• Ever wanted an RSS feed of all your favorite gaming news sites? Go check out our new Gaming Headlines feed! Read more about it here.
  • We have made minor adjustments to how the search bar works on ResetEra. You can read about the changes here.

TetraGenesis

Member
Oct 27, 2017
3,138
I feel like a lot of people are missing the fact that most of us currently running VR Headsets use like, pretty obscene amounts of Super Sampling if we can. So we're already rendering at some really high resolutions. If you just actually have those more pixels such that you don't need to super sample anymore, just present them closer to native res, you'll be having a better experience with resolutions you're already pushing.

I run most stuff on my Vive at 2.0x SS or more. I run Fallout 4 VR at 2.5x. And of course, you can always render in the other direction as well, subnative and it'll still likely look better due to having more pixels.

Okay, sure. But that just reinforces my question? What kind of rig do you need to run this? (This = 4K per eye variety) All I know now is that you have one. Whatever it is.

Well a lot of us didn't get the 4k per eye version. We did the 2560*1440 per eye "5k" variation, I don't know which was tested at CES. But there's also a 4k per eye version who's actual input is still 2560*1440.

Ah, good to know. But yeah, I'm specifically wondering about the 4K per eye version.
 

I KILL PXLS

Member
Oct 25, 2017
11,542
Wow, whole bunch of bogus links. RoadToVR says it's way better than before, and just has a few more issues that would be easy to fix in the next few months. Then you linked to a bunch of random Redditors bitching about how unprofessional the booth was, how they weren't able to try it, and one guy who said it "felt weird". Oh yeah, and one guy who "read a CES assessment report" and said to ask him about it in Chinese.
Ehhh, setting aside the Redditor posts, that RoadToVr impression piece isn't nearly as positive as you suggest. The only improvement they mention is that the distortion on the edges of your view are greatly improved, but then they mention new display issues are there now. Literally everything else they mention is on the negative side outside of the fact that the resolution is still noticeably better than Rift/Vive and even then they say "but maybe not as much as you'd expect if relying on a simple on-paper resolution comparison" and go on to mention the screen door effect is still very much a thing. Also this paragraph:

Above is a lot of detail about what ultimately boils down to this: the Pimax "8K" has potential, but it's still up in the air whether or not the company can sufficiently tune up the headset to please consumers who are pitting the small startup against expectations set by the top companies and headsets in the industry. There doesn't seem to be any unproven hurdles in their way, but it's going to take careful execution to get there.

...doesn't exactly sound like "just has a few more issues that would be easy to fix in the next few months" either.
 

Aztechnology

Community Resettler
Avenger
Oct 25, 2017
14,139
I really want them to work on a lower resolution option for the far peripheries, scaled rendering. I don't need the same details in my periphery as I do in the center of my vision, and it would open up some performance to us.
 

Alo81

Member
Oct 27, 2017
548
Okay, sure. But that just reinforces my question? What kind of rig do you need to run this? (This = 4K per eye variety) All I know now is that you have one. Whatever it is.
I've got a 1080ti as well.

To contextualize it a bit, doing the math - rendering at 3840x2160 per eye would be equivalent to a Steam SS multiplayer value of 3.26 with the current Vive. That's certainly hefty, but not so far future that there aren't already some VR games playable as such, or others that wouldn't be potentially attainable with a theoretical 1180ti.

If we're talking about the 5k model (2560x1440 per eye) that lines up to about 1.45x in current steam SS.

The main point I'd like to get across is that these metrics aren't unfathomably large or unachievable.

Math based off of Durante's post here, minus the inherent Fallout SS.
https://www.reddit.com/r/fo4vr/comments/7jlr1a/full_investigation_of_the_2_supersampling_settings/
 

TetraGenesis

Member
Oct 27, 2017
3,138
I've got a 1080ti as well.

To contextualize it a bit, doing the math - rendering at 3840x2160 per eye would be equivalent to a Steam SS multiplayer value of 3.26 with the current Vive. That's certainly hefty, but not so far future that there aren't already some VR games playable as such, or others that wouldn't be potentially attainable with a theoretical 1180ti.

If we're talking about the 5k model (2560x1440 per eye) that lines up to about 1.45x in current steam SS.

The main point I'd like to get across is that these metrics aren't unfathomably large or unachievable.

Math based off of Durante's post here, minus the inherent Fallout SS.
https://www.reddit.com/r/fo4vr/comments/7jlr1a/full_investigation_of_the_2_supersampling_settings/
Okay, that makes a lot more sense. Thank you for clarifying! They have some future proofing going on here with the highest tier model but it's good to know it's more reasonable than I expected.
 

