Pimax Crystal - status, updates and fixes (Part 1)

I think in this case they will have a harder time getting away with it. Diagonal FOV is a bit more nebulous but horizontal and vertical is pretty clear cut, Pimax can’t really pretend to have calculated the hFOV and vFOV differently. That’s why I’m thinking if they move to making sales directly in the EU then this could land them in serious hot water as it would be breaking the law to advertise incorrect numbers (especially to this extent, I guess like 5% you’d get away with but we’re looking at more like 20% which is huge).

This isn’t me trying to make a fuss even, I genuinely think it’s important for them to address and rectify this as it could really come back to bite them. I’m guessing what we’ll see is the hFOV and vFOV get omitted from the spec sheet and just the diagonal left up, which is not a solution I would find acceptable, but hopefully something better comes from it.

3 Likes

Luckily, your marketing strategy fights against these losses. :rofl:

Ok, any technical News despide the NDA?

This can be quite easy. As there are a variety of tools for calculating FoV; some do so incorrectly. See @philpw99 post and your & @risa2000 response explaing that one tool calculates hmd fov incorrectly.

Then also remember the Vp2’s “legit” 120° horizontal FoV that is rounded up. lol

1 Like

I don’t think that is true as @risa2000 tool collects information for example from steam requests rendering info. So for example some values change due to what FoV setting you choose and whether you have pp on or off.

Now it is true if not mistaken that a simple lens change will not be reflected in Risa’s tool if you don’t change the lens setting in the software.

Oops didn’t see this on reply as you stated more ir less what I said above. :man_facepalming:

1 Like

116 real to 120 advertised is still bad imo, but it’s not going to really cause any issues but 103 being advertised as 125 I think will.

1 Like

Let’s not look to encourage disharmony. With passive aggressive posts like these.

1 Like

Yes, Valve was smarter. They only spoke of 20% more than others.

43 posts were split to a new topic: Overreacting to Reactions Discussion

I will consider closing this topic if folks do not get back on topic. Though with no updates for the time being; not sure there is much to discuss. Obviously not discussing people’s petty retaliatory actions because they don’t like a ‘thumbs down’ Perhaps I should look into removing members ability to see whom has given reactions to align with Redit’s methodology.

I would rather not see reddit like negative behaviors here.

5 Likes

All it would take is for one Pimax employee to get ‘a’ Crystal what ever stage of development it is wether prototype, CES demonstrator or final retail unit and just do a quick FOV test.

Pimax either can’t, don’t want to or are hiding the truth.

It’s fairly bizarre because the truth always comes out eventually.

2 Likes

I can understand silence from the Chinese portion of Pimax right now due to the new year. Although @PimaxUSA has been online a good few times so would have hoped that he could have chimed in by now. I assume he has a crystal at hand, so a quick comment on it (even just a “Looking into it and will give an official response soon”) would be much appreciated. Surely Pimax knows the FOV for the crystal, so very confusing to see these 103 numbers that wildly contradict the official spec sheet numbers.

4 Likes

Here is my best speculation about the silence:

The Pimax crew we engage with are disconnected from the core team and engineers. They get their information either from translated, long distance communication, or via hands on experience with an engineering sample/finished product.

This is all to say, the Western side of Pimax is playing telephone. It also puts them at the mercy of cultural differences and whatever communication bottlenecks the tech side of the company has. (In my personal experience, engineering departments are usually pretty bad at communicating outside of their vertical.)

This is all to say… our side of Pimax is likely stuck in a situation where they either have to say “I don’t know”, or to pass on whatever promises they’re getting second hand/translated from the Chinese based engineering team.

If they say the first, they look ignorant. If they pass on the latter, they end up being wrong half the time and look deceitful.

I would strongly, strongly encourage Pimax to be transparent about this and take the first path. And to work really hard internally to clean up communication paths.

Finding a way to be authentic and accurate with consumers is a huge, necessary step towards building a lasting and reliable brand image.

