Camera position changes car behavior

[mirial]

Good post, and I've given different opinions on the cams in the past.

Bumper view has poorest visibility, but offers the greatest indication that the car is about lose rear end grip by the fact that it is lowest of all and also gives you clear visibility of the track in the foreground and you able to monitor any change in slip angle more accurately than being positioned higher because your line of sight that is fixed to points of reference down the road is closer to the horizontal than the other views. I know many don't consider it real, but in reality, the dash and pillars are barely in the foreground of your peripheral vision anyway when you are looking far ahead down the track, so I consider it the most realistic and best compromise out of all the views.

Quite a few of the top guys use roof cam and I've seen that they make very few mistakes, quite likely because visibility is excellent, but feedback in terms of feeling the rear end is my opinion, compromised slightly.

I know many use chase cam, but I think it's more because they prefer watching the car do skids and all that, so you can't possibly expect to get fast consistent lap times if the car is the central focus which is a common mistake beginners make.

Probably the correct way is to just focus on the track ahead with the car in your peripheral vision.

 

[Nando deBem]

Not directly, but we don't know what they did with the camera/control.

There's a few methods for doing cameras in a game like this.

The first is what we assume they did, which is to mount cameras to the car (or "actor" if this were an action game) and we control the actor. The car has it's physics and we "drive" the car. The cameras get a bunch of parameters to smooth out their movement.

When it comes to games though, this type of control looks odd. The actor doesn't move as naturally so for action games they started implementing a method where the player controls a camera and the actor follows along.

IF (I am saying IF) that were to happen here, we as players would direct the camera and the car would do it's thing following along and looking really natural. The car physics remain the same, but if the camera is a "target" that the car is chasing, then it's behavior following that target would change. I'm not saying that IS what is happening, but that would explain why the car's behavior is different. As a player, you wouldn't be able to easily tell the difference.

Now, that's all speculation. All I am saying is the cars are driving differently. Why? Dunno. But they are.

Please, find an open space and do donuts using different cameras while accelerating, maintaining speed and breaking. Try to minimize the variables. Same inputs, same starting points, Brake Balance, assists... Same everything you can make the same. Upload the videos and share it here.

 

[Ryan Dungey]

Took out the huracan GT3 and all I can say is in all camera views I consistently set 1:35 lap times on Dragon trail using the huracan gt3. The only main difference is visibility and roof cam is the best for that. If you know the track well enough though the camera view makes no difference. The cars are not driving differently at all.

 

[VFOURMAX1]

You do realise that he didn't say what your quote says, right?

I just can't see this being possible in such a high-quality game without a glitch or error on your part being involved.

Lol! I do not care who you are that is funny

With all of the errors, glitches and broken systems within this game saying it is not possible in such a high quality game is funny.

Sorry you do not share the humor. I wish his statement really did have more truth to it.

 

[vvise]

Reading this thread just makes me chuckle and think of the grip bug thread and the lancer abs thread. Just want to point out both of those provided the same evidence as this one.

 

[zzz_pt]

Reading this thread just makes me chuckle and think of the grip bug thread and the lancer abs thread. Just want to point out both of those provided the same evidence as this one.

Except not.

Grip bug, reported directly to PD and, from what I know, they were able to reproduce it. To fix will be the next step (hope it doesn't take too long).

ABS was confirmed too. Just try it yourself.

 

[Voodoovaj]

Took out the huracan GT3 and all I can say is in all camera views I consistently set 1:35 lap times on Dragon trail using the huracan gt3. The only main difference is visibility and roof cam is the best for that. If you know the track well enough though the camera view makes no difference. The cars are not driving differently at all.

Finally.

Ok, what are you using for control? Wheel or pad?

 

[vvise]

Except not.

Grip bug, reported directly to PD and, from what I know, they were able to reproduce it. To fix will be the next step (hope it doesn't take too long).

ABS was confirmed too. Just try it yourself.