Simplex

Member
Oct 27, 2017
405
Wow, whole bunch of bogus links. RoadToVR says it's way better than before, and just has a few more issues that would be easy to fix in the next few months.
Did you already forget that the original shipping date was January?

Then you linked to a bunch of random Redditors bitching about how unprofessional the booth was, how they weren't able to try it, and one guy who said it "felt weird". Oh yeah, and one guy who "read a CES assessment report" and said to ask him about it in Chinese.
I think you misunderstand what word "bogus" means.
http://forum.pimaxvr.com/t/pimax-time-to-answer-the-ces-train-wreck/4955

And one more "bogus" link for you:
1f6cb3084275919011379c5c986391c22b966e74_1_690x463.png


Pimax response:
V5 is the first prototype after the tooling for mass production. It requests lots of tuning works. However, we still decide to bring it to CES to showcase the backers the latest version.

We made several changes with v5 before CES. For example, the lenses and the mechanical design of the lenses are both new tooling version. The change may lead to new issues, e.g. the reflections, but the issues will be eliminated and the benefit of the new design will reveal after fine tuning. To reach approximate 30% increase in terms of brightness level, we added 3 more LEDs in the backlight panel from 12 to 15 and increased the LED current supply to the panel as well. As a result, the current supply in other parts decreases and new issues occurs e.g. tracking performance. It's not hard to resolve, the team is working on it. We use valve tracking technology, the tracking performance won't be an issue at all in the final product.

After hardware and software fine tuning and all the issues resolved, we plan to ship limited number of testing units to selected backers, and may bring the units to more backers to test via roadshows. The details of the issues and analysis, along with shipping schedule will be updated later this month.
 
Last edited:

Aztechnology

Community Resettler
Avenger
Oct 25, 2017
14,139
Maybe Facebook need to buy Pimax..?
The issue is these extra features are not really something those big companies are wanting to focus on. The resolution per eye, expanded FOV etc are interests for them. But currently the focus is on the supply side etc. It would have to be someone like Nvidia or Intel and then it would likely be about licensing the tech. Which would be fine, but it's hard to push competition forward without items in the wild.
 

TetraGenesis

Member
Oct 27, 2017
3,138
PC Gamer has a really positive article: http://www.pcgamer.com/i-tried-pimaxs-5k-and-8k-vr-headset-with-200-degree-field-of-view/
It's still an open question how this will all shake out.

What I know is that this:
ytjgLzzQivPwjXrqjN6VPX-650-80.jpg

looks seriously awesome compared to all the "standard" HMDs.

Genuine question, Durante. I do all my VR on PSVR because I bought into that in a major way before I had any sort of PC capable of running VR.

They say the 8K version was running on a single 1080 (maybe a Ti but they couldn't say). I don't have a gaming PC but I have a 1080 Ti that I use in an eGPU with the spring 2017 HP Spectre x360 15".

The memory is 16 GB DDR4-2133 SDRAM (2 x 8 GB) and the processor is an Intel Core i7-7500U (2.7 GHz, up to 3.5 GHz, 4 MB cache, 2 cores).

Obviously there are limitations to an eGPU. Thunderbolt 3 is max 40 MB/s instead of native PCIe 3.0's 1000 MB/s and the Thunderbolt 3 port for the Spectre x360 15" (not the 13" for some reason) is only 2 lanes PCIe instead of 4 lanes.

Admittedly, I'm not totally sure what that last one means, hahaha.

Anyway. In your expert opinion, how equipped would that setup be to run any VR, let alone -say- the 5k PIMAX?

EDIT: Just realized it only has one USB-B port and one USB-C port when you account for the Thunderbolt 3 USB-C being occupied for the eGPU. Since I imagine there's not enough bandwidth via a USB dock, that puts the kaput on VR, I guess ¯\_(ツ)_/¯
 
Last edited:

Simplex

Member
Oct 27, 2017
405
I am rooting for Pimax because I'm a backer, but I backed with full awareness that this may be a catastrophe. Let's all hope they deliver.

DCS demo:
https://www.youtube.com/watch?v=nfcse_rY5V8

This was probably already discussed, but seeing how wide is the headset, I am a bit concerned about aiming down sights in games like Onward, or in games where you shoot a bow.
 

Dreamwriter

Member
Oct 27, 2017
7,461
I am rooting for Pimax because I'm a backer, but I backed with full awareness that this may be a catastrophe. Let's all hope they deliver.

