P3D V4.5 settings and Scenery

[Noel 4,487]

5 minutes ago, Mace said:

I have found the same (that 6 phys cores isn't enough) thru extensive testing of my rig.   If I were rebuilding now I would not hesitate to get at least an 8 physical core cpu.   Based on what I see in the hardware forum, I'm not sure the HT on or off is quite as important with the 8 physical core cpu's.  It's critically important for us 6-core guys as you know.

 

That's plausible for sure, that 8 will be sufficient.  

Edited January 8, 2020 by Noel

[DellyPilot 109]

On 1/8/2020 at 3:29 AM, Noel said:

That's plausible for sure, that 8 will be sufficient.  

I have 8 core and turning off HT made a big difference. P3D is heavily Core 0 dependent so HT = faster Core 0 = better FPS.

Turning off JobScheduler completely fixed all my blurries and made almost no difference to Frame rate, just dont run lots of other applications in the background.

Edited January 9, 2020 by DellyPilot

[mitsos 6]

I am using P3D v4.5.13 and i am still using this line ... just like the old fsx dayz, you know ? 😉
...and it still works, but my frames sometimes go from like, 50, down to 45 or 43 or even to 18, IF i have a
VERY HEAVY scenery like ... Amsterdam for example.
But NO blurries AT ALL !!! everything loads up correctly. 😊

FIBER_FRAME_TIME_FRACTION=0.48

[Bert Pieke 7,584]

7 minutes ago, mitsos said:

I am using P3D v4.5.13 and i am still using this line ... just like the old fsx dayz, you know ? 😉
...and it still works, but my frames sometimes go from like, 50, down to 45 or 43 or even to 18, IF i have a
VERY HEAVY scenery like ... Amsterdam for example.
But NO blurries AT ALL !!! everything loads up correctly. 😊

FIBER_FRAME_TIME_FRACTION=0.48

That is a very high setting... Most users would choose something between 0.10 and 0.20.

With this high setting, you are definitely prioritizing scenery loading over performance.

[Noel 4,487]

Wake me up when someone with a 9900K can comment on HT on versus off that is my only question.  

1 hour ago, Bert Pieke said:

That is a very high setting... Most users would choose something between 0.10 and 0.20.

With this high setting, you are definitely prioritizing scenery loading over performance.

I never have blurry anything and don't use any custom FFTF setting, but I use a 6 core w/ 12 LPs per HT enabled.

[SteveW 1,792]

Just for a laugh I'm running P3D on AM3=00,11 on the 18 core HT enabled. So that's only one core with HT sharing.

No blurries, same fps. Takes a lot longer to load up.

 

So it seems anyone concerned with the AM somehow responsible for this is wrong.

 

In any case there's no such things as no AM. A six core no HT no AM is AM63=111111. So I can switch HT off on the 18 core and use AM63 and I'm using six cores the same.

So does P3D work the same on a six core with six cores as it does on an 18 core with six core AM. Answer is yes, it works the same.

 

Edited January 10, 2020 by SteveW

[DellyPilot 109]

15 hours ago, Noel said:

Wake me up when someone with a 9900K can comment on HT on versus off that is my only question.  

I never have blurry anything and don't use any custom FFTF setting, but I use a 6 core w/ 12 LPs per HT enabled.

I already commented Noel

No HT gave me good FPS improvement

No Jobscheduler removed all blurries with FFTF 0.17

Edited January 10, 2020 by DellyPilot

[Noel 4,487]

44 minutes ago, DellyPilot said:

I already commented Noel

No HT gave me good FPS improvement

No Jobscheduler removed all blurries with FFTF 0.17

Were you using AM 65524 and no FFTF entry when you had HT on?  This is the comparison I'm after, with all other apps constrained to the core0 or at least off of core1.  The fact that turning off HT gave you an 'FPS improvement' suggests you weren't using AM 65524 but some other values that traded frame rate for terrain texture loading.   If you don't mind try that AM, unlimited frames in-sim, VSYNC to 30 if your screen can do it would be awesome.

Edited January 10, 2020 by Noel

[SteveW 1,792]

I tried AM 65524 = 11,11,11,11,11,11,01,00 which uses seven cores and 13 LPs. The rightmost use of the single LP is good. Loaded in 29.3s.

The excess of LPs to get the background task done pushes out the system resources.

