Debugging with Pimax Support

Patience required…

12 min readOct 27, 2021

Context

I have been having trouble with my brand-new Pimax 8KX. I have tried multiple times to solve issues. Pimax support is nice in that they connect to your computer and try to debug issues. The process they follow, however, seems a bit unstructured, and it does not seem to actually find solutions, or even diagnose problems. This write-up below details the *third* time that I have had one of these sessions. This time, I was there for the whole duration, trying to help. I am documenting this sequence, in hopes it can help some other poor souls when they face issues with their headsets.

In the spirit of transparency, every single word (modulo titles, intros) after this section is on the ticket/shared with Pimax. I am using this write-up as a repository of my interactions, as I am not sure the ticketing system used by Pimax would be a permanent solution). Ultimately, I want the headset to work, my intention is to solve the issues and to help other users. When the 8KX works it is glorious, period. I just wished it worked more often :-)

With this introduction out of the way…

TL;DR (December 9 update)

I’ve given up. Support is very responsive but completely ineffectual. Either that or the product is so random/brittle, that whatever gets “fixed” on a Monday, stops working on a Tuesday, or works whenever it wants to. Support keeps trying, over and over, the same “solutions” which are proven (by now) to not be permanent. In the process, I waste enormous amounts of time, only to end up pretty much where I started.

Thus, as of this writing, the 8KX sits unused nearby. Every once in a while I give it chance to see if it decides to work that day. Sometimes it even does. Those days I am happy. They are about 10% of the time. The rest I am frustrated with.

I’ve bought a Varjo Aero which I have yet to receive. Will try it out. If it turns out to be reliable, and the smaller FOV tolerable, that will be the answer to this Pimax nightmare. Perhaps the 12K will solve these issues, though I am skeptical. While the 8KX is clearly sub-par in terms of quality (especially for a USD2k product!), it was the helplessness of the support team, and the inefficiency of the support procedures that shone a light on the company, and its maturity level, undermining my faith in the company being able to work out their kinks and issues (in a reasonable amount of time)

Here’s to hoping Varjo is better!

My debugging saga (currently on-going)

About 15 days ago I opened a ticket… By then I had had a couple of support interactions where other problems had been solved (notably tracking, which as you will see, was not really solved). I was frustrated, and wanted to return the product.

October 11

Very quickly, as is common with Pimax Support, we got into a conversation. Also typical, they ask you to take certain steps and give them the results, as you can see from the image below.

The same day, a couple of hours later, I respond via email with the information provided, as you can see below.

October 14

For some reason, my response does not make it to the ticket (when before it always has). I let some time go by, assuming they are working on the issue, but after three days have passed, I email them back

October 25

Many days pass by without any response, so yesterday I sent another email and I also updated the ticket manually myself. This is what I sent:

Hello,

It has been a number of days. No response to my note.

I am trying your steps. Now lighthouse_console.exe does not see the LHR device.

I need your support.

When I realized that the LHR was not being seen, I tried the following:

1) I deleted ProgramData/pimax/runtime/config/lighthouse

2) Rebooted everything (pitools, HDM, etc)

3) Tried lighthouse_console.exe, with bad results

Interestingly, piTools shows no error. It even shows the setup as “Lighthouse” but the base station is clearly not being used for tracking!!!

Finally, the base station is located 2.5 meters away and it has a white light on it. The following picture is taken with the iPhone where my headset is, more or less.

And here is a close-up.

In short, I am not sure what else to do. I need this issue fixed.

Thanks

Luis

Within hours, I receive a reply from Yvette, with:

I promptly send the IDs required, and earlier today we had the debug session. After the debug session, I documented everything in the ticket. Below is what I shared.

October 26

This is an update on this ticket. I spent about THREE HOURS watching the Primax Support team try to debug and solve my issue. Unfortunately, they left the system in a worse state than they found it. For instance, when they started their session, the Base Station was sometimes detected. Now, it is not detected. I am providing the data below to help in the debugging. It is UNBELIEVABLE how unstable the system is. Things that work at one point stop working a minute later.

I am attaching the chat (though it is only partial as there’s about an hour that I was not able to capture).

The PROBLEM

There are multiple problems with my setup, in order of importance:

Tracking. Simply stated, the HMD does not see the base station or it loses tracking often. When it loses tracking, the image starts to slowly move diagonally or to one side until it fades to grey and then comes back. With regards to the base station, it is important to note that I have one base station, model 2.0, bought from Pimax alongside the headset. It is correctly positioned (2.67 meters high, at about 2.5 meters from the HMD). You can find pictures of it and its position earlier in this ticket.
90 Hz flicker. When I enable 90Hz, I get flickering on the screen every few seconds.
Blackouts. Occasionally, the screen goes black, piTools shows “Disconnected”. This happens at times, not sure why.

