Jump to content
Sign in to follow this  
GSalden

P3Dv4.3 : my observations

Recommended Posts

Thanks Skywolf.

Is there a downside to SLI? Rob seems to have had some issues with it breaking each driver version - has this been your experience too?

When you say it goes bonkers if GPU usage goes above 90% I presume you mean FPS drop dramatically? I get the same behavior with my 1080 in heavy GPU environments, but from where I sit adding another card can only really increase the amount of data that can be pushed through the GPU(S) before you reach that level right?

 


Kael Oswald

7950X3D / 64GB DDR5 6000 @ CL30 / Custom Water Loop / RTX 4090 / 3 x 50" 4K LCD TVs

Share this post


Link to post
10 minutes ago, KL Oo said:

Thanks Skywolf.

Is there a downside to SLI? Rob seems to have had some issues with it breaking each driver version - has this been your experience too?

When you say it goes bonkers if GPU usage goes above 90% I presume you mean FPS drop dramatically? I get the same behavior with my 1080 in heavy GPU environments, but from where I sit adding another card can only really increase the amount of data that can be pushed through the GPU(S) before you reach that level right?

 

FPS drops happen when SLI link between the GPUs goes out of sync.  That is when chances increase for P3D crash.  At times I have even seen, SLI is broken and entire sim is running on one card.  Usually moments later P3D does crash.

I tested with single 980 and my FPS are way less on 4K.  SLI is the only way I am able to enjoy P3D4.3 with complex addons and sceneries.  It is no magic bullet, but I get the performance boost.

Amount of data is still limited to pci express bus and what speed are your lanes.  With my i7-5820K - I get 28 PCi Express lanes that means I can do 8x/8x in SLI mode.  If I get the next level cpu in the family which has 40 pci lanes - then I could do 16x/16x.  Single use card does run in 16x.

For SLI throughput 16x/16x should help.  8x/8x is good but it is also P3Dv4 graphics engine.  At this stage it has tons of legacy code in it.  Till we get new complete engine - I am hoping for v5...then everything is new again.  Lets see.

Just don't saturate the card to its full limits.

I am having amazing time with 4.3.  Amazing release though

Edited by Skywolf
  • Upvote 1

Active Pattern: MSFS2020 | In Long term Storage: Prepar3d  

How I Evaluate Third Party Sim Addon Developers

Refined P3Dv5.0 HF2 Settings Part1 (has MaddogX) and older thread Part 2 (has PMDG 747)

Share this post


Link to post

Thanks. I've never run SLI so am curious to try it... Might wait until the 1180 arrives and see how that performs first though.

  • Like 1

Kael Oswald

7950X3D / 64GB DDR5 6000 @ CL30 / Custom Water Loop / RTX 4090 / 3 x 50" 4K LCD TVs

Share this post


Link to post
6 minutes ago, KL Oo said:

Thanks. I've never run SLI so am curious to try it... Might wait until the 1180 arrives and see how that performs first though.

Yes...that is what I would do too first.

 


Active Pattern: MSFS2020 | In Long term Storage: Prepar3d  

How I Evaluate Third Party Sim Addon Developers

Refined P3Dv5.0 HF2 Settings Part1 (has MaddogX) and older thread Part 2 (has PMDG 747)

Share this post


Link to post
6 hours ago, KL Oo said:

Hi Gerard,

I understanding from this thread you're not running the virtual cockpit?

I'm very impressed by your SLI results but am interested to see if you get a similar improvement when using the VC. Are you able to post any results using the VC in SLI mode as opposed to VC without SLI?

Cheers,

 

Here I have a Full suze 737-800 home cockpit which is running on Prosim - P3Dv4 . Therefore no VC.

I have a 737 TQ that is only working with that combo. Therefore I can only test like I do now.

But sure you will benefit too : beside the VC you will also see the outside world and there you will see your gains. 

I am not sure but probably the cockpit shadows will benefit too from SLI.


13900 8 cores @ 5.5-5.8 GHz / 8 cores @ 4.3 GHz (hyperthreading on) - Asus ROG Strix Gaming D4 - GSkill Ripjaws 2x 16 Gb 4266 mhz @ 3200 mhz / cas 13 -  Inno3D RTX4090 X3 iCHILL 24 Gb - 1x SSD M2 2800/1800 2TB - 1x SSD M2 2800/1800 1Tb - Sata 600 SSD 500 Mb - Thermaltake Level 10 GT case - EKWB Extreme 240 liquid cooling set push/pull - 2x 55’ Sony 4K tv's as front view and right view.

13600  6 cores @ 5.1 GHz / 8 cores @ 4.0 GHz (hypterthreading on) - Asus ROG Strix Gaming D - GSkill Trident 4x Gb 3200 MHz cas 15 - Asus TUF RTX 4080 16 Gb  - 1x SSD M2 2800/1800 2TB - 2x  Sata 600 SSD 500 Mb - Corsair D4000 Airflow case - NXT Krajen Z63 AIO liquide cooling - 1x 65” Sony 4K tv as left view.

