Bob Scott

That blueish tint is the result of attempting to replicate the Rayleigh scattering effect in the P3D's shaders, and it surely is way overdone much of the time. I've concluded that somebody on LM's P3D team must be the descendant of a clan of Smurfs. I tried and tried with Envshade and even experimented with the shader code myself, and never have been able to mitigate it to my satisfaction. I'm a bit sceptical that RDShade will do any better, but I'd be a happy camper to find out that it does. So I'll be interested to see the results of your trial.

You're gonna need a bigger plane... 😉

I'm closing this thread...discussion to continue now in the release thread:

FWIW, that first "unknown value" in the M4.csv entries looks to be the max elevation of terrain in the vicinity of the airfield...looks like it takes into account terrain within 10nm or so.

The PACOTS routes are fixed routes with semi-permanently defined waypoints. The NATS are not--they are dynamically-defined routes/waypoints based on forecast upper-level winds in the North Atlantic, and they change from day to day (ARINC sends out a daily message that defines the NATS routes for the next day). So where a monthly nav database might be expected to include the PACOTS routes and waypoints, it would/could not include the NATS routes/waypoints due to their constantly-changing nature.

🤣🤣🤣

Ummm...No. "Supersonic" means travelling at an airspeed faster than the speed of sound at that air density and temperature. A high groundspeed due to a 250 knot jetstream tailwind does not mean the jet is travelling through the air mass at faster than Mach 1.

The PMDG birds have never been particularly kind to users when trying to load them after loading other aircraft in the same simulator session...this problem might have nothing to do with your settings, but rather that something loaded by the other plane in your intervening step is left behind and crashing the PMDG panel when you reload it. Instead, I'd recommend you make the changes and then shut down the sim and restart.

I am unaware of a mechanism to do a mass un-flagging of posts already marked as read.

A revised post has replaced the thread starter per OP request, and has also been reposted into the Tips and Tricks subforum.

