kksnowbear
-
Posts
373 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Posts posted by kksnowbear
-
-
12 minutes ago, SharpeXB said:
So your seated eye level is about 44”. A desk is 30”. That leaves 14” for your monitor. Look at the cockpit thread for all the people with bigger screens than 14” high
Where is the law that says the monitor has to be on the desktop, exactly? Or that the desktop must be close to the wall? Or that one must use the reflector type headset?
9 minutes ago, MAXsenna said:And this can be solved by using IR LEDs instead of reflectors. Just sayin'
Sent from my SM-A536B using Tapatalk
Which I acknowledged way earlier. That said, the OP has the issues he does because he's not using LEDs.
-
1
-
-
15 minutes ago, LucShep said:
He uses a 55'' screen with TrackIR and seems to have no issues whatsoever.
Admittedly I haven't watched the whole thing, but again, it's missing necessary detail: Height of the monitor top, desktop/table height, level of chair, how 'tall' this guy is in the torso, whether he's using LED array (OP is not)...
...basically, the same thing I've been saying all along: What is the angle from headset to the reflectors?
-
3 minutes ago, SharpeXB said:
Then adjust the sensitivity or curve so your head doesn’t need to move so much that you lose tracking. Do you have a TrackIR or know how it works? It’s a pretty simple adjustment.
Yes, I've owned several thanks.
As I said above, of course you can adjust the sensitivity - but at a certain point this will mean it's too sensitive; approaching unrealistic (turn head 1/4th inch, view moves all the way aft).
Moreover, and again: Only necessary because camera is too far away from vertical center (too high).
-
1 minute ago, SharpeXB said:
Sure the monitor could slide down a little bit but that’s still higher than what you’re suggesting.
What I'm suggesting is that optimal is level line of eye sight aligned with the top of the monitor. Anything else may work but is not optimal. And the further you go from optimal, the more likely you'll have problems because of angles.
That's what I've said, nothing more nothing less.
-
7 minutes ago, LucShep said:
I never had an issue with usage, and have used it with a plethora of screens (32'', 46'', 55'', 43'' and now the current 50'' screen), be it sitting close or far from the screen (50 cm to 150cm distance).The problem with this as any sort of proof is that it's missing necessary detail. Most significantly, it doesn't say anything about height of the monitor, nor which sized monitor at what distance. And again, if you move away from the camera it changes the angles.
10 minutes ago, LucShep said:forcing a steeper angle of the camera (towards the head of the user) and readjusting the response curves.
As I mentioned, of course you can adjust curves and force angles, but there are limits, and no amount of curve adjustment can overcome the fact that, with the camera on a steep vertical angle, your head doesn't turn in the same level plane as the line of sight between camera and reflectors. At some point the light will not be reflected toward the camera any more.
-
8 minutes ago, SharpeXB said:
Yeah but that’s only feasible with a very small screen. Not a realistic solution in this case.
The option here that doesn’t involve a home improvement project is to try adjusting the response curves. The “Default” preset that TrackIR starts with isn’t actually ideal for what we are doing with this, checking six and so forth. IMO the “Smooth” preset works better as a starting point.It wouldn't require any "project" if the monitor weren't too high. That's a choice that was made when it was mounted, and not by me.
Adjusting the curves etc is really all just trying to overcome the fact that the camera's too high.
TrackIR works fine, even with larger monitors. It's not accurate to say it requires a "very small screen", that's not correct. But as with *anything* there are limits.
-
14 minutes ago, SharpeXB said:
Well that’s hardly possible with any monitor bigger than 20”
The bottom of a large display would be down at your knees.
How about try for the center of the screen at eye level.
No, the convention is exactly as I described; top of the monitor or slightly below. That's been the standard forever and it's all over online. Go check. Even better, cite a reputable source that agrees with your opinion.
In the OPs case having the monitor that high is jacking up the angles enough it gets goofy when he looks down and to the side. That's because light doesn't bend - these are reflectors, and with light, the angle of incidence equals the angle of reflection. Too steep and the light doesn't get reflected back to the camera, and the rest is history.
You can try curves in the software, but it's possible that enough of a curve will be too sensitive (these are inversely proportional).
The active LEDs in a Track Clip may work since they have lenses on the LEDs.
It remains, however, that the problem is the camera's too high. And that's because of where the monitor is.
