Scenery is nice, but the time proportion that you look at the scenery is tiny.

 

For me, scenery is far more important than a fancy looking cockpit. I like both flying heavies, and also light GA. The whole experience for me is taking off from one place and landing in another "different" area, and feeling like you have gone somewhere.

 

With IFR flights, I agree a decent cockpit is important, especially having fluid and clear instruments, however scenery becomes even more important if you enjoy simply getting out a VFR chart, planning a route, and then flying it using visual aids. On my FSX install, I have photoscenery installed for UK, France,Germany,Switzerland,Spain and Poland, and I also have Orbx scenery installed for Australia/NZ and the PNW. Even in a short distance, e.g. A flight from Northern France to the South of Spain, it's possible to see the scenery changing from green fields and villages, to the rugged dry landscape of Southern Spain. Even flying at 37,000 feet, this is important to me.

Not everyone is into the "heavies", and flying at 30+ thousand feet.

true...but even heavies need to come down and land.

 

To me, scenery is about dynamics. I dont want satellite images pasted down cause it always is the same. I want to see it look different depending on conditions. Nothing is more rewarding then to descend through a cloud layer at a new region and think "oh..so thats what winter looks like here".

May be this is off-topic a bit, but the real reason I have not moved over to X-Plane is the the fact that sophisticate plane maker like PMDG and RealAir do not support it yet. 90% of your interaction with the SIM is via the cockpit of a plane. Scenery is nice, but the time proportion that you look at the scenery is tiny.

 

For me, scenery is far more important than a fancy looking cockpit.

 

IMO, a well done cockpit is a must

 

I need good scenery to maintain an interest. Scenery is why I took up real flight

 

Flight dynamics..............of course, they need to be believable.

 

Exterior graphics........also very important.

 

As to who does the above, I don't care.

Tessellation + Shader 5 + PhysX would be enough to produce real time weather including snow accumulation. For example, the base poly for a tree would be more primitive for those trees that are not evergreens, use Tessellation to subdivide the base poly, then use shaders 5 to do the seasons, physX could be used for "active" weather (snow fall, rain fall).

 

It's possible ... not sure why you folks are saying "hire someone" ... I'm sure Austin has people very capable of doing this just as Ace's crew did ... just a matter of time investment (and of course money).

 

BUT (and this is a big BUT), performance may be a big issue as it'll really stress just how good a video card is and it'll certainly need SLI maybe even Quad SLI ... you've been warned, so get out the check books.

 

Unless of course UniLoc has a patent on the "concept of using Tessellation and Shader 5.0 to produce real time weather", then we're all screwed ... again.

...It's possible ... not sure why you folks are saying "hire someone" ... I'm sure Austin has people very capable of doing this just as Ace's crew did ... just a matter of time investment (and of course money).

..what about a third party? Could a 3rd party access the required parts of the sim and sell that capability? It'd be a win/win for both LR and the 3rd party.

 

 

Are we ever gont to see an FSX SP3 that would make it 64-bit, so we can avoid those VAS limit OOMs that tend to happen with high end products combined like ORBX, PMDG, etc? Probably not since the code is now owned by P3D, which is does not follow the FSX customer model.

 

I have XP10 on a separate drive, and I really tink it's promising, and it seems like a natural successor to FSX sine Microsoft will no longer pursue development. I think XP10 gives the most flexibility to add-on developers. I'm going to enjoy FSX as long as I can in the meantime.

..what about a third party? Could a 3rd party access the required parts of the sim and sell that capability? It'd be a win/win for both LR and the 3rd party.

 

Anything is possible, but I would imagine this is part of the core render so I don't see how a 3rd party could do this with any semblance of "performance" ... 3rd party involvement would be to simply provide the base scenery.

 

However, it's still a challenge, most base scenery comes with trees in full bloom (aka summer) ... which you actually would NOT want because tessellation/shaders could be used to produce the "Bloom" ... the object classification along with season and weather would define tessellation/shaders ... for example an base scenery texture would have a different level of tessellation/shader vs. tree vs. building ... but they are all primitive polys with textures so it should be doable - but there are limits on how many polys you can do even with tessellation support.

 

This has been done before so it's not really a "new concept" ... it's just that DX11 & PhysX would make it viable on today's hardware.

It's possible ... not sure why you folks are saying "hire someone" ... I'm sure Austin has people very capable of doing this just as Ace's crew did ... just a matter of time investment (and of course money).

 

...aaaand there's the rub.

 

We'll be lucky to see the platform improvements for XP10 which they planned to do over the course of it's eternal beta...nevermind anything like this. Honestly this sounds more like an X-Plane 12 feature - a great feature nonetheless, but at least 5 years off in my mind (*Barring Laminar Research going on a most welcomed hiring spree!)

