Mobile Driving/Flying Cockpit with Motion and Tactile ( Build )

[RCHeliguy]

After all that audio discussion I've come to the conclusion that there are two reasonable audio options with the Aero.

1. Surround sound or a pair of external speakers
2. IEM's.

I've tried the over the head cans a few times and the fan vibration issue is very real. Unless you are playing with a game that has a pretty consistent volume at mid levels or better all the time, the fan vibrations are amplified by the headphone's structure and what you hear is a lower more annoying pitch than the easy to ignore fan noise without them.

Without any headphones on you can hear the fan a little bit, but it's not remotely annoying even when things are quiet and it's definitely non-intrusive. Surround sound works well and is obviously the most convenient.

With IEM's the fan noise isn't an issue at all. I can barely hear the fan noise and it's even less than without headphones.

I still consider having the headset's fan running a net positive because I'm having zero issues with fogging lenses while it's running which is all the time except when it first starts it up.

I've had a fan on my rig for a while that would blow air through my Index and that has always worked very well to prevent fogging when I am in my rig. However in roomscale it could sometimes be an issue. I've even turned on my overhead ceiling fan to keep air flow to help with fogging. With the Aero that isn't an issue anymore. I still use that fan on my rig to keep me cool, so it's not coming off my rig.

Others may feel differently about this vibration noise, or possibly have a headset that for some reason doesn't conduct vibrations as well. Or maybe their in game sound is alway up high enough that they can't hear the fan vibrations, but for me, the over the head cans are not an option.

 

[Deleted member 197115]

The Aero aspheric lenses are designed with something that I would almost call a physical version of fixed foveated rendering even though that really isn't correct. The lenses focus more pixels into the center of your view increasing the resolution there.

Wondering how that works with geometry correction.
All lenses give that pincushion distortion with more pixels allocated in the middle, but it has to be compensated with image barrel distortion to have proper final geometry.

https://smus.com/vr-lens-distortion/

 

[RCHeliguy]

I think the Aero is doing this to a larger extent than other headsets currently on the market. If you look at the ppd values listed for most headsets there is a horizontal ppd and a vertical ppd. The Aero has an average horizontal ppd, average vertical ppd and maximum ppd which is in the center.

I don't know the math behind the software that maps a display to a VR lens, but what little I understand shows it's extremely complex.

Here's my guess at how this works.

You start with a display with all of it's attributes, resolution, color space, refresh rate, latency,RGB pixel arrangement and other characteristics.

Then you design a lens to make the image on that flat display wrap around a viewing area. This lens needs to allow your eye to naturally focus at many different focal points from near to far and still get a clear in focus image. Lots of math, crazy optical calculations and lots of trial and error.

Then I assume you display a grid through the lenses and see where the pixels actually end up on the other side of the lens and use software to warp the image displayed through the lenses to create a more rectilinear grid for your eye.

But warping the image like this will spread and squish the separate red, green and blue pixels into new orientations which can lead to Chromatic Aberrations and other color problems, so then I'm assuming that you have to remap the the different color pixels independently through very low level software. This has to be done at the lowest level because what comes out of a video card has an RGB value for each pixel assuming they coexist in the same location. So your driver software would need to be aware of your specific color pattern to adjust for that. It's also likely that the different emitters are not all equally efficient and they will each need to be tweaked to be level. But on top of that the emitters are projected into a different sized area on the lens, so you may need to boost the brightness of the emitters going to less dense areas and drop the brightness where they are focused in a denser pattern like the center of the lens.

On top of that people's eyes are not the same distance apart, so the lenses move relative to each other and while I would assume that if the lenses are centered on each eyeball there shouldn't be any additional issues, it appears that there are issues related to supported a range of IPD's. I don't understand the issues behind this.

There have been displays made with all 3 color emitters stacked. This completely eliminates the CA issue, but this is more expensive and triples the number of pixels. So an 8Mp display would have an 8Mp stack for each color totaling 24Mp, but only have 8Mp viewable pixels. I believe XTAL did this a while back. The upside is that because you don't need to do software color compensation, you actually do get a sharper image for the 8Mp that you can see.

On top of all that once you get an image mapped properly then you have to deal with all the temporal stuff. How do you handle updating the displayed information and reconciling the native refresh rate and the incoming signal.

I'm sure that is just scratching the surface of the issues the VR headset companies are having to deal with.

