The overall impression here is that DTGFS is yet another FSX service pack then - 10 years after the fact and counting... 64 bit is good news, but its at least five years too late. 

 

I was sure Flight would be the base for a new sim, but it seems that it was too much work for DTG, and I can understand that. But as a veteran flight simmer that isn't good news, but for someone picking up FS for the first time the FSX/ESP-world is probably good enough. I'm getting pretty tired of the default landclass and autogen. I've been staring at it for so long that I know every field and street in the rural and urban textures, and it's pretty boring when every town in Kansas looks exactly the same as in Ohio, or every town in Sweden look exactly the same as in Hungary. 

 

Flight School might be fun though, and I like missions and challenges, so bring it on! 

Speaking of atmospheric scattering.....

 

 

 

The thing that concerns me more is the weird cyanish color lighting.  I also don't seem to see any shadows, although there is something that looks like weird self shadowing on the trees, and drop shadows around the buildings at the airport.  

 

It looked like an attempt at HDR with scattering ... volumetric fog could also be used for atmospheric scattering with minor adjustments.  P3D V3.x "SpeedTrees" are considerably better than what I saw in the Flight School screenshot ... but adding shadows to an "X" (the tree object) makes the rendered scene look worse (shadows that don't make sense) unless at very specific view angles.  In the DTG Flight School screenshots, the trees weren't blending well with the scattering either ... hopefully DTG will be adjusting as work progresses.

 

Regardless, without 3rd party support DTG wouldn't get far ... that's the ball and chain ... so I can understand why they opt'd for FSX SP2 code base to keep that layer of compatibility even at the cost of performance.  For the time BGL was a good idea, but today there is no benefit to it other than compatibility ... it's a layer not needed and can be accomplished in other more efficient ways.  

 

DX12 is nice but given this market (FS users seem to hang on to OLD hardware and old operating systems) ... the few DX12 titles I've used didn't really meet the hype of DX12.  But regardless Win10 is 11.5% market share, DTG isn't going to code for an even smaller slice of market.

 

Can this market support so many segmented variants of a flight simulator?  Not many people (like myself) that will use just about all flight simulators available for purchase.  I can barely keep up with P3D, XPlane, and DCS (I've tried and set aside many).   DCS 2.0 has the performance edge hands down, but it's a VERY small world to fly/fight and it's focus is combat ... and those that like distance LOD radius will be very disappointed with DCS 2.x.

 

P3D Sim Director with Scaleform and avatar mode can do everything described in DTG Flight School ... instructor sitting next to you and verbalizing instructions (and more).  But my last check with 3rd party, I haven't seen a single 3rd party entity use scaleform (there have been many cases where it would fit perfectly but because it's not compatible with FSX, the feature isn't used).  Only recently have we seen a few scatterings of 3rd party using SpeedTrees (from the default libraries) for P3D. 

 

So we have the issue of 3rd party not using new P3D features because they don't work in FSX - diverging from the FSX path is still a major revenue fear (pending who you want to believe, 3rd party market sales are 40-70% P3D, 30-60% FSX) ... that seems to be the biggest hurdle.

 

So if DTG bring us something "new" (they made it clear Flight School will have no SDK but there other flight product for end of year release will have an SDK) they'll be hitting the same hurdles as P3D.

 

Users have invested so much in FSX, they don't and will not give it up ... not even for progress regardless of who (DTG, LR, LM, etc.) makes that progress.  So the dominant flight sim most likely to succeed is the one that brings "as much" compatibility with it - and hence the performance ball and chain.  And some will not leave their castle because of cost, doesn't matter how good a new product looks, if it requires hardware upgrade then it's a no go.

 

Not suggesting good or bad, just a harsh reality of a this FS market.

 

Cheers, Rob.

It's like Kodak who invented the digital camera, and look at them now... Sticking to ones guns is a good strategy, but only for so long. The tides always turn at some point. I think sticking to FSX after 10 years is risky business. Not just because of the dated graphics (better lighting and shadows doesn't really help when the autogen look like crap), but the performance issues. 30fps is abysmal (and the realistic target in P3D with decent graphics). Works like a charm for those flying in straight lines, but during tight turns and when fine maneuvering choppers, things become rather choppy (pardon the pun). What I see with DTGFS is even more fragmentation of a tiny niche market. Hopefully I'm wrong and DTG blows us all out of the water! 

The lack of atmospheric scattering is why I keep the visibility distance quite low in P3D. This "softens" distant objects and the horizon, and makes the scene look much more realistic. I see screenshots taken in broad daylight with "unlimited" visibility, and they hurt my eyes!!

