FOV and Input Lag/Latency

[Bobby Pennington]

I just made a video of what using the correct FOV per all the calculators and the math and how it can destroy your experience. I've been chasing my tail with my G27 settings and it turns out the biggest problem was the stupid FOV.
Visit my boobtube page MulliganF1 for a full description of my setup and explanation.
This is the only way I can give back to the Sim community since I don't make skins, tracks etc., but I'm working on it.
Please let me know if the video helps. Have a great night and as always Opposite Lock to 'ya.

P.S.: I hope I posted this in the correct section. Still learning.

 

[Brandon Wright]

Ok, here's my video. Of course this is one of those things that's hard to illustrate with video, but watch my steering inputs. At first I start with what would be a high FOV and you can see I use a lot of steering input, at times almost turning the wheel 180º. Next I switch to an average FOV, close to what my "mathematically correct" FOV would be and you can see there's a bit less steering input, most of the time in the ~90-120º range. Finally I switch to what would be a low FOV and my steering inputs are even less, rarely going much past 90º and some turns using as little as ~45º. I tried to use the same braking points to keep things as consistent as possible and the only setting I changed was the distance spanner in the triple screen app which is the setting for the distance between your eyes and the center monitor (first setting was 300, second was 600, third was 1050).

I'm not saying this changes the physics or anything, not real sure what I'm saying other than different FOVs do have quite an impact on how the car controls/behaves and I don't believe that a simple perception difference would result in such changes to my steering inputs (just my opinion, of course). And of course what you can't see on the video is the difference in how my wheel feels, with a high FOV it's spongy and a bit lethargic but with a low FOV it's tight, crisp, and twitchy, it's not a drastic change but it's noticeable.

 

[Wiz]

@Brandon Wright Video is private

 

[Brandon Wright]

@Brandon Wright Video is private

Odd. Try it now.

 

[Rodent]

Always get jelly when I see your setup Brandon. Is that Z1 dashboard you running on the wheel? Really considering a small screen for it though why in the world they don't make an android/iphone app is baffling.

Also, video no longer private.

 

[Brandon Wright]

Always get jelly when I see your setup Brandon. Is that Z1 dashboard you running on the wheel? Really considering a small screen for it though why in the world they don't make an android/iphone app is baffling.

Also, video no longer private.

Thanks. For the screen above the wheel it's just an iOS app called DashPi running on an old iPhone6. I hope to get Z1 someday but this works fine for now.

 

[Rodent]

Dang, my old 3gs is too old for it, and my androids too big to fit behind the wheel. Heh.

Also, to stay on topic, tried going from 30 to 35/40 in fov and hated it. Once you get used to a low fov it's just as hard or harder to go back it seems.

 

[Bobby Pennington]

Ok, here's my video. Of course this is one of those things that's hard to illustrate with video, but watch my steering inputs. At first I start with what would be a high FOV and you can see I use a lot of steering input, at times almost turning the wheel 180º. Next I switch to an average FOV, close to what my "mathematically correct" FOV would be and you can see there's a bit less steering input, most of the time in the ~90-120º range. Finally I switch to what would be a low FOV and my steering inputs are even less, rarely going much past 90º and some turns using as little as ~45º. I tried to use the same braking points to keep things as consistent as possible and the only setting I changed was the distance spanner in the triple screen app which is the setting for the distance between your eyes and the center monitor (first setting was 300, second was 600, third was 1050).

I'm not saying this changes the physics or anything, not real sure what I'm saying other than different FOVs do have quite an impact on how the car controls/behaves and I don't believe that a simple perception difference would result in such changes to my steering inputs (just my opinion, of course). And of course what you can't see on the video is the difference in how my wheel feels, with a high FOV it's spongy and a bit lethargic but with a low FOV it's tight, crisp, and twitchy, it's not a drastic change but it's noticeable.