 

[RCHeliguy]

The more I look at VR headset design the more amazing they are. The headaches involved in designing a headset that comfortably fits a wide assortment of head sizes and face structures and eyeball widths seems nightmarish.

The more I read comments and complaints from people about all the headsets out there, the more I'm glad I don't work in this incredibly frustrating space.

The whole Bigscreen Beyond concept makes a lot of sense in this context. You get a headset that fits your face exactly and because it is custom fitted to your face it can be ultra lightweight and not encumbered by all the adjustability issues other headsets face. I'm not call that the ultimate headset by a long shot, but I understand why a company would go in that direction.

I've considered the Index near perfect from an execution standpoint, and some features were very cool like the frunk that mostly goes unused but just as I finally had a use for the frunk I stopped using that headset. Great concept, bad timing.

In the almost 4 years since the Index was introduced we have seen a lot happen.

1. Much higher resolution.
2. Better color - micro LED
3. Better dynamic range blacker blacks micro OLED
4. Aspheric lenses ( no glare or god rays )
4. Ultra light smaller designs
5. Eye tracking ( dynamics foveated rendering, auto-IPD adjustment )
6. A headset customized for your face.
7. Ultra wide distortion free optics (XTAL)

In the 4-9K range
Varjo's VR-3 was the first headset with a retina display in the center.
XTAL was the first with a distortion free ultra wide FOV years back.

Based on rumors it would appear that the next Index will have:

1. Much higher resolution.
2. Much better dynamic range, now looking like 10 bit color and HDR
3. Eye tracking, auto-IPD, foveated rendering
4. Internal CPU(s) to "hopefully" take the load off the GPU and not just be for stand alone use.
5. Assumed aspheric lenses
6. Assumed off ear speakers
7. FOV unknown
8. Size & Weight unknown
9. Release date unknown

Going forward, my "guess" is that the Crystal will be like an alpha release of what the next gen VR headset should be and it will be interesting to see how it goes. I think it's possible that Varjo is working on an update to the Aero, and if so, I suspect it will be very good. I think the next Index will be what a complete next generation VR headset should be and I think it better be if it ends up with a 5 year gestation period.

I do like the Bigscreen Beyond idea and I assume that the Aero and next Index will both be largish comparatively.

My wishlist has historically been for the "trifecta" (High Resolution + Wide FOV + higher frame rates). Now I would add smaller form factor and light weight to that list.

Got to love a moving target! As companies show what is possible, desires only increase.

 

[Deleted member 197115]

I think the Aero is doing this to a larger extent than other headsets currently on the market. If you look at the IPD values listed for most headsets there is a horizontal IPD and a vertical IPD. The Aero has an average horizontal IPD, average vertical IPD and maximum IPD which is in the center.

Did you mean PPD? I think it's more of a foveated rendering software trickery, not lenses, even that in general all lenses are sharper in the center.

 

[RCHeliguy]

Did you mean PPD? I think it's more of a foveated rendering software trickery, not lenses, even that in general all lenses are sharper in the center.

Yes, thanks, I fixed my text to ppd.

As I understand it, they actually do focus more resolution in the center and it's not simply a matter of the center of a lens being sharper. I believe that they are doing a lot more image warping to achieve this than we "typically" see. Granted we don't have a pile of headsets on the market with aspherical lenses. It's possible that this becomes the norm, who knows?

 

[Deleted member 197115]

Sony dropped aspheric in favor of Fresnel design in PSVR2.
This article outlines some possible reasons.
Looks like next gen will be most likely pancake based once they solve inherent small FOV issue.

What is the Best Lens Technology for Virtual Reality?

What lens technology is best for a VR headset?

www.shaunpoore.com

https://vrx.vr-expert.com/the-difference-between-pancakes-lenses-and-current-fresnel-lenses-found-on-vr-headsets/

 

[RCHeliguy]

These lenses all have their place and have different strengths and weaknesses in terms of size/weight, cost, FOV and light transmission.

I'm all for making these headsets as light and comfortable as possible. Pancake may be the future and hopefully they can find ways to increase FOV. The other issue with pancake lenses is light attenuation and that is an intrinsic part of the design.

It will be interesting to see what Valve has coming next because of the potential 10bit color and HDR support. Using 10bit color doesn't mean that you actually have a full 4x increase in dynamic range, just that it can be represented numerically. Still I would "guess" that might preclude using pancake lenses unless they've made some kind of breakthrough with brighter displays. That also increases the data bandwidth by 25% assuming raw color data. Every time a bit more information is leaked/found the more questions are raised.

