Jump to content

Recommended Posts

Posted

see this image. Barely using the 8 other CPU threads.

 

attachment.php?attachmentid=178618&stc=1&d=1518240605

826772999_singlecore.thumb.jpg.2ea1dcf704e10ff5a5875045e983965c.jpg

DCS Wishlist: 2K11 Krug SA-4 Ganef SAM, VR-TrackIR icons next to player names in score-chart

PvP: 100+ manual player-kills with Stingers on a well known dynamic campaign server - 100+ VTOL FARP landings & 125+ hours AV-8B, F-14 crew, royal dutch airforce F-16C - PvP campaigns since 2013

DCS server-admins: please adhere to a common sense gaming industry policy as most server admins throughout the industry do. (After all there's enough hostility on the internet already which really doesn't help anyone. Thanks.)

Dell Visor VR headset, Ryzen 5 5600 (6C/12T), RTX 4060 - basic DCS-community rule-of-thumb: Don't believe bad things that a PvP pilot claims about another PvP pilot without having analyzed the existing evidence

Posted

Why do you consider this a bug

7800x3d, 5080, 64GB, PCIE5 SSD - Oculus Pro - Moza (AB9), Virpil (Alpha, CM3, CM1 and CM2), WW (TOP and CP), TM (MFDs, Pendular Rudder), Tek Creations (F18 panel), Total Controls (Apache MFD), Jetseat 

Posted

Cause afaik then my CPU is the bottleneck with my GTX 1060 6GB.

DCS Wishlist: 2K11 Krug SA-4 Ganef SAM, VR-TrackIR icons next to player names in score-chart

PvP: 100+ manual player-kills with Stingers on a well known dynamic campaign server - 100+ VTOL FARP landings & 125+ hours AV-8B, F-14 crew, royal dutch airforce F-16C - PvP campaigns since 2013

DCS server-admins: please adhere to a common sense gaming industry policy as most server admins throughout the industry do. (After all there's enough hostility on the internet already which really doesn't help anyone. Thanks.)

Dell Visor VR headset, Ryzen 5 5600 (6C/12T), RTX 4060 - basic DCS-community rule-of-thumb: Don't believe bad things that a PvP pilot claims about another PvP pilot without having analyzed the existing evidence

Posted

Some thoughts:

- until ED confirm otherwise, DCS is still single threaded

- what else are you running in parallel that could have nearly maxed out a core

- from a quick glance, seems that most “cores” are in use, but with the secondary threads less used.

 

So to me, looks like useful data but doesn’t look like a bug to me.

7800x3d, 5080, 64GB, PCIE5 SSD - Oculus Pro - Moza (AB9), Virpil (Alpha, CM3, CM1 and CM2), WW (TOP and CP), TM (MFDs, Pendular Rudder), Tek Creations (F18 panel), Total Controls (Apache MFD), Jetseat 

Posted (edited)
see this image. Barely using the 8 other CPU threads.

 

attachment.php?attachmentid=178618&stc=1&d=1518240605

 

this is how decent multi-threading is supposed to look like. the only definite way to spot a cpu bottleneck would be one or more cores flatlining at 100%. your cpu seems to do fine...

 

Some thoughts:

- until ED confirm otherwise, DCS is still single threaded

By now it should be clear, that DCS is not single threaded anymore. There is room for discussion on how DCS handles multi-core, but it is very obvious that DCS uses more than one core for render-related task. just look at the damn graphs!

Edited by twistking

My improved* wishlist after a decade with DCS ⭐⭐⭐⭐🌟

*now with 17% more wishes compared to the original

Posted

By now it should be clear, that DCS is not single threaded anymore. There is room for discussion on how DCS handles multi-core, but it is very obvious that DCS uses more than one core for render-related task. just look at the damn graphs!

 

The graph does not mean anything concerning single or multi-threaded application. Motherboards/chipsets are dividing tasks (including single thread apps) since the introduction of multi-core CPUs. There is also a misunderstanding about "CPU bottleneck". An application like DCS is designed to run as fast as possible, so to take advantage of all "free time" of CPU, in other words: use 100% of the CPU every time. In this context: everything is a "bottleneck" since everything is intented to run as maximum stage... If the "Graphical part" is quicker, the "graphical part" will wait for the CPU to finish his job, if "CPU part" is more quicker, the CPU will wait for "graphical part" to finish his job... You will ALWAYS have something that wait for another thing, this is limitless.

Posted

the fact that the GPU seems to be maxed at 100 and the CPU not would say to me your GPU is limiting you.

 

now if your GPU was at 50%... and it looked like that. i would say CPU.

My Rig: AM5 7950X, 32GB DDR5 6000, M2 SSD, EVGA 1080 Superclocked, Warthog Throttle and Stick, MFG Crosswinds, Oculus Rift.

