Powertrain Feedback Collection

The topic/issue to bring up is...
some issues with the DCT gearboxes that sometimes tend to behave like manuals plus a certain "manual transmission" bug that affects all types of automatics and two ones that affect the Torque Converter transmissions of the modern ETKs (plus some more I discovered later and edited my post to type them - one for the modern ETK manuals and one for the modern ETK 8-Speed DCT).
6 issues in total (8 issues with the edits). Dunno if they also happen for real, but I think they should certainly be corrected.

1st issue: the DCT gearboxes let off the gas pedal when downshifting.

This happens when...
the car is cruising at a steady speed (say with the Cruise Control set at 100km/h or 62mph) and then the player wants to accelerate by pushing the gas pedal to the fullest (or by pushing the '^Up' button, if they play through a keyboard, like my case). In order for the DCT gearboxes to find the most suitable gear for quick acceleration, they let off the accelerator pedal every time they downshift, just like what manual gearboxes do when the clutch pedal is being pushed.

Instead it should...
just downshift till it gets in the most suitable gear without letting off the gas and accelerate linearly, again without letting off the gas, because valuable time is wasted in this phase. After all, DCTs are known for rapid gearshifts and for predicting what the driver is gonna do by having the nearby gears already selected!



2nd issue: the DCT transmissions push the gas pedal when downshifting during braking to a halt or when just losing speed naturally while coasting.

This happens when...
a braking attempt on a dime is being performed or when the car naturally slows down due to the forces of nature (air resistance and friction). The DCT transmissions push the gas pedal in every downshift because they somehow think the engine is gonna stall.

Instead it should...
just downshift without pushing the gas pedal because DCTs are automatics and they shouldn't do such a thing (and because they're famous for predicting what the driver is gonna do by having the nearby gears already selected)!



3rd issue: the DCT transmissions forget to downshift till 1st gear.

This happens when...
the car is cruising at a steady speed, without pressure, on top gear (say with the Cruise Control set at 100km/h or 62mph) and then a stopping attempt on a dime is performed. The car slows down without the DCT gearbox downshifting all its gears till the 1st one so, by the time the car gets stationary, the classic engine judder is heard and the engine feels like it's gonna stall because there's load on it.
In a few words, the DCT transmission feels like it got caught off guard.

Instead it should...
just downshift all its gears till the 1st one before coming to a halt (the truth about predicting the shifts is true in this case too)!



4th issue: all types of automatic transmissions (Torque Converter Automatics, DCTs and the CVT) respond to the 'Left Shift' keyboard button just like a manual transmission would do.

This happens when...
an automatic car is on Cruise Control and is accelerating or cruising at a steady speed (again, say at 100km/h or 62mph) with no throttle control from the player. You press/hold the 'Left Shift' button only to watch the gas pedal being lifted, the engine slowly going back down to idle (with the gearbox shifting either up or down in the meantime or, if you put it in "N", the revs dropping to idle) and the car slowing down.
This only happens when the Cruise Control is on (without any player-induced throttle control) and you're doing the speed set by the Cruise Control or any speed below that threshold.
Also, when the car is on standstill in "D" or "R" and the Cruise Control is on, it remains stationary (except for the transmissions that roll slightly forwards and backwards - then it's natural even within this weird case) as long as the 'Left Shift' button is held and, also, the engine remains on idle and doesn't redline.
On the other hand (and regardless of the Cruise Control being on or off), when you want to accelerate from a standstill or from any speed below the Cruise Control threshold by pushing/holding the '^Up' button along with the 'Left Shift' button, this slowing down doesn't happen in this case. You can accelerate as if the 'Left Shift' button was never pressed/held.

Instead it (they) should...
just not respond to the 'Left Shift' button at all! They're all automatics and there's no manual clutch mechanic to be controlled by the driver! Not directly at least!




5th issue: in the 8-Speed and Diesel 8-Speed Torque Converter Automatics of the modern ETKs (plus the 8-Speed Torque Converter Automatic of the Bastion), the revs rise while accelerating from 50km/h or 31mph on 8th gear at full throttle, reminding me of the CVTs behavior in one way or another.

