DTG new FS could be blessing for LM

[Kilstorm]

If Dove Tail Games delivers on their release date of a new  x64 sim later in this year, it would give LM a good look at what overall performance it brings.  As of now LM has a very good thing going with plenty of addon community support where DTG will be starting from scratch almost to build its addon library so out of the gate LM has time to determine its options on "whats next" with them beyond what they already have planned.  As well as be able to look at how DTG advances to determine what works and what doesnt saving lots of time and money in R&D.

If they see what they like and want to advance P3D to a 64 bit platform we all will have a very mature FSX:SE, P3Dv3 sim to keep us happy as those two develop the next sim all to our benefit. Normally one would want to be a leading edge of the game but in this case there are advantages and disadvantages to leading when leaving behind a wealth of customers and addons. I know both LM and DTG are marketing to two different markets of game and trainer tool but they are both advancing the same sim engine and the market place can and does cross over.

There is no losers in it but there are unseen gains and all the benefits go to us the customers. 

[Museumkeeper]

A very informative and positive summary of what we all have to look forward to in the Sim world over the coming months/years

 

Kind regards

 

Ian

[pmb]

I fully agree, too. And would even go further.

 

LM might well have *some* 64 bit version in the backyard for some time. There may be several reasons they did not yet release it, one of them wanting to fix memory management in the 32 bit version first before turning to 64 bit. 

 

I could even imagine a scenerio where the yet to be released DTFS and the supposedly yet to be released LM P3D64 would be add-on compatible again. Wouldn't this be a win win situation for all parties: Lockheed-Martin, Dovetail, add-on makers - and us customers?

 

Not sure if both companies are willing to understand this, but I wouldn't exclude it.

 

Kind regards, Michael

[jabble]

 

1 hour ago, pmb said:

I could even imagine a scenerio where the yet to be released DTFS and the supposedly yet to be released LM P3D64 would be add-on compatible again. Wouldn't this be a win win situation for all parties: Lockheed-Martin, Dovetail, add-on makers - and us customers?

 

That would certainly be the best result for customers and add-on developers - if they shared SDK and external data structure, even if the internal engines worked differently.  I'm not too optimistic that will happen, but you never know.  I pity Orbx and others who keep having to jump through hoops to update addons for new versions of FSX/P3D - maintaining them for 2 data-different sims would be a nightmare!

[dominique]

You, guys, remind me of the Middle Ages dream of  some, to recreate the Roman Empire  ! The most probable future is that the two sims will diverge more and more.  Happy will we be if there's no wars !

 

11 hours ago, Kilstorm said:

 the market place can and does cross over.
 

 

Would it ?

 

LM gets its money in selling a sim core to corporate/institutional entities and do not sell addons. The functionalities (including structural changes) which will be added, in the future, will mostly be specifically made for its market. The recent change logs are telling.

 

DTG gets its money by discounting its sim core heavily, several times a year, and selling DLC which are not so cheap for what you get (remember no ObjectFlow in what they subcontract to OrbX) . Furthermore, the train simulator core seems to have a new version almost every year and compatibility of the older DLCs  is often broken if I believe what I  read in the user community comments on Steam.DTG is a machine to make profit out of short lived and light DLCs.

 

Why would  DTG accept to open its SDK to 3rd parties, except against an expensive licensing program ? Any addon made outside their grab, for example coming from a 3rd party company making stuff compatible with P3D, would be money lost.  If I were them, I would differentiate my sim like crazy.  One big casualty in that scenario : freeware is dead.

 

Thats the premises for an evolution. Now will the dynamics of the competition force DTG to evolve their business model ? Also, will LM stay in the consumers market ?

 

 

 

[neucoas]

I think that the fact that LM didn't released a 64bit yet is more "commercial" than anything.
If they release it just after DT launchs their FS, will confirm this.