Define confirmed? I'm consistently faster in roof cam than bumper cam. Does that confirm a roof cam bug? "Try it yourself" is anecdotal, not scientific. I'm not into snake oil. Also I'm pretty sure PD said there's no abs bug, and that everything is working as intended. I'm not convinced that careful and focused braking isn't just faster than slamming the brakes and counting on abs, maybe it's just easiest to control in the lancer and produces good results more consistently.

But no need to start an argument, that's besides my point anyways. I just thought it funny that "my lap time is better when X" is by default a bug. It's also funny to me is how quickly some people chose to discredit this, but back another "bug".

 

[zzz_pt]

Define confirmed? I'm consistently faster in roof cam than bumper cam. Does that confirm a roof cam bug? "Try it yourself" is anecdotal, not scientific. I'm not into snake oil. Also I'm pretty sure PD said there's no abs bug, and that everything is working as intended. I'm not convinced that careful and focused braking isn't just faster than slamming the brakes and counting on abs, maybe it's just easiest to control in the lancer and produces good results more consistently.

But no need to start an argument, that's besides my point anyways. I just thought it funny that "my lap time is better when X" is by default a bug. It's also funny to me is how quickly some people chose to discredit this, but back another "bug".

Confirmed as the issue was found (grip bug) and passed on to the responsible team. I don't know anything more other than that. Just waiting for it to be fixed.

The ABS is working as intended, not a bug. What I told you is that you can test it yourself.

Regarding the camera, I'll need to test it myself, probably with a car I'm more familiar with than the Lambo.

 

[Ryan Dungey]

Finally.

Ok, what are you using for control? Wheel or pad?

Controller. Felt no different to drive in any view. Stock huracan Gt3 with front brake bias at 2 and TC at 2. In each view i set lap times in the 1:35.6xx area. The only things that's different is being able to place your car better on the track and like a said if you know the track well enough that doesn't even matter.

 

[Voodoovaj]

Please, find an open space and do donuts using different cameras while accelerating, maintaining speed and breaking. Try to minimize the variables. Same inputs, same starting points, Brake Balance, assists... Same everything you can make the same. Upload the videos and share it here.

I am waiting on the upload. I did slow runs instead. On the slow movements, the differences look subtle, but you can see them. The pivots look to be in different places than I thought they were. All four look bad.

Controller. Felt no different to drive in any view. Stock huracan Gt3 with front brake bias at 2 and TC at 2. In each view i set lap times in the 1:35.6xx area. The only things that's different is being able to place your car better on the track and like a said if you know the track well enough that doesn't even matter.

It could be that I am reacting to it differently and feeding more input on the different views. After I did the video, I started wondering what the heck I was seeing.

waiting on the upload...I hate my isp at the moment

 

[eran0004]

Not directly, but we don't know what they did with the camera/control.

There's a few methods for doing cameras in a game like this.

There is really only one method: you control the car, and then the camera tags along.

You can't give a physics engine a point in space and time (by moving a camera) and tell it to figure out how to get there. You need to start by calculating the physics, and that will give you the location.

With all of the errors, glitches and broken systems within this game saying it is not possible in such a high quality game is funny.

Sorry you do not share the humor. I wish his statement really did have more truth to it.

It isn't possible. But it has nothing to do with the quality of the product, it's because games doesn't work that way. Bugs are not magical errors that appear for no reason, they are always the result of a weakness in the system. A bug cannot exist if the design doesn't allow it.

 

[Voodoovaj]

So, here's video showing the strange pivots. Notice that the car, in every camera location, pivots at the rear axle MORE OR LESS, which is ok.

First run is cockpit cam, second is roof cam, third is chase cam, last is front cam. On that last run, it could totally be that the conditions are not controlled, but I swear I see the rear end sliding back and forth more than in the others.



Maybe I am reacting poorly to the front cam and giving too much input, but I am doing it every time and I can't seem to not do it.