Brandon thanks for making the video. It's easy to see. Seems some of the guys get it and others get caught up in the semantics.
When I ran to high of an FOV I had to work hard to overdrive the car because, like you said, it felt like an old Buick. The past year and a half to 2 years I've been using 33 fov and I could easily overdrive the car if I wasn't careful. With it set right now turning lap after lap is a breeze.
Anywho, my whole point in making the video was to possibly help someone out. Glad you understand the concept and appreciate the support.
As Mario Andretti says “If everything's under control, you're going too slow.”
Have a great day.

 

[BhZ]

Ok, here's my video. Of course this is one of those things that's hard to illustrate with video, but watch my steering inputs. At first I start with what would be a high FOV and you can see I use a lot of steering input, at times almost turning the wheel 180º. Next I switch to an average FOV, close to what my "mathematically correct" FOV would be and you can see there's a bit less steering input, most of the time in the ~90-120º range. Finally I switch to what would be a low FOV and my steering inputs are even less, rarely going much past 90º and some turns using as little as ~45º. I tried to use the same braking points to keep things as consistent as possible and the only setting I changed was the distance spanner in the triple screen app which is the setting for the distance between your eyes and the center monitor (first setting was 300, second was 600, third was 1050).

I'm not saying this changes the physics or anything, not real sure what I'm saying other than different FOVs do have quite an impact on how the car controls/behaves and I don't believe that a simple perception difference would result in such changes to my steering inputs (just my opinion, of course). And of course what you can't see on the video is the difference in how my wheel feels, with a high FOV it's spongy and a bit lethargic but with a low FOV it's tight, crisp, and twitchy, it's not a drastic change but it's noticeable.

It's just your perception. Speed perception is different as well as the movements of the car. The FFB is the same, it just change the way you see things and the way you react based on that. Why would the devs do such thing? And again: "FOVs do have quite an impact on how the car controls/behaves" isn't this about physics?

 

[Ian Cameron]

Ok, here's my video. Of course this is one of those things that's hard to illustrate with video, but watch my steering inputs. At first I start with what would be a high FOV and you can see I use a lot of steering input, at times almost turning the wheel 180º. Next I switch to an average FOV, close to what my "mathematically correct" FOV would be and you can see there's a bit less steering input, most of the time in the ~90-120º range. Finally I switch to what would be a low FOV and my steering inputs are even less, rarely going much past 90º and some turns using as little as ~45º. I tried to use the same braking points to keep things as consistent as possible and the only setting I changed was the distance spanner in the triple screen app which is the setting for the distance between your eyes and the center monitor (first setting was 300, second was 600, third was 1050).

I'm not saying this changes the physics or anything, not real sure what I'm saying other than different FOVs do have quite an impact on how the car controls/behaves and I don't believe that a simple perception difference would result in such changes to my steering inputs (just my opinion, of course). And of course what you can't see on the video is the difference in how my wheel feels, with a high FOV it's spongy and a bit lethargic but with a low FOV it's tight, crisp, and twitchy, it's not a drastic change but it's noticeable.

So just to clarify, which are you saying is the best FOV correct FOV at around 600mm for triple screens or the 1050 setting.
I have 3x 27" and my correct distance is 620mm which I believe is similar to yours, but in view of what your saying I have increased mine to 820mm which makes a big difference overall. 1050 is too twitchy. Still not sure which I should use.

 

[Brandon Wright]

It's just your perception. Speed perception is different as well as the movements of the car. The FFB is the same, it just change the way you see things and the way you react based on that. Why would the devs do such thing? And again: "FOVs do have quite an impact on how the car controls/behaves" isn't this about physics?

You're right, the dramatic change in steering inputs and how my wheel feels was clearly just all my imagination.

So just to clarify, which are you saying is the best FOV correct FOV at around 600mm for triple screens or the 1050 setting.
I have 3x 27" and my correct distance is 620mm which I believe is similar to yours, but in view of what your saying I have increased mine to 820mm which makes a big difference overall. 1050 is too twitchy. Still not sure which I should use.

