Pimax Interview – An Indie Company’s Effort to Push VR Forward
Pimax is, for all intents and purposes, the proverbial David fighting against the Goliaths of the Virtual Reality market. Whereas companies like Facebook (Oculus), Valve, HTC and Sony (PlayStation VR) have been known globally for decades, Pimax was founded less than six years ago by a group of engineers and VR enthusiasts.
Their goal was to ultimately deliver a high-resolution Virtual Reality headset that could virtually eliminate the so-called Screen Door Effect (SDE) while also allowing a much wider Field of View (FoV) than any other available headsets, in order to surpass the binocular-like vision that VR users are currently accustomed to. After shipping the first product with the Pimax 4K in April 2016, Pimax went all-in on its 8K project, which eventually landed on Kickstarter where it got funded with $4,236,618 on November 3rd, 2017. It still holds the Guinness World Record for the most successful crowdfunded Virtual Reality project to date.
That said, as one could have easily anticipated, the small company encountered a series of production issues which led to several delays. The delivery of Pimax 8K headsets to all Kickstarter backers was only completed about a year ago, and even then the headset itself wasn't quite as good as the customers (and perhaps the company itself, too) had hoped for.
Amidst the initial issues, though, Pimax had found a niche. None of those giant companies were providing a VR headset with comparable specifications, particularly with regards to the resolution and FoV. Thus, the company kept iterating and coming up with new headset designs.
At CES 2020, when COVID-19 still hadn't disrupted the plans of the whole world, Pimax had made a series of announcements of its refreshed lineup, going all the way up to the 8KX, which should meet and even surpass the original promises made with the 8K headset.
Obviously, the company's schedule was impacted by the pandemic just as everyone else's on the planet. Nonetheless, production has now resumed, and we've recently got the chance to discuss everything that's going on at Pimax with Chief Operating Officer Kevin Henderson. Brace yourself for a long, interesting chat, particularly for VR enthusiasts.
How did COVID-19 affect your manufacturing pipeline and production in general?
Well, they wouldn't let anybody in the factory for one thing, and we couldn't go to the office, either. Beyond that, our suppliers had the same problems, so the whole supply chain was impacted.
Even just getting materials and plastics and metals and, you know, all the basics that you need. It was a tough time, but slowly it has been getting back to normal. And now all of the employees are back at work at our two main locations in China, we're fully staffed again and we're up and running. Our daily meetings are all back operational and the whole system is going So, yeah, I can say that, at least for China, it is definitely better than it was, rather dramatically so.
How much of a delay would you say you've suffered in production at Pimax? Is it about a month or more?
Different things have had different effects. For instance, with our headsets, we had a baseline delay of about a month. But then when we got up and running again, there were more delays related to some of the materials and some of the suppliers.
What we did was we manufactured what we had that was in the pipeline with parts that we had at the factory and things already on the way, so we were able to crank out some of the newer headsets, the 8K+ in particular as that one was in the pipeline much more significantly. So we were able to build quite a few of those and now that we are back up and running, the majority of the manufacturing has been that, because that's what we had the most parts for.
Right. And I've read that you have been slowing down production for the Pimax Artisan, right?
Yeah, it was slowed down. In fact, it was almost at a stop for over a month, but now, the Artisan is moving back into the slot again. So I think that the Artisan also will be back into full production in a matter of days.
Do you already have an estimate on when the Pimax Artisan could be shipped?
Well, there's two parts to that. There's the people who had ordered them. And who are we know now we're, we're shipping though, catching up on those. And then there's new ones. We're still about three weeks away from getting it into the pipeline in a nice way. We'll slowly make inventory available.
What we've been trying to do is sync this on the website where it says it's in stock with the warehouse. So now, when it says it's in stock on the site, it should actually be in stock where people can order it right then. The same thing goes with the Pimax 8K+ units, we're shipping those within 24 hours from the order most of the time. That's because we use Amazon's fulfillment center so that orders from our site are processed through Amazon.
Can you share an estimate on how many Pimax 8K+ units you've shipped thus far?
I'd say it's in the thousands.
Alright. Let's grab some questions from the Pimax community. What is the status of the open source program?
It is something that I've always been interested in. I'm a big fan of open source, I think it's something that benefits everyone. And I think in our case, being a small company, it's something that gets us to a place we probably couldn't be on our own. So I'm a giant advocate of it.
