twistking

Single pass stereo would be both realistic and huge! I think currently DCS doesn't even do proper multi-pass and instead renders two cameras completely seperated (everything is calculated twice). At least that's the only way that i could explain the massive CPU overhead from VR. Multi-pass would share some calculated data between each eye's pass, single-pass would render the whole scene in one single pass, theoretically reducing CPU overhead by a lot.

No, there is now the LOD slider in the options, which would be the better option for everyone chasing that sweet extra performance by reducing unit LOD That said, if the newly released AI assets are a sample of the low quality version, then they might be good enough for most VR players. You'd probably need a 4k monitor (or 4k headset) to really appreciate the high quality versions in actual gameplay. By the way, is this actually confirmed? Are the newly released assets the "free" low LOD tier?

To be honest, currently i can only think of one game that uses voice recognition natively, which is only a sim in the broader sense. It's called "tower! simulator". It does exactly what its name suggest and i think it's a bit boring, but voice rec works fine! Mostly i use thrid-party voice recognition. For sim racing there is f.e. "crew chief", which interaces with the simulator to add a spotter and an interactive, fully voiced, race enginner. It has also built in voice recognition to interact with your engineer (or to simply bind keyboard macros). Works like a charm and is free. I think the voice rec component uses MS Windows functionality under the hood, but i'm not sure.

tried again today and it crashed -again- after roughly an hour in the mission. the mission reliably and reproduceably crashes both client hosts and dedicated hosts. it is very lightly scripted, but it uses the new b52 AI models and utilizes the new datalink functionality to add AI vipers as members and donors. no mods. please have a look at the provided logs. i can provide even more crash logs of the mission, but i'd assume the datadump above should provide enough info for now.

I hope they consider voice as an (optional) interface option and integrate it into the VoIP/radio simulation not only for MP but also SP. I use voice recognition in other sims and it works flawlessly, despite my german accent.

Now it's official that DCS will get raytracing in 2024, this issue gets more and more ridiculous. Proper visibility for nav lights and beacons is way more important than sophisticatedly computed penumbrae while basic sprite lights don't even come with a performance hit: Flight sims in the 2000s could already render nav lights and beacons to realistic distances. It's just sprites! Raytracing won't help with that, it just need some good old tuning and tweaking. Tactical night flying is borderline impossible without modern jets' HMCS and buddy tracks. Aircaft beacon lights and landing lights should be visible at ~50km, nav lights (green/red) should be visible (and distinguishable) several kilometers at least. Brightness settings should be respected, so that dimmed lights are less visible (obviously). Come on, ED! This also gets more and more important with the PTO coming to WWII (navigation over vast distances) and the (mid) cold war era getting more fleshed out (no datalinked buddy tracks).

The intro talks about raytracing, but then this is not followed up on. Has this been simply forgotten? I would love to know what RT features the team is considering. Also it would be good to have some clarification on it being vendor agnostic. Apart from that this was a nice, chunky newsletter. Something like that every few month please!

Yes. I don't understand why ED cannot get the thing over the finish line. It's a simple module and there isn't even that much left do do. It will have to be finished at some point anyway, but until then it will only continue to produce discontent in the community.

Realistically both Intel and AMD will have their next generations of GPUs launched before raytracing arrives in DCS. I'd be surprised if those GPUs were not viable for at least running some RT effects. So let's wait and see...

Oh... i see. Sorry! I do udnerstand your concern then, however i'm again fairly certain that maybe Wags himself has been bamboozled by Nvidia marketing to some extend. To be fair: In practice raytracing is not very viable on everything but Nvidia currently, AMD's next generation is not announced yet, so i assume he was just thinking about the current market situation and said RTX without thinking too much about the implications. It would be stupid for ED to artificially limit raytracing to a vendor. Clarification would be nice though.

Raytracing is not a proprietary technology!!! If you think that you'll need an Nvidia card for raytracing, you've been bamboozled by Nvidia marketing. It's safe to say that AMD is working on closing the gap in regards to raytracing performance and we will surely see AMD's next generation of GPUs hit the stage with improved raytracing performance.

Regarding the German Mig-29: Does anyone know what type of ordnance or pod is seen on this image?