I find the 600-700 range to be the best on my setup. It was rather difficult to drive with the first and last options and with the 1050 setting it actually almost made me a bit nauseous. If it's too twitchy reduce it, if it's too sluggish increase it.

 

[Ian Cameron]

You're right, the dramatic change in steering inputs and how my wheel feels was clearly just all my imagination.



I find the 600-700 range to be the best on my setup. It was rather difficult to drive with the first and last options and with the 1050 setting it actually almost made me a bit nauseous. If it's too twitchy reduce it, if it's too sluggish increase it.

Yes think your right been spending so much effort getting FOV mathematically correct, never occurred to me to lower it further. As I said my correct setting is 620mm I increased it to around 740mm and after two laps shaved 3 seconds off my best time. Hard to explain but for me it has made situational awareness better, I can brake later and not loose control in the corners. certainly for me it works, just a matter of fine tuning I guess.

 

[PhilS13]

I'm not saying this changes the physics or anything, not real sure what I'm saying other than different FOVs do have quite an impact on how the car controls/behaves and I don't believe that a simple perception difference would result in such changes to my steering inputs (just my opinion, of course). And of course what you can't see on the video is the difference in how my wheel feels, with a high FOV it's spongy and a bit lethargic but with a low FOV it's tight, crisp, and twitchy

Yes you are saying this changes the physics. And it doesn't.

It's not a semantic issue, it's not some guys "getting it". Your perceptions are being messed up with and you are failing to realize that it's the only thing going on in there.

 

[Bobby Pennington]

It's just your perception. Speed perception is different as well as the movements of the car. The FFB is the same, it just change the way you see things and the way you react based on that. Why would the devs do such thing? And again: "FOVs do have quite an impact on how the car controls/behaves" isn't this about physics?

BhZ, no it doesn't change the physics at all. It does, however, change how we react and the steering wheel behaves in our hands. To close and it feels very "Twitchy" and to far away you get tunnel vision or it's like driving on a sidewalk and it can feel a little "sluggish". Get it right and all is right with the world. No one number will fit us all. Yes I did get comfortable with my FOV at 33, but I had to turn my wheel sensitivity down to 20% in Logitech Profiler and I still had problems with over driving certain cars. Just experiment and see what feels good to you.
I'll make another shorter video with the fov to close, just right and to far away and you can see how it effects the wheel and how smoothly or not I am with it.

 

[Bobby Pennington]

Ok since I opened this can of worms here's one last video. I turned the visible wheel and arms on so you can see the difference in how jerky or smooth my inputs are. Maybe this will translate better. Remember, I never said I was right, just wanted to share with the community.
Video start "Twitchy"
1:49 Normal
3:29 Ok but tunnel vision and driving on a sidewalk.

 

[TrevorGreenfield]

There is definitely a difference in how you drive based on FOV.
You guys are trying to make this all sound so simple, just set it to what the calculator gives you. Which discludes the other variables at play.

The physical reality of trying to get the car as close to real-life proximity as possible to emulate the exact view you would have as the race car driver, works in theory.

In real life, the variables are MANY.