And the biggest drawback to it was that the version that we released was kind of a hybrid between an older one and a new one. But our next plan is to bring that up to date and on top of that, we're going to be adding one of our headset designs into open source and parts of our firmware into open source. Robin, our founder, had defined the three phases for open source and we did phase one, the only thing that we wish we'd had the resources to do, that we haven't done is to take that GitHub release that we did on the for the PI tool and bring it further up to date, which is something we definitely plan to do very soon.
Are the distortion profiles going to be editable at some point in the open source program?
I asked the team that very question when I saw it from you and they said that that is something that we're considering to do. Something that we'd like to do is open up the adjustment, the distortion profile adjustment curve, but I will say, you know, it's not something that we get many complaints about nowadays. It used to be a huge issue. In the early days, when I first started at Pimax, it is something that people used to bring up a lot.
So that's something that we focused on a lot. There have been like 30 releases of PiTool now, and 22 to 23 releases of firmware for the different headsets, each release getting an updated distortion profile or a new one. On top of that, we did many, many experiments with the way the cowling fits on your face. The result of that was our comfort kit, which is not just designed to be comfortable, it also changes the angle and the distance and it does it in a way that dramatically helps with distortion. So between the firmware updates that we've done, the PI tool base we've done and the comfort kit, out of maybe 600 or 700 people that tried the Pimax headsets at CES, only one or two even made mention that they could even see distortion.
So you see it as a problem that's becoming less and less prevalent.
It is dramatically less of an issue than it was when I started, it's something that's, you know, that is really it's something that we are always thinking about, because we don't want to cause that to happen. Now that we've got it down to the degree that people are happy with, generally speaking, we don't want to make changes that go in the wrong way. The other thing is on the new headsets, the panels are smaller and it makes it much easier for us to make the profile.
But yeah, the engineering team says they are interested in seeing if it's technically feasible to open up some tools where Pimax enthusiasts can make those adjustments on their own.
You've got a lot of upcoming add-on modules, from the eye tracking module to the hand tracking module, the wireless module and so on. How many of those can you use at once, though?
Here's what you can do. The modular audio strap connects to the audio jacks and in the recent headsets, we have audio jacks on both sides, that's one of the changes we made. The audio doesn't take up any of your slots, so you can add that one and still have your two USB-C top and bottom module adapter slots. You could use those for eye tracking and hand tracking at the same time. Or you could have, say, eye tracking and wireless at the same time. In effect, what it boils down to is you can have the modular audio strap plus two modules. In the future, we've been talking about multiplexing and having versions of the modules that combine more than one into the same modules, which would let you have three but right now, in the modules that were in limited production, you can only have two plus the modular audio strap, which for most people would be okay.
Good to know. Regarding the wireless module, the Oculus Quest is really popular, it's selling extremely well, and one of the reasons is that it's completely wireless. Can you talk about the wireless module that's coming for Pimax headsets?
When you look at our resolution, which is 4K native per eye with the 8KX, the data transfer rate requirement is real high. That definitely complicates the ability to go wireless. However, we do have a solution for it, there is a solution out there with sufficient bandwidth that we've been experimenting with for a while. And at first, we were telling people that we thought that 8KX would not be able to have full resolution across wireless, but I think we will, and it will also handle all of our Pimax headsets.
Do you expect minimal latency?
It is minimal latency. I can't get into the details of the solution directly, but I'll just tell you the bandwidth on it is very high. And it's actually high enough to encompass greater than 4K per eye, so the solution is actually very low latency. I think people will be satisfied with it. The key is to manufacture enough of them where the price becomes reasonable.
Can you disclose the manufacturer or is that still under wraps for now?
I think the supplier or pipeline of that is fully protected. I can just tell you, it's something that we've been working on for a while. We did switch out some of the parts to encompass the bandwidth of the 8KX. But it makes the solution much, much more enduring and in fact, this solution will work on future headsets as well, so it would be the same solution for years to come.
Do you have an idea if it will be available this year?
I would think it would be available this year. But the problem with it is, again, to manufacture enough of them. We have to have a plan for the actual manufacturing where it's a sufficient volume to hit the price targets that we have. I think we have a good plan to achieve that, working with our partners. Between our newest headsets and the existing ones, there's a pretty good population of Pimax headsets out there. Anyone that already has a headset can pick up this module, so I do think we have sufficient volume to hit our price target, we've run the numbers on that and I think it's gonna be reasonable.
