd appreciate intelligent input into how to make them qualatively better (no flaming if you don't like them tho', I'm sensitive and may cry
).
- 466
Sukerkin, I am fascinated by your equation. Please, if you care to, might you explain how a constant that relates a circles' circumference to its radius figured into your spring rate calculation. It is more like e=mc squared than I had guessed, I would never have expected a constant. Come to think of it, why didn't you use c, everyone knows it relates to speed...sukerkin
Hi Sukerkin,
Just have a question re your cpm calculation. My understanding is that the calculation is
Fn = 187.8 / SQRT (SD)
Where
Fn = Natural Frequency in cpm
SD = Static Deflection in inches
Can't remember my original source (as it was a number of years ago) but try here http://www.rqriley.com/suspensn.htm
The formula I orginally found was
Fn = 187.8 / SQRT ( Load / k )
Where
Fn = Natural Frequency in cpm (use 3.13 instead of 187.8 for Hz, or divide the result by 60)
k = spring stiffeness in lbs/in (for vechiles use Wheel Rate instead)
Load = Weight carried by spring in lbs
If we assume that Wheel Rate is equal to Spring Rate, and as I calculate cpm for Front and Rear seperatly, then a more practical formula is
Fn = 187.8 / SQRT ( WoW / ( SR * 2 ))
Where
Fn = Natural Frequency in cpm
SR = Spring Rate (x2 to take both springs into consideration) in lbs/in
WoW = Weight over Wheels in lbs
The problem I'm having is that although your formula appears to be the calculation that is used to calculate the Resonant Frequency (Natural Frequency) of a Spring Weight system, my limited understanding from what I've read on the web so far is that 1 / 2 * pi defines an oscillating input at the system resonant frequency.
So now the real question...
Is your formula the correct one to apply, instead of the one I've been using?
Any insight into both/either formula/s would be appreciated.
Just have a question re your cpm calculation. My understanding is that the calculation is
Fn = 187.8 / SQRT (SD)
Where
Fn = Natural Frequency in cpm
SD = Static Deflection in inches
Can't remember my original source (as it was a number of years ago) but try here http://www.rqriley.com/suspensn.htm
The formula I orginally found was
Fn = 187.8 / SQRT ( Load / k )
Where
Fn = Natural Frequency in cpm (use 3.13 instead of 187.8 for Hz, or divide the result by 60)
k = spring stiffeness in lbs/in (for vechiles use Wheel Rate instead)
Load = Weight carried by spring in lbs
If we assume that Wheel Rate is equal to Spring Rate, and as I calculate cpm for Front and Rear seperatly, then a more practical formula is
Fn = 187.8 / SQRT ( WoW / ( SR * 2 ))
Where
Fn = Natural Frequency in cpm
SR = Spring Rate (x2 to take both springs into consideration) in lbs/in
WoW = Weight over Wheels in lbs
The problem I'm having is that although your formula appears to be the calculation that is used to calculate the Resonant Frequency (Natural Frequency) of a Spring Weight system, my limited understanding from what I've read on the web so far is that 1 / 2 * pi defines an oscillating input at the system resonant frequency.
So now the real question...
Is your formula the correct one to apply, instead of the one I've been using?
Any insight into both/either formula/s would be appreciated.
- 466
Excuse me for grabbing the mike (nods to Sukerkin), it appears, at this point to the lay observer I am, that these frequency calculations, at least based on the link provided by Kiljoy, are concerned with predicting pAsSeNgEr cOmFoRt.Kiljoy
I took the liberty of browsing to Robert Riley's website, rife with plans for recumbant bicycles and "Urbacars," (din see any F1's lurking in the shadows); but no such calculations or formulas as listed above. Very informative data on basic systems and their functions, as relates to light weight urban car design, the "Rollover Caused by Tripping" chapter triggered a cloying nostalgia for an especially trippable '67 Barracuda, yet the"taughtest" or highest performance system considered was that in a sportscar. Would not the extension of these calculations, assuming you wanted to establish a frequency level equavalent to the "comfort" a race driver would want, provide solely more stability without regard to other characteristics? Also you seem to imply PI has something to do with wheel rotation as a frequency, does speed somehow cancel the diameter leaving half PI standing there?Hi Aarque,
Before jumping to conclusions maby you should take the time to read a little more carefully and, take the time to read up on cpm in relation to suspension tuning.
You will find the reference to the formula at the end of the section on "Ride comfort", just before the heading "Implications of High Payload-to-Vehicle Weight Ratio".
Also may want to check out this http://www.smithees-racetech.com.au/performancefactors.html , I think Sukerkin provided a link to this website.
Edit link didn't post correctly
http://www.smithees-racetech.com.au/performancefactors.html
Before jumping to conclusions maby you should take the time to read a little more carefully and, take the time to read up on cpm in relation to suspension tuning.
You will find the reference to the formula at the end of the section on "Ride comfort", just before the heading "Implications of High Payload-to-Vehicle Weight Ratio".
Also may want to check out this http://www.smithees-racetech.com.au/performancefactors.html , I think Sukerkin provided a link to this website.
Edit link didn't post correctly
http://www.smithees-racetech.com.au/performancefactors.html
- 466
Kiljoy, thank you so much for trying to clarify, surely Heaven holds a star for one such as you with the patience of a saint. I am sorry to seem so
.
`````` 188
NF = ----------
```` /---------
``` V SD
NF = Natural Frequency in Cycles Per Minute (divided by 60=Hz)
SD = Static Deflection in Inches
Which is indeed intended to be used for predicting passenger comfort. No refrence to PI or its implication, I am sure you can provide the connection. It is in resemblance as an equation only to the formula posted by Sukerkin:
(1/(2*PI)) * SQRT ((Spring Rate * 1000)/Quarter Weight)
The Smithee's page appears to be devoted to providing a formula by which a user can predict a suspensions STIFFNESS- the only characteristic considered- by calculating it's resonant frequency. Sadly, this formula does not apply to leaf springs or "large diameter coil springs working on a poor motion ratio." Succulent facts I am confident the engineers took into account when coding GT4. MY $.02, if I could go so far as hazard a guess why Sukerkin linked the Smithee's page, is the information summed in the final paragraph:
If the suspension and steering is (are?) mechanically sound, the greatest gains in performance will come from optimising weight transfer, ahead of any other suspension modifications.*
Another fact we can all rest confident Polyfony hard wired into the game
. the only equation on the page you listed is this:Kiljoy
`````` 188
NF = ----------
```` /---------
``` V SD
NF = Natural Frequency in Cycles Per Minute (divided by 60=Hz)
SD = Static Deflection in Inches
Which is indeed intended to be used for predicting passenger comfort. No refrence to PI or its implication, I am sure you can provide the connection. It is in resemblance as an equation only to the formula posted by Sukerkin:
(1/(2*PI)) * SQRT ((Spring Rate * 1000)/Quarter Weight)
The Smithee's page appears to be devoted to providing a formula by which a user can predict a suspensions STIFFNESS- the only characteristic considered- by calculating it's resonant frequency. Sadly, this formula does not apply to leaf springs or "large diameter coil springs working on a poor motion ratio." Succulent facts I am confident the engineers took into account when coding GT4. MY $.02, if I could go so far as hazard a guess why Sukerkin linked the Smithee's page, is the information summed in the final paragraph:
If the suspension and steering is (are?) mechanically sound, the greatest gains in performance will come from optimising weight transfer, ahead of any other suspension modifications.*
Another fact we can all rest confident Polyfony hard wired into the game