 

[RCHeliguy]

More feedback on the Aero. There is definitely a acclimation period.

Today I was in VR just feeling absolutely freaking blown away!!!! For some reason it just looked gorgeous to me !

Not only that but I'm obviously still learning how to position my IEM's. Today I twisted my 846's just a bit past where they normally press in and they became much louder and sounded even better.

I was only running 26% volume and they sounded solid and loud. Really happy with them.

Reality Check Time: I'll need to see how things feel on average over a number of days to see if I'm actually becoming more acclimated now and did find a better way to position my headphones or if I was just having a good day. Only time will tell, but today I was supremely happy with both the headset and audio solution.

I should mention that I got into the Steam FOV test room and measured ~110H by 85V FOV and felt this way despite that.

 

[RCHeliguy]

I finally started playing Half Life Alyx from the beginning at highest 35ppd. The 4090 is at 93% and the whole computer is running about 475W.

It is absolutely amazing how much detail is in that title and how well it holds up to close scrutiny. I going over every area closely just to see what they have.

I'm still feeling completely blown away. Headphones are still at only 26% and everything sounds crystal clear. They did such a phenomenal job on this game.

 

[RCHeliguy]

I'm not sure if the tactile vibrations are to blame or if these would come loose normally.

The large bolt controls the center spring tension for the H pattern shift. I've had it slowly back off over the years and I've retightened it a few times. I'm not sure if the Locktite will hold here. I didn't degrease the threads. This looks like it could use a second large nut to tighten against the existing nut.

Since I set up my Ultimate+ pedals last summer, I've had the clutch spring bolts come loose a couple times as well. There didn't appear to be any grease or oil on the clutch bolt, so I'm optimistic that this will hold. However a lock washer between the black tension ring and the nut would probably work well too. Or I could take a small pipe wrench to the black tension ring while tightening the nut against it tighter.

FWIW I've verified the bolts holding the brackets supporting the Race Bass isolation under my seat a few times and they have stayed solid since it tightened them well a week or so after installing my D-Box system.

 

[GagarynGaribaldi]

Replacing the existing nuts with nyloc nuts would probably fix your issue.

 

[RCHeliguy]

Replacing the existing nuts with nyloc nuts would probably fix your issue.

Yeah there are a few ways to keep a nut from moving!

 

[RCHeliguy]

The Aero in summary:

The Aero looks absolutely beautiful in the titles that you can achieve 35ppd and still looks very good in titles where you can only achieve 27ppd. At 35ppd the Aero looks substantially sharper than the G2. At 27ppd it is only slightly sharper, but still has a much larger infocus area to look at, so it is still an improvement. While I am feeling like the Aero is a compromise in the FOV department, the image quality is so much better that I can't go back to the Index.

Tracking with the Aero has been nearly as good as the Index, but the Aero either has fewer sensors on the headset or they are all placed facing forward. Also the Index used to remember my position when I opened up iRacing I was normally nearly perfectly centered before pressing the center HMD button. The Aero doesn't seem to remember that. I have seen a very rare and short lived tracking hiccup in room scale games that I didn't have with the Index, but it's usually when starting a room scale game where I've just grabbed or repositioned my headset likely blocking sensors with my heads. However on my rig, with full chassis motion in Rally games etc.. tracking has been perfect. No issues there at all.

There is a bit less stability, in that sometimes SteamVR will initially crash or Steam will not come up. You also have to exit out of any game when changing the ppd value in the Varjo Base software. I've had a couple instances where I had to unplug and plug my headset cable back in when the Base software for some reason couldn't see the headset.

I've gotten comfortable enough with the IEM's not to be overly upset about dealing with them, but they are an inconvenience. I will use surround speakers whenever I have guests. Typically guests want to ask questions, so that audio solution works better anyway.

I am enjoying the sharp images a lot and most of the time I do forget about the somewhat restricted FOV. However, it does still come to mind periodically especially in Room Scale games.

There is a bit of motion blur more noticeable in some titles than others, but it's there. Since this is not a hardware issue, I am hoping that Varjo will eventually update the base software so this issue goes away.

The Aero is at the limits of what a 13900K/4090 system can drive and it will likely take the next generation CPU/GPU before it can be fully harnessed in a lot more titles at 35ppd at least with current software optimizations. However I would still like more FOV at this same ppd, so I'm expecting that the next headset that I buy may require a 5090 to perform similarly to what I'm seeing now or rely on continued software optimizations.