DCS demo:
https://www.youtube.com/watch?v=nfcse_rY5V8

This was probably already discussed, but seeing how wide is the headset, I am a bit concerned about aiming down sights in games like Onward, or in games where you shoot a bow.
I don't understand the concern...how could extra width cause issues?
 

Ferrio

Member
Oct 25, 2017
18,072
I don't understand the concern...how could extra width cause issues?

If the headset is too bulky you can't bring up your controls to your head because it physically blocks them. I already have that issue with Vive sometimes when trying to aim down sights and the controller bumps into the headset.
 

Dreamwriter

Member
Oct 27, 2017
7,461
If the headset is too bulky you can't bring up your controls to your head because it physically blocks them. I already have that issue with Vive sometimes when trying to aim down sights and the controller bumps into the headset.
Gottit, I'm more used to the smaller Rift controllers than Vive. Hopefully the Pimax controllers and/or Knuckles work well.
 

Mascot

Member
Oct 26, 2017
1,706
All quiet on the Pimax front.
Any updates in the past five weeks?
90fps achieved yet?
Shipping dates narrowed down?
Photos of the devs sipping cocktails on the caribbean island they bought with some money they 'found'..?
:P
 

Fafalada

Member
Oct 27, 2017
3,066
To contextualize it a bit, doing the math - rendering at 3840x2160 per eye would be equivalent to a Steam SS multiplayer value of 3.26 with the current Vive.
I'm a bit lost where you're getting that number? That'd be 1.8x multiplier on each axis where I'm seeing 3.5x horizontal scale.
Or to put things into perspective - once you account for Lens Correction supersample, we're looking at 38MPix frame for both eyes. It's pretty clear naive approach isn't going to cut it with existing GPUs.
 
Last edited:

Chaosblade

Resettlement Advisor
Member
Oct 25, 2017
6,596
I don't get the "scam that's never shipping" reactions to this. A scam project doesn't develop multiple fully functional prototypes (much less thier prior "4K" headset) and attend events demoing it or send them out to previewers.

It might not meet expectations but I'd be pretty shocked if they don't actually release anything.
 
OP
OP
Durante

Durante

Dark Souls Man
Member
Oct 24, 2017
5,074
Saying it's "all quiet" just after a large update is a bit silly :P

I'm a bit lost where you're getting that number? That'd be 1.8x multiplier on each axis where I'm seeing 3.5x horizontal scale.
I think Alo's value is based off of the idea that you wouldn't have inherent (lens correction) supersampling on the PIMAX.
(3840*2160)/(1512*1680) = 3.26
And if they ship at 80 Hz, that would actually shrink to 2.9.

Of course, that's not entirely apples to apples, since you'd get fewer rendered pixels per display pixel.
That said, it's also not all that far-fetched, some HMDs already ship with a significantly smaller default correction supersampling factor than the Vive.
 
Last edited:

Fafalada

Member
Oct 27, 2017
3,066
That said, it's also not all that far-fetched, some HMDs already ship with a significantly smaller default correction supersampling factor than the Vive.
Short of killing the FOV completely, there'll always be some SS required to match the curvature though. Also - to be fair, lens-matched rendering only needs about equivalent of HMD's native-resolution worth of pixels if implemented well, but that optimization works for all headsets - so ultimately the delta scales with display resolution, without the fudge factors.

Though I'm curious, of the headsets that need smaller correction - what are the tradeoffs? Of the higher profile ones, FOV and other elements fall into roughly the same range (Oculus, Gear, PSVR) alongside having very similar optimal supersampling targets.
 
OP
OP
Durante

Durante

Dark Souls Man
Member
Oct 24, 2017
5,074
Though I'm curious, of the headsets that need smaller correction - what are the tradeoffs? Of the higher profile ones, FOV and other elements fall into roughly the same range (Oculus, Gear, PSVR) alongside having very similar optimal supersampling targets.
I don't think they necessarily need smaller correction - they just use smaller values. I assume the tradeoff of lower quality at the default super sampling settings is accepted. I.e. all the WMR HMDs render at close to native panel resolution, and the reason for that is likely just MS wanting to advertise lower minimum specs.
 

Fafalada

Member
Oct 27, 2017
3,066
I don't think they necessarily need smaller correction - they just use smaller values.
Ok - but then in context of original post I responded to, we're the realm of arguing "utilizing" 4k TV is perfectly achievable as long as you're ok just using Xbox1-S output with it.
GPUs have some catching-up to do to be able to comfortably utilize PIMAX, I don't think there's any ways around that.
 