Is there a possibility to get a higher than 75 Hz refresh rate for the Pimax 8KX in native mode, or is that too much to expect?
No, I wouldn't say it's too much to look for, I would just say to look at our history on the refresh rates. With the 5K+, we were originally showing it at 80 Hz. And then it was 85. And then finally, when we released it, it was 90. And then today, you can drive a 5K+ at 120 or even 140 Hz. We have the 5K+ Super, which is 180.
We've come a long way with increasing those refresh rates. People even on our very oldest headsets can now do 110 hertz for free. So we're always experimenting with that. And it's something that we are interested in adding back if it's possible. If we do achieve that, we'll send it out as a firmware update.
And by the way, you should know that if you put the 8KX into scaled mode, you automatically get up to 90 hertz anyway and it's still 4K per eye, just scaled, so the SDE (screen door effect) is still small. If you're playing something like VR table tennis where it's very fast, or anything like an action game or Beat Saber and you want to have a high refresh rate, just launch the headset in the scaled mode, bump it up to 90 Hz, you can do that right now. And of course, as we move along we hope to improve that, just like we have with our previous headsets.
On the same software front, can you speak about the evolution of BrainWarp? You've recently discussed the 2.0 version at CES, when you said it can reach 120 hertz on average, surpassing the industry standard.
That's what we're testing with. The base on an 8K+ is 110. And we've been experimenting with faster, you know, with the new version of BrainWarp, we hope to achieve stable refresh rates that are fast and once we do, we'll release them for free updates, just like we have on the older headsets.
There are a ton of new features that are going into PiTool, we have a giant roadmap for it. Now as far as the BrainWarp piece of PiTool goes, driving the panels at higher refresh rates is something that we're always working on. The other thing is integrated DFR (dynamic foveated rendering). With the eye tracking, we're going to be tying PiTool and the eye tracking solution together very, very tightly to achieve the least latency and the least performance impact you can get between the eye tracking solution and the headset solution. We have a version of PiTool that integrates DFR, so if you have the eye tracking module, you can get universal DFR inside PiTool in the future.
We're doing a lot of things with user configuration, we're exposing advanced settings out. We just put the GPU catalyst feature recently, which lets you optimize for flight and racing sims so that you can get the maximum performance out of your GPU. That was something that we exposed in the advanced settings and we're gonna continue to do that, we have a ton of upcoming features aimed at exposing advanced settings to power users. Of course, the more you do that, the more danger there is of having something that's really difficult to understand. Of course, some people on the forums and other places kind of dedicate themselves to learning what all these options are. One of the things that we're working on is to have some sort of automatic settings, where the user doesn't have to meticulously go through, you know, a dozen settings to get the optimal experience for any particular game or software.
Kind of like a Pimax version of GeForce Experience from NVIDIA?
Yeah, we're looking at things like that, hopefully we can dramatically improve the user friendliness factor of it, have an advanced power user section and then something where it's much more automated. The same thing with BrainWarp, the features of BrainWarp that we add, which is quite a list, would be activated automatically for some of those people.
Is there alternate eye rendering among those features?
They're experimenting with a lot of things. That's something that we haven't implemented yet into it, but it's being experimented with. The key is to get it to where it's completely stable and where it's compatible with most things. The problem is there's so many different methodologies that games use to render.
When we introduced parallel projections, the ability for PiTool to compensate for that, that was an attempt to bring in games that didn't automatically calculate the angles, for wide FoV especially, as the wider it is, the further the angles are from your eyes. So we wanted to have that where that was automatic and once you check that checkbox, it all happens in the background and corrects for it. Of course, in the future, we want to have it where it can detect whether it's needed or not. The thing is, for things like alternate eye rendering and a lot of other techniques that we would like to get in there, we hope to implement them in ways where they're compatible with everything. The DFR, the dynamic foveated rendering that we're integrating tightly into PiTool coming up with the eye tracking, does work with pretty much all software. So you'll get a nice benefit across the board and on top of that, we also will let you adjust the strength of it, so you can determine how dramatic the effect is.
Regarding the eye tracking module, do you have any further information to share on its availability?