As I mentioned above, TrackIR clearly was not intended to be used with a big monitor only two feet away. (Or maybe that's why they have the active LEDs solution).
-
2 minutes ago, maverick90 said:
in the Track IR app those 3 green dots all disappear when I look left/right and downwards.
That's because the angle is too great for light to be "seen" off the camera's axis.
-
12 minutes ago, maverick90 said:
Here, I took a pic. The blue line represents roughly where the height of my eyes would be from Track IR's sensor.
Yup. As I said, monitors too high forcing the angles (especially down/sideways) to be too steep.
Everyone can say or do what they want. But the overwhelming convention is your eyes should be aligned with the top of the monitor (maybe a tad lower).
In your picture the blue line should ideally be level with where your eyes fall, which ideally is also level in the same plane as the side-side movement when turning your head (which yes, coincidentally, will be level with the floor). If that's not practical than as close as possible. Further from that you get, more likely you're gonna have problems.
-
10 minutes ago, kksnowbear said:
Turning to the left/right looking outside of the canopy is fine though. It only happens when I turn left/right and look down towards the console that this happens.
When you turn left right *and* look down, the angle is compound relative to centerline. The hypotenuse of the triangle gets longer. You are approaching that distance/angle I described above.
-
1 hour ago, Hiob said:
TrackIR has no way of knowing if the line of sight between the Camera and sensors is tilted relative to the floor.
I never said that.
What I said was that the plane of rotation of your view side to side will be different than the plane of light (which doesn't bend) passing between camera and reflectors.
And it absolutely is. Unless, as I said, the camera and reflectors are level with the plane in which your head turns side to side.
Do I need to post a picture to prove this?
And TrackIR doesn't need to know sh*t about the level of the floor for this to happen. It happens inherently because you rotate your head side to side in a plane that is not that same as the angle of camera to reflectors, if you mount the camera too high.
And although there may be some "compensation" it is also absolute that it will get bad enough at a steep enough angle that software can't correct it. Light doesn't bend. If the reflector is off angle enough, light will not be reflected back where tge camera can see it. If it's approaching that angle, it will behave oddly.
8 minutes ago, maverick90 said:The issue I'm having is when I'm turning my head to the left/right and downward (looking towards the console on the F18) the focusing will get out of whack like it can't pick up on what I'm trying to look at, and then somehow the view defaults back to the rear and I have to kind of move my head around and try it to reset the view back to the front and center of the cockpit. I fly the F18 and it makes it almost impossible to access the side console, especially the further back part of the side consoles is where the problem begins.
Turning to the left/right looking outside of the canopy is fine though. It only happens when I turn left/right and look down towards the console that this happens.
How far from the screen are you sitting normally?
-
2
-
-
5 minutes ago, MAXsenna said:
Exactly why he need antlers
LOL you are correct of course...just not sure if he *wants* antlers
Definitely gets my vote for "most creative" solution!
-
1
-
-
Also, the OP more or less tells us he's close to the TV:
On 4/19/2024 at 2:24 PM, maverick90 said:I have the TrackIR mounted to the top of the tv and the angle to my head is kind of steep.
Further away, shallower (lesser) the angle. Closer, more steep (greater) the angle.
-
2
-
-
Strictly as a guess, the way it sounds to me is OP has a larger screen, which is (probably) closer, thus forcing the larger angles I've discussed. Tilting becomes necessary/greater so the camera can 'see' the reflectors, but that doesn't resolve the rotation problem.
He describes that moving back improves the tracking issues but is too far from the mouse/keyboard.
His statements, taken as a whole, to me sound like his monitor is too high at the distance it's mounted.
PS As I described above, you can get away with the centerline of the monitor being higher if you're further away. But the closer you get, the more important it becomes to have the top of the monitor inline with your eyes (ie same as the reflectors/LEDs).
-
2
-
-
Again, as I already explained: Tilting (either the camera, the reflectors, or both) will absolutely result in rotation in one or more axes not being in a level plane. For example your head and view will normally rotate in a level plane in the X axis when looking side to side - but tilting the camera/reflectors means that TrackIR "sees" a plane that is not level on the same plane. This results in tracking that doesn't follow head movement (and cannot be corrected via the software).
The more you tilt either camera or reflectors, worse it will get. Again, ideally the camera and sensors/reflectors will be aligned on the same level, with zero angle displacement. While this may not be entirely practical, it is optimal, and the closer you can get to optimal the better.
