dsc106

Reflected, you make the best campaigns, and I see you favor older aircraft. The Mirage F1 has really surprised me as a module - much more fun to fly than I had anticipated, and the quality of the module is top notch. Any chance we might see a F1 campaign from you?? I do hope you’ll consider!

Ahh... thanks for the info. Ok, so he's still our point person, but just radio silence for the past 3 months? @The_Fragger it would be great to hear an update as its been a while, especially after with the ambiguous news of Razbams departure, many people would like to know where things are at...

I was suggesting you were confused and thinking of the 2023 video which was tagged as being produced with DLSS and multithreading. I can find no indication the 2022 was created on Vulkan. I will be very happy to be shown otherwise, do you have a source/link/reference? certainly, I hope you are correct as it would bode well for seeing vulkan sooner than later.

Not quite. The 2023 video was done with DLSS and multithreading.

Got it, thanks for all the great information. And good to know about motion reprojection. So essentially, with quad views - even if I am using it already on a G2 - I should expect that moving to a Pimax Crystal will essential be a “free” or “almost free” move regarding performance drop? Since most of the extra pixels aren’t really being rendered at all? (But a much higher performance impact if I was not using my quad views and have to rely on standard fixed foveated rendering, such as in SimRacing or MSFS)?

This is what I was hoping to see - your program replacing other MM apps and having a new custom one built it. That would have been awesome, and made your app a one-stop shop for launching and Mod Management. I don't understand why you abandoned it, but I also don't understand coding that well. What do you mean by millions of mods would have to adhere to your system? I thought it would essentially just copy/paste/cust mods into and out of the right places.

That is great news, I will give Quadviews a go with my G2 soon. If I install this, will I still be able to use the standard FFR in other OpenXR titles which do not support quadviews, such as MSFS 2020 / simracing? Or will it essentially disable the ability to do that elsewhere? Regarding a performance comparison between the G2 & Pimax Crystal... to compare apples to apples, if I am running quadviews on G2 & Crystal, I understand the Crystal will still be a bit more advantageous due to eye-tracking. But in a demanding situation such as an AH-64D in Marianas with high graphics settings where I currently get perhaps 50fps (running 45 repo). Let's say I install quadviews on G2 and can boost that to 55-60fps. Great! But what should I expect on a Crystal in that situation? I am concerned I will have to turn down my graphics settings in order to keep 45fps reprojection. Ideally, if I get the Crystal, I would hope to be able to hold 60fps reprojection on 120hz using Quadviews, high details, and DLSS quality. My rig is a 5950x & RTX 4090. In Q1 2025 I plan to upgrade to a 15900k & RTX 5090 assuming they are out around then. I wonder if I should just wait to upgrade my headset until my hardware can scale up a bit.

Also: i see quad views also works with reverb G2 in fixed foveating rendering mode. Is it worth using? Will it perform better even in fixed mode VS the FFR baked into the openXR toolkit?

I could use more detail… I don’t even get 80-90 on my G2 except in causcaus or high level flight. So I don’t see how that would be possible. open Mariana’s free flight in Huey. Take off, fly just past the carrier on water at about 300-400ft, then turn counter clockwise back around. I will get around 50-52fps there with shadows flat, everything else mostly high, DLSS quality. I fly low level rotor in the most taxing game situations. Even with quarries it seems the crystal should be twice as demanding?

I get the haptic feeeback, I didn’t say it wasn’t there, I said the total feeling - even with all of my hardware - pales in comparison to the other helicopters. There is a lack of aliveness between all forms of feedback… deacceleration and coming to an OGE hover, for example. It might be behaving technically right on FM, but it feels like nothing - feels like playing a video game or flying a toy, vs feeling and seeing the stress on the airframe through a combination of auditory, visual, and haptic cues that fully communicate a sense of “aliveness.” This applies to all maneuvers and scenarios as far as I can tell. A general lack of cockpit vibration, atmospheric turbulence feedback, sound design, etc. You don’t feel the motor chug or work, you can’t “feel” the stress of a hard maneuver, etc like you can in the other modules. I am sorry that my description is not better, but as I said, the tactile feel isn’t immersive like the other modules are, even though the FM is now miles better. I am hoping this element can be improved as well, as it is nearly as important as the FM itself.

Hi all, I am curious about others thoughts here. I think the new FM is a huge improvement but the module still lacks to me. The helicopter just feels dead and lifeless in terms of immersive tactile feedback. playing in VR with a buttkicker and Virpil CM3 dry clutch, collective, etc. I have all of the modules and this is the only one that feels very docile in terms of feedback. i get no sensation of the motor working hard when I put it into challenging positions, no real feedback that lets me know when I’m struggling to maintain lift or about to enter VRS, cockpit shake is not much of a thing. i wish I had better language to describe this but while the FM now feels much better, the module itself does not really feel real in anyway. I can completely lose myself in the Huey, for example, but I am painfully aware I’m playing a video game when it comes to the Gazelle. i am hoping the developer can give us some better feedback and make this thing feel more “alive”. And I am even more hopeful that this lifeless feeling is not present in the Kiowa, because I am extremely excited for that aircraft and will be let done if it feels “blah” like the gazelle currently does and cannot live up to the bar ED has set on the other helicopters.

So was The_Fragger part of razbam? Hence the lack of updates since this switched back to Miltech? It would be nice to hear, after this transition, if things are still on track and going well and what we might expect going forward.

Is there any indication of how many years out this is? Not sure if we should be expecting this anytime remotely soon.

Is it still planned to have mod management support added ala OVGME mod manager?