This happens when...
the car accelerates in top gear at low speeds (put it in "M8", that's the only way this situation can be replicated) and then, at around 75km/h or 47mph, the revs drop to around what the engine would spin at 100km/h/62mph in 8th gear and from then on the engine revs climb up linearly and the car accelerates gradually on 8th gear.

Instead it should...
just accelerate gradually from the bottom of the rev range, without raising the revs!



6th issue: again in the modern ETKs (and in the Bastion), when driving the 8-speed Torque Converter/Diesel 8-Speed Torque Converter Automatics and changing gearbox modes with the 'X' and 'Z' keyboard buttons (not by changing modes through the UI Drive Mode Selector and its "Ctrl + Q" key binding), there are sudden changes in the engine revs.

This happens when...
for example, you drive in "D", in 8th gear, at 100km/h/62mph. The engine spins at around 1400rpm (1400rpm for the 854d/856d, the 854t/856t/ttSport/ttSport+ versions usually spin at around 1700rpm on 8th gear), then, with the 'X' button, you put the TC gearbox in "S8" and, just after that, you notice the gearbox going crazy, downshifting 3 or 4 gears and the revs rising/pinging like the engine was stung by a bee. And this can also happen the other way round, when putting it from "S8" to "D" by pushing the 'Z' button.

Instead it should...
the input by the 'X' and 'Z' buttons should follow the good example of the UI Drive Mode Selector and just not raise/ping the revs at all because it upsets the drivetrain for no good reason!

edit 1 (November 15th 2021) - Just found out something else: issues 1 and 2 also affect the 8-Speed Torque Converter Automatics of the modern ETKs. When accelerating in "D" and upshifting, the gas pedal gets slightly lifted off and when accelerating/decelerating in "S/M" and upshifting/downshifting, the gas pedal gets slightly lifted off/pushed respectively!

edit 2 (December 1st 2021) - Discovered two more bugs, both about Launch Control and its relation with the 6-Speed Manual/Diesel 6-Speed Manual/ttSport 6-Speed Manual and the 8-Speed DCT Automatic Gearboxes of the modern ETKs!
The first one, about the 3 Manuals, is that in the 800-Series they have no Launch Control while in the K-Series they have it and it's functioning properly.
The second one is about the 8-Speed DCT. The 800-Series doesn't have Launch Control while the K-Series has it but it's not functioning properly. When you are stationary and press both the brake and the gas, Launch Control works as it should but when you let off the brake, the car is still stationary (or creeping slightly forwards, sometimes) with the engine revving in Launch Control revs and, after a few seconds, it gets going.

Issues 1, 2, 3 and 4 are also found on the ETK 8-Speed DCT Transmission. The Issue No. 2 is found only in the K-Series when equipped with the 8-Speed DCT. I never came across this issue in the 800-Series with this transmission. But check both the modern ETKs to make sure, just in case.


My specs are...
CPU: Intel B960
GPU: Nvidia GeForce GT620M
RAM: 4GB DDR3
(it's a simple laptop)

 

I agree with everything @McBeamer94 says. The torque converter slip on most regular automatic gearboxes is atrocious, even my dad's 2003 Ford Mondeo doesn't do that. I have also noticed that when driving a gas powered car with the DCT, the shift behavior is far from realistic. A modern DCT will upshift at 2000 rpm to save fuel for example, whereas all DCTs in BeamNG like to ride it out until 5000 in first and then subsequently shifting at lower RPMs once the speed is increased. The Diesel DCTs behave as they should, like their real life counterparts.

 

About the Torque Converter Automatics, it's good that they shift early during calm acceleration (say on cruise control from a standstill) but some of them, just like the gas DCTs you said, think they upshifted too much and then they downshift at least one gear and accelerate more aggressively.
Also, the ETK 8-Speed DCT has a certain problem on Launch Control: you push the gas and brake simultaneously to rev up and build boost but when you let off the brake, the car is still stationary, with the engine revving on Launch Control, and after a few seconds it gets going. Dunno if it's generally a problem with this specific gearbox but I'll test the modern ETKs again. And if it is, I'll edit my post and write this down.

