January 26, 20206 yr 7 minutes ago, divide said: Haha, no. Here's FlyTampa's St Barts: Don't know how you created the blurry FlyTampa screenshot, but that sure is not what it looks like on my system.. Bert
January 26, 20206 yr I got it from a youtube video, so it's indeed a little blurry because of that, but all youtube videos I saw of FlyTampa St Barth looks about the same in term of texture/terrain/tree details: https://youtu.be/Nrf8gxFrAFY?t=28 https://youtu.be/GpmGT5qr0TE?t=14 https://youtu.be/gZS316Gk-r0?t=5 Edited January 26, 20206 yr by divide
January 26, 20206 yr Trust me, it looks a lot better than that on a good system, but I'll grant you that the FSX/P3D default scenery for St. Barths is terrible and the MSFS scenery impressive. The last of your Youtube videos actually gives a pretty good representation of what it should look like.. and a good system will make it look a lot crisper. Edited January 26, 20206 yr by Bert Pieke Bert
January 26, 20206 yr Lets wait to see what the Airports Video brings.... I bet a lot of our questions will be answered. Still fascinated to see if they can model a Waterfall...... Cant Wait. H.
January 27, 20206 yr I’ll take the water coloring on FlyTampa’s version over MSFS one though I will say..
January 27, 20206 yr High alpine areas with ortho imagery already look very good in other sims but what intrigues me about this shot is the combination of great scenery with realistic weather effects. Overall I think this looks absolutely stunning. Here’s some details I observed: 1. Spot the difference! The shapes of some of the crevasses and snowfields clearly don’t match. The big question is why. Is this an indication of MSFS using different (newer?) imagery or is it the result of Azure correcting/retouching the imagery? What is strange is that so far Bing’s imagery has always been exactly represented in the sim and also the rest of the image appears to match (e.g. clouds mentioned in 2.) Whatever the reason, there is a clear discrepancy between Bing Map imagery and the sim. 2. This ridge isn’t covered with snow in Bing Maps. Again, the question arises of whether this is Azure’s doing or because of Asobo using other imagery. The two mountains nearby are partly covered by clouds (which can also be seen in the sim) so one possibility could be that Azure mistook the clouds for show. The longer I look at it the more I believe that these patches of snow are artificial rather than part of the satellite imagery. But I can’t say for sure as the image is too blurry. 3. In Bing Maps this area is completely covered with show (winter shot) whereas in the sim this is not the case. Again, due to the great distance conclusions here should be treated with caution. Hats off if this is the result of Azure’s colour correction. 4. I know I probably shouldn’t jump to a conclusion regarding this last detail as it is too far in the background for proper scrutiny but I’ll do it anyway. Low-res mesh has been a concern in this forum for some time and in this screenshot there is yet another indication thereof: there is no Matterhorn. As in Bing Maps, this mighty mountain is represented as a little pyramid at least 500 metres shorter than it’s supposed to be. I know it’s still in alpha and all, but this is really starting to bother me. I hope I’m wrong and that is has something to do with draw distance or the like. Other interpretations are much welcomed. That’s it folks. Maybe someone with better knowledge of scenery design could clear things up a little. i9-11900K, RTX 4090, 32 GB ram, Honeycomb Alpha and Bravo, TCA Airbus sidestick and quadrant, Reverb G2
January 27, 20206 yr @Shack95 Great effort! I think a lot of it has to do how MSFS2020s scenery works. They have mentioned it quite some times but it also took some time for me to fully digest the extent of this. They are most probably going to apply autogen (grass, trees, buidlings and maybe snow) on the scenery in accordance with what information they extract from image data. So you avoid ugly inconsistencies between tiles and areas with bad resolution don't really matter. In fact, I guess it will look much more realistic in the end. You observations might very well have something to do with that. Happy with MSFS 🙂 home simming evolved
January 27, 20206 yr On 1/24/2020 at 2:07 PM, tweekz said: The biggest problem with MSFS2020 will be peoples expectations. 😄 I think peoples biggest problem is their expectation of other peoples expectations of which they know zero. Simmerhead - Making the virtual skies unsafe since 1987!
January 27, 20206 yr On 1/24/2020 at 9:41 PM, 767lover said: 'slightly wrong hue of the BEAUTIFUL water in the new sim' Not slightly wrong, this color of the water as pointed out does not belong in this region, not at all, even when cloudy. What I have seen in images and clips so far seemed to be correctly colored water for the region so I am sure they will correct this! I don't think that is the fault of the sim, but rather the graphics card. I currently turn off all water enhancements in FSX and P3D as I find them highly unrealistic, especially when the water isn't calm. Simmerhead - Making the virtual skies unsafe since 1987!
January 27, 20206 yr 14 hours ago, divide said: Haha, no. Here's FlyTampa's St Barts: Here's MSFS St Barts: And here's the real St Barts: Which one looks closer to the real deal ? 🙂 The better closer look to the real is my scenery 😉 http://marcoh.gratisim.fr/images/SB-capture_3_after.jpg The scenery is free. The runway is already sloped. But the MSFS scenery is not bad. There will be a little job to change the vegetation, add fews things like local houses, boats, cars, etc.
