VULKAN for Prepar3D v5?

[SteveW 1,792]

The use of cross platform is not driven by the specifics of performance improvements, they are handy but this is rather based on the likelihood of sales across multiple platforms and also important is the easiness of use of the API. These are the important factors. Look at games engines and the programming methods to write them they are a higher level now.

The cross platform engine cannot perform better than a single platform engine, but in fact could perform worse if not implemented correctly as they say on the Vulcan site. Vulcan is only compared to DX12 as similar.

Advances in GPUs define advances in graphics APIs. All the current "competitors of DX12" are not really competitors, rather they are users of the same technology, only able to do their thing because of the work to get GPUs up to specifications available to DX12 type specifications which comes from the work done in the late 90's on MS DX.

Currently most work on the GPU is shaders but spare horse power can be utilised for other functions.

 

[Lorby_SI 2,038]

1 hour ago, Cruachan said:

 I am sensing a shift in the stance taken hitherto by LM who are becoming much more accepting of the wider adoption and appeal of their flight simulation project extending far beyond their prioritised commercial and military commitments

Hello Mike,

I'm curious, where and how did you sense that? The recent interview with Adam Breed seems to suggest the opposite. Also, the clarification of the EULA, especially for the Academic license, IMHO doesn't leave any room for "wider adoption".

My personal view is, that the future of desktop simming lies in X-Plane, Aerofly, DCS, but not in P3D. The addon market has taken quite the downturn in the last 3 years. At least from what I can see. Selling 1000 copies of anything is already considered a success these days, according to other devs I know. Plus, in every version LM invests a lot of resources into features that are of no interest to the flight simmer, like the SimDirector and the SimOperator. I expect the number of P3D users to decline in the next three years, moving to the other platforms, especially if those readily adopt the newer technology like Vulkan - end users just love that. LM probably won't mind, their attention is elsewhere. Maybe, if that happens, they will reconsider their licensing options. But I doubt that too.

Best regards

Edited July 20, 2018 by Lorby_SI

[WotanUK 289]

10 minutes ago, Lorby_SI said:

Hello Mike,

I'm curious, where and how did you sense that? The recent interview with Adam Breed seems to suggest the opposite. Also, the clarification of the EULA, especially for the Academic license, IMHO doesn't leave any room for "wider adoption".

My personal view is, that the future of desktop simming lies in X-Plane, Aerofly, DCS, but not in P3D. The addon market has taken quite the downturn in the last 3 years. At least from what I can see. Selling 1000 copies of anything is already considered a success these days, according to other devs I know. Plus, in every version LM invests a lot of resources into features that are of no interest to the flight simmer, like the SimDirector and the SimOperator. I expect the number of P3D users to decline in the next three years, moving to the other platforms, especially if those readily adopt the newer technology like Vulkan - end users just love that. LM probably won't mind, their attention is elsewhere. Maybe, if that happens, they will reconsider their licensing options. But I doubt that too.

Best regards

I will put money that doesn't happen, at least not in the way you have described.  The only realistic sim that could take the place of P3D is X-Plane, the others aren't really even direct competitors.  DCS is heavily focused on military aircraft and even if it wasn't the area's covered are way to small and release in geological timescales, by the time the entire world is simulated (if ever) i suspect the AC will have answered the last question.  Aerofly is much like DCS, small areas with superb detail, they would also need to add numerous functions to the core sim.

The only way people like me are moving from P3D to X-plane is if LM pull the rug from under us, not because i dislike X-Plane, but because i have so much money invested in P3D.  I also have the Professional License for which one of the allowed uses is Simulation, exactly what i do with it!  Don't forget it's also not entirely a one way street, us simmers are effectively paying beta testers for LM, we often find problems that commercial customers would take much longer to find.

[shivers9 1,174]

3 hours ago, FalconAF said:

Repeat after me:  "P3D is NOT a GAME.  It is a SIMULATOR.  I am NOT a GAMER.  I am NOT a GAMER.  I am NOT a GAMER...……"

I think you are on the right track here. IMHO one of the problems here is that some many of our fellow flight simmers think of P3d or X-plain or AFS2 as a program that runs in our home on one computer Just like  the "Goat Simulator. Much insight can be gained by reading the forums. Hardly a month goes by with out a question like "will P3d run on my laptop?" The simple answer is no it will not run on you laptop and was never intended to. When I say run, I mean run well and as it was intended to. Yes, it will load and boot up and sort of run. That is the real and correct answer. The folks who ask this question are "Gamers" and that is not a bad thing. They very well may be future "Flight Simmers" and that is a very good thing for the survival of the Hobby.

Yet another problem is that  we all have gotten use to the idea that we will have all our problems solved by the next release of say P3d. If that dont work for us then surely a faster CPU or GPU will come out soon. The problem is that all of those "fixes" are getting streached to their limits. I don't think the complete answer answer lies in Vulcan or DX12. AFS2 may be the go to for flight simmers in say 2030 but I doubt it. My guess is that in the end hardware will hold the answer. My vision for say 2020 or 2025 will be a computer that is running say two processors synced and 2 or more Video cards. LM has spent a lot of time in the area of software connections to Prepar3d. P3d would become the space (simulated world) that we fly in either in VR or on a laptop. All the things like weather, ATC, and planes would be stripped out and provided by outside developers and run outside of the "simulated world". We really already have many of these things working today. Most cockpit builders use many computers and monitors to run their simulations. I do believe this direction is more likely than the chance of some new process that will give us a 10 GHz cpu running at 72 degrees. 