The biggest problem is #1, because I need accurate tracking. This problem is the one that the support engineers tried to solve, without success.

Throughout the debugging session, the Support person often (and repetitively) asked about the setup of the HMD and Base Station. In order to establish context, in the picture below you can see the headset, on top of the seat, roughly where my head is usually at. The headset was located there throughout the debugging session, without being moved at all with only one exception, and that is when the support engineer asked me to remove the foam cover to unplug the headset on its end.

In addition, and to allay any concerns, the following picture shows the location of the base station (up to the right). That’s about 2.7m high, and 2.5m from the headset. Rather perfect location.

When the engineer first connected, the system was “sort of working”. In other words, because of some miracle, I am not sure why, the base station was detected and I was actually using the headset.

As a matter of fact, the engineer was able to run many tests while the system could “see” the base station. While the base station was being seen, it was still not working right. While the debugging session was happening, I observed two things (visible on the desktop, which renders what the user is seeing on the HMD)

The picture would at times jerk/move a bit. Every few seconds it would “tremble” a bit
At least on two occasions, the image slid to one side/diagonally, and then came back. This behavior is exactly the same as I experience when in a game, and I see the cockpit of the car start to move upwards to the side as if I was in free fall.

KEEP IN MIND THAT THE HMD WAS COMPLETELY STATIC, FIXED IN POSITION WHEN THESE TWO PROBLEMS SHOWED UP.

When the engineer did some debugging, he found a lot of error/sync messages as a result of the “dis” command, as you can see in the pictures below.

Ultimately, sooner or later he would end up with an “Assertion failed: explicit right sync does not match pending frame”

The problem above, or these error messages, never went away, but the situation got worse.

A couple of times I was asked to unplug all cables, inclusive of the headset, and disconnect everything from the computer except for the Pimax stuff. I complied. Here’s the pic of the cable I disconnected, in addition to those at the back of the PC of course.

The engineer got the system to a point where it was somewhat stable, except for the two points mentioned above. In order to fix them, I suppose, he decided to upgrade the headset’s firmware. He uploaded the P2_DEBUG_8KX_M299_t20210922093913_r6066b70.dfu file to the desktop and then used the DFU utility to burn it into the HMD.

After that was done, however, everything went backward. Now the base station was once again not seen in PiTools or anywhere, not even in lighthouse_console.exe.

At around this time, the lighthouse_watchman_update.exe command was tried, not sure if it worked or not.

Multiple times the engineer tried to download a set of tools/updaters, and after three tries, the download succeeded.

However, I am not sure it was used for anything, as every time that we tried to update the firmware, the following error message would get displayed:

He also tried to use the 123.json script, but for some reason, he was not able to paste the correct LHR id into the script, so I am not sure it was ever used for anything, in spite of trying.

Perhaps these two things ought to be attempted. If you send me instructions on what to do I will happily do it, or help, as I offered in the chat:

Ultimately, the support engineer gave up, and responded with:

“I have sent your log to the R&D team. Please be patient and wait for the result. Please don’t worry. If there is a result, I will notify you immediately”

I have just tried closing and restarting PiTools, and this is what I see.

It looks encouraging, as you have it in “Lighthouse” mode as well as 90Hz. Furthermore….

This image above might lead you to believe the 8KX is being tracked… But that is a mirage, as you can see that the base station icon is not blue, and also when you use the HMD you notice it is not working. Then, if you try lighthouse_console.exe, you do not even see the base station, as you can see below.

I have gone to great lengths to document this in hopes you take some time to read it and you can find a solution. The debugging sessions seem very much “hunt and peck” and repetitive. I can help if you tell me how.

Thanks

— Luis

October 27

October 28

Same day, about 5:30pm PST

October 29

November 1

November 2

November 3

November 10

November 11

November 12

The escalation above had zero effect. I did not even receive a reply from Joan, or an acknowledgment from anybody.

In what follows, for about a month, I continued to work with support, only for support to continue to try the same non-solutions. No new ideas, or fixes forthcoming. I think we ended doing two more remote sessions in this period.

The above message arrived at 10:25 pm my time, inviting me to confirm or deny that the ticket has been resolved. I was sleeping at that time. Not one hour later, at 11:19 pm my time, the email stating that the ticket had been “closed” arrived.