When the camera is on top of the monitor, having the monitor at the proper height (eyes at or near level with top edge) avoids the problems with angles being too great, and also falls within the almost universally accepted ergonomic practices that are proven to reduce strain and fatigue.
-
2
-
-
FWIW I suspect the problem has to do with the location of the TV, relative to where your line of sight is.
The typical recommended arrangement is your line of sight (when at a 'default' relaxed or normal position) should fall at the top of the monitor, or close toward the top - I've seen some places say within a few inches of the top, for example.
What you don't want is your line of sight at the middle of the screen when using a large TV at a short distance (say, somewhere around 24"). This is not only improper for ergonomic reasons (neck/back strain) but will also cause problems with TrackIR. (Note that some people 'get away' with line of sight at mid-screen as they're sitting further back; this is not the same geometric relationship as when you're sitting closer, since the relative angles change. These same people sometimes recommend line of sight at mid-screen without considering that it's inappropriate if sitting closer to the monitor). The reason for the ergonomic recommendations has to do with muscles in the neck and back, and the closer you sit to the monitor, the more 'displaced' from zero all the angles become, causing muscle strain and fatigue. (This is easily corroborated online).
What happens is that the closer you sit to the monitor, the angle from centerline of the camera to the IR reflectors is increased, eventually to the point that the reflections can't be reliably 'seen' by the camera. This is because you can't (practically) mount the camera in the middle of the screen. Since the camera is (typically) on the top edge of the monitor, if you sit closer and align the monitor such the your line of sight is at the center of the screen, it forces a large angle between camera and reflectors.
If you lower the monitor, effectively moving your line of sight higher (toward the monitor/TV top edge) then you reduce the angle between the camera and the reflectors.
TrackIR isn't really designed to be used with a large screen TV at short distance. That creates too large an angle, as discussed. It's designed to be used with monitors that are much smaller than a 55" TV, at a shorter distance, and with your line of sight at/near the top edge of the display (as is the 'best practice' recommendation almost universally, not the middle of the screen). That arrangement reduces the angle to nearly zero. SInce TrackIR works with light and reflections - which don't generally bend lol - that angle needs to be close to zero or you have to compensate/offset somewhere else. You can't just 'tilt' the reflectors up, because doing that means rotations in two of the three planes are not going to be level. Ideally the 'displacement' between camera and reflectors is zero in all three axes, in practice it winds up being a matter of space and distance but should be kept as close to zero as possible to minimize the chance of having problems.
Or at least that's what it sounds like to me.
-
1
-
-
1 hour ago, BaronVonVaderham said:
Thanks, I’ll check that out.
as for the USB affair, I have a North Sea coloured suspicion it’s to do with the panels I’m using. They come with backlighting and the Arduino powers that backlighting. It would totally not surprise me if that results in an excessive current draw (I need to get one of those usb meters you mentioned). They’re not my own design, and architecturally I would power the backlighting separately.
the panels seem to use PCB’s for various reasons, one of them being backlighting, sometimes for routing signals to/from Arduinos.
I might rip those off and redesign the circuits. Hopefully I can reuse the materials.I'd be strongly inclined to agree with your suspicion re backlighting.
The little inline USB meter is nice to have if you're doing anything concerned with USB power. Helped me out more than once, for sure. Not terribly expensive (I paid ~$20 but you could probably do better):
MakerHawk USB Multimeter USB Voltmeter Ammeter Load Tester USB Voltage Current PD Battery Power Capacity Charger Type C Meter Tester LCD Display Cable Resistance QC2.0/3.0/4.0 N10 1.44 Inch Screen https://a.co/d/2n5GSFx
HTH...best of luck to you
-
1
-
-
1 minute ago, Hiob said:
....and the hot mess that windows is. Especially regarding resource management.
Absolutely.
-
To be fair to ED, I don't think it's their fault that Intel developed this goofy-a$$ed "P and E core" nonsense, nor that any individual decided to buy into it.
Just saying.
-
Having thought about this more:
I can't say I've ever tried it, but there might be a small chance you could install two Nvidia drivers, one each for the 4090 and the 730. I'm not sure it would work TBH. Something else you could try is set the 730 (via Device Manager > Properties) to use a generic/'standard' Microsoft driver, which might work alongside the Nvidia 4090 driver. (These drivers are provided by Nvidia to Microsoft anyway, but you never know what may/may not work).