With the Aero I'm no longer concerned about getting more resolution. I would go so far as to say past 35ppd, we are getting into diminishing returns for gaming. We could work to approach a full retina display, but why bother until we have a larger FOV and hardware that can render that much resolution and FOV. 35ppd is already extremely sharp. You can not see the pixels. So I would call that mission accomplished and nothing to focus on until other areas are addressed.

There are a LOT of optimizations that are possible with dynamic foveated rendering and other techniques that are likely to improve between now and the next big GPU release, so it's possible that by the time the next release that we could push a wide FOV headset at this level of detail or push the Aero in more titles with the existing hardware.

Time will tell.

I do think the Aero is the best thing available at the moment at a cost. I'll continue to enjoy mine until something I want more is released.

I don't expect to have much more to say about the Aero.

 

[SafetyThird]

Hi Mark, I found my way here after googling for ideas while building my new sim racing rig as I already had flight sim controls and wanted to incorporate everything into one unit. Well, that's turned into quite the deep rabbit hole

Over the past three weeks I've worked my way through this thread with great interest and no small amount of admiration along with a certain amount of envy, I have to admit. What an amazing system you've built. I've learned a huge amount about both sim racing hardware and also how to design and engineer solutions, something I have no background in. I'm what I think gets called a 'maker' these days, having been building stuff all my life but I'm more of a bodgineer than an engineer

You've gone down several paths I've been considering and ultimately your journey has shown that some of those are not the right way for me, which has saved me time, effort and money already, thanks for that.

It's going to be an interesting journey for me on a much smaller budget but it's been great fun reading through your own adventures so thank you very much indeed for putting in the effort to document everything so very well and being so open about both the successes and failures.

You mentioned that there's a tactile discord that requires an invite, could I ask for one of those please as I've just started to explore that side of things and am currently setting up some budget Dayton pucks as a first step and dipping my toe into simhub. I've posted some of that info over on the tactile thread here on RD.

Many thanks, J.

 

[RCHeliguy]

Just saw something smart by Simucube.

First they have created a mount for their new brake pedal that fits a SimLab P1-X which makes mounting easier. I could see that being challenging for some.

It's also designed to mount HE Sprint or Ultimate throttle and clutch pedals on either side of their brake pedal.

It also allows plugging the HE pedals into the Simucube module. Given the control board for the Sprints is in the brake pedal and how this also simplifies things, also another smart move.

I've not seen it up close, but the concept makes too much sense and addresses many questions that I suspect their potential customers have been trying to figure out.

 

[Deleted member 197115]

It was on FarceBook
Mounting plate is generic thing and I doubt many could benefit from it especially when most rigs and pedals come with own pedal plate, adapter is more interesting but we need to see if it's any better from SW perspective and profiles that HE or other pedals support natively. In the end any serious sim allows binding mutliple controllers to different pedals, so the benefit is questionable.

Having Frankestein monster of the pedal on the other hand is not something I want to see in my rig, but this is personal thing, some might not mind.

 

[RCHeliguy]

Great answer from someone who is not interested in their product.

Now for the people I know of who are interested in this monster of a brake pedal, providing a simple path will help sell product.

I don't think many existing pedal decks would support the new SimuCube pedal. It is very long and likely needs support the full length of travel along the screw.

Anyone who has read this thread knows I would custom build a solution myself, so this isn't for me. But many others who are looking at a $2,000 brake pedal will want a turnkey solution and this will make it easier for them to make that purchase. So I consider this a smart move by SimuCube that will enable sales.

 

[stigs2cousin]

Mark,

I agree with you here.

This kit will make the jump much easier for the "not so DIY" crowd.

Another idea to make it better would be to improve the True.Drive software to Smart.Control level which both parties should be interested in.

To be honest I use very little of the Smart.Control features except for the curve setting on my clutch pedal. But this is a feature I´d miss on True.drive.

MFG Carsten

 

[Deleted member 197115]

I don't think many existing pedal decks would support the new SimuCube pedal. It is very long and likely needs support the full length of travel along the screw.

Not entirely accurate.
Take a look at the picture and dimensions, will bolt to 80/20 profile with ease, can fit P1X pedal tray for sure, just measured, and you can probably even use just first two mounting holes in some cases.
If you check SimExpo videos it was mounted to standard rigs with the back sticking out.