 

Engine thermal/cooling behavior. The engine thermal simulation in BeamNG has always seemed sort of off to me. Some things could just use a small tweak here and there, but other things just actually don't make any sense. I'll start with the issues I find to be more of a problem.

Oil thermal simulation. Mainly, the way many of the vehicles manage it. The aftermarket oil cooler seems almost entirely useless - When I tested it on a track in the 200BX, at most, I'd see 10°C of improvement; even when there's a turbo pumping 30+ PSI of boost into the engine. So, there's no benefit to the oil cooler there... Meanwhile, you can equip it on the I-Series or Sunburst Sport S and the turbos will overheat anyways. Again, it doesn't seem to be of much help in those cases either. Also, in a lot of vehicles, you can track them hard until the fuel runs out and the oil temperature will be completely fine; so it would be nice to see oil thermals updated to put more emphasis on oil cooling - primarily on vehicles with forced induction.
There is also the nitpicky thing of having to choose between a front bumper support or oil cooler in some cars, plus, the lack of a factory/intermediate option (such as external block helical-style coolers) compared to the all-out aftermarket part in the game, but those are smaller issues.

Radiator fan effectiveness and behavior. Like the oil cooler, the radiator fans don't seem to do too much. They're definitely much more effective than the oil coolers, but they don't seem to be as functional as one might expect. Still, the radiator fans in BeamNG seem to be so inefficient, that it's nearly impossible to tell whether they are activated or not. In fact, while testing the Bluebuck (I get that modern shrouded fans and radiators are more effective), the nearly overheating coolant cooled down much faster by simply turning the engine off, rather than letting the radiator fan extract the heat away. While all fans need to be more powerful in general, upgraded radiators should have fans that are even more improved; specifically, heavy duty radiators (high performance rads are included, too, but HD rads compared to standard rads seem to be as useful as oil coolers).
In terms of behavior, all the "modern" radiators in BeamNG seem to act similar to very early and simple electric fan systems. As in: Coolant gets too hot? Fan turns on. The activation temperature is too high, there is no attempt to normalize/average-out the coolant temperature and (specifically on the more modern cars in-game) there seems to be no variation in fan speed to reduce noise and wear (even 2 or 3 levels of speed would be lots, but BeamNG only has on or off).
As for the smaller things that could be added: Fan noise when the fans are spinning and interior warning lights (a CEL or the temperature icon turning red, like on the UI) would be really nice and help make the system seem more integrated into the cars, rather than just something that happens.

Coolant thermostat effectiveness and behavior. Again, just like the previous entries, the coolant thermostat's implementation seems to be no more than "It's there and it does things." Instead of allowing vehicles to warm up faster, the thermostat seems to only be an indication of coolant temperature. You can wait over half an hour for vehicles to warm up in BeamNG, when usually vehicles should take at most 15 minutes to reach operating temperature. When the coolant does start to allow the thermostat to open, it seems to open and close way too quickly. You can throw a coolant thermostat in a pot of boiling water to see that it takes a few seconds to expand, and you can take it out to see it usually takes even longer to close; so the coolant thermostat's behavior in BeamNG of opening and closing almost entirely within a second is just inaccurate. Not only is the thermostat response so fast that it may cause metal fatigue, but it's still surrounded by hot coolant, so it should not close anywhere close to as fast as it does, let alone even as much as it does (this is in relation to how the thermostat opens and closes more with throttle input, rather than coolant temp).
And as for smaller details that would be nice: Having different thermostat temperatures for different vehicles would actually add a lot to differentiate them from each other. Right now, the system seems to be a one-size-fits-all copy and paste - which isn't necessarily a problem - but just including one small detail like that would help a lot in making the vehicles feel more diverse.

