Understanding FOV and its variations in Pimax headsets

General discussion
1

图片
图片1500×844 161 KB

FOV, what is it?

Field of View (FOV) is a crucial aspect of any optical device, including Pimax headsets. While Pimax has advertised the maximum FOV of its headsets, there have been instances where users have experienced a discrepancy between the promoted and actual FOV. In this blog, we will explain why there can be a variance in FOV in Pimax headsets and what factors affect it.

FOV in Pimax headsets

FOV is divided into HFOV (horizontal), VFOV (vertical), and DFOV (diagonal). At Pimax, when we talk about FOV, we use DFOV as a measurement. So this is 200° for the Pimax 8KX and 140° with the Pimax Crystal (when using 35PPD lenses).

  • – HFOV: Horizontal FOV
  • – VFOV: Vertical FOV
  • – DFOV: Diagonal FOV

图片
图片788×519 39.9 KB

In terms of theoretical pure mathematics, it can be seen from the figure that HFOV is determined by Focal Length and sensor width (W), and VFOV is determined by Focal Length and sensor height (H). DFOV is jointly determined by Focal Length, W, H.

Why?

However, there are many factors that affect the FOV number, such as individual differences, external conditions, and design and manufacturing limitations. These affect not only in Pimax headsets, but all VR headsets. Let’s take a closer look at each of these factors.

1: Personal Differences

First of all, there is no denying that FOV varies from person to person. There is a limit, based on what is rendered and optics. As for maximum FOV, the eyes need to be at an optimal distance from the lenses, which can be hard to achieve with many face shapes and head shapes. That’s why some individuals cannot get the number we promoted.

2: External Conditions

In addition, the FOV of the human eye is not fixed, but changes according to external conditions such as light intensity and the distance of objects. Therefore, some assumptions and trade-offs may need to be made when simulating the FOV of the human eye.

The FOV of the actual experience depends on many factors such as the user’s interpupillary distance, the distance between the eyes and the lens, and the testing method. The test data of personal use of three-party tools is only recommended as a relative reference value. Pimax has tested over 100 individuals to measure the FOV of its headsets, and the promotion FOV numbers for the Pimax 8KX and Pimax Crystal are what the company obtained. However, users may experience slightly different FOV numbers based on individual differences and the actual application. In other words, it’s difficult to guarantee that every user can achieve the promoted FOV numbers due to several factors that affect the FOV experience.

3: Design & Manufacturing Limitations

During the design and manufacturing of Pimax headsets, there are different aspects of limiting factors that can affect the FOV, such as resolution, brightness, contrast, light transmittance, image artifacts, distortion, and uniformity. And other non-display issues such as: eye zone, size, weight, battery, cost, durability, etc of the headset. In addition, the improvement of the field of view will also put forward higher requirements on the computing power and power consumption of the equipment.

Conclusion

Overall, Pimax’s careful design offers a more efficient FOV for individual users without sacrificing pixels per degree or any distortion. While it’s challenging to accurately measure and describe FOV, Pimax’s testing and user feedback confirm that the FOV of the Pimax Crystal is larger than other popular headsets in the market.

3 Likes
2

Okay but rendered FoV is fixed and we know for fact pimax made errors calculating P2 series FoV in terms of Hfov which maxes out at 160 wide.

Now hopefully this is not trying to bypass the question why the Crystal from chinese testers that have ran @risa2000 Hmdq tool which collects driver data. In that test it clearly shows the Crystal requesting a FoV to be rendered 103° × 103° with a diagonal around 120ish. So during that published test the Crystal can only display a total FoV of 103 x 103. Is that maybe the 42ppd lenses running small?

It is only hard to measure perceived FoV which cannot be greater than what is rendered. Which is true of all vr hmds. You cannot see more than what is there.

7 Likes
3

Which is why, as has previously been pointed out by users, Pimax do not need to lie.

-When you already exceed the competition by a wide margin: Why untruthfully exaggerate the advantage to sound even greater?

Doubling down on nonsense claims can only hurt the already shaky credibility and reputation of Pimax, so please don’t push them any further – none of us want Pimax to keep shooting themselves in the foot.

There is no mathematics, nor any objective physical measurement reports in the post I respond to, at the time I respond to it.

5 Likes
4

Does this mean all these YTs are going to report 140 degrees FOV? Fantastic, can’t wait to see their videos.

5

How is that true? If FOV changed due to light intensity the world would be a constant wobble especially if you walked from a dark room into open sunlight.

If I take my camera and alter the aperture only the light level changes. FOV always stays the same.

FOV only changes when the distance between the lenses to sensor panel changes. So FOV on a14mm lens would look super wide compared to a 400mm zoom lens.

Pimax need to rethink the OP.

