Jump to content
Search In
  • More options...
Find results that contain...
Find results in...

How is GPU Usage/Load actually calculated/measured?

How do programs actually measure/calucate this? you know 0-100% usage. 
How does it know? like number of active transistors?

Intel i5-7600K  | Cooler Master Hyper 212 Evo  |  MSI H110 PC MATE  |  2x8GB Crucial Ballistix LT Sport 2400MHz  | MSI GTX 1080 ti Gaming X 11GB Radeon HD 7770 1GBDark Base Pro 900 (Orange)  | Toshiba 250GB 2.5" Drive -Game Drive | Crucial MX200 250GB 2.5" SSD - Boot Drive |  

 

Living Room PC: AMD Ryzen 2400G | VEGA 11 | ASUS P106-100 6GB V1 | 2x8 GB Crucial Ballistix 2400MHz | Samsung 120GB SSD | Seagate 500 GB HDD | Node 202 + 450W PSU |

Link to post
Share on other sites

It's pretty stupid actually. More a time load. This is especially important if you have a HT/SMT enabled CPU. Assuming workload is balanced, when you hit 50% you're effectively using all the "real" cores, and the region between 50-100% is the small extra benefit from HT/SMT.

 

Edit: doh! I misread the question as CPU not GPU. I'm guessing it is similar though.

Desktop Gaming system: Asrock Z370 Pro4, i7-8086k, Noctua D15, Corsair Vengeance Pro RGB 3200 4x16GB, Gigabyte 2070, NZXT E850 PSU, Cooler Master MasterBox 5, Optane 900p 280GB, Crucial MX200 1TB, Sandisk 960GB, Acer Predator XB241YU 1440p144 G-sync
TV Gaming system: Gigabyte Z490 Elite AC, i5-10600k, Noctua D15, Kingston HyperX RGB 4000@3600 2x8GB, EVGA 2080Ti Black, EVGA 850W, Corsair 230T, Crucial P1 1TB + MX500 1TB, LG OLED55B9PLA 4k120 G-Sync Compatible
Streaming system: Asus X299 TUF mark 2, i9-7920X, Noctua D15, Corsair Vengeance LPX RGB 3000 8x8GB, Asus Strix 1080Ti, Corsair HX1000i, GameMax Abyss, Samsung 970 Evo 500GB, Crucial BX500 1TB, BenQ XL2411 1080p144 + HP LP2475w 1200p60
Former Main system: Asus Maximus VIII Hero, i7-6700k, Noctua D14, G.Skill Ripjaws V 3200 2x8GB, Gigabyte GTX 1650, Corsair HX750i, In Win 303 NVIDIA, Samsung SM951 512GB, WD Blue 1TB, Acer RT280k 4k60 FreeSync [link]
Gaming laptop: Asus FX503VD, i5-7300HQ, DDR4 2133 2x8GB, GTX 1050, Sandisk 256GB + 480GB SSD [link]


 

Link to post
Share on other sites

As an educated guess: It seems like a simple ratio of active vs idle time over the sample period.

 

It has nothing to do with how many of the total transistors are being utilized. CPU's are composed of many sub-sections that each handle a particular type of calculation/workload. So measuring utilization in that way would not be useful.

BabyBlu (Primary Rig): 

  • CPU: Intel Core i9 9900K @ Per Core 5.2Ghz, 5.2Ghz, 5.2Ghz, 5.2Ghz, 5.1Ghz, 5.1Ghz, 5.0Ghz, 5.0Ghz; Cache @ 4.8Ghz; 0 AVX offset
  • Motherboard: Asus Maximus XI Hero
  • RAM: G.Skill Trident Z RGB 4x8GB DDR4-3200 CL14 @ 4000Mhz 16-18-18-36
  • GPU: MSI RTX 2080 Sea Hawk EK X
  • Case: Phanteks Evolv X
  • Storage: XPG SX8200 Pro 2TB NVME, 3x ADATA Ultimate SU800 1TB (RAID 0), Samsung 970 EVO Plus 500GB NVME
  • PSU: Corsair HX1000i
  • Monitor: Dell S2417DG 23.8" 2560x1440 165 Hz Gsync
  • Cooling: Custom water loop (CPU & GPU), Radiators: 1x140mm(Back), 1x280mm(Top), 1x420mm(Front)
  • Keyboard: Corsair Strafe RGB (Cherry MX Brown)
  • Mouse: Corsair Scimitar Elite/MasterMouse MM710
  • Headset: Corsair Void Pro RGB
  • OS: Windows 10 Pro

Roxanne (Wife Build):

  • CPU: Intel Core i7 4790K @ Per Core 5.0Ghz - 5.0Ghz - 4.9Ghz - 4.8Ghz, Cache @ 4.6Ghz, De-lidded
  • Motherboard: Asus Z97A
  • RAM: G.Skill Ripjaws X 4x4GB DDR3-2133 CL9 @ 2200Mhz 9-11-10-22
  • GPU: EVGA GTX 1080 FTW2 Gaming ICX
  • Case: Corsair Vengeance C70, With Custom Side-Panel Window
  • Storage: Samsung 850 EVO 250GB, Samsung 860 EVO 1TB, Silicon Power A80 1TB NVME
  • PSU: Corsair AX760
  • Monitor: Samsung C27JG56 27" 2560x1440 144Hz Curved Freesync
  • Cooling: Corsair H115i RGB
  • Keyboard: Rosewill RK 9000V2(Cherry MX Blue)
  • Mouse: Glorious Model O-
  • Headset: SteelSeries Arctis 7
  • OS: Windows 10 Pro