[Cruachan 705]

58 minutes ago, Lorby_SI said:

I'm curious, where and how did you sense that? The recent interview with Adam Breed seems to suggest the opposite. Also, the clarification of the EULA, especially for the Academic license, IMHO doesn't leave any room for "wider adoption".

Hi Oliver,

My interpretation of Adam’s response to the question was that LM saw no driver for change and that there was a tacit acceptance of the current situation.....whatever that might be. Probably best to avoid discussing this any further in case the Moderators of this forum become fidgety!

Best regards,

Mike

[Lorby_SI 2,038]

4 minutes ago, Cruachan said:

Hi Oliver,

My interpretation of Adam’s response to the question was that LM saw no driver for change and that there was a tacit acceptance of the current situation.....whatever that might be. Probably best to avoid discussing this any further in case the Moderators of this forum become fidgety!

Best regards,

Mike

Hello Mike,

I'm not sure that said interview is food for moderation, it is all out in the open. Do you mean the question if LM are planning to drop the Academic license? If it is, in the same paragraph Mr. Breed not only says no, but he also spells out who is eligible for that license. I can't see anything that goes beyond their existing EULA in that.

Best regards

[shivers9 1,174]

2 hours ago, Lorby_SI said:

I expect the number of P3D users to decline in the next three years, moving to the other platforms, especially if those readily adopt the newer technology like Vulkan - end users just love that.

Well Oliver, if you are correct then we are headed in the direction of having the hobby of Flight Sims returning to it's roots of being just another flying game. That is sad really. Well at least then the kids will be able to run it on their stupid little cell phones.

[pgde 571]

On 7/18/2018 at 6:03 AM, ckyliu said:

I hope Vulkan gets widespread adoption because I can't stand Windows 10 and it's constant updating breaking things; I was dreading I'd be forced there by DX10.

You aren't forced to use Win10 for DX10. I am on Win7 Pro and DX11. It is DX12 that requires Win10.

Regards!

[ckyliu 1,749]

Yeah I realised my mistake once I was outside the edit post time period, DX12 stopped me getting Forza 7 Or Forza Horizon 7 and I got Assetto Corsa instead

[Noel 4,496]

10 hours ago, Lorby_SI said:

Plus, in every version LM invests a lot of resources into features that are of no interest to the flight simmer, like the SimDirector and the SimOperator.

This has always been my primary complaint and is the foundation for why I am exceedingly slow to pay for new releases.  I really do see it as paying beta testing.   It's still the best total show in town though and I will continue to use P3D 3.4 as long as it will run as it satisfies all aspects of my training program requirements.   I would happily jump ship if X-plane could match the sheer depth of content in P3D.

[FalconAF 213]

15 hours ago, WotanUK said:

Don't forget it's also not entirely a one way street, us simmers are effectively paying beta testers for LM, we often find problems that commercial customers would take much longer to find.

But look at the bigger picture for the developer.  We (home computer simmers) often find (and report...and COMPLAIN about) "problems" that are of absolutely no concern to commercial and military training users.  Then "we" expect LM to provide a fix "for us" in the next update.  Which historically rarely happens for many of "our" problem reports, because many of the problems "we" report are a non-issue when P3D is used as a commercial or military training product.

Yes, we are allowed to use P3D.  Yes, we MIGHT find and report problems that WOULD be an issue for commercial or military trainers.  But the MAJORITY of the problems "flight simmers" find and report would NEVER BE HAPPENING IN THE FIRST PLACE when P3D is used in a realistic commercial or military training environment, run by people who have the knowledge on how to prevent those problems from happening in the first place.

Trust me.  "Blurry textures" are not a "show stopper" problem for commercial or military computer training scenarios.  If some of THEIR textures are blurry when running their training scenarios, it's because those specific textures don't NEED to be crystal clear 4096 resolution to begin with.  And if parts of a scenario DO need crystal clear textures, the people running the simulation will use ONLY the amount of "scenery resources" necessary to meet the training requirements.  They will not be "overloading" the computer system running the simulation scenario with enough "stuff" to choke a pony.  😉

THAT is the reason for the XML scenery and sim objects method used now for P3D.  A commercial or military training environment will develop and use XML-based training "packages" that contain ONLY the amount of things needed for a specific training objective.  They will not be loading the whole darn world and every possible addon airport available all at the same time for a training scenario that is conducted in say the Nevada Test Range for a Red Flag exercise (which, by the way, is why something like DCS DOES run so well on a home computer, even when using massively complex aircraft with it).  Nobody cares...nor wants...nor NEEDS anything "scenery" from any other place in the world in the simulation at the time.

So, you want to fly a 12-hour long haul flight in P3D?  With all the scenery you would need for it already loaded in your flight sim scenery library?  You are a flight simulator HOBBIEST user.  No commercial or military training scenario is ever going to try to do something like that.

