Positional tracking calibration

"Tgaud" wrote:

Even the Oculus Rift DK has such a sticker on the bottom of its controller

well you're blathering about an unbelieviable cost of a FCC hardware certification.
A certification that occulus rift Already need.

So, I dont see why, as long as they already need this certification, they wouldn't include proper RF hardware inside ?

Where did I use such words? I did not use such words. Your slander is not welcome. Go back to spreading your hatred at other forums. This thread is about tracker calibration to allow tracking with software and built-in (existing) hardware. Please take your bad attitude elsewhere so we can continue our discussion without your off-topic negativity.



Why do you persist in belittling those who have more education and experience than yourself? Does you ego really need that kind of self-inflation for you to live with yourself?

"geekmaster" wrote:
Where did I use such words? I did not use such words. Your slander is not welcome. Go back to spreading your hatred at other forums. This thread is about tracker calibration to allow tracking with software and built-in (existing) hardware. Please take your bad attitude elsewhere so we can continue our discussion without your off-topic negativity.

And RF requires FCC (and other) certification, which is incredibly expensive for a small production run, and will add a LOT to the manufacturing cost.

Do you have Alzheimer already ?

I think we can end the debate, with this simple statement.

"Tgaud" wrote:

"geekmaster" wrote:
Where did I use such words? I did not use such words. Your slander is not welcome. Go back to spreading your hatred at other forums. This thread is about tracker calibration to allow tracking with software and built-in (existing) hardware. Please take your bad attitude elsewhere so we can continue our discussion without your off-topic negativity.

And RF requires FCC (and other) certification, which is incredibly expensive for a small production run, and will add a LOT to the manufacturing cost.

Do you have Alzheimer already ?

I think we can end the debate, with this simple statement.

Probably a language translation issue on your part. In the USA, "incredibly expensive" is commonly used to mean "very expensive", and it was qualified with "small production run", making it completely true. However, you said "unbelievable cost" in your previous post, which commonly means that NOBODY can afford it. They are not the same, although your language translation skills may have failed you in this case. It helps to be a native speaker of a language to understand such nuances. And being a non-native English speaker does not qualify you to effectively participate in your argumentative challenges over language semantics that you do not understand.

And does that sarcastic "Alzheimer" comment make you feel better about your youth and inexperience? You made the false comparison, not me...

I will continue my posts in this thread about my progress in this area of research. I will try to ignore your disrespect.


http://mythocreatology.com/Ankle-Biter.jpg

You're the one clearly disrespectful here, doing personnal attacks and insults.
And in a lot of post in this forum. you're very offensive with ppl.
Can you try to let other ppl exprim a different opinion of you ?
Even if you post in every topics this is not YOUR forum.

And i Made my point over your hypocrisy. You can play with words all you want to get back on your feet.

"Tgaud" wrote:
You're the one clearly disrespectful here, doing personnal attacks and insults.
And in a lot of post in this forum. you're very offensive with ppl.
Can you try to let other ppl exprim a different opinion of you ?
Even if you post in every topics this is not YOUR forum.

And i Made my point over your hypocrisy. You can play with words all you want to get back on your feet.

Your claims are false. You have misquoted me. You have removed important context, and you have made claims about me that are inappropriate and discrespectful. I have only disputed your unsupported claims, but other than identifying your behavior as that of a "troll", there were no personal attacks or insults coming from me. That is what you have been doing repeatedly.

Your opinions are welcome if they provide useful information. But your harsh and unsupported criticism of my posts is not welcome. When you call my posts "B.S.", that opinion is also not welcome. When you call me a liar, that opinion is not welcome. When you imply I have Alzheimer's, that opinion is not welcome. When you criticize, condemn and complain without a trace of support or evidence to support your opinions, that is not welcome.

But if you actually provide some sane and lucid comments that challenge the FACTS in my posts, then that kind of debate is certainly welcome. However, you did not make such a challenge, preferring instead to debate language semantics and word definitions that you seem to misunderstand (especially in their common usage by native speakers of that language).

So, if you can express a valid opinion with real facts about the TOPIC of discussion, please do so. But your unjustified negative and disrespectful opinions about the POSTS of others, for which you seem to be famous, are not welcome here.