I also tried AM5460 uses the middle six cores 00,01,01,01,01,01,01,00 and six LPs to do the same job. The CPU used less power = less heat. The time for the fps to stabilise across the first two minutes of the three minute tests was quicker by a couple of seconds. Loaded in 34.0s.

So I would install a good AM and stop worrying about it and worry about what counts, where the addons go.

Edited January 10, 2020 by SteveW

[Noel 4,487]

27 minutes ago, SteveW said:

I tried AM 65524 = 11,11,11,11,11,11,01,00 which uses seven cores and 13 LPs. The rightmost use of the single LP is good. Loaded in 29.3s.

The excess of LPs to get the background task done pushes out the system resources.

I also tried AM5460 uses the middle six cores 00,01,01,01,01,01,01,00 and six LPs to do the same job. The CPU used less power = less heat. The time for the fps to stabilise across the first two minutes of the three minute tests was quicker by a couple of seconds. Loaded in 34.0s.

So I would install a good AM and stop worrying about it and worry about what counts, where the addons go.

Sir, I'm not worrying about anything.  I'm planning my next CPU purchase and as I have repeatedly explained the six core version of this AM (4084) on my 3930K CLEARLY outperforms the HT disabled setting--and that is NOT ONLY with load time, it's with runtime as well.  I have no idea why, but when I disable HT I get a steady hiccough every short interval of time in certain scenarios which I never get w/ HT enabled, and with a superior load time.  I am trying to ascertain if a 9700K is going to be ideal, or it the 9900K will be the better choice. 

Edited January 10, 2020 by Noel

[SteveW 1,792]

Keep your shirt on! Only trying to help.

[Noel 4,487]

I'm sorry I see you did a similar test using AM5460 as a surrogate for HT disabled thank you for that.  The only comment missing is how the sim performed if you can set up for unlimited frames in-sim, vsync on, and monitor set at 30mHz refresh.  As I mentioned there is a clear regular stutter when I do this test on my 3930k in dense scenery so I'm assuming that is having to do w/ less than ideal loading of terrain texture but I'm not sure.  If 8 total cores is gets around this, i.e. is sufficient to manage terrain texture loading in the most complex of scenery (provided the 9900K can manage that scenery to match the vsync to 30), then one can skip out of the heat from HT enabled, and as well as the add'l cost of the 9900 over the 9700.

Edited January 10, 2020 by Noel

[Stephen Simpson 7]

All sorted. Clean cfg file. No stutters or blurries thanks to all who contributed to my issues.

On another topic. I dont have orbx installed anymore seeing as I am using JFNexGen UK but what would be a recommendation for Europe scenery instead of orbx.

Ta.

Steve S

[SteveW 1,792]

OK well done Steve.

@Noel, No worries I understood your question and concern. I hope someone can come forward with the similar specification for you. I was using a much bigger CPU and strong GPU setup. So instead my testing with those AMs was to carefully check to see how much actual difference the performance was on my test bed which is a very accurate recording. Which in those circumstances was barely any difference between them.

The differences we get in the sim with changing around the sim cores is very subtle and hardly impacts the actual way the sim performs alone. I'm saying the big differences when we change the sim AM, even just rearranging cores, must be down to how the remaining systems find themselves operating, and how they become arranged. Also as you mention the important factor when testing is that the locked/vsync/tb/unlimited situation has a wide variety of effects. Then we have the type of monitor, and the cable, hdmi, display port, all play a part in the final outcome. My point is that when we see changes to performance with only rearranging the AM that means there's somehow some other thing or things impacting the outcome. So at this stage it's better to accept one good AM and concentrate on the remaining setup. The AM and corralling of addons is a final step to obtain maybe up to 5 or 8% more stable flow of the sim. We can't add performance.

My advice is that the latest high end CPUs especially 8 cores and more all work very close in performance on P3D because they are drawing from similar base hardware now like the memory SSDs and GPUs. get the better GPU rather than CPU if money permits. Extra cores or LPs only really gain loading speed, which does effect the flow of the sim if we can load it quicker at any time, but that burst of loading pushes out the system resources at the same time. So in short I think you will find more or less the same performance is seen around most those PCs in the long run.

Edited January 11, 2020 by SteveW

[chapstick 2,757]

I’ve tried AM=340 with my 8700k/HT on and got the worst stutters I’ve ever seen. Just panning around was unbearable. What could that signify? What’s the next setting to test?