How did you come to this conclusion?  The AutoGen system is considerably better in P3Dv2.

 

Did you see Word Not Allowed's comparisons? The autogen system is horrible compared to FSX.

Did you see Word Not Allowed's comparisons? The autogen system is horrible compared to FSX.

 

This is an entirely subjective statement - can you qualify how it's "terrible"? ............ From my use of both sims :-

 

• I could not have anything more than "normal" autogen in FSX without taking a big FPS hit.  In P2D I run "Extremely Dense" with very little impact over lower settings.
• In FSX autogen 'popped up' all over the place - in P3D2 it doesn't.
• In FSX autogen was concentrated towards the closer field of view - in P3D2 it appears to scale further back towards the horizon (on like for like settings).

So how is the AG in P3D2 "terrible" compared to that in FSX ?

Did you see Word Not Allowed's comparisons? The autogen system is horrible compared to FSX.

I've seen my own comparison. FSX autogen pops up in clumps right in front of the plane or right off the wingtip, constantly distracting me and knocking me out of "the moment."

 

P3D autogen loads way off in the distance where I rarely ever notice it.

 

Winner: P3D.

This is an entirely subjective statement - can you qualify how it's "terrible"?

 

I can't call it improved over FSX when the performance hit caused by higher settings is much, much worse.

I can't call it improved over FSX when the performance hit caused by higher settings is much, much worse.

 

The performance hit isn't much, much worse depending on how you use it.

 

Many people, including me, are seeing higher performance with virtually no popup in comparable normal use levels vs. FSX. 

 

Is upper level performance worse i.e. 48 FPS vs. 100 FPS in isolated tests with nothing else going on his specific setup? For him it seems to be. But in heavy load situations (i.e. actual use situations) it isn't. I can use max autogen with my normal FSX settings and it's better in P3D. 

 

The other thing to note is that autogen draw calls are done by the GPU now in P3D. He's running a GTX580 with only 1.5gigs of VRAM. So these results are expected when you isolate autogen. His CPU power outpaces his GPU power in his setup. That's better for FSX and how it runs for this specific test.

 

Running max autogen with only 1.5gigs of VRAM is going to bottleneck the GPU and kill performance in P3D. 

 

So is FSX's system better? Sure, if you run nothing else to bottleneck the CPU in the process, but who does that? Practically no one. Most are killing their CPUs in FSX in actual normal use situations with heavy addons, AI traffic, weather, etc. 

 

 

The performance hit isn't much, much worse depending on how you use it.

 

Many people, including me, are seeing higher performance with virtually no popup in comparable normal use levels vs. FSX. 

 

Is upper level performance worse i.e. 48 FPS vs. 100 FPS in isolated tests with nothing else going on for benchmark purposes on his specific setup? Seems to be. But in heavy load situations (i.e. actual use situations) it isn't. I can use max autogen with my normal FSX settings and it's better in P3D. 

 

The other thing to note is that autogen draw calls are done by the GPU now in P3D. He's running a GTX580 with only 1.5gigs of VRAM. So these results are expected when you isolate autogen. His CPU power outpaces his GPU power in his setup. That's better for FSX and how it runs.

 

Running max autogen with only 1.5gigs of VRAM is going to bottleneck the GPU and kill performance in P3D. 

 

So is FSX's system better? Sure, if you run nothing else to bottleneck the CPU in the process, but who does that? Practically no one. Most are killing their CPUs in FSX in actual normal use situations.

 

Autogen in Prepar3D 2.0 still runs on the CPU. Have you checked his previous tests?

Autogen in Prepar3D 2.0 still runs on the CPU. Have you checked his previous tests?

 

Autogen draw calls and instancing are done on the GPU now. This has been stated by LM numerous times.

 

You also can't assume a linear scale, which is what I'm trying to get across. In his isolated tests, are upper level FPS rates different? Yes. What about when the load is increased overall?

 

What's the difference in an actual real world situation with some AI traffic, a few heavy addons, and weather? Does FSX still handle autogen better? No, it doesn't. You get massive popups, it struggles to load consistently, and performance suffers greatly across different autogen levels.

Autogen draw calls are done on the GPU now. This has been stated by LM numerous times.

 

Test results show otherwise.

It seems that your findings (or is it Kostas, that you're conveying?) are very inconsitant with what the likes of myself, Bonchie and others are seeing.   As I said above, that you seemed to miss,  I can run much higher AG settings with P3D2 than I can with FSX.  

 

Pehaps it's an ATI thing!

Come on Dave, surely the opinion of someone reading the opinion of someone else entirely, is more valid than our actual experience over the past few days.

 

Right ?

It seems that your findings (or is it Kostas, that you're conveying?) are very inconsitant with what the likes of myself, Bonchie and others are seeing.   As I said above, that you seemed to miss,  I can run much higher AG settings with P3D2 than I can with FSX.  

 

Pehaps it's an ATI thing!

 

Firstly, I'm talking about Word Not Allowed's findings. Secondly, you might be right that it could be an AMD thing (since NVIDIA drivers are incompatible so far), but since autogen runs on the CPU, I can't see how this would make a difference.

 

If only somebody else did some testing as thorough as Word Not Allowed's. That would clear up a lot.

Kostas article is a good one with actual comparison shots between FSX and P3d showing the density difference

 

"Yet, the FPS are 1/3!!! So we are paying with 1/3 of the FPS in P3Dv2, just to have a little bit more of autogen (just compare the shots!) and have it not pop-up? Sorry, too high a price for me"

Kostas article is a good one with actual comparison shots between FSX and P3d showing the density difference

 

And he's the only one who has done such thorough testing with proof (screenshots). Of course, I'm more inclined to believe him right now.

To what end ? 

 

I'm not interested in proving Word Not Allowed wrong/right/otherwise, it would honestly, for me anyway, be a waste of time that could be better spent enjoying the features that I am apparently imagining exist.

 

From the learning center:

 

Performance Improvements
Object Instancing

• Instancing allows the rendering system to have graphics cards draw multiple copies of the same entity just once. This greatly improves autogen building and tree performance. 

 

Regardless of what Word Not Allowed or anyone else has to say, this bears out for me in my experience with P3D2.

 

I can't even imagine the FPS hit the NY area would take, in FSX, if we attempted to draw as many objects as are drawn in P3D2 at full right sliders. I literally LOLd when I saw it the first time myself. 

 

You act like none of use have any experience with previous versions of the sim. As if none of us have a frame of reference. I most certainly do, and I have no agenda to promote anything either way. P3D2 is much, much better, period. Take it, leave it, believe whatever/whomever you want.

Regardless of what Word Not Allowed or anyone else has to say, this bears out for me in my experience with P3D2.

 

I can't even imagine the FPS hit the NY area would take, in FSX, if we attempted to draw as many objects as are drawn in P3D2 at full right sliders. I literally LOLd when I saw it the first time myself. 

 

You act like none of use have any experience with previous versions of the sim. As if none of us have a frame of reference. I most certainly do, and I have no agenda to promote anything either way. P3D2 is much, much better, period. Take it, leave it, believe whatever/whomever you want.

 

Do you have proof to back up your statements?