Wobbie, on 14 Feb 2016 - 12:08 PM, said:

 

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 Robert...

 

lol..., it's Robin..

Me bad.. spell check on a mobile is quite iffy, I meant 32bit & DX11! (opps I actually got both wrong)

Our goal was always eventually to get it to feature parity with FSX, but didn't get the chance.

 

Yes, it's hard to gather interest when releasing any "new" product that has less features than older existing products ... graphics and performance wasn't sufficiently different enough to encourage use.  

 

Retaining a good level of backwards compatibility is key to success ... and that's the challenge.  P3D's success has been primarily due to LM's ability to retain most (not all) of that compatibility (even at the cost of performance).

 

Even without installers, there is A LOT of older content that still works in P3D.  

 

I thought some of MS Flight developers ended up at LM?

 

Cheers, Rob.

 

 

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

 

I think PMDG makes sense since they focus on airliners, and obviously Flight wasn't there yet.  Unfortunately the way we told developers that we weren't going to have 3rd party dev support at launch without really a reason did not make them very happy, and some were fairly negative.  At least one also broke their NDA and posted information in forums.

 

The challenge for Flight and DFS will have the same, is that the store (whether steam or Windows Store) takes a cut of sales, and most likely DTG will want a cut on top of that.  For GFWL everything had to go through MS, and meant also having proper licensing.  I'm not sure if steam is the same or not, but the question is how much does that leave 3rd parties in the end.  Some people are spectating 3rd parties will make their addons cheaper with an entertainment product, but i'm not really sure if that is going to be true...

 

For Flight we decided that the 3rd parties need better tested tools, documentation and support, but we didn't have the man power to do that at launch so ended the program.  Most of us internally were disappointed that it ended, and I was not really happy with how they messaged it either.

 

 

 

Retaining a good level of backwards compatibility is key to success

The problem is that it also limits innovation and forward progress.  Having to support and maintain the old code paths becomes an ever increasing amount of work.

 

 

 

I thought some of MS Flight developers ended up at LM?

Only one that I know of and he left before we shipped.

the first part of this thread talked a lot of which engine Dovetail was basing there new simulator off of.  They have announced it will be based off FSX.  The question I have is isn't Flight based off of FSX.  If it is cant they take the best of the features from Flight and put in to there new simulator that they will release later this year.  My impression was Flight was always based off of ESP/FSX.  Can anyone here answer that? stonelance?

 

Thanks

Flight was indeed based on FSX as confirmed by Stonelance and ACES at the time they were actively working on Flight. Much of the FSX features could have been implemented into Flight over time as many of the FSX-features were brought along, but not activated.

 

 

if it is cant they take the best of the features from Flight and put in to there new simulator that they will release later this year.  My impression was Flight was always based off of ESP/FSX.  Can anyone here answer that? stonelance?

 

Yes it was, but we completely change the underlying infrastructure, such as the file system, way we manage addons, and references between various files.  Several file formats also changed to take advantage of those features, and we switched more to using middleware than making everything in house, so the model, animation, sounds, collision detection were all different.  We also rewrote large portions of the rendering pipeline to be more flexible, support instancing and do less work on the CPU, also layered on top of those infrastructure changes.  It would be easier to move features from FSX forward to Flight, because systems tended to have less dependency on infrastructure.  Moving backwards from Flight to FSX would not be easy due to the number of dependent systems that would also need to be moved over.

 

I suspect they may have had the same thought you had, but I'm not sure how much they looked into it.  There are some things that would be easy to move over.  The system used on the vegetation to get better lighting, and the rest of the lighting system in general, which is mostly isolated to shader changes.

Thanks Stonelance for the insight

Hi Stonelance. fascinating reading some of your comments!

 

I loved MS Flight, the fidelity of the ground objects, the light scattering and those tree shadows... Despite what you have stated regarding MS Flight as a base simulator for Flight School, as much as I'd like to agree if I were in DTG's shoes I would have definitely also gone with FSX as the starting platform.

 

As you've pointed out Flight has undergone several changes and compatibility would have been more involved on the Flight platform - this alone would have killed the decision for me.

 

But if you need more reasons, well it's simple really - there is something which I'm sure you're very familiar with in software development and that's the iron triangle (time vs quality/scope vs cost). Choosing FSX is a business decision to give DTG a chance of getting a product to market with the most appeal to both newcomers and hard core gamers. This is where DTG Flight School comes in which is essentially a proving ground for some of the new tech and buying more time to refine the Flight Simulator when it eventually is released at the end of the year.