Reposted by admin--original post by @Cpt_Piett and discussion at: Hello fellow MSFS'ers! Here's a guide on improving image quality for those using DLSS. Goal: to get the best image quality without a significant impact on performance. Useful for: those with Nvidia RTX GPUs. A section of the guide pertain to DLSS3 for those with RTX 40 Series GPUs. 1. What is DLDSR? Deep learning dynamic super resolution utilizes the Tensor cores in RTX GPUs, in order to upscale the image to a higher than native monitor resolution. It then downsamples the image to fit the monitor. Whilst doing so it enhances the detail, edges, and shimmering of the image. With the advanced AI-tech included in DLDSR, Nvidia promises that the performance impact is only half compared to the much older DSR technology. There is a caveat though: For G-Sync to work, MSFS resolution need to be set at the same as Windows resolution. 2. What does DLSS have to do with this? Used in combination with DLDSR, DLSS can mitigate performance loss whilst still enabling a very high image quality. Especially when using DLSS super resolution in ultra quality mode (see below). Opinions of which is the better alternative - DLSS vs TAA - differ wildly, both on this forum as well as out there on the world wide web. I find that there's very little perceivable difference. According to an oct '23 article from PC Guide, DLSS could provide better results in some scenarios: Besides, DLSS is a relatively new technology which is continuously updated by Nvidia. 3. How to enable DLDSR: Enable the first two options under DSR - Factors in global settings in Nvidia control panel: Restart the PC, start MSFS. It's now possible to use two higher resolutions than the native monitor's resolution. If you have a 4K 3840x2160 monitor, the two new available resolutions are 5461x2880 and 6144x3240. The difference in total pixels is huge: 3840x2160 = 8,3 megapixels 5461x2880 = 15,7 megapixels 6144x3240 = 19,9 megapixels I've found that the highest resolution (2.25x native) will increase the GPU load significantly, increasing temperature and fan noise. Thus, I use 5461x2880. 4. How to enable DLSS ultra quality in MSFS? In MSFS graphics settings under DLSS, it's possible to choose from several alternatives, ranging from ultra performance to quality. The latter option downscales the image to 0.67x the rendered resolution. However, when using the latest DLSS version, another alternative can be chosen: ultra quality, which only downscales to 0.80x. This process of implementing this is a two-step process that involves 1) updating the DLSS version and 2) changing some settings in Nvidia profile inspector. Besides using DLSS ultra quality, there are other good reasons for updating to the latest DLSS version. Even though MSFS was featured by Nvidia when introducing DLSS3 in September 2022, the DLSS version in MSFS is outdated. Current version in MSFS is 2.4.12.0, if I'm not mistaken. The most recent Nvidia version is however 3.5.10.0. Why download the most recent version? Because newer versions of DLSS usually bring improvements in image quality, performance, or both. 5. How to update DLSS version? The easiest way is to use the app DLSS swapper that can be found on github. This will only give benefits when not using frame generation though. For those that want FG in combination with the latest improvements in DLSS, the method is to manually replace the current MSFS DLSS version with the most recent one, which can be found here. Step by step process: 5.1) First make a backup copy of C:\XboxGames\Microsoft Flight Simulator\Content\nvngx_dlssg.dll. I just renamed the file to nvngx_dlssg.dll.old. 5.2) Then extract the file from the downloaded .zip into C:\XboxGames\Microsoft Flight Simulator\Content\ and choose replace. 5.3) To double-check if you have the latest version, download GPU_Z, navigate in the top menu to advanced - NVIDIA DLSS, then start scanning. When finished, it will show the active DLSS version for MSFS (and other games/apps). Note: This procedure has to be repeated when updating MSFS (sim updates etc.). 6. Using Nvidia Profile Inspector to enable DLSS ultra quality in MSFS This step involves changing some settings in profile inspector. For those who haven't used inspector before, it might seem a little overwhelming at first. I suggest watching a tutorial first to get familiar with the interface. Step by step process: 6.1) download version 2.4.0.4 of inspector on GitHub 6.2) go to issue #156 - DLSS related setting added - and download CustomSettingNames-DLSS.zip, or just use this link 6.3) extract into inspector folder (the folder where nvidiaProfileInspector.exe resides) 6.4) set the following settings as follows in inspector global profile under section 5 - Common: 7. How to use force resizable BAR on in MSFS? While in inspector, I'd also recommend forcing rBAR on. This does not improve image quality, but can improve performance: Resizable BAR is available for those with RTX30 GPUs (or later) combined with either an AMD Ryzen 5000 (or later) CPU or Intel 10th-generation (or later) CPU and corresponding motherboard. How to force rBAR on? Set these as following (also section 5 - Common): 8. Some added tweaks 8.1 While I'm at it, I'd also recommend setting anisotropic filtering to 16x in Nvidia control panel (MSFS-specific section), then setting it to off in the sim. This might lead to a better result than using the default sim option. 8.2 Tweaking UserCfg.opt settings. The file can be found here: C:\Users\<username>\AppData\Local\Packages\Microsoft.FlightSimulator_8wekyb3d8bbwe\LocalCache. Open the file in notepad or Notepad++ then navigate to the PostProcess section. For those struggling with pixelated/grainy images (especially seen in clouds), I'd suggest changing FilmGrain from 1 to 0. To minimize the risk of seeing a halo effect around objects (trees, buildings), change Fringe from 1 to 0. Finally, for much better shadow quality, go to the section called Shadows and change from default 2048 to either 4096 or 6144. Conclusion: I find that I get a very high image quality using these techniques in combination. Cockpit instruments are crisp and easy to read and I don't have any ghosting on speedtapes etc. Hopefully someone finds this guide useful. It's lengthy and quite wordy as it's basically a summery of various youtube videos from IslandSimPilot and OverKillSimulations put into one place on this forum.

You might want to go look at PMDG's 24 Feb 2024 update (there's a repost of it here in this forum), as it seems some PMDG 737 users have been having this same problem (plane stuck in the sky), which they've narrowed down to an issue with TCAS getting overrun with data. Might be related, or might not, but I had the same issue with the Fenix A320 some weeks ago as well.

And I'm saying that managing the VSpeed has the effect of managing the attitude. The two are not independent of one another. I know that if I select a VS of 3000 fpm or less, it will result in an acceptably moderate deck angle for a climb. And I know that if I were to set an acceptably moderate deck angle (e.g. hand-flying in a FBW acft like a Scarebus), it would result in a VSpeed of less than 3000 fpm. Either way, if I am climbing at 5000 fpm, it's gonna be accompanied by an excessive deck angle, and it will be out of the comfort zone. Both high attitude and high VSpeed are valid and complementary indications, and directly controlling either attitude or VSpeed indirectly controls the other.

If you post the edited version here, I'll do an admin edit to replace the original post for you.