 

Editing to add: Don't forget X-Plane is not a DirectX platform - it's based on the features of OpenGL... anything exclusive to DX11 and therefore the Windows platform will be quickly rejected, as it ruins their cross-platform capabilities.

It's possible ... not sure why you folks are saying "hire someone" ... I'm sure Austin has people very capable of doing this just as Ace's crew did ... just a matter of time investment (and of course money).

 

I would hope your kidding, somehow I don't think you are?

 

I could easily elaborate on what Gregg has said, I won't. I will add one thing though, don't "hire someone" hire a bunch, or source it all out, just get it done. I'm not going to buy, Version 11, 12, 13 and so on, before Laminar Research provide us with what a "complete" simulator. I'm done at Version 10 with giving him another dollar, hopefully I'll be able to buy some addons to make version 10 complete. I'm not going to continually give him money, so he can give me a few crumbs version to version. Finish a version, allow developers to make addons, and I'll decide what I want to purchase to make it better. That model has worked very well with the Microsoft series, no reason it can't work here as well. We all know why it hasn’t......

 

Glen

Regarding adding these features vs. LR "in-house" work, another point that's been brought up is the wariness of a 3PD to add to or improve on something in XP for fear that LR will do it in an update. This is especially valid for payware, since sales depend on having a need for the feature the payware offers.

 

Maybe LR needs to publish some kind of high level development roadmap to give folks a general idea of what to expect with future updates. It would make planning for addon development much easier, I think.

Regarding adding these features vs. LR "in-house" work, another point that's been brought up is the wariness of a 3PD to add to or improve on something in XP for fear that LR will do it in an update. This is especially valid for payware, since sales depend on having a need for the feature the payware offers.

 

Maybe LR needs to publish some kind of high level development roadmap to give folks a general idea of what to expect with future updates. It would make planning for addon development much easier, I think.

 

Very true, and a good idea.

 

Glen

Anything is possible, but I would imagine this is part of the core render so I don't see how a 3rd party could do this with any semblance of "performance" ... 3rd party involvement would be to simply provide the base scenery.

 

However, it's still a challenge, most base scenery comes with trees in full bloom (aka summer) ... which you actually would NOT want because tessellation/shaders could be used to produce the "Bloom" ... the object classification along with season and weather would define tessellation/shaders ... for example an base scenery texture would have a different level of tessellation/shader vs. tree vs. building ... but they are all primitive polys with textures so it should be doable - but there are limits on how many polys you can do even with tessellation support.

 

This has been done before so it's not really a "new concept" ... it's just that DX11 & PhysX would make it viable on today's hardware.

 

I'm seeing a lot of terms being thrown about here, so I just want to clarify how these technologies are actually used for everyone involved. This is not a criticism; it is pretty hard to keep up with all the new goodies card makers bring to the table.

 

Firstly, PhysX merely uses the GPU as a general purpose parallel processor for object and world physics calculations; it has no effect on graphics or the generation of graphics. A GPU is very efficient at doing these types of calculations, so this is a sensible way of doing it. However, this comes at a cost to your graphics card, as now the physics engine competes with the rendering pipeline for horsepower.

 

Secondly, tessellation is not a process that is used to create new geometry. Tessellation merely interpolates existing geometry and chops that geometry up into smaller polygons, giving more smoothness. This is most applicable to complex curved surfaces. For example, a flat runway is easily represented as a single polygon, a rectangle. Tessellating that 128x just gives you the same flat runway, except now it's 128x more costly to draw. However, consider a cylinder. If you give it only 4 sides, it becomes a square column. But if you tessellate 2x, now it's a octagonal cylinder. 2x that and it has 16 sides, etc. That's the benefit of tessellation. However, it cannot be used to create geometry that didn't already exist. The downside is it takes real GPU juice.

 

Honestly, there's no good reason to leverage those technologies for some trees and a little snow. I can think of a fairly easy way of doing an equally convincing job using some simple particle systems and a few vertex and pixel shaders. You would still need all the seasonal textures, but you could blend them using an Unreal Engine style multi-material surface and alpha channels. And you wouldn't need Quad SLI to do it.

 

X-Plane's lack of traction is due solely to Austin's lack of cohesive vision, low interface quality, and the inability for X-Plane to stabilize and iterate instead of feature grabbing every year or so. The XP10 platform looks to be good, if it can get stabilized in the 64bit realm and then bring the art assets up to date with the code.

I would hope your kidding, somehow I don't think you are?