[Maurizio Giorgi]

I can speak only about train simulator because i have it. This year, the version of TS 2016 is not a core remake but the same structure with more difficult tasks for the driver, plus some contents (routes and locos). i had TS 2015 and all the addons previously bought, work perfectly because nothing changed basically. The thing that people is maybe waiting is a simulator 64 bits structured, and this would be a real change for all the machines. Performances and problems on train simulator are not so distant respect to the flight simulator. I would say that flight simulator performs better on my machine, considering the many more polygons used and displayed.

 

 

[JacquesBrel]

25 minutes ago, dominique said:

Whyt would  DTG accept to open its SDK to 3rd parties, except against an expensive licensing program ? Any addon done outside their grab, for example coming from a 3rd party company making stuff compatible with P3D, would be money lost. 

 

Why would they choose to restrict 3rd party development as much as possible? Train Simulator already has a large amount of freeware content on Steam Workshop - not just scenarios, but actual routes created by freeware developers using the tools that come with Train Simulator. Anyone with a minimum, of simple, basic business sense knows that P3D and FSX are only alive today due to the massive amount of 3rd party content made by commercial and freeware developers. And the few official statements from DTG after their recent press releases strongly indicate that they are quite aware of this. 

 

"We recognise the passion that exists in the flight simulation community around enabling players to create user generated content. We will have more information to share on this as we get closer to the launch of Dovetail Games Flight Simulator."
 

[Kilstorm]

42 minutes ago, dominique said:

 

44 minutes ago, dominique said:

  12 hours ago, Kilstorm said: the market place can and does cross over.

 the market place can and does cross over.

 

Would it ?

Yes. I totally stand by that. There are simmers here who are pilots in training and there are pilots who like the entertainment of simming.  Thus as we have here in these forums as almost every other flight sim forum users of both P3D and FSX:SE with many users having both sims on their computer.

[dominique]

31 minutes ago, Kilstorm said:

Yes. I totally stand by that. There are simmers here who are pilots in training and there are pilots who like the entertainment of simming.  Thus as we have here in these forums as almost every other flight sim forum users of both P3D and FSX:SE with many users having both sims on their computer.

 

Of course we have users in both categories but a market is not driven only by the demand but also by the supply side : where do DTG and LM see where is their main profit source, users and products wise. If I had one wish, in any case,  is that LM which has a great potential to bring new technologies to us, stays in the consumers market on the long run. Which is not entirely sure.

 

55 minutes ago, JacquesBrel said:

 

Why would they choose to restrict 3rd party development as much as possible? Train Simulator already has a large amount of freeware content on Steam Workshop - not just scenarios, but actual routes created by freeware developers using the tools that come with Train Simulator. Anyone with a minimum, of simple, basic business sense knows that P3D and FSX are only alive today due to the massive amount of 3rd party content made by commercial and freeware developers. And the few official statements from DTG after their recent press releases strongly indicate that they are quite aware of this. 

 

"We recognise the passion that exists in the flight simulation community around enabling players to create user generated content. We will have more information to share on this as we get closer to the launch of Dovetail Games Flight Simulator."
 

 

Not saying that you don't have good points here and adding that I'd love so much me to be wrong and you right but, being old, I tend not to entirely trust what we call, in French, CareBears marketing rhetoric (Bisounours). I indeed hope they understand why Flight was a failure (besides the fact that MS wanted to be out of that kind of business). Again, where does their profit come from will be the key.

 

 

 

 

.

 

 

[Kilstorm]

3 minutes ago, dominique said:

 

Of course we have users in both categories but a market is not driven only by the demand but also by the supply side : 

 

I disagree. A market is made up of supply and demand but a market is only driven by demand and fueled by supply.  There is no successful market where there is supply without a demand. Thats why we always say "theres a market for that" when there is a known demand and if we see a new product without a known demand we ask "is there a market for that?"

[dominique]

 

1 minute ago, Kilstorm said:

I disagree. A market is made up of supply and demand but a market is only driven by demand and fueled by supply.  There is no successful market where there is supply without a demand. Thats why we always say "theres a market for that" when there is a known demand and if we see a new product without a known demand we ask "is there a market for that?"

 

This is not what I meant and hence not contradictory with what I wrote but  I am not sure that a public discussion on market economics is what this forum is made for ! So let's go flying, we agree there...

