P3Ds future: I prefer facts - the more, the merrier!

[rocketlaunch 614]

3 hours ago, d.tsakiris said:

Personally, I prefer LM's approach.

Well it worked for a while but the ESP engine is really beginning to show it's age.

[Ray Proudfoot 8,961]

37 minutes ago, rocketlaunch said:

Well it worked for a while but the ESP engine is really beginning to show it's age.

32-bit DX9 7 years ago and now 64-bit DX12. Please explain what aspect is showing its age.

Later: as an example I have to scale back scenery density at EGLL in P3Dv3 to prevent VAS going above 3.8Gb. Little of my i7-8086K CPU is used. A GPU with 11Gb VRAM is total overkill.

Now, in v5, airports like LatinVFR Miami can look great with up to 10Gb of VRAM used. Specific P3D core actions can be assigned to individual CPU logical processors.

LM has made huge strides in performance and compatibility.

Edited January 2, 2023 by Ray Proudfoot
Extra info

[d.tsakiris 418]

40 minutes ago, rocketlaunch said:

Well it worked for a while but the ESP engine is really beginning to show it's age.

I concur with Ray. I believe so much has been rewritten (auto-gen, lighting, clouds, rendering pipeline, ...), it's not appropriate to say it's the old ESP engine anymore.

[rocketlaunch 614]

13 minutes ago, Ray Proudfoot said:

32-bit DX9 7 years ago and now 64-bit DX12. Please explain what aspect is showing its age.

Ray, I know you are one of the biggest champions of P3D but really? A vanilla install of P3D looks hideously outdated compared to modern graphic engines.  

[rocketlaunch 614]

On 12/16/2021 at 11:38 AM, Ray Proudfoot said:

I’ve just read his post and this was my reply...

Gary, your post needs to be read by LM. It should serve as a wake up call that tinkering with the ESP engine cannot be justified any longer.

They need to up their game and use modern software methods to make scenery designing as easy as it now appears to be in MSFS.

You have given me a compelling reason to consider switching to MSFS at a future date once it’s a more stable product. Clearly it is for airport design but has some way to go before the same can be said for aircraft.

I wish you well in both your personal and professional life and hopefully one day in the not too distant future I can buy your products for MSFS.

It’s a real eye-opener highlighting how difficult it is for him to use an ageing engine to design scenery. The ESP engine has lasted well but maybe it’s now time to draw a line under it and switch to a modern system. We all know airports from some developers can look great in P3D but at a significant performance hit.

Maybe this will be the kick up the backside that will convince LM to consider how P3D development should go.

 

 

[Ray Proudfoot 8,961]

1 minute ago, rocketlaunch said:

Ray, I know you are one of the biggest champions of P3D but really? A vanilla install of P3D looks hideously outdated compared to modern graphic engines.  

That’s a subjective opinion you’re perfectly entitled to. I, on the other hand, am very happy with the default scenery plus ORBX FTX Global. I also use FS Global to enhance mountains. That’s it. Far less required than in earlier versions.

There is far more to a sim than looks. The completeness of the SDK for one allowing third parties to create applications other sims are incapable of. That’s why there’s a free 3rd party utility that taps into the Active Sky weather SDK giving the user a weather radar that can be used with any aircraft.

I don’t want this topic to descend into another “my sim is better than your sim” argument as it will be locked.

I don’t believe any sim can run without third party addons. To believe so is fantasy. All require enhancing hence why there is an active third party market.

Do you still fly using P3D?

[d.tsakiris 418]

15 minutes ago, rocketlaunch said:

 

That, I believe, was about the UK2000 developer (Gary) stopping development for P3D, because he liked other provided tools better. Some then said he was using old tools, and I can't really comment on that, other than that I have a few of his add-ons, and they do seem to be missing modern technologies.

Edited January 2, 2023 by d.tsakiris

[Ray Proudfoot 8,961]

@rocketlaunch, in response to your post apparently quoting something I said a year ago if all 3rd party airports were the same as UK2000 then P3D would lack airport quality.

The fact remains that there are plenty of higher quality airports - FlyTampa, LatinVFR et al that produce higher quality. Perhaps it was the software Gary used that was lacking in advanced features. Or more likely, he struck a balance between complexity and price. His were some of the cheaper airports available.

Still waiting for your answer. Do you fly P3D?

[DChockey08 299]

P3D will continue to live long on my hard drive for all the reasons already described in this thread. The biggest reason being: there is no alternative (in my opinion). 
 

I was firmly in the camp of “wait-and-see” for MSFS (I was a day-one purchaser and use it occasionally for VFR) and figured (like FSX when I was on FS9) that it was a matter of time before I switched over: “give it a couple years when PMDG and FSLABS and ActiveSky are on there and I’ll be over”

 

The problem is now 2.5 years after release, not only are not all the problems fixed, but it’s looking like they won’t be. I won’t go through them because they are personal to me (historical weather with AS, AI traffic, lack of seasons, etc). 
 