1. Triple screen vs Single - This alone can be a reason why the FOV calculators don't work so well. If you have a small single screen, you lose so much peripheral vision due to the narrow FOV angle, it makes most racing very challenging, even when just hotlapping. That peripheral vision makes it more comfortable to drive, easier to look ahead into corners and easier to judge traffic on the road especially next to the car. If you have a larger single screen (look at 4k 65" videos on YT), this is far less of an issue.
2. How close the monitors are - While the FOV compensates for this, there is simply no way to illustrate - the further you physically move the monitors away from yourself, the more you are separating yourself from the racecar, and proportionally the more you will have to make up with lower FOV angle. Hence, a triple screen at a reasonable distance may be fine to race with but a single screen will have to have a 3x lower FOV to achieve the same from the same distance. At a certain point, the FOV becomes so low and exclusionary, most single screens are simply unusable unless they are VERY close to your face. Closer than recommended viewing distances by manufacturers certainly.
3. Correct FOV for your setup is one thing, correct FOV for your brain, which is trying to replicate reality, is another. Your brain wants to see more like when you're in a real car. The farther disconnected from this you are, the more your brain must adapt to this alternate reality. This makes it more difficult to feel connected. When you find the right FOV compromise for the setup, you feel more connected.
4. FOV is indeed a ratio. Not surprisingly, the steering device also has a ratio. If I change what I see on screen but don't change the steering wheel, it is definitely going to make a difference in how I drive the car. The more zoomed into the view I am (lower FOV), but the less I feel connected to the rest of the racing environment, The more twitchy and immediate the car feels, which I might react to. The more connected I feel to the environment, the less intensely I will need to react to the car because it becomes more about challenging the corner as a whole, which leads to more smooth lines and aggressive planning of setting the car for entry, etc. The perfect setup, large triple high resolution screens placed close to the driver, or a large single curved screen, etc. is obviously only afforded to those willing to part with serious money. It would seem to me too that physically, if my FOV is 50 degrees and my wheel has a 540 degree ratio, that if I set my FOV to 35 degrees or 90 degrees, that is simply going to cause a change in the amount of input I am required to make. Maybe someone can show me that this isn't the case.
5. I didn't want to make the above point even longer, but at times I will race with a dual stick gamepad. This is where the biggest difference is most visible. I simply lack the ability with my thumbs and the small 180 degrees of movement on the stick to be able to control the twitchy movements required at smaller FOVs. As you back off, higher FOV, movements get smoother and smoother and it gets easier to control the car with the controller. This is a recent discovery because I have been only racing with a wheel/pedal setup for 15 years, but it really illustrates what I was trying to say on point 4.
6. VR/Rift bridges this gap, but is still limited by overall resolution and no real peripheral input. Hence, it gets you closer and does some other basic trickery to get you closer to being in the car (it dissolves the gap to the screen!), but you still have an inescapable tunnel vision of sorts, and you need to physically move your head instead of just flicking your eyes.

 

[Bobby Pennington]

There is definitely a difference in how you drive based on FOV.
You guys are trying to make this all sound so simple, just set it to what the calculator gives you. Which discludes the other variables at play.

The physical reality of trying to get the car as close to real-life proximity as possible to emulate the exact view you would have as the race car driver, works in theory.

In real life, the variables are MANY.

