@Noel Yes this is pretty much it and FS2020 seems to be behaving the same.

Compare this to Intel TBB (an example of a Work Task Scheduler):

https://software.intel.com/content/www/us/en/develop/documentation/onetbb-documentation/top/onetbb-benefits.html

oneTBB enables you to specify logical paralleism instead of threads

Most threading packages require you to specify threads. Programming directly in terms of threads can be tedious and lead to inefficient programs, because threads are low-level, heavy constructs that are close to the hardware. Direct programming with threads forces you to efficiently map logical tasks onto threads. In contrast, the oneTBB run-time library automatically maps logical parallelism onto threads in a way that makes efficient use of processor resources.

oneTBB emphasizes scalable, data parallel programming

Breaking a program up into separate functional blocks, and assigning a separate thread to each block is a solution that typically does not scale well since typically the number of functional blocks is fixed. In contrast, oneTBB emphasizes data-parallel programming, enabling multiple threads to work on different parts of a collection. Data-parallel programming scales well to larger numbers of processors by dividing the collection into smaller pieces. With data-parallel programming, program performance increases as you add processors.

 

PS: XP11 is not using Intel TBB though, or not that I know of, they have implemented their own AFAIK.

[edit] I've tried finding some illustrations to explain this but it seems it is harder than I expected! Here is one showing concurrent tasks for example:

Edited May 14, 20215 yr by RXP

6 minutes ago, RXP said:

@Noel Yes this is pretty much it and FS2020 seems to be behaving the same.

Compare this to Intel TBB (an example of a Work Task Scheduler):

https://software.intel.com/content/www/us/en/develop/documentation/onetbb-documentation/top/onetbb-benefits.html

oneTBB enables you to specify logical paralleism instead of threads

Most threading packages require you to specify threads. Programming directly in terms of threads can be tedious and lead to inefficient programs, because threads are low-level, heavy constructs that are close to the hardware. Direct programming with threads forces you to efficiently map logical tasks onto threads. In contrast, the oneTBB run-time library automatically maps logical parallelism onto threads in a way that makes efficient use of processor resources.

oneTBB emphasizes scalable, data parallel programming

Breaking a program up into separate functional blocks, and assigning a separate thread to each block is a solution that typically does not scale well since typically the number of functional blocks is fixed. In contrast, oneTBB emphasizes data-parallel programming, enabling multiple threads to work on different parts of a collection. Data-parallel programming scales well to larger numbers of processors by dividing the collection into smaller pieces. With data-parallel programming, program performance increases as you add processors.

 

PS: XP11 is not using Intel TBB though, or not that I know of, they have implemented their own AFAIK.

Sorry, could you please repeat that in English?

 

34 minutes ago, RXP said:

@Noel Yes this is pretty much it and FS2020 seems to be behaving the same.

I guess, but streaming data seems different in that I'm not seeing quite the hits on the non-main threads I always see in busy areas in P3D.  But as I say I do see significant activity on the non-main threads, just not the bursts to 100% I see for example flying out of KLAX in P3D.  I use Process Lasso FWIW to mask Core0 off of MSFS, and put other Windows processes that might be more demanding on Core0.

Edited May 14, 20215 yr by Noel