I'm sorry if any of this came off as mean or rude, I understand a lot of work has been put into the engine thermal simulation to get it where it is today. And it's actually the most in-depth vehicle thermal simulation of any game - Which is exactly why I made this post. BeamNG is so close to hitting it right on the nose. It would be awesome to have BeamNG be so hyper-realistic that you could learn from it (it already has that in terms of driving dynamics), and with these changes, you might actually be able to. Heck, even including an overheating light on the dash could help people. This might be stretching hope a bit, but having that light in the game and seeing what it eventually leads to could allow people to correlate that in real life (if it ever happens), and maybe even save their wallet. Again, that might be a stretch, but it's coming from a guy who avoided sliding into a ditch one day because I learned in BeamNG to ease off the brakes if they lock up.
Sorry for the long post like always, but it would be really nice to see those changes be implemented, even if it's not exactly how I pointed them out.

 

I legit didn't realize the coolant temperature warning lights didn't come on in game. Good catch.

 

I agree that oil temp feels a relative unimportant statistic as it on many cars it only becomes an issue long after the coolant has boiled away. Which for an halfway modern production car is probably even realistic, Keep in mind that most cars dont even have an oil temperature readout because outside of significant abuse or technical problems, most modern engines are so well designed that they dont run into the issue.

I think you might underestimate just how much heat energy an running internal combustion engine (ICE) creates, especially under load. For example, an 100 KW = 136 HP ICE petrol engine will waste arround 2/3 of the fuels energy as waste heat. So at peak power = 100 KW engine power it will take in fuel worth 300 KW of which 200 KW will be plain heat instead of useful mechanical power.

200 KW or 200000 Watts is an enormous amount of heat energy.... equivalent to 200 1000 Watt electric grills.



A significant amount of that heat leaves the vehicle with the exhaust but what remains is still very significant and can destroy an engine within minutes in case of a failing cooling system.

The ability of 1 or 2 fans to cool down such a heat source is in comparsion relative modest. Which in usual operating conditions isnt that bad of a news because normally a car is driving, deliving thousands of liters of fresh cool air over the radiator. Problems arise when load and speed are not rising at the same time. Like driving uphill with a trailer or heavy load..... relative low speed limits radiator air flow and a labouring engine creates large amount of waste heat that the radiator fans just may be unable to remove in time.

To add insult to injury, classic cars like the Bluebuck dont use an electric fan system but one coupled to the engine directly. Fan speed an engine RPM are directly linked which worsens the problem when heavy load is coupled with relative low engine speed. Such a fan can at best compensate only slightly more than the engine idle heat load because it will spin relative slow at idle. Adding an horrible ineffcient classic V8 to the construction and an automatic transmission which torque converter losses further increase idle load and it is easy to see why shutting down the engine actually improves cooling speed.

The is also a limitation to BeamNG engine thermal simulation as that it appearantly always circulates the coolant even with the engine off. As a result, cooling capacity stays the same engine on or off, further helping a shut down engine that does not have to rely on hydrodynamics alone.

Not quite sure about this. For the cars I have tested it usually took a while for the fans to start up?

- Fully open thermostat
- Temperature still increasing
- Fans start up

Start up temperature is often higher than shutdown temperature which is also realistic. The idea being to reduce on off cycles of the fan and only use it when passive cooling of the radiator is not enough.

My real life car starts the fans at 104 degree C of coolant temperature. Below that, the fans will never start up. Once active however they will not stop at 103 degree C but instead cool down the engine to 94 degree before stopping again.

They have usually only one operating speed and a second one which I would call emergency use only.... you can hear it briefly when you restart a recently stopped hot engine... the coolant temperature sensor detects very hot coolant and the fan switches to maximum speed for half a second before switching back to regular speed as fresh regular hot coolant.

Most fan stress nowadays come from AC running..... most cars have at least one fan running all the time when AC is active which usually even overcompensates the increases engine load (in english, the coolant overall temperature is lower due to the fan being forced on even though the temperature is not that high).

Some fans do make noise, but just like in reality it is pretty faint in comparsion to the running ICE.

