TrackIR 5 - Important observation

[MIghtymoo]

Hi all TrackIR 5 users.

Get buttery smooth tracking and smooth image if your set your fps to 60 (or 120 fps, if your system can manage )

I have been flying for two years with G-sync enabled and head tracker and have been a happy camper, but did not know what I was missing out by matching the frame rate of DCS with the frame rate of the TrackIR (120 fps). Works perfectly in (120/60/40/30 fps).

Personally I do this by setting the max frame rate to 60 in Nvidia Control Panel. 

If you own TrackIR, I highly recommend you to test it. 

Now I fly low level over Syria cities and image/headtracking is super smooth.

 

 

Edited May 7, 2023 by MIghtymoo

[SharpeXB]

I don’t think TrackIR’s sampling rate has anything to do with screen refresh rate. Any more than using your mouse or keys to move your view. It’s simply mouse-look attached to your head. Think about it this way; does your mouse in a non-TrackIR game like an FPS need to have a sample rate synced to your monitor? No. TrackIR has always seemed smooth to me.

Edited May 7, 2023 by SharpeXB

[MIghtymoo]

44 minutes ago, SharpeXB said:

I don’t think TrackIR’s sampling rate has anything to do with screen refresh rate. Any more than using your mouse or keys to move your view. It’s simply mouse-look attached to your head. Think about it this way; does your mouse in a non-TrackIR game like an FPS need to have a sample rate synced to your monitor? No. TrackIR has always seemed smooth to me.

 

If I have a frame rate of e.g 45 in DCS and the TrackIR has 120 fps, moving your head you are still getting 45 fps, but the head movement appears a tiny bit choppy/struttery as the timing of the position of your head sent to the PC is not matching the frame being drawed by the GPU. 

If you have a IR-camera with lower refresh rate this is more noticable. 

You have to try it out to experience it for yourself…TrackIR goes from goes from good (with variable refresh rate on the monitor) to super smooth (when locking the frame rate either (120/60/40/30)

Hope someone with better tech skills can explain this in a better way.

 

Edited May 7, 2023 by MIghtymoo

[Hiob]

1 hour ago, MIghtymoo said:

Hi all TrackIR 5 users.

Get buttery smooth tracking and smooth image if your set your fps to 60 (or 120 fps, if your system can manage )

I have been flying for two years with G-sync enabled and head tracker and have been a happy camper, but did not know what I was missing out by matching the frame rate of DCS with the frame rate of the TrackIR (120 fps). Works perfectly in (120/60/40/30 fps).

Personally I do this by setting the max frame rate to 60 in Nvidia Control Panel. 

If you own TrackIR, I highly recommend you to test it. 

Now I fly low level over Syria cities and image/headtracking is super smooth.

 

 

 

You're late to the party. This topic has been discussed countless times here. Some propose the same as you others think it is snake oil. Personally I think the latter.

[MIghtymoo]

Better late than never, though

After flying for hundreds of hours in DCS, I believe that «snake oil» when I saw the difference with my own eyes tonight.

Does not cost a penny to test for yourself…

 

[SharpeXB]

1 hour ago, MIghtymoo said:

If I have a frame rate of e.g 45 in DCS and the TrackIR has 120 fps, moving your head you are still getting 45 fps, but the head movement appears a tiny bit choppy/struttery as the timing of the position of your head sent to the PC is not matching the frame being drawed by the GPU. 

If you have a IR-camera with lower refresh rate this is more noticable. 

You have to try it out to experience it for yourself…TrackIR goes from goes from good (with variable refresh rate on the monitor) to super smooth (when locking the frame rate either (120/60/40/30)

Hope someone with better tech skills can explain this in a better way.

 

 

TrackIR isn’t doing anything different that a mouse or keyboard would do to change your view. Do those need to sync their polling rate to your monitor? No. Imagine how crappy a mouse would be with a 60Hz polling rate. 
Input devices like these don’t have frame rates, they simply send inputs.

Edited May 7, 2023 by SharpeXB

[Py]

5 hours ago, SharpeXB said:

Input devices like these don’t have frame rates, they simply send inputs.

... at a certain USB polling rate.
If you've ever used an extremely crappy mouse, it's actually quite noticable. Decent ones use a polling rate of 500 Hz or higher so it feels smooth regardless of the montor refresh rate.