And really...if a commercial or military training scenario needs to "teach" someone what it would be like landing at KSFO in a complex aircraft (which would really have the goal of teaching the AIRCRAFT systems...not what the scenery looks like down to being able to count the rivets on the terminal buildings, other aircraft, or how many blades of grass are blowing in the wind), the KSFO airport and surrounding area do NOT need to be 4096 resolution quality in the display.  Commercial and military training scenarios concentrate on displaying what is NECESSARY to meet the training requirements.  That's the difference between how a commercial or military training user would use P3D vs us "hobbiest" users. 

Our flight sim forums are full of commercial developers and users.  We ask for their opinions on our "problems", and when they give us realistic answers, way too many of us still want to "argue" with them.  That's sad.  It's sad because when you ask someone who is actively involved and successful as a "commercial" or "professional" entity, then after they answer you you tell them THEY are wrong, that is normally not a reflection on their qualifications.

Edited July 21, 2018 by FalconAF

[WotanUK 289]

3 hours ago, FalconAF said:

😉

Our flight sim forums are full of commercial developers and users.  We ask for their opinions on our "problems", and when they give us realistic answers, way too many of us still want to "argue" with them.  That's sad.  It's sad because when you ask someone who is actively involved and successful as a "commercial" or "professional" entity, then after they answer you you tell them THEY are wrong, that is normally not a reflection on their qualifications.

I am not sure that i agree entirely with your post, you make some excellent points but if what you are claiming is true then my question is why do LM still let us buy the product and get the support, they could stop us tomorrow if they wanted to?

You say that "blurry textures" aren't a problem to commercial users, i agree if you are only looking in the civilian spectrum, but if i am wanting to simulate a mission for the RAF i want sharp textures all the way for my target and package assets.

Yes the DCS model i mentioned in my post, small areas with superb detail, but LM didn't go down this route with P3D, i guess they could have but they choose to simulate the entire world at a lower level of detail; i suppose if you want to use DCS to simulate a potential attack on the DPRK you have to wait for DCS to release a module by which time i am sure that the Earth will have formed the federation and we will be battling Klingons.  Using P3D what we have is exactly what you say, areas that are low detail, and areas that we are interested in are high detail, often with add on packages.

Not sure, i can see why a military mission might want to simulate a 12hr mission profile with all assets on route.

Nope sorry i disagree entirely here, if your point was correct the big Level-D sims would just consist of a runway, why else is anything needed, but in fact they have been getting better graphically along with the home sims.  I mean, i guess you have a point when it comes to things like GSX, but i would suggest Active Sky (or something like it) is very welcome even in the commercial space.

Not sure what your point is here, commercial customers are far more demanding than we are and indeed LM are probably more responsive to their requirements than to ours but it doesn't mean that our desires don't cross on occasion.  People said that LM would never go 64-bit because that was only a desire for simmers, that commercial customers wouldn't be interested because (as you say), they only ever run small areas in low detail; but LM went 64-bit.

I don't think we are (or even should be) a priority for LM, but equally i don't think they want to stop us using the product and they do fix things that we flag up, they even attend flight sim conferences with the same simmers that buy the product, they have done nothing so far that says you shouldn't be buying this product.

[Cruachan 705]

Excellent post, Ian!

i suspect there’s a lot more going on than we are told directly. LM own the code and all intellectual rights which gives them exclusivity to react in any way they choose to their evolving user base and feedback. Like the rest of us, they must be enthusiastic flight simmers, and will be sharing our appetite and passions for positive changes in their simulator. I bet each and every developer has a fully spec’d installation in their respective homes. 

Employing the Star Trek analogy, their ‘prime directive’ may be to concentrate on their commercial and military customers’ requirements, and yet we have witnessed the sim evolve far beyond that as they actively encourage third party participation in the project. To me, this speaks volumes about how they are viewing and accepting our desires to exploit the software and hardware technology to better their simulator in any way that is practicable. Why would they not? Also, this is underlined by their decision to progress to a new version.

We have been and continue to be an important, if not vital, part of the equation. Without us, it is more than likely the project would have matured long ago without having realised its full potential. To me, this has demonstrated great wisdom on the part of the Project Developers. In the end we can only speculate about company policy, but I would argue that the results to date speak volumes about what us actually happening and our shared visions for the future. I can only hope that Version 5 continues to embrace this vision rather than employing features that may threaten to undermine it.

Regards,

Mike

[shivers9 1,174]

Mike....thanks for starting what has turned out to be a very interesting thread. Lots of good opinions and best of all no one is yelling!

[DylanM 1,069]

3 hours ago, Cruachan said:

We have been and continue to be an important, if not vital, part of the equation. Without us, it is more than likely the project would have matured long ago without having realised its full potential. To me, this has demonstrated great wisdom on the part of the Project Developers.

Couldn't agree more - every interaction I've had with the devs has reinforced the idea that they are just as passionate about simming and creating an environment, as they are about building a stable commercially viable product. I'm hopeful this mindset will carry through into the evolution of their engine...