The problem with some of the releases where we're working with partners, like 7invensun and Ultraleap and others, is that some of that supply chain was also affected just like ours, so it's hard to nail down exactly when we'll be able to ship those. For the eye tracking, we were at the very beginnings of limited production on that. The problem with that kind of thing is that we're tooling how we're manufacturing housings for the headsets in the same plant where we manufacture the housing for the eye tracking and the hand tracking. It has to be where we can get those done and get those finally tested and assembled and everything. But generally speaking, the designs themselves are completed and even the drivers are now complete for both of those. So, I would say that we're right at the end game. Our plan was to actually have those things out and be able to show some units at GDC, which, of course, isn't happening now. But we wanted to announce some rapid shipping at some point very soon after that, but I think now we're just trying to work with the suppliers and the partners to get the production numbers from them on some of the pieces that we need. I think in the next weeks we're gonna announce a hard timeline on that, but it's all ready to go as far as completing the engineering for them. I will say this, we're also working with some software companies on the hand tracking, and we've got some software that we'll be able to talk about that's optimized for it and I think people will be really happy with some of that.
I think that some of the games that are gonna take advantage of that will be really slick, what I've seen was very impressive, and I think that that will be released at the same time as the hand tracking module.
Are you trying to work a lot with game developers to optimize their games for Pimax hardware and software?
That's one example with the hand tracking module. Of course, working with people to eliminate the need for parallel projections is another thing and we do that all the time. On top of that, we have our B2B solutions, which are for experiences that are not necessarily consumer oriented. For those, you know, we help them on the engineering side and the software side, the driver side. Part of that is things like optimizing the frame rate, we do that every day. And there's quite a few that we work with on that. The key to that is, many of them know that while right now we're the only wide field of view headset out there, they know that in the future there will be other companies that make headsets that have a wide field of view and they want their games to be compatible with those. So they're planning for the future to be forward-looking. We get a lot of requests saying 'Hey, we want to optimize for a Pimax headset because we know that at some point in the future, there'll be other headsets that have the same challenges'. In our case, we're compatible with everything you can, most things now don't need parallel projections but some things do. We have a little checkbox in our driver where our software will automatically adjust for it. But if you want the maximum frame rate for each piece of software, they have to be the ones to optimize. And every day, there are new ones that are doing it, especially in the simulators like driving and flight, those are the ones that are really focusing on that kind of thing.
I also have to ask about your own controllers. How are they shaping up? I know that the community wants to know more, mostly with regards to the reasons behind all the delays.
We have two controllers, the Sword and the Sword Sense, that we're working on. It boils down to a matter of resources, it took Valve about three years to turn the Knuckles into the Index controllers. It's a difficult and expensive thing. We've gone through many, many iterations of it at one point.
Now we're back pretty much to what it originally looked like. That said, I think in the coming months we'll be releasing more information about that. But it's something that is in active development, we just haven't had the resources to attack it to a degree that Valve did with the Knuckles and there's a lot of testing that has to go into that. We didn't want problems, like controllers that could be damaged on impact, we didn't want to have it where they could easily fail for any reason, which requires a lot of extra testing. And we're continuing to do that, we are still committed to releasing the Sword and the Sword Sense controllers. But until then if someone does want controllers now, if they previously ordered a Sword or Sword Sense controller, they can apply to get the Index controllers instead if they want to.
As you mentioned earlier, we are still relatively in the infancy of VR, there's a lot of room to grow yet. Do you think that cloud based rendering could be one of those things that can help?
I'm a huge fan of that. I think that that is a big piece of the future where you could have a fully cloud based render or a hybrid render, where you use some of your GPU plus an assist. There are a lot of levels of that that are possible. I've seen all kinds of demonstrations and behind the scenes on for things like that. We've experimented with things like that, it's something that we are definitely interested in. That is potentially a big piece of the future for VR, both cloud rendering and hybrid cloud rendering, and it could reduce the cost of entry into VR rather dramatically, you could have a $500, $400 laptop theoretically and a very high end VR device using methods like that.
But that's the beauty of PiTool, by the way. Since we make this piece of software that we have control over, that we've developed, it's full of all kinds of capabilities. We can build those kinds of things into and we don't have to wait for other companies to package a unique mix of things into it. We can do it ourselves, we have full access, and we have our own firmware on the headsets so the whole design is custom. That's one of the reasons we're able to release so many headsets, people joke about we have a lot of headsets. That's why we have that capability, because it's all custom and it's all internal. And we can do short runs of things where we do 1000, 2000, or 5000 units or something.
Just following up on that, you've said it yourself, you've got a lot of Pimax headsets on the market now. A colleague of mine in the press told me that on one hand, this is good for customers as they have lots of options, but on the other end, the less discerning customers could maybe find it a bit confusing. So, what do you think about that?
