Jump to content
Oovee® Games

Inaccurate Truck Stats


Recommended Posts

We're all for realism here, right? Well, one thing I have recently noticed is the quite unrealistic torque values for two trucks. Now, I don't know how the game calculates torque, but the values in the XML files and the torque in real life are quite mismatched.

How do I know this, you ask? Well, I just did some basic maths to find out the relationship between the torques of the trucks in question, which happen to be the B-131 and the C-4310, to find out how much more powerful the C-4310 is, or in my case, how much weaker the 131 is. So, I looked in the XML files:

...and found out that the torque of the B-131 is "54894", whatever that means, and the C-4310's torque is 140000 unnamed Spintires units. Now, from that we can calculate how much weaker the B-131 is (if you instead want to find out how much stronger the C-4310 is, just swap the numbers around the division sign):

 

This means that the B-131 has about 0.39 times as much torque as the C-4310:

Now, let us compare this value to that of real life. For this I went onto Wikipedia to find their engine specifications - the 131:

...has 402. No need to worry about the units because we will just be ending up with a decimal again. And the C-4310:

637, at 1600rpm*. Now, finally, to divide these numbers in the same manner as before:

This means that the B-131 has about 0.63 times as much torque as the C-4310.

And here is the result:

 

Now I wouldn't expect any developers to go into such meticulous detail, but I think this would be a great thing to make accurate in the game, as the trucks would be much more realistic and would feel exactly as they do in real life. All you need is this simple maths method to calculate the relative torque of every vehicle in the game perfectly. Then, a way to equate the Spintires torque unit to real life. And this is rather simple, too: 

  1. Get a truck, note its real life weight and torque in first gear.
  2. Use that data to calculate how fast it would accelerate in real life.
  3. Accelerate in the game.
  4. Compare the acceleration ingame to your calculation and calibrate accordingly.
  5. Use my torque-difference calculation method to accurately portray every truck in the game.

Anyway, thanks for reading this mini-essay, and feel free to correct any mistakes I have made, as I am sure there are some.

*It would be an interesting addition to have varying engine power output based on the RPM, although I am not sure how difficult it would be to replicate the power curve of everything, but then again, that's a little over the top (as if anything else in this post isn't).

Edited by DeltΔV
  • Like 1
  • Thanks 1
Link to post
Share on other sites

Wow, you spent some major time on this.  I've been saying this for a long time, that some torque values are wrong in this game, but not in the detail that you have.  I believe they are still work in progress, as Zane has updated some values using a torque calculating tool while other values haven't been updated (for a long time), and even some of those that have been updated, for example the C-6520, seems wrong as it's almost the same as the torque of the D-535 while other C-type trucks have much lower torque values.  It would be nice if all the torque values were corrected and finalized.

Edited by Unster
Link to post
Share on other sites

Although I do agree with the sometimes wonky feel of the trucks torque wise, I sort of wonder if it was intentional or if axle ratios, transmission ratios and transfer-case ratios are also somehow calculated into the mix? Example

A truck with 450Tq with a 2.73 axle ratio wont tow as good as say a truck with 450Tq but a 5.86 axle ratio

you could add first-gear or a creeper gear or even a transfer-case low range and that would massively alter the outcome and allow 2 seemingly incomparable trucks to become comparable towing/pulling wise. If that makes any sense lol.

  • Like 2
Link to post
Share on other sites
On 8/31/2019 at 12:00 PM, Sync_THE_BLACK_WIDOW said:

if axle ratios, transmission ratios and transfer-case ratios are also somehow calculated into the mix?

I doubt that that kind of thing could cause such a difference, and I'm pretty sure the values are of engine torque, considering that they are in the <Motor/> section.

But it's definitely possible.

 

[EDIT]: Why am I stupid in this post? Of course it makes a difference!

Edited by DeltΔV
past me was an idiot
Link to post
Share on other sites
  • 4 weeks later...

I used the Gearbox Calculator kindly provided by PressureLine, who seems to have disappeared from here. I've inputted data meticulously collected for each truck into this spreadsheet in order to obtain the results found in the vehicle XMLs respectively. Such data includes torque, RPM, axle ratios, wheel size and individual gear ratios. 

https://spintires.nl/gearbox-calculator-spreadsheet/

As a result, you'll likely find "strange" torque settings because as far as I know, it uses a mixture of all of the data. Notwithstanding, the gear ratios have a lot of say in the end result shown in-game.