There is really only one method: you control the car, and then the camera tags along.

As you can see in the video, that is not always the case in a video game. The front of the car is following something, and it isn't the front tires.

 

[Nando deBem]

What just happened?

 

[Ryan Dungey]

So, here's video showing the strange pivots. Notice that the car, in every camera location, pivots at the rear axle MORE OR LESS, which is ok.

First run is cockpit cam, second is roof cam, third is chase cam, last is front cam. On that last run, it could totally be that the conditions are not controlled, but I swear I see the rear end sliding back and forth more than in the others.



Maybe I am reacting poorly to the front cam and giving too much input, but I am doing it every time and I can't seem to not do it.



As you can see in the video, that is not always the case in a video game. The front of the car is following something, and it isn't the front tires.

Watching that video I can say that the weird pivots your experiencing is all down to speeds and steering input. Definately nothing to do with different views it's all down to your steering input amount and speed along with vehicle speed. As from testing I experience no difference between all cams my laps were consistent. I put down your laptime differences to you finding different camera views easier to drive with.

 

[afrojcactair]

@Voodoovaj May not be controlled enough to know, as car is not pointed in same direction to camera. Watched several times, and at first blush thought turning followed front tires as should. Jumping back and forth for comparison, chase cam does look a little different in the video.

Can you tune first gear for a really low speed, and just hold the accelerator, start from same grid marker, so that at the start line you can turn to get from one side of track to a set fixed point. What ever the test, if you think that vid looks odd, please share.

Interested in playing with my 911 and cameras, but half the world is getting connection fail.

 

[Highlandor]

I was looking through the forum and didn't find this mentioned anywhere, so I felt it needed to brought to light. This all started for me when I saw a video of ORMA_Snow driving in roof view and I wondered if it would help my lap times. I assumed, the better visibility would help me find the line better, and it does, but it does far more than that.

Agree 100%

I think people will experience it more with cars with no aero and sport tyres.

Was trying some setups with friends on RX7 with sport mediums, all aids off, (I use chase cam), back end was sliding all over the place, changed to cockpit view and felt like a different car, went 4 secs+ a lap faster.

Difference was like driving on ice to having superglue for tyres.

It was a 6 lap race @ Nurb Gp, weather fine - did 3 laps going sideways with normal chase cam then 3 laps in cockpit view chucking the car around...

It was the same on previous GTs, cars were always more stable in cockpit view compared to chase cam. Chase cam has that horrible twitch (turning in after braking) that means you have to wait a tiny bit extra for car to settle. Worst case scenario is the sideways rotation, the car starts to slide after outside rear wheel gives up and it's gone.

 

[Nando deBem]

So... if you guys wanna test that with even more controlled variables... cockpit view, and change the POV settings. Or chase cam, and Chang it’s settings. But!
My guess is still not enough control to replicate the same inputs in different tests. One of the runs you were more than double the speed of the others.

 

[SkylineR88]

So I can literally press square to change cam to go around corners better in the middle of a race if your theory is true ?

 

[TheGeologist]

Didn't Kaz say he thought the roof view was easier, this the reason they decided not to give it a rear view camera?

 

[Highlandor]

So I can literally press square to change cam to go around corners better in the middle of a race if your theory is true ?

It'll depend on what car, tyres etc etc you're using - but essentially, yes.

The more you're suffering from rotation, the more you'd benefit by changing view.

I don't have this problem with GR3 & 4 cars as they have a huge amount of grip (tyres + aero), this amount of grip exceeds the affect of over rotation, but use a FR road car with no aero and sport tyres (mediums or less) and it becomes alot more apparent.

GT6 had a fairly easy fix, just reducing the fr tyres on a MR car by 1 compound would give the car natural balance, but that doesn't seem to work as well on GT sport (for the cars that get affected by this the most).

