Physics discussion thread

I

I have seen plenty of commercial / esport level setups for ams2 for bunch of different cars and vast majority of them, engine braking was set to be 1, so even more engine braking than on the default. If the setup is otherwise well balanced, more engine braking can help with performance just as described above

 

When it comes to overall braking force, sure, that’s true in theory. But the pragmatic reality is that the rear tires can only handle so much braking. The rear brakes are built to cover that (and a bit extra) without any help from the engine. Heat does play a role, but that’s usually easy to keep under control with a little extra cooling.

 

How does this actually work with the diff regulated classes like GT3, do you have the info on how the diff should be set for the different cars and set it according to that within the game (IMSA regulations and is it different from what it is for WEC and LMGT3 for example what comes to differential)?

If so, I would assume other sim devs would do the same and not set based on their preference, especially if highly different from IRL values (also assuming that the diff is simulated well enough in the sim of course).

For default setups in cars where power and coast ramps can be adjusted, it's quite clear that at least on rF2 and LMU, in a lot of cases, the setups are not exactly set up for turn in and mid corner rotation (main reason I tend to dislike those default setups and end up always opening the diff for those scenarios in those sims almost always and also the reason why I dislike driving fixed setup series on those sims). Though I have to say I find the coast ramp often AMS2 also a bit too locked to my taste and open often it by few degrees.

 

Still on the "floaty" subject, the other thing that change this perception is a more direct steering wheel. Changing steering sensitivity (let's say to 75%*) makes the car seem much more in line with other sims (LMU, rF2, RR comes to mind).
AMS2's steering is very numb near the center (lack of "bite") and with some really floaty camera parameters (user dependent) can make it a lot worst.

I wonder if this could be a related.
And BTW, kudos to Reiza.

* just a side note: the virtual wheel behaves the opposite way - it turns less when steering sensitivity is increased

 

It's impossible to say what people mean by "floaty"... for some, it might be because of the camera or ffb (maybe something specific in ffb) for others it might be actually how car handles in certain situation(s) etc.

 

Anyone using an actual wheel as a controller, in any "sim," should use fully linear steering (i.e., no fudging of the curve by the game or your own wheel control panel). Why add distortion to something so basic as steering unless you are playing a drift video game or similar?

 

By "floating," people mean that the car doesn't feel connected to the road. You can reduce or mask this with setups on the car, with camera angles, or with FFB. But with certain cars, it will remain, and in my opinion, that's due to the physics of Ams2. It can be improved, as the comparison of GT3 to GTE, for example, shows. GTE are significantly more connected than GT3. Maybe not everyone can feel it, and maybe not everyone wants to feel it. But it's definitely there.

 

For sure. I can only speak for myself, and I expressed just that. With the same car, the same wheel amplitude (i.e., 540º on a GT3), and default setups, another sim can manifest better ‘directional control’ than AMS2. The action around the center is one that stands out tall in that landscape.

I’m very results-driven - anything that resembles reality to a higher degree is, by definition, ‘better.’ Sim racing is different from driving a real car: a true simulation would require modeling the exact same number of real particles in quantum space… everything else is just a shortcut. AMS2/ME handles graphical shortcuts quite well and is considered a benchmark in that regard.

But how do I know if the slider at 67.5% is actually more linear than at 50%? How do I know how AMS2/ME implements this? And how do I know if other sims operate in the same ‘shortcut’ arena - perhaps applying a curve that feels closer to rack forces but is mathematically inaccurate and further from reality?

I use a formula wheel, and I can think of a few reasons to set the slider above 50% - but not when driving formula cars.

 

Are GT3's in turn also more connected than GT4's? (Yes, IMO.)

Does the Advanced Trainer feel more connected to the road than the basic one?

I hope so, because that's how they should/would feel in real life based on their tires and design of the cars.

 

Agree that this is an issue--the UI in many games does not help to find or identify the linear/neutral setting that allows the game to be played as the designers intended (without external distortion).

Even worse, trying to find a neutral, distortion-free setting on a DD wheel control panel. I want to feel the FFB from the game, not some filtered or boosted generic settings provided by the wheel manufacturer. At a minimum, there should be an easy way to one-click to get rid of all that so you can test it and then add-in anything if you are not happy with the "real" FFB provided by the game.

 

I think both positions are correct. Marc is right; the goal should be to remain as linear and neutral as possible, like in other simulations. Joachim is right, because sometimes this linearity or neutrality feels wrong, and who says it's really neutral? The best example is the new VNM bases; forget everything you previously thought was right about DDS. Here, there's linearity, but there's also the opposite. And it's precisely this mix that makes it extraordinary and feels right, or closer to realism. I think we should learn to think outside the box without explicitly insisting on one thing.

 

Unfortunately, I have to disagree with you here. All bases have filtering, and all bases have filters, and it can even be beneficial. I think you should reconsider your preconceptions here, because it doesn't work that way anymore. Software and hardware have changed a lot, and so have the settings. A Simucube, for example, has a reconstruction filter, the most important control for this base. And there are other examples of filters and their effectiveness.

 

I honestly don't think everybody who uses that term about AMS2 means that exact same thing by it.

I personally don't get that kind of "not connected to road" feeling, even from the GT3. There is something different there though for sure, but my brain interpretes as somekind of "flex" or "play", which would maybe be combination of tire flex, chassis and suspension movement. I tend to get this kind of "soft" feel (I guess some call it "boaty"), which I think is partially related on how the suspension works and is often setup (very different from something like rF2/LMU and ACC)

Then there is that what I feel is very big window where you can have this neutral steering kind of behavior, that is definitely different compared to the sims I have played. Usually the faster the player, the more obvious that is.

GTE are definitely more connected and less "boaty". Not sure is it the tires (wider as well), higher aero, less weight, but there there is definitely less of that "play" there, even in slow corners where the aero is not a factor.

 

I get this feeling of flying very strongly in GT3. And I'm not the only one who has this "problem." Try playing a different sim and then switching back to Ams2. Then you'll become even more aware of this feeling. But as I said, there are other great cars and combinations for me, which is why GT3 is less important to me. This is just my personal opinion of these cars, nothing more and nothing less.

 

But for those you love gt3. It very important. And there are not really other game with imsa a d 70 tracks to drive in multiclass.

 

My best bet/hypothesis is that the issue is related to car/world rendering in cockpit view, since the car seems to pivot around the driver’s seat (the ‘center wheel sensation,’ as some call it). But cars aren’t obligated to rotate around us (egocentric simracers) .
External views show no evidence of disconnection, so I conclude that the problem is not related to car physics but to the graphics engine itself.

 

I'm not saying DD wheels don't need filters and settings designed to mimic a mechanical (real world) wheel--of course they do because without them they feel nothing like reality. I am saying that a "neutral" or base line setting that minimally interferes with the FFB in a game should be a single-click option, or the default setting.