[Kilstorm]

Agreed!

 

[Rimshot]

1 hour ago, dominique said:

Not saying that you don't have good points here and adding that I'd love so much me to be wrong and you right but, being old, I tend not to entirely trust what we call, in French, CareBears marketing rhetoric (Bisounours). I indeed hope they understand why Flight was a failure (besides the fact that MS wanted to be out of that kind of business). Again, where does their profit come from will be the key.

 

The creative director of the DTG flight products also said the following about third party content. That's quite a different approach compared to the one MS took with Flight, don't you think?

 

" Dovetail Games Flight Simulator will be a 64-bit simulator using DirectX 11 and take full advantage of modern GPU and CPU performance. The advantage of this updated technology is that we have been able to introduce a range of globally applied techniques such as dynamic range rendering (HDR), atmospheric light scattering and physically based rendering (PBR). This means that aircraft and the environment will have the potential to look better than they have ever done before.

The downside of all this new technology is that the code bases are quite different so add-ons from FSX will not just work “out of the box”. 

Naturally we will provide developers with all the tools, documentation and support they need to create great new content for this new platform and will be working closely with them during that process."

[JacquesBrel]

By the way, the community manager from Dovetail has started a thread on the AVsim forums in order to answer questions from the community. So there's a chance to ask him and perhaps try to get some more solid information, in case you have concerns about where Dovetail will stand in relation to the flightsim market and the community.

 

http://www.avsim.com/topic/483642-ask-dovetail-games-about-dtg-flight-school-flight-simulator/

[Jordan King]

58 minutes ago, JacquesBrel said:

By the way, the community manager from Dovetail has started a thread on the AVsim forums in order to answer questions from the community. So there's a chance to ask him and perhaps try to get some more solid information, in case you have concerns about where Dovetail will stand in relation to the flightsim market and the community.

 

http://www.avsim.com/topic/483642-ask-dovetail-games-about-dtg-flight-school-flight-simulator/

 

Thats a bold move but one that I think other Sim developers could learn from. I hope he brought some RedBull with him!

[Mickel]

Good luck to them... they will find avsim a Kobayashi Maru exercise though.

[NZ255]

Just changing the sim from 32bit to 64bit wouldn't improve performance, in fact it might make it slightly slower as now the computer has to crunch 8 byte numbers instead of 4 byte ones.

The only advantage of 64bit (apart from marketing, which is doing a fantastic job ) is allow the application to use way more memory. That's a good thing for the users as then the OOMs won't be showing up anytime soon. Except the same could be achieved with memory management and you get the feature of retaining backwards compatibility (whether that's a good thing is another story).

Memory management would be a lot harder as you have to find where you're losing memory, or rather where you could be freeing it (e.g. that scenery that was loaded in at the beginning of the flight that is now hours away and out of view), which in a code base as big as FSX/P3D (I can only imagine) isn't an easy thing. LM are making headway with this and are probably still working on it.

 

In terms of putting it into an analogy, memory management is like plugging holes in a boat taking on water, going 64bit without fixing memory leaks is like buying a bigger boat.

 

To be honest I'm not bothered one way or the other, I want advances as much as the next guy.So bring on the new sims, I guess.

[Jon Clarke]

Quite honestly I only see the basic Sim as a tool to operate addons. The default Sim alone has always been very basic and I think unrealstic, unless you are a committed "in cockpit only" flyer of sophisticated planes that require a lot of in cockpit attention to Systems etc. I know i buy scenery addons to make the flying landscape more realistic, adventurous and immersive. There is no real immersion with any of the Sims only out of the box. If you consider all the types of addons people buy from high quality mesh, sceneries, airports, weather engines, replacement textures, landclass and planes, then it shows exactly how basic the original sim is.

A bit like making sure you buy a good quality pair of shoes. Once you have the good quality, comfortable shoes you are then able to go out and see the world full of natures "addons". it is the addons that create the entertainment, immersion etc, etc. If the sim can handle all the addons and their respective complexity then all is great.