 

There is also the perception of image. Microsoft Flight was controversial when it came out, and to be frank about it, a lot of people without your insight would have raised a question or two about building a platform for the next "FSX 11" on Flight technology - if somebody would have raised that after the demise of Flight people (as in perception with the brand) this venture would have ended before it started. 

 

DTG have been unwavering with there commitment to simulation technologies and in particular with reaffirming there commitment to flight simulation. The changes that we know of may seem cosmetic to you, but they do one thing and that is push what was the FSX platform forward - for me that's enough. And I'm sure in the couple of years they've been developing, that systems have been iterated on to allow updates and upgrades to be provided more easily and it's a large part of there business model to maintain the sim and it's content going forward.

 

For the 64bit question, I can't really see how a new venture like this wouldn't involve 64bit at some point and if this was going to happen, then better they have it now. Compatibility could be compromised a little but there should at least be a reasonable 'path' for at least the third party developers to follow.

 

So there you have it, FSX a choice based on common sense, not very romantic but DTG need to earn a buck like the rest of us.

A really silly thing i didn't notice before, but still interesting that i noticed on Flight... planes projecting their shadows on the trees

it's the little details that makes you like something for me

Flight is full of little detail and that makes it just incredible. I personally like the scenery and clouds reflecting from water, just amazing. (Am I of topic, hope not...)

 

-eelis-

As you've pointed out Flight has undergone several changes and compatibility would have been more involved on the Flight platform - this alone would have killed the decision for me.

 

I agree, it would have been harder, but not impossible.  Look at what I have been able to achieve with just working a bit in my spare time for free.  Also, you are assuming they let existing addons work on their platform.  There is also a huge difference in usability of the SDK of Flight compared to SDK.  There is a post from Gibbage somewhere that you can get someone else's opinion other than mine  The advantage that gives in ability to quickly create content is way more important than compatibility with old stuff.  Of course they will eventually have similar functionality, but they could of had it two years ago, instead of two+ years from now.  Backward compatibility is great from the users perspective, but it causes a lot of extra work for developers and makes it difficult to move the tech forward which hurts performance and visual quality.  They changed the lighting and the shaders to be PBR, so unless they moved away from the uber shader that FSX used, any old content is going to look wrong in DFS.  Also, the change from 32 to 64 bit is already going to break some compatibility, so I think your argument is valid but a little weak here.

 

There is also the perception of image

Definitely, but they could have just not mentioned MS Flight at all, and said it was based on FSX, which would still be true if it was based off of Flight.

 

And I'm sure in the couple of years they've been developing, that systems have been iterated on

You assume they have been working on it for a couple of years, but that is not the case, unless you want to count the work they did for FSX:SE as part of this work.

 

 

Obviously DTG made their decision for business reasons.  Based on what i know, I just don't think it was the right one.  They have a different set of info to make their decision based on, so it makes sense they might choose a different path.  The most telling thing to me is they didn't appear to even investigate MS Flight as an option, so they intentionally made a decision without having all the info.

..

The key to success for moving forward will be compatibility with existing 15 years of content (100's to 1000's of 3rd party devs over 15 years have produced A LOT of content that can't just be ignored) ... that's A LOT of content (it's the same challenge Laminar Research have been facing for many years).  Any company that wants to move forward WILL need to provide good/accurate conversion tools for existing content (just provide the tools).  

 

It's ok to have the "old legacy" content not look as good as "new" content, "new" content will replace old content over time ... but they both MUST be able to co-exist (once converted).  I would venture that MS Flight would be around today if they had allocated development resources to provide extensive content conversion even if it didn't look "as good as" native content.  It's all about content conversion ... the missing key element that no one wants to address because it's a huge challenge.

 

The other option is to provide both a 32bit product and a "new" 64bit product ... keep the legacy base intact providing the opportunity to get that content eventually moved over to 64bit over a long period of time.  

 

This market is like no other, how it's approached needs to be different than any other sim/game market.

 

This is the ONLY "long term" successful path in my humble opinion.

 

Cheers, Rob.

Any content relying on DLLs from FSX will not work in DFS as those are compiled as 32-bit. Anything dependent on SimConnect or FSUIPC will not work for the same reason. This eliminates most (if not all) aircraft.

 

Some scenery could work (at least terrain), as it is unlikely Dovetail would go to the effort to change the BGL file format for terrain. However, that might obscure the scenery by using zip files or encryption. Scenery objects might work, but I have my doubts. Anything that is distributed with a 'console' to enable features of scenery will probably not work.

 

Generally speaking, DFS will not be backwardly compatible. They definitely took the easiest path to release a new sim, and I predict it will be very profitable. Most simmers just want the series to continue.

 

Dick