6

All they mean is that the FOV you can see may be limited by the lens size itself. If you have a reverb G2 you would know what that means. That’s why the frankenFOV mods are so popular. Fresnel lenses on an 8KX effectively didn’t have the problem. Also, the canted displays provides a wider total FOV (or at least should) than other displays. This reduces binocular overlap and also increases need for Parallel projections. We will soon see if the FOV is truly better than the competitors. 103 degrees is not impressive to me personally. I’m coming from their 8KX however. For someone coming from Aero or G2 it will seem nice though.

7

Hmmm no they literally said the human eye FOV changes with light intensity.

8

yeah that would have to be demonstrated for me to believe…

3 Likes
9

The human field of vision actually changes depending on the situation. For example, if you focus on something, the field of view is reduced. But if you’re 24/7 on amphetamine, you’ll will always get the full program. xD

3 Likes
10

@hammerhead_gal
You may want to look at my take on the topic (VR headset rendered FOV calculation | VR Docs), kind of idea precursor to HMDGDB, where I was trying to explain the same things you are trying here (or on your blog). You could notice there that I defined the HFOV and VFOV across the FOV centerline (frustum axis) as is common in any technical literature and not along the borderlines as you do on your picture.

Apart from that “detail” however there is not much subjective about it (the rendered FOV) as I explained further down in my doc and it can be easily “measured”, which is what led to HMDGDB.

5 Likes
11

Everyone’s already said it, so I don’t need to, but while it is true personal differences can change perceived FoV, in this case the response is honestly a rather shameful display from Pimax and comes across as a bit condescending when the whole reason this FoV thing started was because we got the rendered values that aren’t gonna be effected by subjectivity, which Pimax is just ignoring. They themselves, in this very reply, even note there are limits based on what is rendered. Like, there have been multiple conversations across the community beating this point in already.

Does seem like this is kind of the final nail in the coffin for the Crystal’s FoV, however.

6 Likes
12

I appreciate @hammerhead_gal 's write-up/post on the forums regarding FoV.

However, I would like to see something re-iterated or even iterated more strongly regarding the MINIMUM FoV the Crystal can do in the expected optimal use environment. Without all of the subjective variances thrown in there to yield potential skepticism regarding the Crystal’s FoV capabilities.

“Technically speaking, given all factors are at optimal setup, the Crystal Display is capable of a MAX Diagnal FoV of ###°. Based on the design of Crystal’s displays, lenses and other hardware capabilities. This is +/- ###° as compared to the PIMAX 8KX’s Diagnal 200° FoV.
:spiral_notepad: User experience and results may vary due to subjective factors (like distance of eyes to lenses).”

I mean, something like the above would be a more appropriate statement IMO.

2 Likes
13

140 degrees diagonal was advertised for the crystal long long ago…should be 98 degrees horizontal if the displays are square…but the 5 degree canted displays(total) should offer more, at the cost of binocular overlap, so 103 degrees horizontal 98 degrees vertical? This corresponds exactly with the leaked HMDQ that appeared online. Its confusing to me why people thought it would be more…is it just a disappointment because people want this to be a 12k ?

14

103°x103° may be the visual field of one eye, and the parameter of Pimax should be the visual field of both eyes.

If it is the field of view of both eyes, calculated according to HFOV120°, the overlapping area of the two eyes is 86° x 103°, and there are 17° non-overlapping fields of sight for the left and right eyes.

1 Like
15

It is both eyes from the screenshot as it has Vertical, Horizontal and diagonal values.

Vive Pro offset square res to iirc get 116wide? FoV but advertised it as legit 120. No Canting as I recall.

1 Like
16

The Data does not support that. As hmdq gets all that data from the hmd driver no magic just pure data collection and math.

iirc the pic 103Ă—103 with a diagonal of around 120.

Read here.

1 Like
17

No, it should not. Check the simple calculation for “square” view here (Pimax Crystal - status, updates and fixes (Part 1) - #511 by risa2000).

2 Likes
18

im thinking maybe people tought it would be more cause they claim so on their store when you preorder it “FIELD OF VIEW 35PPD：
Horizontal：125°
Diagonal：140°
42PPD：
Horizontal：110°
Diagonal：120°”
thats a whole insane 22 degree less field of view than what they advertise,any credibility on that 35PPD claim also?who knows.

2 Likes
19

I wasnt aware of that. Possibly, they simply used the wrong factor in converting the diagonal to horizontal. If the displays were 3:2 or 16:9 that would maybe make more sense.

But yeah that’s false advertising if 103 is real.

1 Like
20

Oh yeah non-euclidean…once again? Still if crystal is hotizontally 103 degrees fov. They possibly knew it was 98 each eye and simply extrapolated incorrectly on their own to say 140 diagonal, incorrectly…in the same way I did. I dont doubt the math is wrong, just that they did that math wrong too. Can you blame them (8kx 12kx)