Well, I fundamentally disagree...this is a semantic argument that loses sight of the forest for the trees, and disregards the practical reality of how a pilot actually manages flight parameters. There's a clear physical relationship between vertical speed and deck angle when flying normal climb or descent profile airspeeds. I don't believe there exists a fixed-wing passenger aircraft that can climb at 5000 fpm without the pointy-end aimed significantly skyward--in other words, with a high and uncomfortable deck angle. When the pilot lowers the nose (IOW reduces the deck angle) the result is a lower vertical speed. Conversely, when you climb in selected vertical speed mode and reduce the target VSpeed, the result is the nose is lowered--a lower, more comfortable deck angle. Do you propose some other means of controlling deck angle? So yes, I advocate avoiding high vertical speeds during climbs and descents where possible in the interest of pax comfort, because there absolutely *is* a connection between vertical speed and deck angle, and by extension pax comfort.

The 6 degree-of-freedom motion systems in high-end simulators take advantage of the human somatosensory system to, in essence, "trick" the body into an artificial sense of fore/aft acceleration by tilting the motion platform to change the direction of the 1g gravitational vector. People notice and in fact do sense what their bodies interpret as vertical speed when an aircraft climbs or descends, not because the acceleration changes, but because the direction of that 1g gravitational vector is not straight down through their butts when seated. Hence the age-old flying term "seat of the pants"... Many people don't find large deck angles comfortable. Even if it isn't uncomfortable, to most it still does not feel natural. Since (usually) a high rate of climb or descent in a fixed-wing aircraft is accompanied by a significant deck angle (pitch), a pilot trying to maximize comfort will try to avoid high deck angles (up or down) when not necessitated by safety or operational considerations--and that means limiting vertical speed in a climb/descent when possible, as well as managing configuration/speed to accomplish the same outcome.

Ausgezeichnet!

"Safety-Comfort-Reliability"...SAMFOX, baby!! 😉

In my r/w Gulfstream flying, when departing with a light jet, after clearing any obstacles and at/above altitude restrictions I'd fly the climb in VS mode at 3000 fpm until the power required to hold that rate was close to the FMS computed climb power, and then switch to VNAV for the rest of the climb. It's not just about cabin pressurization--at light weights, climb rates at full climb power could be 5000 fpm or more, which results in an uncomfortably high deck angle. The Gulfstream V could carry enough fuel for a 12-hour leg, but most of my flights were far shorter, so really lightweight departures were common. I suspect it's less often an issue with airliners, as they tend to fly pretty full of pax these days, but something like a short leg in a 777 with a super light fuel load for a 1-2 hour leg (Air France used to fly Santiago Chile to Buenos Aires enroute to Paris as one real-world example) would pose the same considerations.

It was a 6.5 hour flight reduced to ~1.5 hrs in real time using Quantum Leap. I wouldn't consider a single CTD a trend--I also had trouble with a couple CTDs this morning and when I finally got it started up there was a new update waiting, so I'm wondering if maybe we got caught in the database changeover on the servers (?)

With this 1.07 update, we now have the ability to change the fuel load inflight, which makes abbreviated flights using a skip-ahead repositioning tool like Quantum Leap or the DevMode teleport tool possible. I just did a 6.5 hr flight from Amsterdam to Doha...took off heavy, flew the SID up to cruise, then used Quantum Leap to reposition the jet to a forward waypoint ~200nm short of Doha, reset the fuel load and flew the descent, arrival and landing. FedEx from Memphis to Anchorage this afternoon... It'd be nice if they'd fix the fuel burn and autopilot stability issues in accelerated flight, but with repositioning now in play, this bird is a good fit for long haul legs without having to endure hours and hours of droning at cruise.

Could be a wonky set of PID parameters for the autopilot, or an inherently unstable flight model, or a combination of the two. It can also be caused by a CG that's too far aft (too close to the CL). You might try loading the plane with a more forward CG and see if that stabilizes things.

There's the free Milviz Piaggio P-180, which is a fast turboprop with performance at-par with a VLJ and the same well-done (and complex) Collins Proline 21 avionics suite as in their B350i King Air. Flight1 has a Cessna 510 Mustang port-over that has a patch to make it compatible with P3Dv5. My personal go-to in P3Dv5 is the PMDG BBJ...its wingspan is only 14 ft wider than a Gulfstream 700/800, and with the extended range tanks it has 6500nm legs. When flying it on short legs with a typical light bizjet payload it's quite the rocket.