Will for example a 64 bit Sim allow me to fly the LA Basin with my normal setting(High) without stutters, slight pauses and so on unless i turn down most things in the settings to ensure i get a reasonable flight in that area ? I don't get OOM's so have not included that aspect as a potential advantage of 64 bit.

[pmb]

24 minutes ago, jjaycee1 said:

Will for example a 64 bit Sim allow me to fly the LA Basin with my normal setting(High) without stutters, slight pauses and so on unless i turn down most things in the settings to ensure i get a reasonable flight in that area ? 

Contrary to frequent believe, this is probably not going to happen. JV already elucidated that 64 bit means larger numbers to crack and addresses to handle which may actually *lower* performance, at least without proper countermeasures.

 

I can add an experience from my life. I had and old (E-something based) 2008 laptop running Win 7/32, At a point, I thought I'd do it a favor and installed Win 7/64. After re-installing the software (just office and  a few networked sim addons) I found it to react somewhat more sticky, reluctant. Notably Simlauncher eating resources like fresh apples was more hesitant to respond. The difference wasn't huge and it didn't become unusable, but I found it noticeable.

 

Kind regards, Michael

[jabble]

13 hours ago, NZ255 said:

Just changing the sim from 32bit to 64bit wouldn't improve performance, in fact it might make it slightly slower as now the computer has to crunch 8 byte numbers instead of 4 byte ones.

...

 

I'll disagree with this - writing for x64 doesn't require use of large data types - we can still quite happily use 8-bit unsigned ints, for example.  The wider addressing is mainly managed by hardware, and with modern CPUs there should be no performance issue there.

 

Programmers usually optimise code for some priority, and if there's a tight memory model, optimisation will prioritise that at the expense of performance.  With far more memory available, the optimisation could be different, and likely boost performance considerably.  Related to that, there would not be the same need to read/write data from disc, something that slows apps down significantly.

 

But that also brings up the main issue - if you'd rewrite and re-optimise the same functionality, you might as well design the whole data model differently, and how it is processed, for much better performance.  In other words, you'd come up with more advanced sim that does things the old one just couldn't do.  Advantage could be taken of newer hardware, such as threading models for simultaneous processing.  However, all that would be at the expense of back-compatibility, as the data structures (like current .bgls) would be completely different.  That's simply the cost of progress.

 

5 hours ago, jjaycee1 said:

Quite honestly I only see the basic Sim as a tool to operate addons.

...

Will for example a 64 bit Sim allow me to fly the LA Basin with my normal setting(High) without stutters, slight pauses and so on unless i turn down most things in the settings to ensure i get a reasonable flight in that area ? I don't get OOM's so have not included that aspect as a potential advantage of 64 bit.

 

I quite agree - a flashy new sim with no addons would die a quick death, so the acceptance of customers and developers must go hand-in-hand.

 

If a new 64-bit sim is well designed and written, it should have considerably better performance in the LA Basin, Seattle, Vancouver, and all the other areas we tend to avoid just now.  Assuming the new data model is efficient, scalable and copes with such density, that is.  There would likely be a higher minimum hardware spec than for FSX/P3D, though.

 

4 hours ago, pmb said:

Contrary to frequent believe, this is probably not going to happen. JV already elucidated that 64 bit means larger numbers to crack and addresses to handle which may actually *lower* performance, at least without proper countermeasures.

...

 

Again, I'll disagree that x64 code is a worse performer in general, but it does depend on 1) just how well the application is written, and 2) the hardware.  I upgraded my 2009 laptop to Win7 x64, and it initially performed more sluggishly.  I then added more memory, and it performed fine - it just needed the extra RAM.  Most of my applications remained 32-bit, with a few new 64-bit versions.  If you are using 32-bit software on a 64-bit O/S, you won't see any benefit over a 32-bit O/S, and indeed the new O/S may need more RAM to perform well.

