New Nvidia driver Version: 382.05 and Pimax 4K



I know that the ghosting issue has been mentioned a myriad of times. I’ve updated to the 382.05 Nvidia drivers released last week and it looks as if the ghosting is worse than before. Where it mostly manifested in horizontal head movement, now it seems vertical movement is also affected and the ghosting seems more pronounced… So far I haven’t rolled the driver back as I’m still messing about with settings etc. Have any of you perhaps noticed the ghosting being worse than before?

Also is there any way of lowering the persistence mode of the display using the debugger codes like we currently need to do to increase brightness?

Lastly, I know your engineers are aware of the ghosting problems. What do they see as the root cause of the problem and how can we as Pimax customers manage this better until a fix to the problem is found or at least officially confirmed that a fix is not possible and we have to live with it that way?

I received my Pimax 4k about two weeks ago, so I am assuming that my firmware version is the latest build.

Thank you in advance

Some of the ghosting can be attributed to the less than adequate gyroscopes that the Pimax 4K utilizes or perhaps even the driver communication with the gyroscopes.

But how do you guys even define ‘ghosting’ ? It’s just that ‘blurry’ effect when you move, right ? I think that’s entirely caused by the LCD panel’s needed time to change a pixel (the display panel latency). It’s just a downside of using LCD for VR.

1 Like

No, my definition of ghosting is the strange visual image that appears upon the turning of head on X/Y axis as if the image is trying to catch up with your head movements.

Don’t believe the LCD is fully @ blame.

Through my testing with phone gyroscopes/opentrack, the ghosting is minimized by a good margin.

Here’s to hoping NoloVR will concur once it arrives.

Agreed, additionally if it were purely panel latency then the updated Pimax or nvidia driver would not have must of an impact on the ghosting to begin with. It could be that the processor in the HMD does not have enough horse power to drive the display and still process head tracking properly or it could be an optimization issue (which I am hoping for). I’m waiting for my Trackhat IR tracker to arrive (currently at customs). Once that is set up, I will also test and see whether using an external tracking solution will improve my experience as well and report back.

That’s if goldfish memory doesn’t screw it up for me LOL.

I’ve read a number of your posts and you definitely seem to know your stuff when it comes to the hardware, most certainly more than most. Lets face it, firmware flashing tools don’t build themselves :slight_smile:

The effect is exactly as CMM explained. The LCD display is definitely not the optimal solution and I am convinced that latency is part of the problem. But with the latest Pimax driver build, the ghosting effect became more pronounced and even more so with the new 385.05 NVidia driver which suggests (at least to me that is), that there is definitely a software component at play here. Initially the ghosting was quite prominent with horizontal head movement but with the new NVidia driver combined with the 1.2.xx PiPlay build, it is just as bad with vertical head movement. The 1.1.92 Pimax driver does not work in PiMax mode with the latest NVidia driver on my system and having spent considerable time last night not being able to get the HMD to work using 1.1.92 I eventually reverted back to 1.2.xx just to get it working again. I’ve decided to stick with the current 1.2.xx build however for now as a headset with ghosting is better than a headset with no working display :frowning: . It seems that I spend more time troubleshooting than actually using the HMD at the moment which is somewhat frustrating to say the least and for the time being it is working so I guess that counts for something.

I’m not playing too much games, don’t have too much time for it and I don’t use v1.2.x drivers because of the degraded image quality, so I can’t really compare. But in my definition what you guys are talking about is not ghosting, but of course that’s just my definition :wink: In VR they use the term ‘motion-to-photon latency’, that’s probably a much better concept in this case and that’s defined as the time between a head movement and the time it displays on the screen. In that case there are indeed several factors that can degrade that, including graphics drivers, HMD drivers, games, Steam, and HMD sensors.

Increased latency indeed might be caused by v1.2.x drivers, but as I said, I’m not using them so can’t really judge about it.

Using Piplay 1.1.92 here as well.

Will never be able to accept reduced visual quality and until that’s addressed, 1.1.92 is what will be in use.

I agree, problem for me is getting back onto the 1. 1.92 build. I’ll take another stab at it when I need my joy killed and have the energy though as 1.2.xx sucks.