The engine repeatedly stalls when I shift into gear!

[RJARRRPCGP]

The engine will repeatedly stop when I shift then press on the accelerator button! Even when I held down on the clutch button!

The workaround is to use the automatic clutch.

It appears that the manual clutch is broken.

I also noticed that the engine won't start again until I reload the game!

[thelusiv]

Try this: Rev the car up as high as it will go in neutral (N). Shift up and drop the clutch. You should get a smoky burnout and forward movement. Now lower the RPM where you drop the clutch until you get a reasonably controlled launch.

As for the engine not starting, the engine must be in neutral (N) to start. Shift the car back down to N and then press the start engine button/key.

All the autoclutch does is tell the "driver" to push the clutch in whenever you are about to stall the car.

The cars in the game should behave more or less just like a real manual transmission automobile. Some have stronger clutches than others, making it a little hard to get them going sometimes, because the clutch stalls the car...

[RJARRRPCGP]

That worked.

The clutch seems to be unrealistically dufficult!

[thelusiv]

Moved to Help...

Glad that worked for you. Yes the clutch is rather stiff on many cars, but that's good, for performance. Helps a little with drifting, makes it easier to break the wheels loose on a downshift. I think the XS clutch is still pretty unrealistic though, and maybe one or two other cars.

[RJARRRPCGP]

You can avoid this problem by modifying the engine torque to a higher value.

It can be caused by the engine torque being insufficient.

[thelusiv]

Well, most of our cars' torque curves are based on real torque curves for real cars similar to the ones in the game. So rather than change the torque, which I think is right, it would be better to adjust down the clutch force, to a more realistic value. The problem is, it's a lot harder to find out how much clutch force a real car has on its clutch plate, than it is to figure out the torque curve for a given car. So we have to make an educated guess at the clutch force, and play with the value until it works like we want it to.

In fact, before the last release (2006-10-06), I adjusted the clutch force on the XS to make it easier to launch, and less likely to stall. It still may not be right, but it's better than before...

If you find problems with certain cars, please tell us which ones need work, and we'll try to update those with better clutch force.

[cotharyus]

I've tested all the cars included in the 2006-10-06 release and not had any problems with clutches. Now there is one there that doesn't have any kind of final drive at all - tops out at about 30 mph in 5th gear....but no clutch problems.

[thelusiv]

I've tested all the cars included in the 2006-10-06 release and not had any problems with clutches. Now there is one there that doesn't have any kind of final drive at all - tops out at about 30 mph in 5th gear....but no clutch problems.

Thanks for testing them, I appreciate it. I try to choose a different car each time I do some play testing but with 21 cars it's hard to keep up with them all. Plus, I really enjoy certain ones and it's hard not to choose the T73 every time

The car you're referring to as having no final drive is the FE (Ford Escort MKII Rally)...we were talking about this car in IRC yesterday I think...I think it has problems both with its gear ratios/final drive, and with its torque curve (because it accelerates from 0 to 7000 RPM very fast...but maybe that's because the transmission's not putting much load on it). I'll see if I can fix this by the time we do another release.

On that note can anyone find me some specs on the Ford Escort MKII Rally? I'd like to see a torque curve from a real dyno pull, a nice graph would be best. The manufacturer's spec sheet will probably do for the gear ratios and final drive.

[cotharyus]

No problem. I enjoy testing all the cars - running the laps looking for consistancy and times, I also am actually trying to see if they react differently on different tracks.

I will do some research and see if I can find specs on the car you're talking about - I have conections here and there that may be able to provide these things. I suspect that the perceived torque issue is related to the gear ratios, but we'll have to get some good ratios in there to figure out for sure.

[reece146]

My four year old picked this car (FE) to drive tonight.

The gear ratios on it are fubar'ed, hence the "bizarre" performance even with auto-clutch.

Here is what the FE.car has in the transmission section:

Code:

[ transmission ]
gears = 5
gear-ratio-r = -3.446
gear-ratio-1 = 11.455
gear-ratio-2 = 6.788
gear-ratio-3 = 5.207
gear-ratio-4 = 4.384
gear-ratio-5 = 3.818
shift-delay = 0.4

I wasn't able to find valid tranmission ratios for this car on the 'weeb. Most tranmissions have fourth be a direct drive (1:1). So, if we divide all the ratios (except reverse) by the fourth gear we get:

Code:

[ transmission ]
gears = 5
gear-ratio-r = -3.446
gear-ratio-1 = 2.613
gear-ratio-2 = 1.548
gear-ratio-3 = 1.188
gear-ratio-4 = 1.000
gear-ratio-5 = 0.870
shift-delay = 0.4

Which are pretty real world values for a five speed transmission. Second, third and fourth are really close to one another but this may be a rallye config for a specific track (speculation).

Most tranmissions have first and reverse be the same - or within a few hundredths as well. That would make 1st gear 3.446 and proogate all the way down but would screw up the 4th gear ratio. If you have ever had a "normal" four or five speed longitudinal trnamission apart you know why fourth is typically 1:1 (direct drive).

Anyway, using the ratios above the car is much more drivable. The first gear feels tall. The drivetrain runs out of steam in fourth gear. Whether it is correct, I dunno. Who made this model for vdrift? Maybe they can comment?[/code]

[reece146]

the torque-curve values for the FE are also screwed up. "torque-curve-11" repeats six times... Looking a tthe values I'm guessing the last value is supposed to be #16.

The gear ratios I suggested above are definitely wrong... too tall.

[joevenzon]

Take a look:

http://svn.vdrift.net/viewvc.py/trunk/da...r?view=log

seems like R1098 added a bunch of fubar values:

http://svn.vdrift.net/viewvc.py/trunk/da...27&r2=1098

we should probably just revert to the R1027 version of FE.car....

[thelusiv]

I've found some better info on old Escorts and I've updated the FE. Check this page out:
http://www.dooyoo.co.uk/cars/ford-escort...l/1020005/

It lists gear ratios for the MkI Escorts, I used the values from the "Type E" transmission. I'm guessing the MkII Escorts used this one since this was the last MkI transmission made, and the article seems to imply that what it doesn't mention about the MkII was the same as the MkI. Also fixed the torque curve by changing all the 11s to the right numbers.

[reece146]

I've found some better info on old Escorts and I've updated the FE. Check this page out:
http://www.dooyoo.co.uk/cars/ford-escort...l/1020005/

Damn, I saw that page last night and glossed right over the tranmission ratios. I figure use those ratios until someone that knows better comes along. I don't know Brit cars that well.

That FE does handle nice though. Fun little car.