[Culley44]

I think anyone waiting on dovetail to release there new flight sim, most are gonna be extremely disappointed. If it does in fact follow what microsoft flight was, it will be

addon after addon on steam. The cost will be as if you had to start all over.As far as Lockheed, there user base is not entertainment, so i don't think dovetail will make

a difference one way or the other on how Lockheed update's it's platform in the future. Just my opinion.

[jabble]

5 minutes ago, Brett Culley said:

I think anyone waiting on dovetail to release there new flight sim, most are gonna be extremely disappointed. If it does in fact follow what microsoft flight was, it will be

addon after addon on steam. The cost will be as if you had to start all over.As far as Lockheed, there user base is not entertainment, so i don't think dovetail will make

a difference one way or the other on how Lockheed update's it's platform in the future. Just my opinion.

 

IF there's an open SDK AND the sim is good, then there's an excellent chance of it being accepted as the successor to FSX, and it will eventually dominate the addon market. Those are big IFs though. You're dead right about LM - who knows what they have in the pipeline?  What if, the day before DTFS launches with a closed SDK, LM announce an upcoming P3D-64 with an open one?

[Rimshot]

1 hour ago, Brett Culley said:

I think anyone waiting on dovetail to release there new flight sim, most are gonna be extremely disappointed. If it does in fact follow what microsoft flight was, it will be

addon after addon on steam. The cost will be as if you had to start all over.As far as Lockheed, there user base is not entertainment, so i don't think dovetail will make

a difference one way or the other on how Lockheed update's it's platform in the future. Just my opinion.

 

The DTG flight sim will not follow what MS Flight was. That has been stated over and over again already. It will be a successor to FSX. And yes, there will be addon after addon eventually on Steam, maybe even in other webshops (just like now actually). There is a q & a thread over at Avsim where the community manager of DTG answers questions. Very interesting. And about having to start all over again, well, aren't we all used to that?

[NZ255]

[  

I'll disagree with this - writing for x64 doesn't require use of large data types - we can still quite happily use 8-bit unsigned ints, for example.  The wider addressing is mainly managed by hardware, and with modern CPUs there should be no performance issue there.

 

Programmers usually optimise code for some priority, and if there's a tight memory model, optimisation will prioritise that at the expense of performance.  With far more memory available, the optimisation could be different, and likely boost performance considerably.  Related to that, there would not be the same need to read/write data from disc, something that slows apps down significantly.

 

But that also brings up the main issue - if you'd rewrite and re-optimise the same functionality, you might as well design the whole data model differently, and how it is processed, for much better performance.  In other words, you'd come up with more advanced sim that does things the old one just couldn't do.  Advantage could be taken of newer hardware, such as threading models for simultaneous processing.  However, all that would be at the expense of back-compatibility, as the data structures (like current .bgls) would be completely different.  That's simply the cost of progress.

 

]

Thanks for the insight, always good to learn more.

Would part of that come from the fact that (if larger numbers were required) CPUs are just so good at calculations on numbers?

I didn't think about the io to and from disk. Good point. Although with the sim allowing you to fly anywhere how would you know what to cache in memory? Or would you just throw everything in?

[jabble]

21 minutes ago, NZ255 said:

Thanks for the insight, always good to learn more.

Would part of that come from the fact that (if larger numbers were required) CPUs are just so good at calculations on numbers?

I didn't think about the io to and from disk. Good point. Although with the sim allowing you to fly anywhere how would you know what to cache in memory? Or would you just throw everything in?

 

Sure, modern CPUs have more capabilities than their predecessors; for example if you're writing 64-bit Intel code, you can use internal CPU registers (faster) for parameter passing (a common action), instead of using the "stack" in normal RAM (slower).  Each new generation of CPUs introduces new abilities, even apart from the clock speed increase.  The version of Windows can matter too, as the various editions have different capabilities and limitations.  If you're at all programmery-minded, this is quite informative.