Posted
The graph does not mean anything concerning single or multi-threaded application. Motherboards/chipsets are dividing tasks (including single thread apps) since the introduction of multi-core CPUs. There is also a misunderstanding about "CPU bottleneck". An application like DCS is designed to run as fast as possible, so to take advantage of all "free time" of CPU, in other words: use 100% of the CPU every time. In this context: everything is a "bottleneck" since everything is intented to run as maximum stage... If the "Graphical part" is quicker, the "graphical part" will wait for the CPU to finish his job, if "CPU part" is more quicker, the CPU will wait for "graphical part" to finish his job... You will ALWAYS have something that wait for another thing, this is limitless.

 

i agree on the bottleneck part, but i disagree on the thread splitting. maybe i misunderstand you there. i know that threads can get shifted from one core to another for better heat management, but i don't know a way to process one thread on two cores at the same time.

i know a lot of "single-thread" applications (games mostly), that will flatline one core at a 100% without any load on another thread. that is the reasons why - until recently - people would advise to go for higher single-thread performance against core-count when buying a cpu specifically for gaming...

My improved* wishlist after a decade with DCS ⭐⭐⭐⭐🌟

*now with 17% more wishes compared to the original

Posted
the fact that the GPU seems to be maxed at 100 and the CPU not would say to me your GPU is limiting you.

 

With multi-core CPU, you will VERY rarely see a core maxed at 100%... This behaviour would be very strange. I can launch an infinite loop of matrix multiplication, i will never see my CPU maxed to 100%... tasks are divided between cores...

Posted (edited)

I have a few screenshots spanning from 2.5 initial release to current 2.Hotfix. Sometimes the CPU is really using ALL real cores, sometimes it's somewhat easing off.In general, imho, more cores are better.

 

54% CPU usage ( reproducable btw ) on a 6c/12t CPU means something.

 

 

https://forums.eagle.ru/showthread.php?t=198518

 

https://forums.eagle.ru/showthread.php?t=197555

 

I cannot double post the screenshots, so please look in the original threads.

Edited by BitMaster

Gigabyte Aorus X570S Master - Ryzen 5900X - Gskill 64GB 3200/CL14@3600/CL14 - Sapphire  Nitro+ 7800XT - 4x Samsung 980Pro 1TB - 1x Samsung 870 Evo 1TB - 1x SanDisc 120GB SSD - Heatkiller IV - MoRa3-360LT@9x120mm Noctua F12 - Corsair AXi-1200 - TiR5-Pro - Warthog Hotas - Saitek Combat Pedals - Asus XG27ACG QHD 180Hz - Corsair K70 RGB Pro - Win11 Pro/Linux - Phanteks Evolv-X 

Posted (edited)
i agree on the bottleneck part, but i disagree on the thread splitting. maybe i misunderstand you there. i know that threads can get shifted from one core to another for better heat management, but i don't know a way to process one thread on two cores at the same time.

 

As far as i know, you don't have control on such thing... You simply create a thread, and the OS's (Windows in this case) kernel manage it for you. Assigning a specifical thread to a particular core is not allowed in common API. Look by yourself:

 

https://msdn.microsoft.com/en-us/library/windows/desktop/ms682453(v=vs.85).aspx

https://msdn.microsoft.com/en-us/library/windows/desktop/ms686277(v=vs.85).aspx

 

Exploiting multi-core require using a specific library which work at low-level close to hardware, and this is used for very specific tasks which can be "parallelized"

 

 

i know a lot of "single-thread" applications (games mostly), that will flatline one core at a 100% without any load on another thread.

 

Never seen that on my life... but why not.

 

that is the reasons why - until recently - people would advise to go for higher single-thread performance against core-count when buying a cpu specifically for gaming...

 

This still true... 2 cores probably helps to separate game from OS background tasks. What will really make the difference is :

- CPU design

- CPU frequency

- CPU cache/registry management...

 

The same CPU design with same clock, I doubt you will see any difference between a 4-cores vs 8-cores... Except the price and marketing...

Edited by sedenion
Posted

Nah 6/12 cores 10 to 15% total CPU usage for me with DCS even less I expect without task manager or msi afterburner or any other software running.

 

DCS still tanks at 2 cores at 30 to 50% and one virtual core at 90 to 100%

 

Sure there is more other core activity but this could be other software or drivers using more threads .

Control is an illusion which usually shatters at the least expected moment.

Gazelle Mini-gun version is endorphins with rotors. See above.

 

Currently rolling with a Asus Z390 Prime, 9600K, 32GB RAM, SSD, 2080Ti and Windows 10Pro, Rift CV1. bu0836x and Scratch Built Pedals, Collective and Cyclic.

Posted
this is how decent multi-threading is supposed to look like. the only definite way to spot a cpu bottleneck would be one or more cores flatlining at 100%. your cpu seems to do fine...

 

No lol, Microsofts task manager is wrong, even in ArmA 3 I get only 20-25 fps on some mods while all CPU cores only at 50-60% and GPU at 30%... sooo... Well done, Microsoft...