So the future for me really is looking more and more like incremental improvements to P3D. I would love an improvement over the land class+autogen system we have. I would love an improvement of the TrueSky engine to give us localized weather. 

[simfan1983 185]

One of the issues with P3D is the lack of error reporting (unless I am missing something). When it CTDs without any indication of the issue, or won't get beyond the splash screen, it would be nice to have some kind of detailed log when it encounters an issue. And sorting out DXGI Device Hung once and for all would also be a nice addition! It's probably too late now to attract back a lot of non-commercial customers, unless of course there is a major step improvement in EA and vol clouds and default scenery and MS either plug the plug or start charging properly for their streaming costs.

[ywg256 75]

who knows maybe Aerosoft and the rest of them only meant they would stop up to V5.3 😄

At this point until we see v6 it is really hard to say what the future holds. It could be as easy as an SDK revamped to the point that it makes developers lives easier. Hopefully we find out in the next month or two.

 

[GCBraun 2,392]

For me, a new P3D version would only make sense if it is based on a new engine, and not only because of the outdated looks. I still have nightmares with all those legacy DLLs that have caused me immense pain over the years. ESP has run its course imho…

[neilhewitt 494]

20 minutes ago, ywg256 said:

At this point until we see v6 it is really hard to say what the future holds. It could be as easy as an SDK revamped to the point that it makes developers lives easier.

There are two main factors that I can see driving third parties away from P3D: the cost of making add-ons, and the potential market size. If, for example, it were possible to build scenery or an aircraft and deploy the same, or substantially the same, code to both MSFS and P3D (which is not totally far-fetched in theory given that they are both ultimately based on FSX/ESP) then it would reduce the development cost for P3D, and sales for P3D would be desirable additional revenue, provided the support costs are in proportion to those of MSFS per-sale on average. But it's not possible because the two have diverged too far. If v6 were to make P3D compatible with MSFS scenery, or at least make it easy for developers to build for both at the same time, that could potentially turn the tap back on for airports etc. I could see UK2000, for example, opting to make their new MSFS stuff P3Dv6-compatible if it was easy and enjoyable for Gary to do so without needing to operate two sets of tooling and doing two lots of testing. 

The potential return per unit of investment on a product needs to be high enough to justify the investment. That depends on how much it costs to build and support it vs the number of sales you make. Because the audience is just much larger on MSFS (10 million players vs maybe hundreds or even tens of thousands for P3D), the potential return is much greater and so given limited funds to invest and limited people to do the work, companies are voting with their wallets. I don't believe you'll see this trend reverse unless it becomes possible to develop products for both sims at the same time with one set of tooling. There is simply no conceivable shape for a v6 which will increase the sim's user numbers to even 10% of what MSFS has right now, and make it commercially worthwhile to develop for P3D as a completely separate platform. 

It's not that P3D is dead. It clearly isn't. It's that P3D is more expensive to develop for. Mass-market beats niche every time unless you can charge much more for niche.

[UrgentSiesta 930]

11 hours ago, d.tsakiris said:

I also am convinced v6 is underway, and, personally, I am hoping for comprehensive backward-compatibility, so I can use the thousands of add-ons available out there.

 

Quote

Orbx hasn't canceled OpenLC Asia, but development is on hold, as the developer is from Ukraine.

Product was scheduled to be released over a year BEFORE the invasion started...

 

Quote

I don't know about “scores“...

Again, it goes back to ones definition of "abandoned", but all you have to do is go look for new v5 scenery vs the same devs/production houses output for MSFS. And, of course, it's further compounded by the decided lack of NEW devs for P3D scenery...

And it's not a knock on P3D, same thing is happening for X-Plane, just to a lesser degree.

[UrgentSiesta 930]

8 hours ago, Ray Proudfoot said:

Respectfully, I disagree. They do consider us by maintaining backward compatibility. If their commercial customers just used the default simulator LM could change it with every new version. But they don’t. I can only guess they respect the many loyal customers who have stayed with them.

Yes, fewer than a few years ago for obvious reasons but there’s still a significant number of serious pilots who like stability and continuity. The Navigraph survey is not the whole picture.

I respectfully disagree with your disagreement 🙂

The fact is that LM themselves remain highly dependent on addons developed by the likes of MilViz (yeah, for .Mil, that's still their name 🙂). But at the same time, the USAF (at least) is clearly concerned with visual fidelity, evidence the T-38C, which was specifically dev'd on behalf of the USAF.

And once you get into the types of large scale, highly networked scenarios they're running, compatibility across all nodes is paramount.

The video I linked has a few tidbits about the benefits of this strategy (e.g., trainee ATC working with trainee NFOs over the network, etc.), and a few others.

So in this regard, the push for backwards compatibility vs new visual fidelity features is just as strong in the .Mil / .Com space as it is for consumers.