Not sure if any production car made for mainstream usage is ever going to inform you on the dashboard about fans spinning or not. It seems even more unlikely when you consider that basically any vehicle today is proud user of " The lying temperature gauge" that displays a needle firmly centered in the middle between 40 degree C - 100 degree C.

They are not even telling you the real temperature of the engine... why would they bother to tell you that the fans are spinning?

Temperature warning lights do exist but they usually only come on long after all available cooling capacity has been exhausted. Fans will spin at maximum RPM long before you get a warning, because the - debatable -mainstream usage idea is that you should not be bothered about things that dont require your attention.

You only get warning lights when you need to take action very soon/without delay.

It should be noted that BeamNG engine thermal simulation works most realistically when driving with medium to heavy loads. Long term idling/very low load sees reduced temperature rise compared to real life cars.

There are probably multiple factors in play:

- No cold start penality. In real life, starting up an cold ICE results in increased friction, wear and tear as lubrication is limited and tolerances are off due to parts being cold. In BeamNG, an engine operates at peak capacity right after start up, and does not suffer from any such real world issues.

- No auxilarly loads: In real life, you will be using lights, heating and numerous more things that cause increased engine load = faster heat up.

- No emission control: A cold petrol engine ECU has one goal.. one goal only.... heat up the catalytic converter asap to meet emission standards. BeamNG does not and the engine is only really concerned with sustaining itself.

- Heat transfer:In a real engine, heat from the exhaust will radiate back to the engine as will heat from friction in the transmission/clutch assembly. Nope for BeamNG.

Some high performance radiators fully open the thermostat at 82 instead of 92 degree, granting you an additional 10 degrees of time for flooring it before it overheats.

ICE efficiency raises the hotter the engine is, thus the point of any modern cooling system is to keep the engine as hot as possible without causing material issues. That temperature seems to be between 90 - 100 degree C. Not below... nor above.

My real life car maintains a temperature of arround 87 degree C in inner city driving and arround 91 at highway speeds.

Yes, it is a one size fits all solution in BeamNG, but that to an extent is necessary. For example, the tachometer UI app turns the cooling icon red at 115 degrees. Which for most cars means you can drive on without damage.... but not the Vivace 1.2 I3 which at high load as arround 5 degrees more block temperature than coolant temperature.

The result is block damage besides the coolant not being fully overheated.

You would need to fine tune the UI app for every car.... that seems like a lot of work for very little gain.



I am not saying that engine thermals cannot be improved in BeamNG, but I wanted to explain some factors behind it.

 

I mentioned this since 2019 (not the exact words)

and this

However, the problem is not fixed
--- Post updated ---

Any sources about this information? I can't find much. And if that's really true, is that even legal?

My opinion is the temp reading is only from the coolant inside the engine
IRL vehicles have three "parts" of coolant: coolant inside the engine, coolant inside the radiator and coolant inside the reservoir tank.
When you cold start the engine the thermostat is fully closed and it's too cold for the coolant to expand into the reservoir. Thus your coolant is only circulating inside the engine as temp increases, and both the radiator coolant and reserved coolant are at ambient temperature, until the coolant inside the engine reaches 90C. After the coolant inside the engine reaches 90C, the thermostat starts to open, this makes the coolant circulation shifts from inside the engine to between engine and radiator, also warm coolant starts to expand to the reservoir tank(that's the reason you should check the coolant level and add coolant when the engine is cold). As you drive further, the coolant inside the radiator gradually increases because of thermal convection. The needle is centered in the middle until the coolant in the radiator reaches 90C±5C where the thermostat fully opens. Long story short, after the coolant circulating inside the engine reaches 90C and the coolant needle will stay centered when the coolant inside the radiator is anything between ambient and 90C. Since most of the time we drive non-aggressively, the coolant inside the radiator is significantly below 90C most of the time and the coolant inside the engine is always near 90C after warming up(unless your thermostat is broken), we may perceive the needle is lying.

In BeamNG, the fluid in the radiator and in the engine is not simulated separately, there's no "parts" of coolant. When driving in BeamNG, you need to heat up the whole coolant as opposed to only in the engine IRL. Thus the warming up is much slower in BeamNG.