Unfortunately this is indeed a recurring experience.

@Djonko: I thought you guys were done with the thumbs down etc. stuff, and you had already made your point?

3 Likes

Do you mean a COO, Sweviver and Operations in three different continents?

Either way any one of them can grab a Crystal because they’re all in different time zones and Crystals been on several roadshows. I can’t believe this is as controversial as it is. Lol

1 Like

Well there prepping for holidays and those whom are not would still need to consult the team before simply honoring community requests.

I agree that it shouldn’t be so hard. I’m just speculating on why it is. :stuck_out_tongue:

I can easily imagine scenarios where the unfinished software side leads to a lot of unknowns for the English speaking Pimax crew. We don’t know how long it takes them to get an answer when requesting clarification internally. There may also be cultural barriers affecting the communication they do receive.

That’s not meant to excuse the present reality. Things need to get better for Pimax to step up as a serious player in the VR world. I sincerely hope that they do.

Well, if nothing else, we can always imagine it not impossible that this thread may have prompted an internal discussion on the issue, within Pimax, looking over aspects of truthfulness and liabilites. I figure were such a discussion taking place, it might take a few days – and then some more time might be needed, to prepare public messaging and corrections, should such action be decided on.

1 Like

Regular Season Ugh GIF by NFL

3 Likes

Well, how about starting off with some more news and reports from the tester to get things back on track? Including a rather interesting one about the FoV.

So, they did not show their exact settings or what degree they offset it, but the tester said that they played around with the software IPD offsets and ended up getting these results in Risa’s tool

H 109.74 V 93.77 Diagnal 119.29 overlap 77.71

They also said that after this and playing around with the face foam’s fit/moving it around, they said that they were getting around 102 horizontal visible to them in TestHMD. I asked about vertical, but he didn’t give me a number, stating that the vertical FoV let him see a lot more below his eyes than he could above them, so sounds like TestHMD wouldn’t quite give the right impression there. These(That is, what he visibly saw) are ridiculously higher numbers than he initially reported and a few degrees higher than he reported getting without any face foam. That cost to stereo overlap though is… Not good, very not good. It does make me a bit uncertain on if the lenses were truly wrong, but I still find the whole thing perplexing as while I could totally believe people at the roadshows wouldn’t notice a couple degrees less than the Index, I can’t believe they wouldn’t notice an over 20 degree and 30 degree difference. It could perhaps be that this person just has a particularly bad faceshape for VR and we only have a sample size of 1 for TestHMD, and maybe the lenses are scuffed, so I’m focusing more on the Risa2000 numbers, since those are less subjective. I distinctly remember some buzz around the VP2’s FoV at its launch cause one reviewer seemed to have a very poor faceshape for it’s design and got low FoV, so could perhaps be a similar case here until we get a greater sample size on the visible FoV to the end user.

Oh, and to break from the FoV briefly, they said the inside-out tracking was fine and they “used it a lot”. Compared it to the Pico 4 and sounded like it was about on par. Described it as fine for most games, but obviously, LH is still more accurate with better volume.

2 Likes

So, that’s pretty much equal horizontal and vertical per-eye FOV, which strikes me as a likely situation.

I guess with a negative offset, your frustum would skew for a wider FOV, as you look out toward the sides through the lens, whilst sacrificing stereo overlap – as long as the lens spacing is adjusted to match the rendered projection… Obviously, when this is the case, one is no longer aligned with the lens axis (EDIT: …which is optically sub-optimal).

Has this tester tried without parallel projections? -It would be interesting to see some non-zero view rotation numbers, which would also give a hint about what the lens c-c spacing should optimally be, for a given combination of user IPD and eye relief…

EDIT: More FOV downward could be either a deliberate downward bias to the frustum shape, or the HMD worn tilted downwards, in which case the facial interface could use some padding at the cheekbones (unless the user chooses to favour seeing more ground than sky) – the fact that previous numbers have identical top and bottom values points to the latter.

3 Likes