Another idea is to use DP-HDMI cables. I have some and they work, but it is only 'one way' (i.e. converts DP output to HDMI, but won't convert HDMI output to DP), however, it could work for connecting your 4090's DP outs to HDMI inputs. This would mean your 4090 is driving all the displays (not optimal IMHO, but could work). Not sure about your 4090; mine can only drive 4 displays simultaneously though it does have 5 outputs (2xHDMI, 3xDP).
Finally you don't mention what motherboard/CPU you use...but one other possibility is using CPU iGPU via motherboard to gain one more display. This, with using DP>HDMI cables as above, might yield the 5 displays you want.
It's probably better IMHO to just go with a 1030 card...but these other ideas I had might be of some help.
Good luck.
-
1
-
-
Oh, incidentally, the Armory Crate Uninstall Tool should be with your motherboard drivers on the Asus website. Asus makes a number of different X670E boards; you don't specify which one you have, but when I look at mine (X670E-F), it is there:
-
I think it's likely this is a driver issue.
See below: 4080's aren't supported by drivers that also support a GT730, and vice-versa. Drivers that are new enough to include support for 4080s (since ~NOV 2022) don't appear to also include support for GT730 cards.
I went back to find the oldest 4080 driver after it's November 2022 release, and that driver doesn't include support for GT730 cards.
I could be wrong, but I think this is accurate.
You'd need a card like a GT1030, which is supported in drivers that also support 4080.
-
1
-
-
Interesting development, though I'm not at all surprised AURA would be causing problems. It's known for being problematic; as I said, I finally just all but quit using it after quite some time, in part due to problems like this.
The USB thing: It seems unrelated to the RGB issue. It may have been going on before the RGB thing, you'd know before I would.
What I can tell you is this: USB ports limit the amount of current that can be drawn in total from the port (see table below). USB 2.0 is 500mA (1/2 amp), which might not be enough when too many things are connected - it depends on the sum total of what each device draws (usually this can be determined via documentation; I have an in-line meter that tells me how much current is being drawn from a USB port). This was one of the major changes in the next USB revision: Current was increased to 900mA.
The 127 devices you mentioned is the limit of separate devices that can be connected to a single port, but that's not considering power. (Mathematically, a USB 2.0 port could only support power for 127 devices if the average power for each device was less than 4 mA and I'm not sure that's very likely TBH.)
I would tend to agree about powered hubs, however it can be tricky. I believe that not all 'powered' hubs are created equal. But, in general, yes, I would think a powered hub would itself draw less than the allowed 500mA on a USB 2 port, provided that's all that was plugged in (no other devices that use USB port power). Also, the 'downstream' side of any hubs cannot exceed what *those* ports are capable of, both the limit of USB ports *and* the limit of the hub's power supply. Again, you'd have to add all the devices to make sure you weren't drawing too much.
Also, it could be that the hub you're using has issues, or that the hub just doesn't work well with your motherboard ports. It's not impossible; people take for granted USB "just works" but sometimes USB can be a pain in the ass, particularly when using lots of devices from different manufacturers. USB is not the 'universal fix all' that people tend to see it as - in spite of the name.
2 hours ago, BaronVonVaderham said:From what others told me, mobo USB ports can’t really deal with lots of connections
BTW I suspect this is more about the increased odds of conflict or overload when using more and more devices, rather than a specific rule that says you can't plug in X devices. As you mentioned, USB 2.0 ports can handle 127 devices, and up to 500mA current per port if I'm not mistaken. Obviously, one device is far less likely to cause problems than 10; 10 less than 50, and 50 less than 120
-
18 minutes ago, BaronVonVaderham said:
in general windows is notorious for poor file housekeeping
Absolutely - but I'm referring more to the known issues with uninstalling the various RGB control apps from different mfrs. If it gets to the point that Asus themselves provide a special clean up app for removing their lighting software...well, that's a sort of tacit admission of itself, IMHO
Good luck - it could still be a hardware issue. TBH I'm not sure if the multiple headers on a board all go back to one hardware controller or not. Wouldn't surprise me if they do, but wouldn't surprise me if they don't either.
Issues with TrackIR placement on 55" TV's
in PC Hardware and Related Software
Posted
Funny...I've had the same one for going on 20 years now it seems ...and it's still just fine. And, to be accurate, I bought it used.
I guess it comes down to handling.