  • Like 1
Link to post
Share on other sites
11 hours ago, Zane (Oovee) said:

I used the Gearbox Calculator kindly provided by PressureLine, who seems to have disappeared from here. I've inputted data meticulously collected for each truck into this spreadsheet in order to obtain the results found in the vehicle XMLs respectively. Such data includes torque, RPM, axle ratios, wheel size and individual gear ratios. 

https://spintires.nl/gearbox-calculator-spreadsheet/

As a result, you'll likely find "strange" torque settings because as far as I know, it uses a mixture of all of the data. Notwithstanding, the gear ratios have a lot of say in the end result shown in-game.

I question the accuracy of that software though, given some of the torque values that you ended up with in the xml files.  For example as I mentioned before, the C-6520 has huge torque and it does truly feel overpowered compared to other trucks.  I prefer setting torque relative to other trucks, if the data is available, then also test drive it in the Proving Ground including one particular test I do.  On the slope next to the dirt track, in manual mode in 1st gear in idle, I test how far up the hill the trucks creeps up (with all wheels + diff lock if available).  If a larger truck idles up the hill, then I consider it overpowered.  With cars I'm a little more forgiving, but even with those I don't like to see them climb walls.

Link to post
Share on other sites
  • 3 weeks later...

If you want this kind of realism you need to associate torque with correct RPM and simulate different engines in details and Im not sure Spintires game engine even support all that. Hardly it simulates anything, but probably just use something like "power levels" range from 0 to 9 and torque values distributed in this range by some primitive math.

This means that even if you will do all these adjustments your model not going to turn into realistic instantly. In reality KamAZ turbodiesel engine will apply its maximum torque almost instantly at very low RPM just above idle values and will form an effective torque plateau in whole RPM range, while ZiL engine will build up torque much slowly until it will reach peak and will start to loose some of its torque at even higher revolutions. 

I seriously doubt this game goes that deep, while its visually immersive and called a hardcore simulation by many its actually just a very simple and arcade game developed from havok technology demonstator and designed to be playable using keyboard. Back then no one took concept of driving rusty steel dinosaurs trough the the Russian mud seriously, I bet profit and success was a surprise for everyone )))

To make it at least somewhat realistic like you wish, it needs to be redesigned completely as an OFF-ROAD TRUCK SIMULATOR in mind.

Edited by grifo
  • Like 1
Link to post
Share on other sites
55 minutes ago, grifo said:

I seriously doubt this game goes that deep, while its visually immersive and called a hardcore simulation by many its actually just a very simple and arcade game developed from havok technology demonstator and designed to be playable using keyboard.

No, this game doesn't go that deep.  It never did and I haven't heard anyone call it hardcore simulation.  On the other hand, it's not an arcade game either, not even close.  It simulates physics better than many of the other so-called simulators, and it doesn't even call itself one.  So let's be reasonable here and not stretch the truth.

Redesigning the physics from the ground up sounds good in theory, but it's not feasible for such a small team, at least not for now.  So let's keep the criticism constructive and reasonable and work with what we have.

Link to post
Share on other sites

Nope, its an arcade, it can be played casually with no knowledge and with gamepad or keyboard. Like Gran Tourismo and Need for Speed, they both have good looking cars, some physics and some adjustable settings, but one requires steering wheel and pedals and have enough details to be simulator-lite and other is ends up to be a dumb arcade for keyboard. Spintires is even more simplistic in this department.  Do you mean arcade games cant have good physics? They can! Anyway you are free to think its not an arcade and Im free to think that it is, doenst change the things I said.

Im not sure what exactly is not reasonable and constructive in my comment, to implement realistic torque in this game like OP suggests, changing some value in XML is not enough unfortunately, it probably needs a heavy redesign of whole system since this system is simple and arcade. Thats what im talking about. It doesnt mean its bad, game is still very immersive and fun anyway.

   

  

Edited by grifo
Link to post
Share on other sites
11 minutes ago, grifo said:

Nope, its an arcade, it can be played casually with no knowledge and gamepad or keyboard.

Trucks can easily be driven by keyboard in this game because the speeds are low and the controls are keyboard-friendly, not because the physics are simple.  Even BeamNG, which has a more advanced physics model than any driving game that I can think of, can be played with a keyboard, though not very well.  It doesn't really mean anything.  Arcade in this game genre means easy / unrealistic physics.  That's far from the case here, especially with my tweaks for more difficult deep mud.  It's discussed in the 1.3.7 update forum if you're interested.