OP
OP
Durante

Durante

Dark Souls Man
Member
Oct 24, 2017
5,074
I've thought about this a while ago, and I believe that it's more straightforward to consider the increased system requirements in FoV terms rather than hardware pixel terms.

Because it's very hard to get an "objective" idea of which rendering resolution you need to fully utilize a given device.
Apparently, Microsoft think that you need a supersampling factor of 1.1 or so, HTC think that it's 1.44, and after playing around with supersampling in many games I'd happily make the case for at least 2.0 supersampling over the native panel resolution being required to "fully utilize" a given hardware resolution.

On the other hand, it's very clear that the ~2x increase in FoV for the PIMAX (any model of it regardless of panel/transfer resolution) will necessitate at least a 2x increase in pixel throughput just to match the rendering quality of a game on a traditional "low-FOV" HMD. (Barring any fundamental technological shift like foveated rendering)

Of course, that's just the rendering quality -- the display quality is then another can of worms entirely, and dependent on hardware resolution, panel subpixel layout and fill rate, and of course again the FoV you are traing to cover. And then there's optical quality to consider, and how much of the displays you are actually using. All put together it's rather involved.

This is also why I don't think that your TV comparison is completely applicable (though I also don't disagree completely with it) -- it's not like the subpixel visibility and panel fill rate of 1080p TVs is something a lot of people notice or that is a real drawback, quite unlike current HMDs ;)
 
OP
OP
Durante

Durante

Dark Souls Man
Member
Oct 24, 2017
5,074
What I find frustrating about all these videos is that they don't ultimately tell you much.
Taking a simple high-res photo with locked focus, at equal lens distances, with different HMDs would show the actual result much better.
 
Oct 27, 2017
1,724
USA
What I find frustrating about all these videos is that they don't ultimately tell you much.
Taking a simple high-res photo with locked focus, at equal lens distances, with different HMDs would show the actual result much better.
Perhaps there is a reason they haven't done this.

The cynic in me says they aren't showing it because of problems or that it isn't as good as they promised. The optimist hopes that they are just holding off on doing it because they want to wait until they are on final consumer hardware and not prototypes.

We'll probably have to wait until 3rd parties get one in hand for those kinds of tests.
 
OP
OP
Durante

Durante

Dark Souls Man
Member
Oct 24, 2017
5,074
They posted a "M1 assembly" photo:
609d6935d5a7ba30aebfcb5c531e7844a8096a17.jpg


Doesn't really tell us much, but the small batch of M1 units should probably ship out soon. This is in accordance with the timeline posted in the beginning of March.
 

Aztechnology

Community Resettler
Avenger
Oct 25, 2017
14,139
They posted a "M1 assembly" photo:
609d6935d5a7ba30aebfcb5c531e7844a8096a17.jpg


Doesn't really tell us much, but the small batch of M1 units should probably ship out soon. This is in accordance with the timeline posted in the beginning of March.

They sent out emails today for 8k backers who are active in the community, reviewers and other notables to test these sets out. Some people here might qualify, I don't know. But I'd love to hear what you guys think if anyone gets it and doesn't have to sign an NDA lol.
 

Zephy

Member
Oct 27, 2017
6,168
With the current GPU prices crisis, I'm in no hurry to get the headset. Delays were to be expected (although they're still disappointing), I won't be able to use the headset with my current hardware and I'm not willing to pay a fortune for current gen GPUs so I don't really care if it takes longer than they initially announced. At least they seem concerned with making a better headset, so in the end it's worth waiting, for me at least.
 

Aztechnology

Community Resettler
Avenger
Oct 25, 2017
14,139
I'm ok waiting at this point for a polished product. I'm fairly happy with what I have now. I do want it, but I'm also waiting on the Vive wireless solution and knuckle controllers still. So all of those components are part of the "next gen" of VR for me.
 

Chaosblade

Resettlement Advisor
Member
Oct 25, 2017
6,596
We're getting closer to what I expected to be the actual release/launch at this point, I don't think I've ever backed a project that didn't miss its target, and the more ambitious the project the greater the delay. I'll start feeling iffy on it if it pushes into next year, delays for improvements are fine but there is a point where interest starts slipping. I've had that happen for a few games and another hardware kickstarter, where it does eventually deliver but it just takes so long I barely care anymore.

For what it's worth I fully anticipate another "delay" following the small beta test, to incorporate feedback. They said Q3 if things are perfect, so at this point I'm hoping for fall.
 