In regards to trackIR, if it's using raw position data then yes syncing it to the monitor refresh rate will eliminate inconsistant update timing due to aliasing, but personally I haven't noticed a difference using G-sync.

If the trackIR software is doing things properly then it should be interpolating/extrapolating the instantaneous position at the frame time and this won't be an issue (but I have my doubts).

[SharpeXB]

1 hour ago, Py said:

... at a certain USB polling rate.
If you've ever used an extremely crappy mouse, it's actually quite noticable. Decent ones use a polling rate of 500 Hz or higher so it feels smooth regardless of the montor refresh rate.

In regards to trackIR, if it's using raw position data then yes syncing it to the monitor refresh rate will eliminate inconsistant update timing due to aliasing, but personally I haven't noticed a difference using G-sync.

If the trackIR software is doing things properly then it should be interpolating/extrapolating the instantaneous position at the frame time and this won't be an issue (but I have my doubts).

I’ve never noticed TrackIR to stutter or out of sync. This all seems hypothetical. Currently I’m using a 120Hz G-Sync monitor and it seems just fine. Other sims like MSFS I had once set limited to 30Hz and TrackIR didn’t seem better or worse there either. 

Edited May 8, 2023 by SharpeXB

[MIghtymoo]

I guess we have different experiences…

I have a G-sync monitor running from 20-100 Hz (Asus Predator 34 inch). 

Have fun flying and see you in the skies.

[Hiob]

11 hours ago, MIghtymoo said:

Better late than never, though

After flying for hundreds of hours in DCS, I believe that «snake oil» when I saw the difference with my own eyes tonight.

Does not cost a penny to test for yourself…

 

I did - extensively. But as I said, no point in oppening yet another discussion about the same topic.

[SharpeXB]

If this theory was true it would be impossible to get smooth gameplay using G-Sync without a variable refresh rate mouse and keyboard. Yet I don’t see that anyone makes those. Nor are there G-Sync joysticks and game pads. I’m sure Nvidia would be happy to sell those but they don’t appear to. 
So I’m going to call this one busted. 

[unlikely_spider]

1 hour ago, SharpeXB said:

If this theory was true it would be impossible to get smooth gameplay using G-Sync without a variable refresh rate mouse and keyboard. Yet I don’t see that anyone makes those. Nor are there G-Sync joysticks and game pads. I’m sure Nvidia would be happy to sell those but they don’t appear to. 
So I’m going to call this one busted. 

Well hopefully your mouse refresh rate is much higher than your monitor refresh rate. Here's an article that explains the relationship between the two and how stutter can be introduced by input devices:

https://blurbusters.com/faq/mouse-guide/

[MIghtymoo]

Great to get some supportive science into this thread. Much appreciated.

[SharpeXB]

Testing this for myself. I think 89-92ish fps G-Sync looks better than a fixed 60. I don’t think it has much to do with TrackIR, just that the higher frame rate looks better. My test is to turn my head from side to side and look for any blur or trails from the white letters on the Hornet UFC. I’m using what’s probably the highest spec monitor today too, a 120 Hz OLED with a 1ms respond rate. So if there’s anything to see I would notice it. 

[MadKreator]

I’ll have to admit, after using track IR for quite a while, pushing as much fps as I could( usually in the 90’s) with 120hz monitor. After I started using Helios and capped my fps at 60 so the helios gauges run smoothly, I really did feel like it smoothed track IR out. Don’t get me wrong it’s always been fast and snappy, never laggy or anything, but I swear on my life I noticed a difference as well Equally as smooth when running capped and solid at 120 fps. Anything in between seems to give slightly more small jitters. Maybe its true on some setups, maybe a placebo, i don’t really know

[silverdevil]

On 5/7/2023 at 2:24 PM, MIghtymoo said:

Hi all TrackIR 5 users.

Get buttery smooth tracking and smooth image if your set your fps to 60 (or 120 fps, if your system can manage )

I have been flying for two years with G-sync enabled and head tracker and have been a happy camper, but did not know what I was missing out by matching the frame rate of DCS with the frame rate of the TrackIR (120 fps). Works perfectly in (120/60/40/30 fps).

Personally I do this by setting the max frame rate to 60 in Nvidia Control Panel. 