The game does have some variable torque based on the speed within the speed range of a given gear.  Sure it's simplified and could be improved, for example by adding the ability to customize the torque curve for each engine, but it's not bad either.

  • Like 1
Link to post
Share on other sites

I liked the idea of @grifo to better simulate the power curve of each truck. 

Years ago I was addicted to GTR2, rFactor, iRacing...owning and selling high end aftermarket wheels and pedals. They were considered (at the time) hard core racing simulators, each one with its own pros and cons, but each one of them had a very nice physics model and an overwhelming quantity of settings.

ST doesn't get even close to such level of simulation of detail so, to be honest,  I don't think that saying it's an arcade is far from the truth, especially because we all know why  and how it was born. 

Improvements are always  welcome, as new and different truck's sounds, graphical improvements, better mud and water...power curve,  but  the small team already did a great job in few months, adding new features and content, making it even more fun to be played and hopefully they will surprise us with even more new feature in the near future (dynamic mud on top of the list).

Arcade... simulation...does it really matter which label we put on it? It's a fuc*ing fun game in any case.

 

 

 

 

Link to post
Share on other sites

For me arcade in this genre means ability to play without pro peripherals, specific knowledge on subject and good skills and has nothing to do with physics of mud.

Point is one who is able in simulators is very likely able to repeat that in reality with very surprising efficency. It was demonstrated not once that pro virtual racers or pilots, demonstrate that they are able to operate real machinery if they given a chance, they can pass pilot tests and show good results at tracks. I doubt that some average Spintires player is ready to face similar situation in reality. Thats another feature of arcade experience. Pressing wasdq doesnt teach you anything. But yes, it doesnt matter after all.

BTW another terrible change, transparent mud pools no doubt was made to benefit casual playstyle (increase slaes) doesnt matter what excuses were given for it. So one can see that this game tries to allign to arcade range of spectre not to simulation one. Its naive to expect more complex simulation and more hardcore mechanics in the future

Edited by grifo
Link to post
Share on other sites
9 minutes ago, grifo said:

I doubt that some average Spintires player is ready to face similar situation in reality.

I disagree to some extent.  The challenge in ST or MR isn't to keep a vehicle on the track at high speeds, obviously.  The challenge is navigating the uneven road without flipping over, make use of available tools to make it through the mud, and competently load logs onto the trucks.  I remember when I first played MR it was a challenge to make the route without flipping over and loading the logs.  Now it's a lot easier.  It's a different kind of skill from the usual racing game.

  • Like 1
Link to post
Share on other sites
9 minutes ago, Unster said:

I disagree to some extent.  The challenge in ST or MR isn't to keep a vehicle on the track at high speeds, obviously.  The challenge is navigating the uneven road without flipping over, make use of available tools to make it through the mud, and competently load logs onto the trucks.  I remember when I first played MR it was a challenge to make the route without flipping over and loading the logs.  Now it's a lot easier.  It's a different kind of skill from the usual racing game.

This means you are now used to specific game mechanics of Spintires. This doesnt mean you can jump into crane fill your truck with logs and cross the Siberia )) In KAmAZ you have 8 gears, you use 1st only if its overloaded and going uphill, you can easily start from 3rd. Bunch of levers for transmission and misc equipment. Even downshifting in such trucks is a sequence you must master before, its not an Volvo trucks.. Game is not giving you a clue about actual hardware. KrAZ is more capable than Ural for gameplay reasons, while its not like this in reality..

If someone can fly like a boss and complete missions in MiL Mi--8 in ARMA with wasdqz buttons its cool, but lets place him in real chopper, he will only be able to open window and smoke a cig with pokerface maybe.

If someone can fly like a boss and complete missions in MiL Mi--8 in DCS, he will be able at least to take the real chopper in to the sky. It takes 20 minutes and lot of switching, turning on\off and checking.

Big difference. Simulators simulating whole picture, hardware, situation, operational sequences. Arcades just emulate general experience, but thats it. 