 

 

the fact that the GPU seems to be maxed at 100 and the CPU not would say to me your GPU is limiting you.

 

now if your GPU was at 50%... and it looked like that. i would say CPU.

 

As I wrote above, GPU usage might be wrong too, as CPU usage is reported wrong too in ArmA 3 for example (Microsoft Windows 10 problem I guess)

DCS Wishlist: 2K11 Krug SA-4 Ganef SAM, VR-TrackIR icons next to player names in score-chart

PvP: 100+ manual player-kills with Stingers on a well known dynamic campaign server - 100+ VTOL FARP landings & 125+ hours AV-8B, F-14 crew, royal dutch airforce F-16C - PvP campaigns since 2013

DCS server-admins: please adhere to a common sense gaming industry policy as most server admins throughout the industry do. (After all there's enough hostility on the internet already which really doesn't help anyone. Thanks.)

Dell Visor VR headset, Ryzen 5 5600 (6C/12T), RTX 4060 - basic DCS-community rule-of-thumb: Don't believe bad things that a PvP pilot claims about another PvP pilot without having analyzed the existing evidence

Posted (edited)

if you think the arma engine is a good comparison, then i can't help you.

it's a well known fact, that the arma engines won't utilize a cpu to the fullest. specifically arma is a good example, that the crude concept of the weakest part in the chain a.k.a. cpu/gpu bottlenecking does not apply to all software in the same way.

 

if you were to measure your cpus tdp, while playing arma, you would find that it would correlate to the 50% load you see in task manager.

i could go into detail here, but i won't since this is not about arma performance...

Edited by twistking

My improved* wishlist after a decade with DCS ⭐⭐⭐⭐🌟

*now with 17% more wishes compared to the original

Posted
Nah 6/12 cores 10 to 15% total CPU usage for me with DCS even less I expect without task manager or msi afterburner or any other software running.

 

DCS still tanks at 2 cores at 30 to 50% and one virtual core at 90 to 100%

 

Sure there is more other core activity but this could be other software or drivers using more threads .

 

Not really FragBum,

 

DCS uses it's 2 cores, correct, it then needs additional cores, if you have them, to do the DX11 thingy, afaik, DX11 can take up to 4 cores, Skatezilla knowsd that more precise I guess.

 

I was also "with a bold !" NOT talking about DCS alone BUT the whole PC, all that you need to run DCS at a high standard. MSI AFterburner or equivalent is a standard toolset, so are many other tasks and apps. But they dont use that much CPU, if they would you would see that all the time and that is not the case.

 

I use Process Lasso to force DCS and DX11 to use real cores only, no Hyperthreaded cores.

 

I am also not the only one with a 8700k that has this experience, the forum has numerous posts from such users stating the same, look around.

 

I am not stating something I cannot reproduce or only comes 1 out of 5 times or such, this is a constant behaviour.

Gigabyte Aorus X570S Master - Ryzen 5900X - Gskill 64GB 3200/CL14@3600/CL14 - Sapphire  Nitro+ 7800XT - 4x Samsung 980Pro 1TB - 1x Samsung 870 Evo 1TB - 1x SanDisc 120GB SSD - Heatkiller IV - MoRa3-360LT@9x120mm Noctua F12 - Corsair AXi-1200 - TiR5-Pro - Warthog Hotas - Saitek Combat Pedals - Asus XG27ACG QHD 180Hz - Corsair K70 RGB Pro - Win11 Pro/Linux - Phanteks Evolv-X 

Posted
Not really FragBum,

 

DCS uses it's 2 cores, correct, it then needs additional cores, if you have them, to do the DX11 thingy, afaik, DX11 can take up to 4 cores, Skatezilla knowsd that more precise I guess.

 

I was also "with a bold !" NOT talking about DCS alone BUT the whole PC, all that you need to run DCS at a high standard. MSI AFterburner or equivalent is a standard toolset, so are many other tasks and apps. But they dont use that much CPU, if they would you would see that all the time and that is not the case.

 

I use Process Lasso to force DCS and DX11 to use real cores only, no Hyperthreaded cores.

 

I am also not the only one with a 8700k that has this experience, the forum has numerous posts from such users stating the same, look around.

 

I am not stating something I cannot reproduce or only comes 1 out of 5 times or such, this is a constant behaviour.

 

Perhaps DX11 is using more threads but I am still only seeing 15% CPU utilisation running DCS which for me is pretty much the same as DCS 2.x has always used. On a brighter note I see now after this latest update I see more GPU and video RAM utilisation, that's a plus.

Control is an illusion which usually shatters at the least expected moment.

Gazelle Mini-gun version is endorphins with rotors. See above.

 

Currently rolling with a Asus Z390 Prime, 9600K, 32GB RAM, SSD, 2080Ti and Windows 10Pro, Rift CV1. bu0836x and Scratch Built Pedals, Collective and Cyclic.

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...