Having more options generally is a good thing, but on the other hand, if you have too many options, it can be a confusing thing. My opinion on it is that there's so much room in the VR industry. I also think it comes down to how much effort it takes to evolve a new SKU versus trying to do something every five years or so. I mean, HTC has a huge number of SKUs as well, they actually have more than we do. But I think ours differentiate themselves more with the resolution and the types of panels that are available and various other things. With our headsets, you really do get to pick. And our modular audio strap, you just snap it, remove it and in the future, we could have different iterations of that. I think that because our headsets are modular, that's part of the value. That lends itself to having more SKUs as you move forward and lets us get things to market faster than the next guy.
When you have something that's a direct replacement for a previous item, like the 8K+ which is effectively a direct replacement for the 8K, eventually you discontinue your models. Sometimes you don't discontinue something because you just happen to be focusing on the next one and you don't outright discontinue it, you know? That's the business model that we've kind of attached ourselves to. I think being a small VR company, it works for us.
Of course, you don't have the resources of Valve or Facebook. The big question then is, do you wish to remain an independent company going forward or would you possibly consider acquisition proposals from big companies?
They got, I'm sure, different levels of partnership that are on the table, we do get some interest along the lines as we do have some pretty novel technology. But our focus right now is just trying to drive VR, push the envelope on the technology itself and bring VR into a place that...Personally, I think VR has a lot more running room on the resolution, on refresh rate, on comfort, on all the aspects of it. Being able to use VR in an everyday environment, working on everyday things and watching everyday media and having it where that's a place that you go naturally to see and do things. Right now, it's a kind of a novel thing that you do for a brief time. Somebody will say 'Hey, I'm gonna play for a while now', they'll get the VR headset and they'll play for an hour or two. Sometimes three, some people go great lengths of time, but generally speaking, it's not the go-to device for your average everyday activities. That is the beauty of VR for the future, as we move forward, it can impact other industries that it doesn't right now. If the question was, is Pimax willing to work with partners for different kinds of things, I'm sure they would look at those kinds of things, but I don't think the team is focused on anything like what you've suggested.
One last, big question, perhaps the hardest of the interview. I don't think anyone in the VR space disputes that Pimax is producing some of the most cutting edge headsets in terms of technology. If you go to places like Reddit or forums the main complaint is always the lack of communication and lack of transparency, lots of product delays that haven't been communicated in the best way, at least in the past. Is that something you are intending to address in the future? Because it seems to be by far the most common complaint people have with Pimax.
No, no, I agree. It's important to me that customers get taken care of, it's a big deal of any company's success. A lot of it is going to be based on how you communicate with customers and how you provide information about what's going on to them. So we made a couple of statements recently about that that are on our forums, we've been doing weekly updates on what's going on. Our founder (Zhibin "Robin" Weng) has taken personal control of the customer service aspect of our company, he is dramatically added to it. And that's the number one area we've expanded, that's been probably 70% of our expansion. And he's been on top of that personally, it's his personal focus. Most of the changes in the company recently revolved around communication, transparency and service. He is all over that, that's actually become kind of his thing. And our daily meetings mostly revolve around that, so I think that we've improved rather dramatically when it comes to that. And we've got hotlines now active 24 hours a day, we've got a chat service that also runs 24/7, we've added countless new people to that area. So I think between those things, we've really, really improved. Also, Amazon does our fulfillment now and they've done a great job, we've done been doing that more and more and more. That provides customers with, you know, info on how long something's going to take to ship and how many days it will take to actually get there and quick, almost instant tracking number.
So it's been great. And we're going to keep expanding that, again, we're a small company, it takes time to get these things implemented for such a small company. I think that the transparency, communication and service aspects will not be the number one thing people ask about Pimax anymore. It'll be, what's next? What features? It will revolve around those kinds of things a lot more.
So the goal is to improve also, customer service response times.
Which we have, the response times are way down from even two, three months ago.
Very well. Is there anything else you'd like to add for the Pimax community?
I just want to thank everybody for sticking with us, we really appreciate it. We are working really hard to get these things out. Our thoughts go to those that might be sick or have been otherwise afflicted by COVID-19, it's a really tragic situation the world is in right now, and we know that's your top priority but we just want you to know that we are here trying to push VR into the next level and get it to bigger, better places for everyone. At Pimax, we're fans of all VR, not just our own.
Thank you for your time.