FOV : 190 degrees

My flightsim vids :  https://www.youtube.com/user/fswidesim/videos?shelf_id=0&sort=dd&view=0

 

Share this post


Link to post
6 hours ago, Skywolf said:

Yes...that is what I would do too first.

 

For myself I choose to buy a second 1080 Ti instead of waiting. 

A new 1180 /2080 will be faster but 30% faster in P3D is a lot. My experience is that it might be 10-15% max with each new , same type  of card ( 1080 —> 1180 )

The advantage with SLI is having to GPU’s working together ...

Surely an 1180/2080 SLI setup would be faster 🙃


13900 8 cores @ 5.5-5.8 GHz / 8 cores @ 4.3 GHz (hyperthreading on) - Asus ROG Strix Gaming D4 - GSkill Ripjaws 2x 16 Gb 4266 mhz @ 3200 mhz / cas 13 -  Inno3D RTX4090 X3 iCHILL 24 Gb - 1x SSD M2 2800/1800 2TB - 1x SSD M2 2800/1800 1Tb - Sata 600 SSD 500 Mb - Thermaltake Level 10 GT case - EKWB Extreme 240 liquid cooling set push/pull - 2x 55’ Sony 4K tv's as front view and right view.

13600  6 cores @ 5.1 GHz / 8 cores @ 4.0 GHz (hypterthreading on) - Asus ROG Strix Gaming D - GSkill Trident 4x Gb 3200 MHz cas 15 - Asus TUF RTX 4080 16 Gb  - 1x SSD M2 2800/1800 2TB - 2x  Sata 600 SSD 500 Mb - Corsair D4000 Airflow case - NXT Krajen Z63 AIO liquide cooling - 1x 65” Sony 4K tv as left view.

FOV : 190 degrees

My flightsim vids :  https://www.youtube.com/user/fswidesim/videos?shelf_id=0&sort=dd&view=0

 

Share this post


Link to post

The question now is : how will P3D behave regarding losing FPS over time when flying over heavy scenery...

I have the feeling that that is still happening. Only now with more headroom I might lock FPS just slighly higher instead of 30% higher.

Also I am wondering If locked vs Unlimited has an influence.

Anyone of you experiencing the FPS drop has ever tested that ?


13900 8 cores @ 5.5-5.8 GHz / 8 cores @ 4.3 GHz (hyperthreading on) - Asus ROG Strix Gaming D4 - GSkill Ripjaws 2x 16 Gb 4266 mhz @ 3200 mhz / cas 13 -  Inno3D RTX4090 X3 iCHILL 24 Gb - 1x SSD M2 2800/1800 2TB - 1x SSD M2 2800/1800 1Tb - Sata 600 SSD 500 Mb - Thermaltake Level 10 GT case - EKWB Extreme 240 liquid cooling set push/pull - 2x 55’ Sony 4K tv's as front view and right view.

13600  6 cores @ 5.1 GHz / 8 cores @ 4.0 GHz (hypterthreading on) - Asus ROG Strix Gaming D - GSkill Trident 4x Gb 3200 MHz cas 15 - Asus TUF RTX 4080 16 Gb  - 1x SSD M2 2800/1800 2TB - 2x  Sata 600 SSD 500 Mb - Corsair D4000 Airflow case - NXT Krajen Z63 AIO liquide cooling - 1x 65” Sony 4K tv as left view.

FOV : 190 degrees

My flightsim vids :  https://www.youtube.com/user/fswidesim/videos?shelf_id=0&sort=dd&view=0

 

Share this post


Link to post

i am not experiencing tissues but have tested frames locked vs unlocked,
since i run the test my frames remained unlocked!

old FSX tweak and tricks no longer apply in P3D,
if i lock my frames it somehow becomes worst than unlocked,

flying over the L.A. grid at night i will get around 19 frames with frames locked at 30,
with frames unlocked i am getting a much smother experience specifically over dense and harsh area vs locked,
the same test over the L.A. grid with unlocked frames yields a close to 30 frames,

1080Ti results, on the left with frames locked at 30, on the right Unlimited,
(credits: test results compliments of Mike @Cruachan)

Locked frame 30 Unlimited

 

Edited by Chris Bell

Share this post


Link to post

If I am correct Unlimited means a FFTF of 0.01 and Locked 0.30

With FFTF Dynamic the FFTF value can be changes so Locked vs Unlimited will have the same value to test with..