The engine in BeamNG is very good as sustaining itself, may be too good. Two examples:
1. Load the engine at idle without touching the accelerator.
Like feathering the clutch without stalling the engine, when reduce clutch pedal input the engine load increases, in BeamNG the engine gets louder but the RPM stays firmly at idle RPM. If you do this IRL the engine RPM will drop, the more the RPM drop the more air/fuel is commanded by the ECU to prevent stall, then the RPM gets stabilized at an RPM slightly below idle at a given clutch input.
2. Engine braking
All ICE vehicles in BeamNG have DFCO(deceleration fuel cut-off), when the accelerator pedal is not pressed, the instantaneous fuel consumption goes to zero even if the engine RPM is only very slightly above idle. And when the engine RPM reaches idle eventually, the instantaneous fuel consumption increases from zero to idle fuel consumption seamlessly.
IRL vehicle differs. Most if not all classic vehicles don't have DFCO, and modern vehicles' DFCO re-engage fuel flow at above idle RPM(mostly 1200~2500RPM) they don't wait until the engine reaches idle. Also the throttle valve has a minimum opening and cannot be fully closed, thus there's also a minimum fuel flow. The actual fuel flow can be either at or above the minimum, or fully zero. So it's not possible for dis/engagement of DFCO to be seamlessly.

 

I get this isn't every vehicle in the world, but my family's F-150's radiator fans are much louder than the engine itself, mostly because of how much air they push. They activate every couple of minutes to regulate the temperature. Also, my '89 Sprint's fan acts similar to the cars in BeamNG, but the fan turns on much sooner, and if the engine is warm, the fan will always run for a few minutes when the ignition is set to the run position (as long as the engine isn't on). Also, the Sprint's fan is almost as loud as the engine as well and when it turns on the temperature needle drops, lol (though, it does have an aluminum rad which cools it too well).
And yeah, a fan warning light doesn't make sense, but a coolant overheating light does. In my Sprint, it doesn't have an overheat light, but the CEL has come on when it overheated before. As for warming up, my Subie will reach 88° - 90°C if it idles in the garage for 20+ minutes, but it usually goes anywhere from 86° to 89° while driving. At most, it'll hit 92° while pushing it hard.
Also, thanks for taking the time to explain all that, by the way.

 

Radiator fans in some vehicles turn on earlier than thermostat, which might be helpful in preventing "heat spot" from accumulating in the engine bay.

 

Any Vehicle I ever drove does this. Seems to be pretty standard nowadys, at least in europe.

The - imho dubious - reasoning behind this is that drivers far away from the technical understanding of ICE operation shall not be confused by the temperature gauge moving up and down but instead shall only be informed about

• Engine cold - Avoid high load and high RPM
• Engine at operating temperature
• Engine hot - overheating may occur soon

So the gauge moves up quickly to the medium position, indicating an water temperature of arround 90 degree C while in fact it barely has gone above 45 °C.

This is bad because its indicates the engine can be floored with minimal wear and tear when in fact it is still warming up and the oil temperature may lag behind quite significantly.

It also means that the needle wont move the next 300 km unless the engine is getting really really hot.

Even worse however, it denies you a pre warning of the cooling system slowly failing. My cars usually operating temperature is between 85 - 92 °C. Needle in the middle.

Lets assume a cooling system with issues and the actual temperature now being 102 °C instead. The gauge doesnt tell you.

But when you add a trailer or go uphill... your chances of getting a hot surprise increase significant.

Ford 1.0 I3 Ecoboost engine suffered from a loose coolant hose in some models. The hose came loose, usually at higher speeds as coolant pump pressure is higher then and pumped out the coolant onto the street instead of inside the engine, resulting in catastrophic overheat within minutes at best.

And again, the lying temperature gauge denies you valuable time of pre warning... only beginning to move upwards when it is almost to late. Instead of maybe 90 seconds of warning time... you get maybe 30 seconds. Thank you very much... not.