If you own TrackIR, I highly recommend you to test it. 

Now I fly low level over Syria cities and image/headtracking is super smooth.

hello mightymoo. i too have gsync enabled and max FPS at 120 in the NCP. i do not notice any stutters using TIR. i also get over 90 FPS minimum on any map. even if your tip is a placebo effect for you, its an effect nonetheless. if you think it helps, then it really does not matter what anyone else thinks. some users on the forum will appreciate at least the suggestion. it costs nothing to try it.

Quote

“The road to success is paved with doubters and haters.”
― Matshona Dhliwayo

 

Edited May 10, 2023 by silverdevil

[SharpeXB]

1 hour ago, silverdevil said:

it costs nothing to try it.

Except time…

[1BRAVO9]

It could be just a case of confirmation bias.

And I'm not being insulting or critical.

Everyone is subject to it, it's just human nature.

[some1]

On 5/8/2023 at 2:52 PM, SharpeXB said:

If this theory was true it would be impossible to get smooth gameplay using G-Sync without a variable refresh rate mouse and keyboard. Yet I don’t see that anyone makes those. Nor are there G-Sync joysticks and game pads. I’m sure Nvidia would be happy to sell those but they don’t appear to. 
So I’m going to call this one busted. 

You don't need gsync in your mouse, but you need high enough mouse sampling rate. If you have a gaming mouse with adjustable sampling rate and try to use low sampling rate (like 125 Hz) on a 120 Hz screen, you'll notice that your mouse control is also slightly stuttering. Even moving windows on the desktop is not exactly as "smooth" as with 500Hz or 1000Hz polling rate.

And it's the same with Trackir. 120 Hz sampling rate of the camera is actually not that much and with variable framerate above 60 FPS range you sometimes get two consecutive game frames without head position update. That's what causes perceived stutter. Of course some people may find it more noticeable than others. 

For keyboards and joysticks this issue is irrelevant, because they don't work in absolute coordinates like mice and Trackir do. But I've recorded a similar phenomenon in a different game once. Here you can see the game in slow motion at 2 FPS, but the camera does not update it's position every game frame, or at even intervals. When played at full speed, this creates a weird effect where the game itself is smooth, but the camera movement is not. 

 

Edited May 10, 2023 by some1

[SharpeXB]

10 minutes ago, some1 said:

You don't need gsync in your mouse, but you need high enough mouse sampling rate. If you have a gaming mouse with adjustable sampling rate and try to use low sampling rate (like 125 Hz) on a 120 Hz screen, you'll notice that your mouse control is also slightly stuttering. Even moving windows on the desktop is not exactly as "smooth" as with 500Hz or 1000Hz polling rate.

And it's the same with Trackir. 120 Hz sampling rate of the camera is actually not that much and with variable framerate above 60 FPS range you sometimes get two consecutive game frames without head position update. That's what causes perceived stutter. Of course some people may find it more noticeable than others. 

For keyboards and joysticks this issue is irrelevant, because they don't work in absolute coordinates like mice and Trackir do. But I've recorded a similar phenomenon in a different game once. Here you can see the game in slow motion at 2 FPS, but the camera does not update it's position every game frame, or at even intervals. When played at full speed, this creates a weird effect where the game itself is smooth, but the camera movement is not. 

 

 

Technically this makes sense. In reality when testing this on my system I see no difference or see that the high 90ish frame rate looks better. It wouldn’t be worthwhile IMO to turn down your frame rate to 30 in a game where you’re getting 45-55fps. I’ve tried locking my frame rate to 30 in MSFS since it still taxes hardware like mine. But a higher 40-50ish rate still looks better. I think it’s possible TrackIR would only really skip frames a lot if you tried running the game above 120fps. 

[some1]

I don't play DCS with trackir, but it does skip a lot of frames in MSFS if the framerate is above 60. However, with DLSS3 Frame Generator it becomes smooth at any FPS up to 120, because then the actual game works below 60 FPS and the other half of the frames come from GPU itself. The change in Trackir smoothness is quite noticeable there when you turn on frame generation, but that requires one of the latest GPUs.