  • Like 1

13900 8 cores @ 5.5-5.8 GHz / 8 cores @ 4.3 GHz (hyperthreading on) - Asus ROG Strix Gaming D4 - GSkill Ripjaws 2x 16 Gb 4266 mhz @ 3200 mhz / cas 13 -  Inno3D RTX4090 X3 iCHILL 24 Gb - 1x SSD M2 2800/1800 2TB - 1x SSD M2 2800/1800 1Tb - Sata 600 SSD 500 Mb - Thermaltake Level 10 GT case - EKWB Extreme 240 liquid cooling set push/pull - 2x 55’ Sony 4K tv's as front view and right view.

13600  6 cores @ 5.1 GHz / 8 cores @ 4.0 GHz (hypterthreading on) - Asus ROG Strix Gaming D - GSkill Trident 4x Gb 3200 MHz cas 15 - Asus TUF RTX 4080 16 Gb  - 1x SSD M2 2800/1800 2TB - 2x  Sata 600 SSD 500 Mb - Corsair D4000 Airflow case - NXT Krajen Z63 AIO liquide cooling - 1x 65” Sony 4K tv as left view.

FOV : 190 degrees

My flightsim vids :  https://www.youtube.com/user/fswidesim/videos?shelf_id=0&sort=dd&view=0

 

Share this post


Link to post

What FFTF value ?

  • Like 1

13900 8 cores @ 5.5-5.8 GHz / 8 cores @ 4.3 GHz (hyperthreading on) - Asus ROG Strix Gaming D4 - GSkill Ripjaws 2x 16 Gb 4266 mhz @ 3200 mhz / cas 13 -  Inno3D RTX4090 X3 iCHILL 24 Gb - 1x SSD M2 2800/1800 2TB - 1x SSD M2 2800/1800 1Tb - Sata 600 SSD 500 Mb - Thermaltake Level 10 GT case - EKWB Extreme 240 liquid cooling set push/pull - 2x 55’ Sony 4K tv's as front view and right view.

13600  6 cores @ 5.1 GHz / 8 cores @ 4.0 GHz (hypterthreading on) - Asus ROG Strix Gaming D - GSkill Trident 4x Gb 3200 MHz cas 15 - Asus TUF RTX 4080 16 Gb  - 1x SSD M2 2800/1800 2TB - 2x  Sata 600 SSD 500 Mb - Corsair D4000 Airflow case - NXT Krajen Z63 AIO liquide cooling - 1x 65” Sony 4K tv as left view.

FOV : 190 degrees

My flightsim vids :  https://www.youtube.com/user/fswidesim/videos?shelf_id=0&sort=dd&view=0

 

Share this post


Link to post

FSPS FFTF is Dynamic AFAIK :rolleyes:
these are not my own TI 
results; i believe no custom FFTF used in first pair presented,

3 hours ago, Chris Bell said:

(credits: test results compliments of Mike @Cruachan)

Edited by Chris Bell

Share this post


Link to post
37 minutes ago, Chris Bell said:

FSPS FFTF is Dynamic AFAIK :rolleyes:
these are not my own TI 
results; i believe no custom FFTF used in first pair presented,

You can use fixed FFTF values too .

Eg : ground = 0.01  /  2500 ft = 0.30 

So flying at 2500 ft at 230 KIAS is 0.30. Using Unlimited and Locked can be tested this way..

  • Like 1

13900 8 cores @ 5.5-5.8 GHz / 8 cores @ 4.3 GHz (hyperthreading on) - Asus ROG Strix Gaming D4 - GSkill Ripjaws 2x 16 Gb 4266 mhz @ 3200 mhz / cas 13 -  Inno3D RTX4090 X3 iCHILL 24 Gb - 1x SSD M2 2800/1800 2TB - 1x SSD M2 2800/1800 1Tb - Sata 600 SSD 500 Mb - Thermaltake Level 10 GT case - EKWB Extreme 240 liquid cooling set push/pull - 2x 55’ Sony 4K tv's as front view and right view.

13600  6 cores @ 5.1 GHz / 8 cores @ 4.0 GHz (hypterthreading on) - Asus ROG Strix Gaming D - GSkill Trident 4x Gb 3200 MHz cas 15 - Asus TUF RTX 4080 16 Gb  - 1x SSD M2 2800/1800 2TB - 2x  Sata 600 SSD 500 Mb - Corsair D4000 Airflow case - NXT Krajen Z63 AIO liquide cooling - 1x 65” Sony 4K tv as left view.

FOV : 190 degrees

My flightsim vids :  https://www.youtube.com/user/fswidesim/videos?shelf_id=0&sort=dd&view=0

 

Share this post


Link to post
Guest

Dynamic FFTF?

changing FFTF will lead to blurries.

unlimited is not FFTF = 0.01, that code is completely bypassed when unlimited that’s why Vsync lock is needed

Chris/Gerard, you gents have been around a long time and I know you know FFTF changes can result in blurries.

Cheers, Rob

 

Share this post


Link to post

indeed Rob (bit surprised here too!),
i have not suggested any changes to FFTF,
i believe Gerard's inquiry is just for bench marking purposes?

Edited by Chris Bell

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...