Edited by grifo
Link to post
Share on other sites
39 minutes ago, grifo said:

Big difference. Simulators simulating whole picture, hardware, situation, operational sequences. Arcades just emulate general experience, but thats it

Well there sure is a variety of implementations when it comes to "simulators".  There are some, like Bus Simulator and American Truck Simulator that simulate the cockpit well as well as general procedures, but fail miserably in the physics department.  ST doesn't simulate cockpits or even controls that well, but it makes up for that in physics, which overall are pretty good actually.  To me the physics matter more, and that's why I'm not calling ST an arcade game.  Arcade to me is something like the modern Need For Speed with its crazy speeds and sticky tires.  The first one in the series was actually pretty decent, a simulator-lite I'd say.

Edited by Unster
Link to post
Share on other sites
  • 2 months later...
On 9/30/2019 at 2:23 PM, Unster said:

I question the accuracy of that software though, given some of the torque values that you ended up with in the xml files.  For example as I mentioned before, the C-6520 has huge torque and it does truly feel overpowered compared to other trucks.  I prefer setting torque relative to other trucks, if the data is available, then also test drive it in the Proving Ground including one particular test I do.  On the slope next to the dirt track, in manual mode in 1st gear in idle, I test how far up the hill the trucks creeps up (with all wheels + diff lock if available).  If a larger truck idles up the hill, then I consider it overpowered.  With cars I'm a little more forgiving, but even with those I don't like to see them climb walls.

N-n-n-n-necro!

The accuracy of the output calculator is related directly to the accuracy of the information that is put in! For the example of the C-652x trucks, here's the relevant info (based on the Kamaz website)

  • Engine: KAMAZ-740.63-400 1255Nm@1313-1500rpm (go modern turbodiesels, nearly double the torque for the C-43xx trucks)
  • Transmission: ZF16 S 151 1st Gear @ 16.47:1
  • Transfer Case Low: 1.536:1
  • Differential: 5.11:1 (one of several options)

To work out the at-the-wheels torque (which is what Spintires uses) we multiply engine torque by the total geared reduction in the drivetrain: 1255*16.47*1.536*5.11=162236.825. Now, for the 652x trucks this number would be perfectly adequate, but when you do the same for the B-130 it comes out at a pathetic number that is often unable to turn the wheels (especially from a standing start, more on that later) so the calculator adds a 'fiddle factor', we multiply the torque number obtained earlier by 1/wheel_radius (1/0.528=1.89ish) to get the theoretical max tractive effort of the vehicle (done this way so that it favors the smaller (A-469/B-130) vehicles without giving a massive unneeded boost to the D and E class trucks) which comes out at 306626ish, which is pretty close to the Spintires vanilla '339279'. Of course the vanilla C-6520 appears to have lost 11 of it's forward gears, so it's anyone's guess as to what numbers were put into the calculator. (For reference the C-255=882*5.26*2.28*8.21=86842 then 86842*(1/0.55)=157894)

@Zane (Oovee) @Localhost (Oovee) If you guys could change the 1st/Reverse AngVel calculation so that the AngVelMin equals 0-AngVel (rather than AngVelMin=AngVel*0.5 - 2.0), then 50% of the vehicle's max torque would be available, regardless of the 'optimum' angvel, from a standing start then the 'fiddle factor' could probably be removed from the calculator and vehicle with relatively high 1st gear ratios (E-167, A-469 for example) would actually work without having to artificially reduce the AngVel speed for 1st/Reverse.

Thanks for listening!

  • Like 1
Link to post
Share on other sites
On 4/14/2020 at 3:58 AM, PressureLine said:

get the theoretical max tractive effort of the vehicle (done this way so that it favors the smaller (A-469/B-130) vehicles

Well, do you remember that issue where the A-469 had more torque than the E-7310? I'm starting to question the accuracy of this 'calculator'...

Edited by DeltΔV
Link to post
Share on other sites
2 hours ago, DeltΔV said:

Well, do you remember that issue where the A-469 had more torque than the E-7310? I'm starting to question the accuracy of this 'calculator'...

No I don't, but that sounds hilarious 🤣 I'll have to look it up!

According to the numbers I ran up (for myself) when Mudrunner was released the A-469 should be 27546, and the E-7310 should be 285880, which is a literal order of magnitude difference. Of course if you mess up your input units (input Nm as kgm or lb-ft for example) your outputs will be all sorts of wrong. Likewise putting in strange wheel radii (inputting them in inches?)

Link to post
Share on other sites
21 minutes ago, PressureLine said:

No I don't, but that sounds hilarious 🤣 I'll have to look it up!