Get back on my feet? You have not gnawed them off yet! :mrgreen: I was never "down". :lol:

No, this is not my forum, though I am welcome to participate in it in a friendly and helpful manner as much as I choose. And so can you, if you are able. So, is it the quantity of my posts, or the breadth and depth of my knowledge and experience, that bothers you the most? My hobbies and interests are vast and deep, and you have not even begun to understand them. I have more than ten-thousand (mostly technical) books in my collection, and I have read almost all of them. I have been creating electronic devices and computer programs for more than fifty years. Some of my (ancient) code is still in use around the world. Robots running my code may well have handled your food. If you live in Greece, my code may have controlled the flow of oil that drives your machinery. If you use products from a huge list of major companies, my code may have helped manufacture that product. What have you done, besides getting your jollies by annoying people?

Why are you more concerned about HOW I said things than about WHAT I said? Why are you more concerned about trivial facts than about important ideas?

When presented with a fact a small mind will care about who said it, and examine how it is expressed, rather then simply accepting it. This is a fact.
-- Oscar Wilde

"Great minds discuss ideas. Average minds discuss events. Small minds discuss people."
-- Eleanor Roosevelt

@Tgaud: You have not answered my many questions scattered throughout this thread. Before you make another post containing misquotes, personal attacks, or general semantic argumentative nonsense, could you PLEASE answer my questions?

Thanks...

Or, if you wish to post something useful or interesting about the TOPIC of this thread, that would be nice too...

"Tgaud" wrote:
... Can you try to let other ppl exprim a different opinion of you ?

What does "ppl exprim" mean? Did you mean "Can you try to let other people express an opinion that may be different from your own?". If so, yes I can. I like a variety of opinions from which to extract the best ideas. That is why I own so many books from so many different authors. I value the (useful) opinions of others. You can see that in my posts here and elsewhere, if you actually READ them before you form an opinion about them or about me. I keep an open mind, and I value all sorts of knowledge and information, even from fringe and taboo areas that others prefer to ignore. But what about you? Do you have preconceived notions that make you automatically reject what I have to offer? You can learn from me and others if you open your eyes and your mind to new ideas, just as I can learn from you if you choose to share some useful information. In fact, I am looking forward to learning something interesting from you in the near future. I hope you do not disappoint me in that respect...

"geekmaster" wrote:
I have no problem with you telling me you are not convinced. I have a problem with people telling me my claims are wrong, and that I am spreading B.S., when they provide no supporting evidence nor convincing arguments with which to dispute my claims

Basically, not a single project that has attempted to perform positional tracking of arbitrary motion (which is what the head does, as there is no stationery anchor position like the tracking of gait via foot sensors) using consumer accelerometers and gyroscopes has succeeded. Every one has either failed to work in a satisfactory manner, on been given up on due to lack of success in the early stages. Claiming that you have a successful method to do so falls into the "extraordinary claims require extraordinary evidence" camp.

So why attempt to discourage me?

1) To avoid you wasting your time as many others have done before you in the expectation of success. Performing the research in the expectation of demonstrating just how bad the problem is is a reasonable, though slightly redundant, pursuit. Expecting to get a usable system out of it will just lead to disappointment.
2) There is continuous clamouring from the peanut gallery that 'Oculus should enable positional tracking' as if it were possible with the existing hardware and Oculus just need to flip a switch. Continuing to encourage this mistaken perception means more people complaining about Oculus failing to deliver the impossible.

"edzieba" wrote:

"geekmaster" wrote:
I have no problem with you telling me you are not convinced. I have a problem with people telling me my claims are wrong, and that I am spreading B.S., when they provide no supporting evidence nor convincing arguments with which to dispute my claims

Basically, not a single project that has attempted to perform positional tracking of arbitrary motion (which is what the head does, as there is no stationery anchor position like the tracking of gait via foot sensors) using consumer accelerometers and gyroscopes has succeeded. Every one has either failed to work in a satisfactory manner, on been given up on due to lack of success in the early stages. Claiming that you have a successful method to do so falls into the "extraordinary claims require extraordinary evidence" camp.

So why attempt to discourage me?

1) To avoid you wasting your time as many others have done before you in the expectation of success. Performing the research in the expectation of demonstrating just how bad the problem is is a reasonable, though slightly redundant, pursuit. Expecting to get a usable system out of it will just lead to disappointment.
2) There is continuous clamouring from the peanut gallery that 'Oculus should enable positional tracking' as if it were possible with the existing hardware and Oculus just need to flip a switch. Continuing to encourage this mistaken perception means more people complaining about Oculus failing to deliver the impossible.

Oh, ye of little faith. I have solved many problems that others claimed were "impossible". In fact, anything that my intuition told me that I COULD solve, I did solve. My intuition tells me I can solve this problem SUFFICIENTLY for it to be a valuable and useful tool to use for more immersive VR experiences. If I cannot succeed, this will be the first time in more than HALF A CENTURY that my intuition will have failed me. I believe in myself, and for good reason.