January 27, 20206 yr No matter how good the 3rd party developers are, their sceneries for FSX/P3D will look cartoonish when compared to the MSFS2020 sceneries. The airport terminals will look more accurate, sure, but in general, I wouldn't say they look better. I would no doubt pick the MSFS2020 default St. Barts scenery over Flytampa's for FSX/P3D. But sure, if Flytampa decides to do this scenery for the MSFS2020, it would look bettter than default, I guess, but still would make me think it would worth to pay around 30 dollars to have this airport terminals just a little bit more accurate and have some landmarks/points of interest around it. About the Water, I think we better wait to have more details about it. It could be the clouds, illumination effects or just orthoimagery being replaced by default textures. If FSX engine had the ability to render the water colors based on some data, I bet they will implement something like that on the sim (if it's not already implemented). We've already seen in the areas the imagery is good, the water colors will be pretty good. 9800X3D@H150i // Msi RTX 5090 Trio OC // 64GB DDR5 6000mhz CL30 // 2TB + 1TB Nvme Dell 27" 2127DGF - 1440p - Gsync - 165hz Thrustmaster TCA Sidestick Airbus // TCA Quadrant Airbus // TFRP T.Flight Rudder Pedals // Logitech Flight Multi Panel
January 27, 20206 yr 14 hours ago, Shack95 said: High alpine areas with ortho imagery already look very good in other sims but what intrigues me about this shot is the combination of great scenery with realistic weather effects. Overall I think this looks absolutely stunning. Here’s some details I observed: You have a sharp eye 🙂 Well I think with this upcoming hyper realistic world engine that is no less than a high fidelity copy of reality, many of us (hopfully not me!) will develop an obsessive habit to compare every detail with reality 🙂 No seriously, I think (and I could be wrong) we think that Asobo took the data and processed it for their world engine as we do with these 3rd party tools that let us create Photorealistic / Orthophoto scenery coupled with some elevation data etc. I believe and as far as I can see in the screens and videos, their engine does way more than that. The data seems processed, corrected, refined and most probably layers of overlays and decals are applied. Not mentioning the variating level of details that produce more refinement on ground level. We can clearly see this in the comparision shots between Bing and some areas they previewed. It's clear that the Bing data is not implemented as is... Also we don't expect the data to produce a 100% match from one corner to another on the planet. The algorithms that correct and refine can produce some unexpected results here and there... After all, it's a data processed with algorithms and QA could take forever to detect the little weirdnesses on the planet. As for snow vs Bing, well hopefully there won't be static snow coverage but rather a dynamic one as they showed us in the "Let it snow video". So I suppose it's a much better option to make the world "dry" and render any snow coverage dynamically using factors such as weather, altitude, location etc. ________________________________LEBOR SIMULATIONSScenery for Flight Simulators since 1998
January 27, 20206 yr Thanks for your insightful comment. 38 minutes ago, Claviateur said: Well I think with this upcoming hyper realistic world engine that is no less than a high fidelity copy of reality, many of us (hopfully not me!) will develop an obsessive habit to compare every detail with reality 🙂 😅 Well, I‘m not comparing details out of pedantry but because I‘m hoping it might yield some clues as to what the ai is doing here and so get a better idea of what I can expect from the final product. 39 minutes ago, Claviateur said: It's clear that the Bing data is not implemented as is... I‘m aware of that. That‘s why I think comparisons are interesting. 48 minutes ago, Claviateur said: As for snow vs Bing, well hopefully there won't be static snow coverage but rather a dynamic one as they showed us in the "Let it snow video". So I suppose it's a much better option to make the world "dry" and render any snow coverage dynamically using factors such as weather, altitude, location etc. This would be great indeed. As long as it looks realistic, especially transitions. i9-11900K, RTX 4090, 32 GB ram, Honeycomb Alpha and Bravo, TCA Airbus sidestick and quadrant, Reverb G2
January 27, 20206 yr The fact that we are going over a new, pre-released DEFAULT sim with a magnifying glass to find some little flaws compared to the real world, has me even more excited.
January 27, 20206 yr Here's a thought. When the satellite takes photos of the planet, it probably takes multiple exposures of the same area; it's just a digital exposure after all and the satellite probably has multiple cameras. Bing maps appears to use one image. Who knows what criteria they might use for selecting the one they use? But what's to say Azure isn't combining multiple images into a better quality average or, if that's not the case, perhaps it simply uses a different image than Bing maps of the same spot, hence the differences noticed in comparisons. In oe interview (I think it was the September preview) the MSFS lead stated that Azure takes about a week to identify all trees on the planet. It's always possible that just before release, and periodically afterwards, they run a refresh with the latest images, techniques and AI learning. It's therefore not unreasonable to expect the quality of the graphics to ratchet up year on year.
Archived
This topic is now archived and is closed to further replies.