 

[Skiddy]

Watching that video I can say that the weird pivots your experiencing is all down to speeds and steering input. Definately nothing to do with different views it's all down to your steering input amount and speed along with vehicle speed. As from testing I experience no difference between all cams my laps were consistent. I put down your laptime differences to you finding different camera views easier to drive with.

What he said.

 

[Nando deBem]

You’re already treating this as a fact. I still haven’t seen proof of this bug or whatever.

 

[Highlandor]

You’re already treating this as a fact. I still haven’t seen proof of this bug or whatever.

Who said it was a bug?

It's just a 'difference', same as any of the other variables in the game that affect grip / stability.

Try for yourself, you need 2 things:

1) to be able to lap at the same pace in both chase cam and cockpit view i.e. to a good level - something like the equivalent of gold in licenses.
2) a scenario (car, track, tyres etc etc etc) where the over rotation exceeds the grip of the rear outer tyre.

 

[Dose555]

Bumper view has poorest visibility, but offers the greatest indication that the car is about lose rear end grip

I completely agree with this.... I use both views, chase cam probably 70% of the time, and bumper cam 30%. I find it way easier to feel the car begin to fishtail while in bumper cam vs. chase cam. I just can't get dialed in on most tracks while in bumper view though. It's so low that it's hard for me judge/see the apex on a lot of turns that sweep downhill (Brands Hatch turn 1 for ex.) I could never get dialed in with roof cam either, but maybe I'll experiment now after viewing this thread.

 

[Nando deBem]

Who said it was a bug?

It's just a 'difference', same as any of the other variables in the game that affect grip / stability.

Try for yourself, you need 2 things:

1) to be able to lap at the same pace in both chase cam and cockpit view i.e. to a good level - something like the equivalent of gold in licenses.
2) a scenario (car, track, tyres etc etc etc) where the over rotation exceeds the grip of the rear outer tyre.

I did and I already stated that.

 

[SkylineR88]

I consider this nonsense.
Until you have proof of existence, I think this is no more than a "percieved" effect.

 

[Sojiira]

I'd guess the differences felt would be determined by skill level. So the closer you get to say a top10 time the less there would be any perceived differences.

Because at the top times you should have found an optimal line and would be pushing the car to the limit at every corner. So if at one corner you can take it at 50 mph, changing your "view" shouldnt make the car be able to take it any faster as it should be close to the optimal limit of how fast you can take the corner.

If on the off chance that it does and is not based on skill, that would definitely be a bug.

 

[Voodoovaj]

Watching that video I can say that the weird pivots your experiencing is all down to speeds and steering input. Definately nothing to do with different views it's all down to your steering input amount and speed along with vehicle speed. As from testing I experience no difference between all cams my laps were consistent. I put down your laptime differences to you finding different camera views easier to drive with.

It may just be as you say.

I am doing it without the TC. I noticed that when I did it your way, with TC on, the Huracan and NSX were both equal regardless of view, but the TC was helping to keep the rear in check.

Without TC, the rear would step out more in bumper cam than cockpit and I could be more aggressive with the throttle. Again, could be perception, but it sure feels like more.


And what's with the alternating camber on the front wheels when they go from left to right? I have never seen front wheels do that? Is that something new?

 

[Tristan Jones]

It'll depend on what car, tyres etc etc you're using - but essentially, yes.

The more you're suffering from rotation, the more you'd benefit by changing view.

I don't have this problem with GR3 & 4 cars as they have a huge amount of grip (tyres + aero), this amount of grip exceeds the affect of over rotation, but use a FR road car with no aero and sport tyres (mediums or less) and it becomes alot more apparent.

GT6 had a fairly easy fix, just reducing the fr tyres on a MR car by 1 compound would give the car natural balance, but that doesn't seem to work as well on GT sport (for the cars that get affected by this the most).

This feels like a perception issue...glad you mentioned MR cars, thought noone would pick up that NSX and Huracan were both MR!

I will have to give this a shot and see if my perception is the same...