It is legal, as there is no law mandating a temperature gauge similar to as there is no law mandating a tachometer. Those two things where built in cars by customer demand. So if you add a temperature gauge, there is nothing stopping you from making it lie throughout most of its lifetime. In fact some modern car models have removed it altogether and instead only offer an overheat warning light.

It is one of my many frustrations with people with zero technical knowledge making buying decisions and the resulting offers from manufacturers.

In real life, there seem to be various ways where manufacturers have decided to mount the temperature sensor.

The most popular seems to be directly at the engine coolant outlet (where the hot coolant leaves the engine block) which is mostly sensible. As you have explained, temperature elsewhere can be much lower and usually isnt relevant because there is a feedback loop in the end.

Assuming the radiator is overheating, the "cold" coolant reaching the engine will be hotter than usual and as such the coolant leaving the engine will be even hotter which then will be displayed/at least taken into account by the ECU which may trigger fans)

Reading the coolant temperature with the secret menu or an ODB2 Scanner at least for my Ford Focus MK 2 indicates that the temperature sensor is in the coolant outlet of the engine. It sees a pretty fast temperature rise that slows down above 80 °C and then stays between 85 °C - 92 °C as temperature is regulated by the thermostat mixing in cold/cooled coolant from the radiator.

The temperature gauge meanwhile is embedded in concrete between 45 °C - 104 °C at which point the fans start up. I dont know when it will raise above middle, because I am not willing to do the stupid things necessary to bring the engine temp higher than 104 °C with the fans running.

104 °C is not enough for it to move. Which when you follow the dubious logic even makes sense. The fans are more than capable of cooling the engine down from 104 °C to 96 °C even in the hottest of summers so why bother informing the driver when there is no actual risk of overheat. (Generously assuming the cooling system to work properly of course)

So yes, the needle is lying... or "calmed" as they say. It displays the same value between 45 °C and 104 °C which is firmly centered in the middle.

 

Automatic transmissions have a habit of not upshifting when they really should, making them unusually sporty at the cost of efficiency. For instance, the Bastion's 8-speed automatic tends to keep the engine at 2000 RPM when driving at 55 MPH - accelerate slightly then decelerate and it will actually go to the gear it should have been in to begin with, dropping the revs to a comfortable and efficient 1500. This is less a fundamental problem with the powertrain simulation as it is with the Bastion's shift points - I have made a trivial mod that fixes it by setting the low-throttle upshift point to 1700 RPM.

And realistically, a transmission for one engine will not have exactly the same settings as one for another engine, even if they are technically the same model.

 

This is simulated in the vehicleController.lua, via the "calculateOptimalLoadShiftPoints" jbeam

 

To add to this, if your car even comes with an oil pressure gauge, sometimes it is nothing more than an idiot light with a needle. My old Ford Mustang (1988) literally had a piece of plastic in the way of the needle to stop it from moving any more than 50% of the way up the gauge. If you ever did have low oil pressure though, it would tell you... once it dropped below 8psi... which, for anyone wondering, is basically on the verge of catastrophic engine failure. The reason they do this though is because your oil pressure can very from 20 all the way up to 70 psi or even higher depending on oil temperature and the speed of the engine. So when people would see the oil pressure gauge moving around they would freak out and take the car in for a service... when it didn't need one. So... Ford decided to do this silly thing instead.

lying gauges have been a thing for a LONG time. If the real life value fluctuates under normal driving and there is a factory gauge for it... more than likely it is lying to you. If you have a sports car it is less likely, but still possible it might lie.

My Corvette for instance has a digital readout for EVERYTHING. Oil pressure, oil temperature, coolant temperature, voltage, speed, rpm, mpg... like... you name it, its got it and none of the gauges try to hide anything from you. You wouldn't believe the amount of people online that are constantly freaking out about low oil pressure at idle or high coolant temperatures or high voltages when running when in reality... there is absolutely nothing wrong with the car at all. Its almost kind of sad.