BlackBox (HTPC):

  • Intel Pentium G4600 (Kaby Lake)
  • Asus H110T/CSM Thin Mini-ITX
  • G.Skill 1x8GB DDR4-2400 SO-DIMM CL16
  • iGPU
  • Akasa Euler Fanless Solid Aluminum THIN Mini ITX Case (passively cooled)
  • Samsung 850 EVO 250GB
  • External 120W Power Adaptor
  • 4K TV
  • Passively cooled by the Akasa case. No moving parts!
  • Handheld QWERTY keyboard and trackpad.
  • Windows 10 Pro, Kodi autostarts on boot

NAS:

  • Synology DS216J
  • 2x8TB WD Red NAS HDDs in RAID 1. 8TB usable space
Link to post
Share on other sites

There are MANY different ways to calculate load. You can see how many units are currently in use over the total number of units, you can see how much bandwidth is currently being used over the theoretical maximum, or use the number of gates, etc etc.

 

There are even some stupid ways that some people use , such as comparing the base clock vs the current clock speeds (such as MS with CPUs).

 

In the case of AMD on Linux with AMDGPU, here's how it's done:

Quote

The amdgpu driver provides a sysfs API for reading how busy the GPU is as a percentage. The file gpu_busy_percent is used for this. The SMU firmware computes a percentage of load based on the aggregate activity level in the IP cores.

Basically the firmware in the GPUs accounts for the activity on the internal IP cores in some proprietary way and reports it back to the system. (that's similar on how most GPUs work since they're all proprietary IP cores).

FX6300 @ 4.2GHz | Gigabyte GA-78LMT-USB3 R2 | Hyper 212x | 3x 8GB + 1x 4GB @ 1600MHz | Gigabyte 2060 Super | Corsair CX650M | LG 43UK6520PSA
ASUS X550LN | i5 4210u | 12GB
Lenovo N23 Yoga

Link to post
Share on other sites
23 minutes ago, Edgar R. Zakarian said:

like number of active transistors?

Pretty sure that would be quite impractical (and complicated). You'd need a ton of additional circuitry to measure and count which transistors are active, which would then need additional space, require additional power and produce more heat.

 

The CPU load is determined by the OS and how it does this depends on the particular OS you're using. If you wanted to go really in-depth, you could probably have a look at how the Linux kernel does it (provided you can read and understand its code).

 

Here's an answer that is a bit more in depth: https://stackoverflow.com/a/3748172

Remember to quote or @mention others, so they are notified of your reply

Link to post
Share on other sites

i would assume it's the ratio of your cpu having work queued vs having nothing queued for the cycle across a fixed amount of time

same goes for gpu -shrug-

 

as for HT, a cpu can process more than 1 type of instruction at once so HT can queue different types of workload for a core to process at once.

Things I need help with:

Spoiler

none atm

 

I hate Intel's pricing, Ryzen's weird quirks, Nvidia's pricing, and Radeon GPUs in general

 

Spoiler

 

Products I like:

Spoiler

Sony Xperia Z1 / Z2 / 10 ii, Asus Strix 970 / 1070, Samsung SSD, WD HDD, Corsair PSUs (AX, RM, CX(grey)), GeForce GPU, NZXT N450/S340, be quiet! Coolers, G.Skill Trident RAM, Logitech M525, Logitech G440, Razer Deathadder Elite

 

Products I hate:

Spoiler

Xperia Z3, XiaoMi 5c, Radeon GPUs, Razer Audio Products, any bloatwares

 

Companies I absolutely adore: (and hope it stays that way)

Spoiler

be quiet! - sent me AM4 mounting for my DRP3 even though it's way past the timeframe stated, no questions asked

Corsair - very good RMA experience, absolutely recommend

 

Companies I hate:

Spoiler

Nvidia, Intel, Apple, TMT (Thundermatch, a retailer)

 

Personal Blacklisted Companies:

Spoiler

Acer: shit tier quality products, shit tier customer service thus far, they "tried" to solve my issue but they arent really doing anything but delaying and delaying. (on-going case since July)

Gigabyte: horrible customer service (gigabyte had literally 0 customer service, asked me to go to retailer with NO WAY to email them about a question) but at least they fixed my shit in ONE MONTH (would probably take me 1 hour to fix if they let me email them)

XiaoMi Phones: built like a tank but the software is buggy as all hell

Seagate HDD: had too many dead seagate drives

Kingston SSD: 300V controller swap thingy

Razer (except their mouse)

 

Remember, just because I had good/bad experiences with these companies/product, doesn't mean you will have similar experiences too. I would still recommend these products if they made sense for your needs, but I'll add a disclaimer of my experience if it's relevant. Feel free to DM me asking why they are where they are.

 

 

Link to post
Share on other sites

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

×