C++ gauges for FSX aircraft are 32-bit dlls. So unless some runtime conversion is used, there won't be backward compatibility for those aircraft. And runtime conversion would be CPU intensive.

 

Dovetail did not buy the terrain from Microsoft, as Microsoft licensed it from 3rd party sources. They have bought the water, roads, mesh from Navtech or other 3rd party sources... so if it looks the same, that's because Navtech's data is the same. Not much Dovetail can do about that... they do not have the resources to create their own data. That would take years of man-hours.

 

The shame about loosing Flight as a base is that it just was much more versatile, and more quickly rendered.

 

64-bit is not quicker than 32-bit... in fact it runs slower in my estimation. And good memory management would solve most (if not all) problems associated with memory exhaustion. We're getting 64-bit because so many forum posters are wanting it, as if it is a cure-all for rendering problems. I don't think it is... it's a marketing pitch. As far as Windows 10, I believe they are only guaranteeing the new sim to work for that OS, and not supporting any problems that may arise with other operating systems. DirectX11 is a nod to system owners that do not have a DirectX12 card... and perhaps Windows7 users will get a pleasant surprise to find Flight Simulator runs on their systems.

 

All that said, I'm very much looking forward to DTG's new Flight Simulator. I'm glad the series continues. I'm glad 3rd Party developers will get a chance to bring their skills to the game. I think a lot of 3rd party FSX content will not work 'as is' in the new game, so I hope the 64-bit, Directx11 incarnation blows us away, rather than being a retread of FSX. Otherwise there is little incentive for large numbers of users to switch to the new sim, and invest in new content.

 

Meanwhile, as this is the Flight forum, I think it's 'so long' to a great sim. Such a shame it is never going to see it's full realization. Thanks to Steve's efforts, we can still bring it on once in a while and even enjoy a bit of new content.

 

Dick

  Courchevel - That thar's one scary airport!  Have to try it. That beats landing on Catalina by miles ...

 

 

You make it sound like a bad thing, but as I demonstrated in my screenshots, a properly implemented HDR+atmo scattering can improve the visual appearence a lot more than other features.

 

Oh yeah, i'm not saying they don't make a difference.  Just that they are easy to implement, sound cool, have some improvement.  To be fair your screenshots were not a good comparison though.  The top one didn't even have any standard depth fog turned on.  You could easily get something resembling the bottom picture in Flight just by having the right combination of lighting and fog.  Atmospheric scattering really benefits at higher elevations though, and in Flight that was not really the goal due to the focus on GA.

 

I was more referring to the fact that it sounds like they focused on visual bang for the implementation effort buck.  IMO shadows and lighting are way way more important to make an engine realistic in general, but also require a better architecture in the rendering code path.  FSX has a scalability problem on the CPU, and that really limits how far you can push object counts, and terrain mesh and texture resolutions.  Flight did a lot of work to reduce that on the object side, but limited work on the terrain side.  Without those improvements we wouldn't have been able to get the perf we had while also being able to do 4 levels of cascaded shadow maps.

 

 

but adding shadows to an "X" (the tree object) makes the rendered scene look worse (shadows that don't make sense)

 

We did it in Flight, and I think it looked pretty darn good (but I'm biased ).  It does require some care over the shadow settings to limit self shadowing on the trees.  If I remember correctly I think i just made the trees sample a lower cascade in the shadow maps, so the shadows were not as sharp.  I think we used speed tree for some of the tree textures, but I don't think we looked at the run time libraries since at the time at least it didn't scale well to that many trees.  If I remember correctly I think Flight can render upwards of 100k trees on a high end system.  On some of the Hawaiian islands I think that is far enough to cover the entire island...  Something more complex like speed tree is great for higher LODs when you get close though.

 

 

 

Can this market support so many segmented variants of a flight simulator?

 

In my opinion I think this is one of the reasons 3rd parties were hesitant about working with us on Flight.  We tried to get them involved, but my impression was the transition from FS9 to FSX almost put many of them out of business because they started making all their addons for FSX, but then a lot of people stuck with FS9, and they had to react and switch back to FS9 compatible addons.  The Flight tools in many regards were very different from FSX with no compatibility between content, so it was an even scarier proposition.  Now they have had a chance to make their pipelines more flexible for a varying set of target platforms, so that probably helps.

 

 

Anyway, it is good to have another option for flight simmers.  Whatever this first release happens to be, I hope people look at the potential, give it a chance, and not just what it is at release.  If more people had done that with Flight it may still be around.  Our goal was always eventually to get it to feature parity with FSX, but didn't get the chance.  In development sometimes it is good to narrow the focus to make sure features get the attention they need, instead of spreading resources thin, but that means other features have to wait until they get the love.  Flight simmers are a diverse crowd demanding an equally diverse set of features, and sometimes people feel left out.