 

Guys, you are probably aware of this but are you planning on improving transmissions anytime soon? This is wanted by a lot of people and this thread shows it. Normal torque converter autos need it the most, driving at anything different than 100% throttle is weird

 

I don't believe they are, I think this thread has more to do with mod creation and/or damage simulation. Kind of a shame as it is much needed, but it doesn't seem to have gained any traction... that's probably as good as it's going to get, although I suspect the reasons are technical (e.g. not enough data to develop an accurate simulation, engine is not responsive enough for real-time simulation, not possible without completely rewriting large parts of the codebase, etc.) rather than due to any lack of interest on behalf of the devs.

I could be totally wrong and would love to hear an official statement on the matter

 

Honestly, I'd be happy if they'd just give us the ability to lock out overdrive in 4-speed autos, or else make them less eager to rush top gear as soon as you back off from 100% throttle. (The Gavril 6-speed is a unique type of "joy" in this respect as it still only does PRND21, so once you're above second-gear speeds you basically have no control over the gearbox at all. On the other hand, the Gavril 4-speed seems to be set up much more aggressively than the other "traditional" autos in the game and is thus the only one that isn't completely painful to drive in the twisties.)

EDIT: Also, I just discovered that basically every car has its "maximum safe G" for oil starvation calculations set at 1.2. Yes, even the Sunburst that was found to oil starve in normal cornering when the mechanic was first introduced. Considering that my old Escort would show an oil pressure light in right-hand corners well before the (garbage) tires even squealed - granted I didn't keep it completely full, but still - and that my Veloster is apparently known to suffer from oil starvation in hard use, with the workaround being to slightly overfill the oil if you can't afford a baffled oil pan, I'm not sure how realistic this is.

Well, actually, considering how well modern performance tires stick, 1.2 may not be far off for the Veloster, but still.

 

How can this be a thread about mod creation or damage simulation in any way? The name is Powertrain feedback collection and this is what powertrain means - "Powertrain parts’ refers to the group of components that generate power and deliver it to the road surface, water, or air (e.g. crankshaft, clutches, torque converters, gear parts, drive shafts, wheels, etc.)".

 

Based on my older post in this topic:

I have been thinking about an better implementation of regenerative braking:

The Capsule Bus Mod

https://www.beamng.com/threads/capsule.84077/

is equipped with an retarder:



This retarder has five different power settings and as such the braking power of it can be fine tuned depending on the driving situation.

Maybe a similar system could be used to allow for different regenerative braking settings for electric vehicles. Several real life electric vehicles allow you to change the regenerative brake effect similarly.

Suggestion:

Add selectable regenerative braking for when electric vehicles coast down, similar to the retarder operation in the video.

Grant access to min and max value in the tuning menu.

This would greatly increase the realism of electric vehicles which usually favour regenerative braking over friction braking outside of emergency situations. As an added benefit, it would also allow keyboard users much more regenerative braking capability.


An less elegant but maybe easier to implement way would be the way Arcanox Hybrid System ( https://www.beamng.com/resources/arcanox-core.6060/ ) handles it:

You can upshift from D to L and by switching into this gear, the electronic automatically uses the highest possible coast down regenerative braking torque.

Be in gear D = normal low regenerative braking
Be n gear L = maximum regenerative braking





Less elegant due to lack of steps but still a reasonable way to implement it as you can stay in L without issues for the duration of the race.


Thank you for consideration.

 

Every single automatic in the game, from the Burnside Dual-Matic to the Vivace DCT, will refuse to upshift with even the slightest brake input, is this intentional?

 

I watched today's Motorweek Retro Review video (@2:00) and I just had an idea: how about adding driver-selectable shift programs on the 80s/90s 4-Speed Torque Converter Automatics, for economy and performance? Naturally, the Performance setting can have higher shift points than the Economy one and they can also have their own key bindings.
The same can be said about adding a Tow/Haul mode on the automatic transmissions of the trucks (the 4-Speed and 6-Speed Torque Converter units of the Gavril Roamer and the D-/H-Series are good candidates for this), with its own key bindings of course.