 

I don't know LR's staff or financial situation, so I really can't commit. I was mearly suggesting they already have the technical knowledge, I'm not their resource manager and don't have access to their financials

 

I do agree that more resource will probably help move them forward into a state that is more 3rd party friendly.

 

PhysX merely uses the GPU as a general purpose parallel processor for object and world physics calculations;

 

This is where snow fall and rain fall could benefit ... to do exactly those calculations. Not suggesting to be used to produce the objects geometry, only to move it.

 

Tessellation merely interpolates existing geometry and chops that geometry up into smaller polygons, giving more smoothness.

 

Yes it produces subdivisions but when combined with an alpha and a displacement map the effect could be "close enough" ... I haven't done this, just tossing out possible ideas ... tessellation + shader 5 is a very powerful tool that can be used in many different ways, it's not just for "smoothing".

 

But to be clear, I'm not suggesting this is the ONLY way ... per my link to and older PDF on how to produce real time snow. There are many ways to approach the problem ... the best way will most likely be decided by current computing power vs. realism.

 

We can agree to disagree ... I don't think "seansonal" textures are needed any longer and indeed isn't viable in the future.

 

In defense of LR/AM ... it's hard to "stabilize" and yet also leverage all the latest computing hardware ... hopefully DX11 and 64bit will be the start of that some stabilization. It's a double edge sword ... a flight sim and 3rd party need to be able to move forward together. FSX was partly held captive to 3rd party's not wanting any change (short term thinking IMHO).

 

The solution is that both 3rd party and flight sim need to figure out how to change together, because change is inevitable.

I don't know LR's staff or financial situation, so I really can't commit. I was mearly suggesting they already have the technical knowledge, I'm not their resource manager and don't have access to their financials

 

To be honest, neither do I. Do they have the technical knowledge, sure they do. They have created something far beyond what "the average person" is capable of, you don't do that without possessing a great deal of knowledge and expertise.

 

All that being said, it sometimes comes to a point in ones life where they may have to admit they have bitten off more than they can chew. I understand the situation that gets bantered about here all the time, the team is small. Well that always was, and still remains a conscious decision. One that's a mistake as far as I'm concerned. If he doesn't act, and act soon, a great opportunity is going to be lost. Someone else WILL come along, it's just a matter of time. He can carry on like he has for the last 20 years, never being the leader in the industry, or he can try something different, and perhaps rise to be number one. It's up to him, I will be here to tell him what I want as a user, hope he listens.

 

We just have to look at the facts. Xplane is not new, it's been around for many years, and has never emerged from the shadows of the bigger player. Now's the chance, but the old strategies that kept you in "second" (maybe lower, FSX - FS9 - XP ? mobile markets have no bearing here) place need to change, or someone else will step in.

 

Look what happened to RIM (Blackberry), they didn't have the foresight to see where the markets / technology were going, what people wanted, what and who needed to be catered too, and look what happened. Now even Apple is feeling the pinch. The same will happen here.

 

Glen

...it sometimes comes to a point in ones life where they may have to admit they have bitten off more than they can chew.

 

At the risk of touching off a whole different kind of debate... there's a well-known phenomenon in business, and especially tech business, known as "founder-itis." Short version: the person who's brilliant enough to conceive the technology and start the enterprise often isn't the right person to guide it from pure start-up to the next level of business maturity. At a certain point the business challenges become different - you're no longer solely concerned with driving the technology, you have to start thinking about alliances, a growth plan that provides for partnership, creating business systems, and so on. It's often desirable to have someone with a pure business background come in to oversee that. The transition can happen in different ways, some of them pleasant (when the founder realizes his or her limitations and brings in a head of business development) and others unpleasant (when the investors "persuade" the founder to become "chief scientist" or "chief innovation officer" or something like that while they bring in their own CEO). It's often very difficult to persuade the founder to accept a more limited role. Some founders can't make the move and their business hits a growth wall. And of course there's always the exception - the Apple board brought in John Scully to supplant Steve Jobs, and we all know how that ended. But in the case of Austin and LR, a lot of the discussion here suggests that they might be arriving at some sort of inflection point where it would help to have a business strategist on board- someone who could create a software development plan designed to give 3PD's more confidence, set levels of investment that would provide for more resources, that sort of thing. I have absolutely no knowledge about whether this is a realistic take on LR or not - I'm just speculating based on what people say here and elsewhere.

 

It's worth noting also - and here I'll risk reviving an old firefight - that different founders have different limitations. JV seems to be a founder who's really good at driving the strategic growth plan, but who possibly shouldn't be the main public face of the company. Different situations call for different tweaks. The point is to make sure that the founders wind up in a position that lets them draw on their very real talents without inadvertently driving the business over a cliff.

 

Stuff to think about at any rate.