 

As for disc I/O, think of FSX when you're flying along, and suddenly there's a stutter as it loads data from the disc.  So, you ask yourself, why didn't it load it earlier, more gradually?  Almost certainly because there's not enough free space to hold it in RAM - a compromise made to cope with a smaller VAS, so it loads at the last moment.  With a larger VAS and a better multi-threaded program, it could predict what data you need long before it's used (e.g. from direction of travel, layered 'bubbles' of increasing detail), and load it gradually via 'queued' functions that execute when the CPU is less busy.  The user wouldn't have to do anything - it would just happen.  Maybe it could be configured, though, if available RAM is an issue.

 

That's just one example of how performance could be improved with more VAS available.  To optimise that new ability, you might like to arrange world data on the disc in a certain way - the new equivalent of bgl files.  As the requirement is different (including how it draws/renders the world), the data structure would likely be different from just now.  And so as the performance increases, back-compatibility is lost.  Plus there's probably a zillion other improvements and new abilities that the developers would like to include.  To keep the same data structures as now would severely limit that - hence the argument that breaking compatibility could bring us a much better sim.  Well, we hope!

 

[pmb]

Hi jabble,

 

over there at the AVSIM forum was just a remarkable remark by Martin from DTG:

 

"Things like scenery shouldn't, as you say, need updating"

 

http://www.avsim.com/topic/483642-ask-dovetail-games-about-dtg-flight-school-flight-simulator/page-5 (#65)

 

The question (#59) was on FSX photoscenery, nothing ORBX-like, and Martin obviously doesn't guarantee it at this point. However, I would read it the way they are at least not intending to completely break the present .bgl file structure. Would you agree?

 

Kind regards, Michael

 

 

[jabble]

18 minutes ago, pmb said:

Hi jabble,

 

over there at the AVSIM forum was just a remarkable remark by Martin from DTG:

 

"Things like scenery shouldn't, as you say, need updating"

 

http://www.avsim.com/topic/483642-ask-dovetail-games-about-dtg-flight-school-flight-simulator/page-5 (#65)

 

The question (#59) was on FSX photoscenery, nothing ORBX-like, and Martin obviously doesn't guarantee it at this point. However, I would read it the way they are at least not intending to completely break the present .bgl file structure. Would you agree?

 

Kind regards, Michael

Hi Michael,

 

For those of us who want a best-possible performing sim, this is worrying!  Still, I don't think they have yet decided exactly what they will do.

 

That one phrase could be interpreted in different ways.  Maybe the new sim will allow compatibility (or partial compatibility) with FSX .bgls, or perhaps it will have an external method of converting FSX scenery to the new format.  Or it could just be empathy for a concerned user, without any committed answer.  "Looking to" implies they're still working out their best solution, and haven't reached it yet.  I'd like to see a more precise answer, but if they haven't decided, we must wait.

[be77solo]

1 hour ago, jabble said:

Hi Michael,

 

For those of us who want a best-possible performing sim, this is worrying!  Still, I don't think they have yet decided exactly what they will do.

 

That one phrase could be interpreted in different ways.  Maybe the new sim will allow compatibility (or partial compatibility) with FSX .bgls, or perhaps it will have an external method of converting FSX scenery to the new format.  Or it could just be empathy for a concerned user, without any committed answer.  "Looking to" implies they're still working out their best solution, and haven't reached it yet.  I'd like to see a more precise answer, but if they haven't decided, we must wait.

That's my take away so far... no real answers on what they'll be doing, or if they even know themselves.

 

Curious to watch, but if it has all the same basic FSX file structures with the same basic FSX code, how are we to expect a sim worth dumping all our existing beloved addons for?

 

Exited to see what they come up with, hope they succeed, but it will take a lot for me to switch.

[Jon Clarke]

It is going to be very difficult to judge the new sim's advantages without detailed addons initially. All the flight sims i have had from FS2000 upto P3D have worked flawlessly in their "out of the box" default state. Very high, smooth FPS even on maximum settings. Mainly due I suppose to mediocre ground textures, 70-120m mesh, limited weather options, basic autogen and built in ai traffic. It is when you start adding on to the sim highly detailed addons that the performance issues start to arise.

That therefore begs the question as to how will it perform with orbx type scenery if indeed orbx will develop scenery for 64bit. I hope they will although my bank manager hopes they won't !!!!!!!!!