Dreamwriter

Member
Oct 27, 2017
7,461
The lenses are really the most important thing to get right, as they effect the entire experience and are the hardest thing to change later, so I'm perfectly fine waiting an extra couple months for better lenses. Note that even the Vive Pro didn't get new lenses, though it definitely could have used a larger sweet spot; it's one of the reasons reviews tend to be "best VR you can get...but only by a little bit, not really worth the price". Good lenses can be the difference between Pimax becoming widely known as the luxury next-gen VR headset, the best you can buy, and a random Kickstarted device that only a few enthusiasts use.

The only risk in a delay is that Oculus suddenly announces a high FOV headset.

Here's Pimax's take on it:
Xunshu at Pimax said:
The core team is specialized in display thus somewhat picky on the visual quality. Comparing with more costs, time, and extra works, it's even harder for the team to move on with the lens design that is not good enough from their point of view, even though many people were happy with the previous versions.

He said theoretically they could have chosen to keep the original lens design, start mass production and ship 8K units right away, making enough people happy to generate cash flow. But they couldn't bring themselves to do that.
 
Last edited:

Zephy

Member
Oct 27, 2017
6,168
Yeah my only fear at this point is that a bigger company puts out something better in the meantime... But seriously, two 4K screens... I don't expect that any time soon from Oculus or HTC.
 
OP
OP
Durante

Durante

Dark Souls Man
Member
Oct 24, 2017
5,074
The only risk in a delay is that Oculus suddenly announces a high FOV headset.
I really think that Oculus and other "large-scale" companies aren't interested in releasing something with the form factor required for 200+° FoV with current technology. It's too "geeky" / doesn't have enough mainstream potential.

Regarding the lens issue, I doubt that they will manage to create a lens system that works perfectly for everyone with this massive FoV based on flat screens.
That said, the lenses in basically every single consumer headset are just a bundle of trade-offs:
larger sweet spot vs less distortion vs glow effects vs sharpness vs screen-door vs artifacts/ridges vs ...
I'm sure they'll have their own advantages and disadvantages.
 

Dreamwriter

Member
Oct 27, 2017
7,461
That said, the lenses in basically every single consumer headset are just a bundle of trade-offs:
larger sweet spot vs less distortion vs glow effects vs sharpness vs screen-door vs artifacts/ridges vs ...
I'm sure they'll have their own advantages and disadvantages.
People have been raving about how awesome the lenses are in the Oculus Go
 

1-D_FE

Member
Oct 27, 2017
8,260
The lenses are really the most important thing to get right, as they effect the entire experience and are the hardest thing to change later, so I'm perfectly fine waiting an extra couple months for better lenses. Note that even the Vive Pro didn't get new lenses, though it definitely could have used a larger sweet spot; it's one of the reasons reviews tend to be "best VR you can get...but only by a little bit, not really worth the price". Good lenses can be the difference between Pimax becoming widely known as the luxury next-gen VR headset, the best you can buy, and a random Kickstarted device that only a few enthusiasts use.

The only risk in a delay is that Oculus suddenly announces a high FOV headset.

Here's Pimax's take on it:


He said theoretically they could have chosen to keep the original lens design, start mass production and ship 8K units right away, making enough people happy to generate cash flow. But they couldn't bring themselves to do that.

It only highlights how difficult optics are. I think the Vive has an amazingly large sweetspot versus the HMDs I've tried. So much of this is tied to a person's physical characteristics. It's why Oculus and HTC both went fresnal with gen 1 (to try and create a large sweetspot for the largest group of users).

I think it's funny reading all those impressions of people doing the Vive/Gear VR lens mod. Most of the people doing it complained of really small sweetspots with their original lenses. And most of those who said the mod made their sweetspot smaller, were people who didn't have issues with the original sweetspot.

There's also been a lot more complaints about that mod as time went by. One of the guys who made a guide recently posted he switched back because he realized it was the thing that had started giving him eye strain. Which can be very subtle and not obvious at first. It's such a complex problem. You start pushing that out to 200 degrees, and you're just magnifying every single one of the issues with lenses.
 

EvilBoris

Prophet of Truth - HDTVtest
Verified
Oct 29, 2017
16,684
You start pushing that out to 200 degrees, and you're just magnifying every single one of the issues with lenses.

I'd imagine that the Optics in these headsets and others are super rudimentary injection mounded off the shelf products. They are likely totally rotationally symmetrical in design , this just won't cut it when your lens optics are placed at such an angle to the eyes.