Hi there! I. Enjoying my early access preview of the map so far. Regarding phase 3, I could use some clarification on those items. Are improvements to the feel of cities planned? On one hand, I appreciate the scope and density, on the other hand, the city areas feel so cookie cutter/procedurally generated as to feel fake at times. The cities in Syria map tend to have a much more hand crafted feel that makes them more believable. i do love the start, here. I think there is something special about the sprawling density of these cities, but the repetitive nature can make them feel less exciting to explore in a helicopter, for sure. full circle… when you mentioned improving city textures and adding more unique objects, how exhaustive will this process be? Are you imaging subtle tweaks, or do you see the current state of the cities as a starting draft before really dialing them in? Im hoping for the latter and that they will evolve into areas that really feel unique and interesting to explore. cheers on the good work so far.

Yes, I have been enjoying 2.9 and DLSS. The last post mentioned that once MT was getting sorted, they could move onto DLSS & Vulkan. Now with DLSS here and MT 7+ months into preview, wondering how long the road to Vulkan might be. I'd love to see that in Q1 2024 but I fear that may be a bit optimistic, unless they have already been working on it in a separate branch for some time now.

For those with the Crystal, how are you doing this in DCS with decent frames? I have a 4090 with G2. Even with 2.9 DLSS, I have to run reprojection 45fps on my G2 (rendering at 3k per eye, then DLSS quality applied), most settings on high except shadows. I do a lot of low level flying in helicopters, around Marianas, or Syria cities, etc. My frame times in those areas can get to 18-20.5ms pretty easily sometimes very close to the max 22ms before I am not longer holding 45fps to interpolate back out to 90fps reprojected. I am also running Fixed Foveated Rendering in the G2 via OpenXR toolkit, not as good as DFR no doubt, but I imagine it closes the gap at least a little in terms of how much net benefit I could expect from DFR? As I understand, the G2 at 3k per eye is around 9 million pixels vs 21 million pixels of the higher resolution Pimax. Even with DFR enabled, I don't see how you could run the game with at a sustainable frame rate without dropping graphics settings heavily, or staying at high altitude or in maps like Causcus? Am I missing something? Does the Crystal not actually incur as big of a performance loss over the G2? If so, why not?

I understand SLI is dead, and splitting frame rendering was iffy business - at least in pancake mode. Many do not realize that while SLI is dead, multiGPU is not. SLI was merely a bandaid solution until PCI buses and low level apis could support this natively with no temporary solution of a hardware bridge. One use case that always worked very well for multiGPU was VR - because rather than splitting rendering of every other frame (like in flat screen rendering) you instead could render in parallel - rendering both eyes simultaneously on separate GPUs, nearly doubling performance. Because VR renders the same image twice, slightly shifted for sterescopic, multiGPU in VR was one of the few truly viable gaming use cases. Currently Vulkan and DX12 has full support for explicit multi gpu... they have had this for maybe 5 years or so now. If developers want it, they can implement it as they wish. Since DCS uses separate render parsers per eye and builds the frames per eye independently it should be even easier to implement. Some suspect that the pro version of DCS (MCS) already supports this (it uses Mutli-edge which enables multi projection systems). For VR, parallel rendering for each eye is the way to go. It could double performance for those with deep enough pockets and accelerate the future. Imagine twin next generation RTX cards being able to play DCS, maxed out graphics, on an ultra high res PiMax headset at 90fps or even 120fps. This could be reality for users willing to splash for 2 top tier GPUs. The tech is here today, and is already mature... can we use it? Perhaps the reason that developers will not/have not added this feature is fear that if they do, Nvidia or AMD may not answer developer help line questions, or fail to optimize drivers to work with their game, etc. - since such a thing could discourage new card sales (why buy a 4090 when you can get 2 used 3080s second hand?). However, if this idle speculation is indeed the case, perhaps ED could leave the support somewhere in the engine to be easily found and enabled by modders unofficially...

Back in April, ED commented that once MT Preview was getting sorted out, they would move focus to things like DLSS and Vulkan. I too would be curious to know what the development timeline for Vulkan looks like today?

I believe we’ve all been patient and at this point the delay on fixing this module is downright offensive. I would have expected a fix within 1-4 weeks maximum, given that a fix could have constituted a reversion to the prior state until fully corrected. There has been no word at all on forthcoming changes and it has essentially ruined my favorite module for several months now. I understand that some tweaks are indeed an actual improvement in accuracy, but the negative net effect has been thoroughly discussed and demonstrated.

What are you talking about? Last stable release was May 5, Huey new engine profile was introduced to open beta on May 18. Am I missing something?

Tim, you are awesome, thanks for your dedication to this. A lot of us LOVE the Huey and are bummed by the current state of things, and no doubt you are providing information quite helpful to getting it all sorted. ED, I understand we are all on the open beta branch and this is part of the beta process! With that in mind, I am wondering if there has been further progress on getting this sorted out for the many passionate Huey pilots here?

Does anyone know if I can plug the MFD button frames into a non-powered USB 3.0 (A) hub? (I am not using screens as I am in VR). I would like to connect the MFDs into an anker slim 4-port USB 3.0 A hub, that daisy chains back into a POWERED USB 3.0 10-port Sabrent hub (so as to only take up 1 port on my main Sabrent hub labelled "MFDs"). Then I could cable manage these frames into the small/slim 4 port hub and have a single USB cable running off of them back to my main hub. Would that work?

Thanks for the update! So I presume we will in short time see a mod manager built into a feature release then?

Does this offer a mod manager? It would be great to see ovgme functionality baked in since that app is no longer in development, and to have everything under one roof!