1. Triple screen vs Single - This alone can be a reason why the FOV calculators don't work so well. If you have a small single screen, you lose so much peripheral vision due to the narrow FOV angle, it makes most racing very challenging, even when just hotlapping. That peripheral vision makes it more comfortable to drive, easier to look ahead into corners and easier to judge traffic on the road especially next to the car. If you have a larger single screen (look at 4k 65" videos on YT), this is far less of an issue.
2. How close the monitors are - While the FOV compensates for this, there is simply no way to illustrate - the further you physically move the monitors away from yourself, the more you are separating yourself from the racecar, and proportionally the more you will have to make up with lower FOV angle. Hence, a triple screen at a reasonable distance may be fine to race with but a single screen will have to have a 3x lower FOV to achieve the same from the same distance. At a certain point, the FOV becomes so low and exclusionary, most single screens are simply unusable unless they are VERY close to your face. Closer than recommended viewing distances by manufacturers certainly.
3. Correct FOV for your setup is one thing, correct FOV for your brain, which is trying to replicate reality, is another. Your brain wants to see more like when you're in a real car. The farther disconnected from this you are, the more your brain must adapt to this alternate reality. This makes it more difficult to feel connected. When you find the right FOV compromise for the setup, you feel more connected.
4. FOV is indeed a ratio. Not surprisingly, the steering device also has a ratio. If I change what I see on screen but don't change the steering wheel, it is definitely going to make a difference in how I drive the car. The more zoomed into the view I am (lower FOV), but the less I feel connected to the rest of the racing environment, The more twitchy and immediate the car feels, which I might react to. The more connected I feel to the environment, the less intensely I will need to react to the car because it becomes more about challenging the corner as a whole, which leads to more smooth lines and aggressive planning of setting the car for entry, etc. The perfect setup, large triple high resolution screens placed close to the driver, or a large single curved screen, etc. is obviously only afforded to those willing to part with serious money. It would seem to me too that physically, if my FOV is 50 degrees and my wheel has a 540 degree ratio, that if I set my FOV to 35 degrees or 90 degrees, that is simply going to cause a change in the amount of input I am required to make. Maybe someone can show me that this isn't the case.
5. I didn't want to make the above point even longer, but at times I will race with a dual stick gamepad. This is where the biggest difference is most visible. I simply lack the ability with my thumbs and the small 180 degrees of movement on the stick to be able to control the twitchy movements required at smaller FOVs. As you back off, higher FOV, movements get smoother and smoother and it gets easier to control the car with the controller. This is a recent discovery because I have been only racing with a wheel/pedal setup for 15 years, but it really illustrates what I was trying to say on point 4.
6. VR/Rift bridges this gap, but is still limited by overall resolution and no real peripheral input. Hence, it gets you closer and does some other basic trickery to get you closer to being in the car (it dissolves the gap to the screen!), but you still have an inescapable tunnel vision of sorts, and you need to physically move your head instead of just flicking your eyes.

Trevor, that is exactly why I created the post. Years and years ago I used a very high fov. Never had any issues with the controller being twitchy or over sensitive so making it around a track was pretty easy once I got my braking points down. Then when AC came out and I got a large tv to use for a monitor, I used the fov calculators and did the math. Somewhere between 28 and 35 is what I got depending on the calculator or the math. I finally settled on 33. Wheel felt overly sensitive so I started turning down my steering sensitivity. Finally at 20% I could drive the car, but it just bugged me. Then by accident I used the hood view one day just to see what it was like. Could drive it like I stole it. Went back out turned the sensitivity up to 100% on my wheel and went even faster. That was the ah ha moment. I knew I'd never use the hood view, but I had to get my cockpit fov where the wheel felt the same as hood view. That's how I arrived at where I am now and I've never been happier. If others get something out of it great. That's what I was trying to do. If others don't and are happy, well I'm happy for them. Wasn't trying to persuade anyone over to my way of thinking. Just giving them some insight into a different approach they may or may not want to try.

 

[Brandon Wright]

I think I understand the perception comments now. I did some tests at Road America with bumper cam last night thinking that removing the cockpit from my view would mean I only see the track and my perception should not be affected. But, changing my distance spanner (FOV) radically changed how the track looked. With a low value (high FOV) the track was stretched out and had a very narrow width, braking zones felt like they were twice as long. But everything felt like it had been stretched longitudinally and squeezed laterally which made it a bit disorienting. A high value (low FOV) did the opposite, the track was super wide but felt scrunched towards me and the braking zones felt tiny, to the point that I was constantly overshooting turn 5 (slow left-hander at the bottom of the hill). So, yes, my perception is different because the track is visually different and this alters how I attack the track.

I then put it at the actual measured distance from my eyes (725mm which equates to a somewhat low FOV) which is close to what I've normally used and things felt normal but I was still overshooting turn 5 a bit and the car felt a bit twitchy. I then dialed it back to 450mm which was a happy medium between the stretched view and the scrunched view and that felt almost perfect. I was much better in the braking zones, I could take some turns in a higher gear, I had no problem at all with turn 5, and I shaved over three seconds off my best lap time. So, call it what you will, but changing the FOV can have more of an impact on the experience than I would have imagined.