According to the numbers I ran up (for myself) when Mudrunner was released the A-469 should be 27546, and the E-7310 should be 285880, which is a literal order of magnitude difference. Of course if you mess up your input units (input Nm as kgm or lb-ft for example) your outputs will be all sorts of wrong. Likewise putting in strange wheel radii (inputting them in inches?)

Sorry, I remembered wrong. I had a thread on this and I just found it. It was the 4320 which the A-469 had more torque than:

Apparently Zane made a mistake with your calculator, something about the 'torque converted' feature.

Edited by DeltΔV
didn't capitalise an 'i'
Link to post
Share on other sites

Yes, that feature can give strange results, especially with a relatively high rpm engine. It's actually meant to be "Torque Converter" for simulating how automatic transmissions work (the torque converter multiplies torque when it is 'slipping'), but I was never particularly happy with how the implementation worked out in the later versions of Spintires.

Link to post
Share on other sites
On 4/14/2020 at 3:58 AM, PressureLine said:

N-n-n-n-necro!

The accuracy of the output calculator is related directly to the accuracy of the information that is put in! For the example of the C-652x trucks, here's the relevant info (based on the Kamaz website)

  • Engine: KAMAZ-740.63-400 1255Nm@1313-1500rpm (go modern turbodiesels, nearly double the torque for the C-43xx trucks)
  • Transmission: ZF16 S 151 1st Gear @ 16.47:1
  • Transfer Case Low: 1.536:1
  • Differential: 5.11:1 (one of several options)

To work out the at-the-wheels torque (which is what Spintires uses) we multiply engine torque by the total geared reduction in the drivetrain: 1255*16.47*1.536*5.11=162236.825. Now, for the 652x trucks this number would be perfectly adequate, but when you do the same for the B-130 it comes out at a pathetic number that is often unable to turn the wheels (especially from a standing start, more on that later) so the calculator adds a 'fiddle factor', we multiply the torque number obtained earlier by 1/wheel_radius (1/0.528=1.89ish) to get the theoretical max tractive effort of the vehicle (done this way so that it favors the smaller (A-469/B-130) vehicles without giving a massive unneeded boost to the D and E class trucks) which comes out at 306626ish, which is pretty close to the Spintires vanilla '339279'. Of course the vanilla C-6520 appears to have lost 11 of it's forward gears, so it's anyone's guess as to what numbers were put into the calculator. (For reference the C-255=882*5.26*2.28*8.21=86842 then 86842*(1/0.55)=157894)

@Zane (Oovee) @Localhost (Oovee) If you guys could change the 1st/Reverse AngVel calculation so that the AngVelMin equals 0-AngVel (rather than AngVelMin=AngVel*0.5 - 2.0), then 50% of the vehicle's max torque would be available, regardless of the 'optimum' angvel, from a standing start then the 'fiddle factor' could probably be removed from the calculator and vehicle with relatively high 1st gear ratios (E-167, A-469 for example) would actually work without having to artificially reduce the AngVel speed for 1st/Reverse.

Thanks for listening!

We are thinking about implementing engine mapping so you can manually adjust the torque curve. :)

Link to post
Share on other sites
On 4/18/2020 at 8:12 AM, Zane (Oovee) said:

We are thinking about implementing engine mapping so you can manually adjust the torque curve. :)

Better (at this point, so as not to horrendously break every existing mod) would be to allow the parameters to be adjusted per-gear (much like Spintires was originally) like this:

<Gear	
	Torque="47933"
	AngVelMin="0.48"
	AngVelOpt="2.89"
	AngVelMax="8.1"
	AutoNextLinVel="2.27"
	MaxDeltaAngVel="0.05"
/>

For even more fun, adding in parameters for AngVelOptMin and AngVelOptMax so you can have a flat spot in the 'curve' (for example the engine in the C-652x trucks makes 1255Nm of torque from 1300rpm to 1500rpm) which would allow for loads of cool things:

<Gear	
	Torque="47933"
	AngVelMin="0.48"
	AngVelOptMin="2.49"
	AngVel="2.89"
	AngVelOptMax="3.29"
	AngVelMax="8.1"
	AutoNextLinVel="2.27"
	MaxDeltaAngVel="0.05"
/>

Any values that get 'left out' of the truck's .xml get automatically calculated as they do now (AngVelOptMax/AngVelOptMin get set to == AngVel) which would preserve backwards compatibility in a fairly elegant way.

  • Like 1
  • Thanks 1
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...