I hope people look at the potential, give it a chance, and not just what it is at release.  If more people had done that with Flight it may still be around.  Our goal was always eventually to get it to feature parity with FSX, but didn't get the chance.  In development sometimes it is good to narrow the focus to make sure features get the attention they need, instead of spreading resources thin, but that means other features have to wait until they get the love.  Flight simmers are a diverse crowd demanding an equally diverse set of features, and sometimes people feel left out.

 

It went both ways. Microsoft didn't give users enough time to get adjusted to the new sim either. The Alaska pack showed potential, but with no "real" aircraft to fly in, there was nothing in there for the veteran simmer. 

It went both ways. Microsoft didn't give users enough time to get adjusted to the new sim either. The Alaska pack showed potential, but with no "real" aircraft to fly in, there was nothing in there for the veteran simmer. 

 

Well it was kind of a perfect storm. I've heard that there were other planes that were intended to be released but never were because aircraft manufacturers were being perhaps a bit greedy with the licensing fees, figuring that MS was ripe for milking. Instead, they got nothing.

 

Nobody won.

 

Which brings up another point. Will some manufacturers that have looking the other way for years, feel quite so lenient of DTG is making a mint on Steam?

 

Who knows.

In my opinion I think this is one of the reasons 3rd parties were hesitant about working with us on Flight. We tried to get them involved, but my impression was the transition from FS9 to FSX almost put many of them out of business because they started making all their addons for FSX, but then a lot of people stuck with FS9, and they had to react and switch back to FS9 compatible addons. The Flight tools in many regards were very different from FSX with no compatibility between content, so it was an even scarier proposition. Now they have had a chance to make their pipelines more flexible for a varying set of target platforms, so that probably helps.

 

I think Robert Randazzo (PMDG) would have something to say about this, as that was definitely not the reason  he gave, why FLIGHT wouldn't work for them. John Venema over at Orbx as well

Well I am happy that the FS series is continuing through DTG. A few years back the bottom dropped out of a lot of people's world when MS announced the death of FS11, yet here we are with what is effectively FS11 arriving in a few months. Surely that's cause for celebration?

 

There's no reason to suspect that DTG will stifle the add-on market, as they know well how the FS world works. I think the market will work pretty much how it does with FSX:SE, with DTG pushing some stuff through Steam for casual users, and the hardcore folks will continue to buy as they do today through the usual online distributers and developers.

 

DTG are doing the same for flight simmers as what they did for MS train simmers; they effectively produced a Train Simulator 2 from the ashes of Kuju and called it Rail Simulator, and still going today with TS 2016. They certainly haven't stifled that market as 3PDs develop routes and rolling stock and sell them how they like (freeware is still there too).

You expect a freebie for your stuff, written for Microsoft's 33bit DX9 software to now work on a 64bit DX12 engine from a totally different developer?

It will be DX11, not DX12 Robin...

Edited February 15, 201610 yr by n4gix
corrected Robin's name!

Well isn't this an interesting little topic... stonelance, please keep talking  :smile:

You make it sound like a bad thing, but as I demonstrated in my screenshots, a properly implemented HDR+atmo scattering can improve the visual appearence a lot more than other features.

Currently all the major flight simulators (X-Plane, DCS, IL2) implement it except for FSX/P3D/Flight, and that's one of the reasons they look more realistic especially at altitude.

I think if DTG will add those features, it will be a lot more than a cool item on the feature list. :smile:

 

The difference scattering made in DCS is a big part of the reason I spend so much time in it now and just couldn't get into it before. X-Plane also became much more immersive after the atmosphere tweaks but now continuing lack of decent AI / ATC and the difficulty hooking it up to VR mean I'll probably do a Jose on it soon.

 

Even slightly updated FSX environment with better performance and atmospheric scattering would be a big draw for me.

 

 

What I see with DTGFS is even more fragmentation of a tiny niche market. 

 

I can see what you mean; for the first time in my 20 years of simming life I'm splitting my time between 3 sims regularly (CloD, BoS, DCS) and occasionally another (X-Plane) whereas previously it was only MSFS series.

 

OTOH I can see the niche market getting a big boost from VR. I can imagine most people who get a headset will probably want to try it out on a flightsim, even just casually. DTG Flight School sounds like a good intro for those people, and making a fuller flight sim for the percentage who want to take it further after learning the basics sounds like very good business sense.