 

[Cote Dazur]

I used to be a knight in shining armour always defending a calculated 1/1 FOV, at least for people with big enough and close enough screen(s)
I was driving using a 98 inch diagonal projected screen and felt very comfortable with my set-up, at least until I tried VR.
I realise now, that I am using VR, that FOV on 2D screen(s) is a compromise at best, driver should use what they fill confortable with, at the time, since most FOV user end up playing back and forth. The solution to "perfect" FOV is using VR goggles. A few second is all it takes to realize it and it feels part of the Ha! Moment the first time you start driving in it.

 

[Bobby Pennington]

I think I understand the perception comments now. I did some tests at Road America with bumper cam last night thinking that removing the cockpit from my view would mean I only see the track and my perception should not be affected. But, changing my distance spanner (FOV) radically changed how the track looked. With a low value (high FOV) the track was stretched out and had a very narrow width, braking zones felt like they were twice as long. But everything felt like it had been stretched longitudinally and squeezed laterally which made it a bit disorienting. A high value (low FOV) did the opposite, the track was super side but felt scrunched towards me and the braking zones felt tiny, to the point that I was constantly overshooting turn 5 (slow left-hander at the bottom of the hill). So, yes, my perception is different because the track is visually different and this alters how I attack the track.

I then put it at the actual measured distance from my eyes (725mm which equates to a somewhat low FOV) which is close to what I've normally used and things felt normal but I was still overshooting turn 5 a bit and the car felt a bit twitchy. I then dialed it back to 450mm which was a happy medium between the stretched view and the scrunched view and that felt almost perfect. I was much better in the braking zones, I could take some turns in a higher gear, I had no problem at all with turn 5, and I shaved over three seconds off my best lap time. So, call it what you will, but changing the FOV can have more of an impact on the experience than I would have imagined.

Brandon, your first paragraph sums this thread up perfectly. Exactly what I've been through in 20+ years of racing games and sims (over 35 years if you want to go back to arcades, Sega, oh Lord help me, and the Commodore 64). Yes, I'm that old lol. If I ever get triple screens I'm going to hit you up for some advice on a starting point. Have a good one.

 

[BhZ]

BhZ, no it doesn't change the physics at all. It does, however, change how we react and the steering wheel behaves in our hands. To close and it feels very "Twitchy" and to far away you get tunnel vision or it's like driving on a sidewalk and it can feel a little "sluggish". Get it right and all is right with the world. No one number will fit us all. Yes I did get comfortable with my FOV at 33, but I had to turn my wheel sensitivity down to 20% in Logitech Profiler and I still had problems with over driving certain cars. Just experiment and see what feels good to you.
I'll make another shorter video with the fov to close, just right and to far away and you can see how it effects the wheel and how smoothly or not I am with it.

It's here that you are wrong. The controller doesn't change at all. It has no reason to. It seems different to you because of the visual perception, but in reality it's always the same. If you don't want to accept this reality, just ask to any programmer if it's possible and why a dev would do that. It's like saying "i changed the resolution in the options and now the car is faster". No way the FOV changes your ffb or controller. I also think you are not telling us something. The kind of problems (perception problems) you are having only show up if you were used to a higher FOV and changed to a lower FOV (and viceversa ofc). In one of your comments you said you *were used* to a lower FOV, which i don't think it's true, since we can get used to almost any reasonable FOV with some practice and what you showed was an incapability to drive with your own FOV (the one you were used to). I think you were used to a higher FOV before than posting this.
I'm used to a very high FOV (the default one) and i tried to lower it to 35. It was undriveable, not because of the FFB or anything, but because every bump, turn, oversteer and oversteer were visually enhanced. So every time i tried to correct oversteer i ended up correcting it too much, spinning. This is exactly what i saw in your video, i just saw your brain getting ****ed by the NEW FOV (and not the one you are used to, as you said).
Btw, i don't think the calculated FOV is actually the correct FOV to use. I prefere to have decent proportions than a tunnel vision (my calculated fov would be 15, i said all).