Jump to content
Sign in to follow this  
737ngx99

Awful AA in P3D v3

Recommended Posts

Hi Steve,

 

could you give an example airport or post a screenie? I always fly with HDR enable and had never have problems in ORBX country or other add ons -freeware nor payware.

 

 

 

Sure, this is an example at an ORBX airport. If I were to make it into a .jpg the dots would be removed due to compression. You will have to open with paint as a browser will display it in lower quality. 

 

PIC

 

This is with 8x AA and 16x AF - No external things were used like nvidia inspector.

 

Note: it's not fixed white spots. The objects shimmer all along the surfaces and it becomes way worse when the scene is in motion. The screenshot also degrades the quality.

 

I've heard it called many things in many threads over the years. "fireflies" ,"camera flashes", "sparkles" etc.,  another guy though it was from the headlights of the cars , but it's not.

 

A video showing this is in order so you can see the magnitude of it, I'll try to make one and see if youtube's compression filters it out as an artifact.

Share this post


Link to post

ORBX airports drive me nuts. I love them, but yes they sparkle like nobody's business on my setup too. I always wondered if it was a mip mapping thing.


Chris

Share this post


Link to post

I'm not sure I completely understand your explanation of how FSX and Prepard3d ver 3.1 differ, if indeed they do, on how they work with Hyper-Treading activated.  According to Nick N who seems to be an acknowledged Guru on FSX he states in his FSX Bible that: "The real terrain 'guru' Adam from Aces as well as Phil Taylor specified years ago that FSX does not support any logical core threading (hyper thread).  The reason you SEE logical core activity in the Windows CPU monitor window is because the physical cache is used for each logical (hyper thread) core.  Logical core means there is no 'physical' hardware core but the hardware cache in use is there and FSX is DUMB so a thread is spawned due to the physical cache in use, but the data is never processed by FSX!".

 

In my experimenting with my dedicated flight simulator computer utilizing a Haswell i7 4790 processor with hyper threading on and off I never saw any indication of FSX improvement with hyper treading on and it fact it appeared to be detrimental. 

 

What prompted my question is at the Gatwick website dealing with how to calculate the proper Affinity Mask depending on what cores to use, there is a statement that says "Please be aware that P3D V2.x & V3.x by default will use all Cores and most importantly all Threads also on CPU's that have Hyper Threading enabled.  On machines with four or more cores, it will dedicate logical core 0 and thread 0 to rendering tasks.  To this end, it is suggested that the Expert Mode calculator be used to tune your P3D V2.x or V3.x Affinity Mask".  That statement suggested to me that Nick's claim that FSX cannot effectively use hyper threading is accurate and more importantly that P3D V2.x or V3.x can use it.  Since I've just recently installed P3D V3.1 on it's own dedicated SSD drive on my flight simulation computer this statement implies that even if I choose not to run an Affinity Mask I may want to activate hyper threading when I run my P3D simulator.  Thus the reason for my initial question, is the statement that later versions of P3D do in fact effectively use the additional logical cores created by activating hyper threading accurate regardless of whether one chooses to use an Affinity Mask?

 

With regard to the use of an Affinity Mask it would seem to me the only likely reason one would need to use one is if they have too many other programs or services running that are interfering with FSX or P3D operating effectively.  Giving up one or more cores to either FSX or P3D to use seems counter productive to me.  The better solution might be to use Task Manager to assign a program (TrackIR for example) to using just the last available core rather than taking cores away from the use of FSX or P3D.  I also find the use of Alacrity a very helpful initial program to use to shut down unnecessary programs and services prior to running FSX or P3D.

Share this post


Link to post

I think you'll find that's just all old stuff nowadays. Here's what 4 v 5 cores HT on and off does for P3D. HT on 4 cores showing up better than 5 cores, and HT on better than HT disabled, P3D, 6 core 4GHz.Just need to know how to test it, and how it works:

P3D4v5core.jpg



Here's 4 v 5 jobs. Also note 116 outperforming 212, then a fifth core introduced at the bottom:

4v5jobs.jpg

Steve Waite: Engineer at codelegend.com

Share this post


Link to post

seconds. Understanding why 116 would outperform 212 should bring a measure of enlightenment for those interested.


Steve Waite: Engineer at codelegend.com

Share this post


Link to post

I'm not seeing dramatic differences between the samples you posted.  There are some subtle differences in the average FPS over time.  It appears that the 116 example doesn't have its average FPS dropping quite as much as the 212 example but the difference to my eye doesn't seem dramatic.  Can I assume these are being tested on P3D and if so what version and are you using the same flight or something else?

Share this post


Link to post

We are fast reaching the point when attempting to compare FSX with P3D as to it's CPU usage, thread optimizations etc is an exercise in futility.

 

Folks - FSX is NOT P3D.  FSX is NOT ESP.  There are similarities but LM has made enough significant code changes to make comparison as to function useless.

 

Nick's references to FSX are correct - FOR FSX - but even he has acknowledged there have been significant changes in P3D, enough to make them quite different.

 

The comment "well, it works in FSX" is completely useless - about equal to saying "it works in Xplane" - Apples and oranges.