As I mentioned previously, I already have "proof-of-concept" code working well enough to convince ME that I can do this. It is not ready for publication though, especially with hecklers in the "peanut gallery"...

And besides, you are not paying me enough to decide FOR ME what I can and cannot work on in my spare time, or what I can or cannot post here or elsewhere. Oh, wait... you are paying me NOTHING! You have balls the size of Texas though, to think you can boss or bully others around like that! Naysayers have not stopped me in the past, nor will they this time.

There have been people like you in the past, discouraging every great invention (large or small). If listened to, we would still not have mastered fire or the wheel. "Nobody has mastered fire! You are wasting your time trying to perfect it! Do not give others hope that will only lead to disappointment!". Bah, I say! You have a sad lack of faith in unexpected progress... really sad, really...

EDIT: While seated or standing in a fixed position, your anchor position is a dynamic combination of your median range-of-motion position and your long-term mean position. Your disagreement proves nothing but your own disagreeable attitude. You will get your "extraordinary evidence" when I am ready. Good things come to those who wait. I could accomplish this faster if it was part of my job, but I have other demands upon my limited time. Sorry to disappoint you.

In the OculusRoomDemo (one of the scaffold programs for my positional tracking experiments), I am seeing the gravity vector magnitude vary significantly as I statically support my Rift at different orientations, even though it should be calibrated in this environment. This tells me that I may need to develop my own per-axis calibration and compensation, so that I can have a stable gravity compensation vector with which to extract pure motion-based acceleration. Perhaps others failed in their prior attempts at this area of research because they neglected to do accurate-enough per-axis accelerometer calibration and gravity vector compensation. For accuracy, the gravity vector must be MEASURED, because it various with geographic lattitude and with geographic crustal density, and the measuring instruments themselves affect such gravity vector measurements. Each unexpected impediment must be solved as it is encountered. But discovering and recognizing such minor problems is the key to solving them. Progress will be made. Googling shows that others have made some recent progress at contrained accelerometer-based positional tracking as well, so it is time for one of us to make a breakthrough in this area soon.

Sadly, toxic comments from the "peanut gallery" occupy my thoughts when I should be resting my mind, and divert my attention away from this project to my many other projects that give me personal satisfaction. The only truly rewarding compensation that I derive from these research topics (besides a pure joy of learning) is the respect of my peers and the admiration of my fans, so if you wish to see faster progress made in this area, please give me some rewarding encouragement, to replace the current attention-repelling negative criticism in this thread. Like Mozart before me, the theme of my public performance is driven by audience applause. Thank you for your (positive) attention.

After measuring the gravity vector magnitude in both directions for each axis of my Rift accelerometers. Not only do the values differ significantly between axes, but they also differ significantly depending on which direction of each axis is facing down. This indicates that we have significant offset errors and range calibration errors, while running inside the OculusRoomDemo app. I thought those things were supposed to be calibrated during initialization of the SDK. Apparently not (or at least not correctly).

Without correct calibration, my gravity vector magnitude varies from about 9.7 to 9.9, depending on orientation. I need to make that much less orientation-sensitive for effective gravity compensation required for approximate estimated positional head tracking.

Even so, my test results within defined contraints for seated or standing still are quite satisfying to me, so with correct calibration it can only get better.
:D

And I still have that problem with my magnetometer X value being useless (stuck at some "much too high" max value except for some very small range of orientation where it suddenly jumps to a lower range of values). This problem exists both in the OculusRoomDemo app and in the raw HID packets right after the Rift is plugged in and powered on.

I was hoping that the accelerometer and magnetometer values in the SDK apps would be more useful to me than the raw HID packets, but apparently not. I guess I need to devote more attention to my own (unfinished) sensor fusion implementation again, to get the accuracy that I need for this project.

If (when) I do succeed at getting accurate and correct accelerometer calibration and gravity vector compensation working, I will attempt to add those "fixes" back into the SDK, where it can become part of a future release. There is already a lot of good stuff in the SDK, so I do not want to abandon it completely.
;)

EDIT: Okay, perhaps OculusRoomDemo is using uncalibrated values. Looking at the LibOVR source code, I see that it does contain gravity compensation code. I need to see how it is being done and used inside the SDK, and why the accelerometer values in the demo are not calibrated. I still have a lot of learning to do, so my understanding of this stuff is subject to change as I learn more details. I will post what I discover about how this works, or is supposed to work. I have a lot of unanswered questions...