Jump to content
Sign in to follow this  
jmig

FPS and CPU/GPU loading in PD3 V4.1

Recommended Posts

They are the same, but there is a very subtle difference as I explained:

26 minutes ago, SteveW said:

Things can pan out slightly differently but not that it's going to make any difference. Since if the jobscheduler is looking at the next LP to start a process when we have HT enabled the next LP may be on the next core or just on the next LP of the same core.

 

 


Steve Waite: Engineer at codelegend.com

Share this post


Link to post

What you're asking is there any reason to do 11,11,11,01 rather than 11,11,11,10.

There's not, maybe.


Steve Waite: Engineer at codelegend.com

Share this post


Link to post
32 minutes ago, SteveW said:

Since if the jobscheduler is looking at the next LP to start a process when we have HT enabled the next LP may be on the next core or just on the next LP of the same core.

 

And so what? I know what - I feel totally miserable!! ))

Really need a practical example of what is said above. What is the subtle difference in that?

Does anybody besides Steve understand the above quote?

Many thanks!!

Edited by Dirk98

Share this post


Link to post

Depends on what's already running before and what's running after.

Often you will see someone say 11,11,11,01 worked better than 11,11,11,10, and then there's someone else that will say it worked better changing the other way around when in fact it should be more or less the same.

Why's that happening? Reason is because the jobscheduler picks slack places to fill with process and that might pan out slightly differently with the sim on 10 or 01. When the sim loads up the main process might be slacker than the other processes during loading a scenario and there's a weather engine loading up...

So if we don't bother starting addon exe apps off of the first core the sim encounters then those addons will most likely use up cores the sim lays itself out on so set it up solid and the uncertainty is reduced.

 

  • Upvote 1

Steve Waite: Engineer at codelegend.com

Share this post


Link to post
15 minutes ago, Dirk98 said:

And so what? I know what - I feel totally miserable!! ))

What you're asking is like; how a handful of beans will be arranged after dropping on the table. I'm saying: depending on the hardness of the materials and gravity, you'll get a varying spread of beans.

:laugh:

  • Upvote 1

Steve Waite: Engineer at codelegend.com

Share this post


Link to post

By 'slack' you mean less busy, don't you? 

Let's take 11 11 11 10 as an example. 

31 minutes ago, SteveW said:

When the sim loads up the main process might be slacker than the other processes during loading a scenario and there's a weather engine loading up...

What main process do you mean here? The one that some call "P3D's main thread"?

Anyway:

Case 1:

Let's take the main process is slacker than the other processes during loading a scenario and there's weather engine loading up. We have 11 11 11 10 allocation, so what happens?  

Case 2:

Let's take the main process is slacker than the other processes during loading a scenario and there's weather engine loading up. We have 11 11 11 01 allocation, so what happens?

I really need to dig it at last, once and for all.

Thank you!!

  

Share this post


Link to post

Not much - when you boil it down like that - but that's not realistic, those beans...


Steve Waite: Engineer at codelegend.com

Share this post


Link to post
Just now, SteveW said:

Not much - when you boil it down like that - but that's not realistic, those beans...

but you always suggest 10 rather than 01 (first core)

Please!! ))

Share this post


Link to post
22 minutes ago, SteveW said:

What you're asking is like; how a handful of beans will be arranged after dropping on the table. I'm saying: depending on the hardness of the materials and gravity, you'll get a varying spread of beans.

:laugh:

It is called the "The Uncertainty Principle" and it has driven generations of Physics researchers mad..

What Steve is saying is: "It depends..."  :cool:


Bert

Share this post


Link to post

The behaviour generally is to move to the next LP when the LP looks busy so as to find a slacker place to start a process. Depending on where you are the next LP might be one assigned to the sim on the same core or actually the LP of another core.

  • Upvote 1

Steve Waite: Engineer at codelegend.com

Share this post


Link to post
Just now, Bert Pieke said:

It is called the "The Uncertainty Principle" and it has driven generations of Physics researchers mad..

What Steve is saying is: "It depends..."  :cool:

I got that part very well. The Uncertainty Principle rules our universe.

But please why does he rather suggest 10 than 01??

Thanks,

Share this post


Link to post
Just now, SteveW said:

The behaviour generally is to move to the next LP when the LP looks busy so as to find a slacker place to start a process. Depending on where you are the next LP might be one assigned to the sim on the same core or actually the LP of another core.

Finally got it, thank you so much!!

Igor.

  • Upvote 1

Share this post


Link to post
4 minutes ago, Dirk98 said:

but you always suggest 10 rather than 01 (first core)

Please!! ))

That's so funny.

Which ever way I put it there's a merry band of followers - "SW says 01", "no SW says 10"

Hehe!


Steve Waite: Engineer at codelegend.com

Share this post


Link to post
4 minutes ago, Dirk98 said:

I got that part very well. The Uncertainty Principle rules our universe.

But please why does he rather suggest 10 than 01??

Thanks,

I do not know, but I have a suspicion that the Windows scheduler will put its housekeeping work on the first logical processor that is available (and it may just possibly start counting at 0).  So leaving the first logical processor free invites Windows to put it's work there..

Just a guess... like where do I put this mouse trap so that the mouse is going to eat the bait..


Bert

Share this post


Link to post

It treats all cores and LPs in a round robin way and core zero might not be the first considered by the algorithm. In other words each core is equal.


Steve Waite: Engineer at codelegend.com

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...