I'd say that ED modules tend to show little buffeting. I think Heatblur has good effects on their F-14. A bigger issue in DCS is that turbulence does not really exist: There is a turbulence setting in ME that technically works, but that is more random gusts and does not give the bumpiness and roughness that you can experience in reality.

Interesting feature for the upcoming DTC/mission planner...

I'm generally optimistic about RT, but i doubt that it will give better performance, honestly. I'm really interested in how the clouds will work with RT. I think they currently already use a type of software ray marching trickery, which might not even be compatible with RT. Hardware raytracing can -generally- not do proper volumetrics afaik. It would be super cool though, to have realistic super soft shadows under overcast conditions which would sharpen up as you climb through the clouds towards the sun.

Shadows generally have a big CPU impact indeed, but raytracing comes with CPU overhead too. Maybe it will still be a net positive on CPU limited systems. Obviously RT shadows will look nicer, than everything that we have now. I'd assume that RT shadows could also be viable in VR, since it should be safe to calculate them only once for both eyes. I did not think of RT reflections for mirrors. Interesting point. It will surely be more expensive than the simple render-to-texture, but would allow "realistic" (perspective correct) mirror reflections (maybe even depth in VR).

in theory DLSS/DLAA could be negatively affected by openXR tubo mode, since i assume that reliable and precise frame time prediction are needed to minimize blurring from inaccurate motion vectors. might be worth to test for those seeing a lot of blur.

Hello, in a recent podcast Matt Wagner confirmed that raytracing would come to DCS in the wake of the vulkan API retooling. On the one hand it was only a question of time until raytracing would come to DCS, as it's now safe to say, that this is where the industry as a whole is going in regards to rendering tech. On the other hand i was a bit surprised about the "announcment" because i feel that the biggest issues of DCS (on the rendering side) will not be alleviated by raytracing. Or put more precisely: It's not obvious how raytracing would help with: Performance, night lighting - especially visibility of sprite based lights (carrier ball, aircraft nav lights etc.), volumetrics (fog)... Wags mentioned raytraced shadows specifically, which seemed odd to me, since screen space shadows were just recently introduced, which work quite well for a flight simulator (would probably struggle more in a typical action game). So what can we expect from raytracing? I'm mostly excited about raytracing in combination with the new clouds, but i'm unsure if those are even comaptible with raytracing. Afaik volumetrics like fog cannot be effectively raytraced currently. Normally i'm a huge fan of realtime GI, but i don't think that it would add a lot to a flightsimulator. Maybe light pollution over cities could be done with RT GI, but i doubt that RT GI would be able to handle atmospherics. I also wonder, if RT could actually help performance on modern cards - for example by freeing up frametime from the multiple legacy shadow calculation. Is it reasonable to expect raytracing in VR? I think that RT shadows and RT GI could be done in a single pass for both eyes. Reflections not so much. Are there games that successfully do RT in VR? What do think could and should be done with RT and what do you expect actually will be done with RT in DCS? Please don't start a flame war against RT. It was clear that, it would eventually come to DCS. If you for some weird reason dislike that, please make a seperate thread. Also remember: Raytracing is NOT proprietarily Nvidia.

I agree. I think it's just useful knowledge if you pick up bits and parts of other languages even if you'll never be able to properly understand or speak it. The vocabulary you need to "read" an analogue cockpit is so minimal, that you can easily learn it while learning the proper operations. Digital cockpits are surely something different though. I would appreciate if the manual could help a bit with understanding cyrillic russian. For example in the cockpit overview part, it could name the buttons not only in cyrillic, but also include romanizations of the word, translation (obviously) and pronounciation with international phonetic alphabet.

I see. Would there be any other benefits of using HF in a rather short range fighter? Did western aircraft keep using HF after VHF/UHF became widely used?

I can't follow. Can you elaborate on those "resons", please? Do you think it's possible to bounce datalink radio against the ionosphere with 80s radio tech?

I was referring to the mystery picture that's said to have been posted earlier but been deleted since. I've not seen it, but @Tarres did identify it as 9.13, which would make it a C. Since then @BIGNEWY has -again- confirmed and assured that we'll only see an A model though...

Wouldn't the enlarged spine make it a Fulcrum-C? ED could just have teased a "Fulcrum", but they made it very clear to expect an Fulcrum-A. I'm confused.