[SharpeXB]

I gave this another try. This time looking specifically at the edge of the canopy bow for any motion blur as well as how smoothly the view is panning side to side. 60Hz definitely shows more motion blur compared to 120Hz although this is as expected. 60 even shows a slight hiccup in smoothness. 120Hz looks better overall with less motion blur and no pronounced stutter. Then comparing that with a 1000 Hz mouse using mouse look. There’s no discernible difference that I can see between TrackIR and the high rate mouse. So I don’t see any smoothing difference between 60 and 120 that’s worth dropping the frame rate for. I’m using NCP all default settings Vsync in DCS set to on. G-Sync on, no frame rate lock, Low Latency to Ultra 

[silverdevil]

1 hour ago, SharpeXB said:

I gave this another try. This time looking specifically at the edge of the canopy bow for any motion blur as well as how smoothly the view is panning side to side. 60Hz definitely shows more motion blur compared to 120Hz although this is as expected. 60 even shows a slight hiccup in smoothness. 120Hz looks better overall with less motion blur and no pronounced stutter. Then comparing that with a 1000 Hz mouse using mouse look. There’s no discernible difference that I can see between TrackIR and the high rate mouse. So I don’t see any smoothing difference between 60 and 120 that’s worth dropping the frame rate for. I’m using NCP all default settings Vsync in DCS set to on. G-Sync on, no frame rate lock, Low Latency to Ultra 

interesting. i always thought one should use gsync OR vsync. but i did read a few things about it and there are a lot of things to read.

gsync obviously will only work with a monitor that supports it. vsync is essentially a frame limiter. one post i read suggested enable gsync and vsync (application controlled) in NCP. AMD (freesync) user would have something similar in the graphics driver control panel. limit the FPS to something lower than the monitor refresh rate (maybe even a few FPS below). then in game (DCS) enable vsync.

with that said. in my setup i have all of this except i do not have vsync enabled in DCS. like i had said, i do not have stutters at all. perhaps in my case vsync enabled in DCS is irrelevant. either way i will take a look and maybe get comparison videos of the two.

[SharpeXB]

30 minutes ago, silverdevil said:

interesting. i always thought one should use gsync OR vsync. but i did read a few things about it and there are a lot of things to read.

gsync obviously will only work with a monitor that supports it. vsync is essentially a frame limiter. one post i read suggested enable gsync and vsync (application controlled) in NCP. AMD (freesync) user would have something similar in the graphics driver control panel. limit the FPS to something lower than the monitor refresh rate (maybe even a few FPS below). then in game (DCS) enable vsync.

with that said. in my setup i have all of this except i do not have vsync enabled in DCS. like i had said, i do not have stutters at all. perhaps in my case vsync enabled in DCS is irrelevant. either way i will take a look and maybe get comparison videos of the two.

What I understand is setting V-Sync to off will disable G-Sync if the refresh rate goes higher than your monitors rate. Some people (super competitive twitch gamers) prefer this for reduced latency and don’t care about tearing. Getting above 120Hz in DCS isn’t likely to happen nor is the latency thing an issue. So having V-Sync on serves as a frame rate limiter. Seems to work ok either way for me. 
https://www.nvidia.com/en-us/geforce/news/g-sync-gets-even-better/

Before I was using this G-Sync monitor and an older PC I found Fast V-Sync in NCP worked really well with DCS at lower frame rates, 40-50ish. On this test above I’m currently running ~100fps. 
 

PS it goes without saying that DCS or any game looks super at 4K 100+ FPS. So taking this away for an indiscernible benefit (or in this case actually worse) isn’t a good trade IMO. 

Edited May 11, 2023 by SharpeXB

[SharpeXB]

Having experimented with both Vsync On and Off using G-Sync I came to the conclusion that Vsync Off is best, indeed that seems to be the consensus on how to use this. Vsync adds latency and you simply don’t need it with G-Sync. You do need to cap your refresh rate 2-3 frames below your monitor’s rate otherwise you’ll get tearing if your FPS exceeds that. I gotta say the results in DCS are fantastic. Very nice and smooth anywhere between 50 and 118 (my screen limit). 
I have the NCP options all on default which means Low Latency and Triple Buffering Off, since those are meant to improve Vsync performance they aren’t needed. I capped the frame rate in NCP since it allows fine tuning as opposed to the in-game one (that feature might have problems too) Everything just looks super nice and smooth. I had tried capping FPS at 60 but the drop in frame rate is noticeable without any real improvement I can see.