 

Vic

  • Upvote 2

 

RIG#1 - 7700K 5.0g ROG X270F 3600 15-15-15 - EVGA RTX 3090 1000W PSU 1- 850G EVO SSD, 2-256G OCZ SSD, 1TB,HAF942-H100 Water W1064Pro
40" 4K Monitor 3840x2160 - AS16, ASCA, GEP3D, UTX, Toposim, ORBX Regions, TrackIR
RIG#2 - 3770K 4.7g Asus Z77 1600 7-8-7 GTX1080ti DH14 850W 2-1TB WD HDD,1tb VRap, Armor+ W10 Pro 2 - HannsG 28" Monitors
 

Share this post


Link to post

Can I assume these are being tested on P3D and if so what version and are you using the same flight or something else?

That's the result from professional research carried out with a sophisticated test harness.

  • Upvote 1

Steve Waite: Engineer at codelegend.com

Share this post


Link to post

We are fast reaching the point when attempting to compare FSX with P3D as to it's CPU usage, thread optimizations etc is an exercise in futility.

 

Folks - FSX is NOT P3D.  FSX is NOT ESP.  There are similarities but LM has made enough significant code changes to make comparison as to function useless.

 

Nick's references to FSX are correct - FOR FSX - but even he has acknowledged there have been significant changes in P3D, enough to make them quite different.

 

The comment "well, it works in FSX" is completely useless - about equal to saying "it works in Xplane" - Apples and oranges.

 

Vic

 

 

Well .... It has taken a while - but I hope that the umbilical has finally and irrevocably been severed. Thank you Vic. A long overdue statement.  :Applause:

  • Upvote 1

Share this post


Link to post

Regarding the AA, generally setting 4xMSAA in the P3D display settings is good with aircraft such as the B58. But other aircraft models like the A36 may not look properly anti-aliased with only this setting, especially when zooming out the view. In this case the infamous SGSSAA with NI can be added (in the same multiple as MSAA) to good effect but practically halving the fps on some GPUs.

Problem I'm referring to here is to do with the scale of pixel sizes in the transparency. Adding mipmaps doesn't help because pixels are still the same size irrespective of how many mipmaps are provided.


Steve Waite: Engineer at codelegend.com

Share this post


Link to post

As someone who has had FSX for several years and only a couple of weeks ago added P3D v. 3.1 into my dedicated flight simulation computer it became immediately obvious that there are significant differences or improvements in this particular version of P3D versus FSX. A brief comparison of the two configuration files, while similar, also have some striking differences with many settings in P3D that don't exist in FSX. After my initial flight in P3D the most obvious visual difference on my system was the inherent better smoothness of P3D. I don't pretend to be an expert in why this is the case but I suspect it's the addition of Hardware Tessellation if you have a graphics card that can support it. It also became evident fairly quickly that the settings that many of us used in NV Inspector were not needed or relevant. We no longer needed to have Anisotropic filtering set in Inspector as P3D v.3.1 has multiple settings available within the program. AA control is also different inasmuch as P3D has various multisampling settings available that do not exist in FSX. There's also HDR lighting, shadow reflection and distance controls, and mipmapping of VC cockpit panels that are not in FSX. There is also a P3D pdf file entitled Settings that contains a section entitled Advanced Settings that indicates a number of P3D settings in the configuration file that can be altered that are not in FSX.

It is, however, worthy to note that two popular "tweaks" that many use in FSX (LOD_RADIUS and FIBER_FRAME_TIME_FRACTION) are still relevant in P3D. Many use 6.5 for the LOD_RADIUS setting in FSX which had to be done by changing the default value from 4.5 in the FSX config file. That is no longer required as P3D has changed the Level of Detail Radius slider in the Scenery setting tab to take the setting to 6.5 if set to Max, 5.5 if set to Ultra, etc.. FIBER_FRAME_TIME_FRACTION which doesn't exist in either the FSX or P3D configuration files can also be added to both. By default this setting is 0.33 but for upper end computer systems it can be lowered which will improve frame rates. Typically many use 0.15 as starting setting. At some point a low setting will start to show the "blurries".

I'm much less concerned with the similarities between P3D and FSX as I am with what's different, new, and improved in P3D V3.1. Thus the question regarding hyper threading and it's relevance to P3D. I'm also curious if P3D has improved the default mesh and landclass that existed in FSX among other things. In my desire to increase my knowledge of the inner workings of P3D V3.1 and what various settings can do for increased quality and/or performance I have found many conflicting suggestions and recommendations which make it harder to determine what's correct and what's not.

Share this post


Link to post

 I'm also curious if P3D has improved the default mesh and landclass that existed in FSX among other things. In my desire to increase my knowledge of the inner workings of P3D V3.1 and what various settings can do for increased quality and/or performance I have found many conflicting suggestions and recommendations which make it harder to determine what's correct and what's not.

 

I dont think anything has changed from fsx in regards to mesh and landclass hence we still have to use Our payware mesh and landclass placement  in P3D


Rich Sennett

               

Share this post


Link to post

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  
  • Tom Allensworth,
    Founder of AVSIM Online


  • Flight Simulation's Premier Resource!

    AVSIM is a free service to the flight simulation community. AVSIM is staffed completely by volunteers and all funds donated to AVSIM go directly back to supporting the community. Your donation here helps to pay our bandwidth costs, emergency funding, and other general costs that crop up from time to time. Thank you for your support!

    Click here for more information and to see all donations year to date.
×
×
  • Create New...