I reckon if it ain't broke, no point messing with it. I've done this since optimal was introduced and had zero problems. Well, modern resource hogging games at least.Īs for everything else I have my global setting to Optimal so only gaming likely uses full power. That is, IMO it's just that bit better (possibly in extreme cases) to set prefer max performance in games. One other thing that solidified this opinion is that a small amount of very graphics demanding games already have the default setting on prefer maximum performance in the NVCP. But if a sudden heavy action/visual screen comes in after sometime not moving much or being in a simple menu I prefer not to take the chance (at the cost of power). My only reason for keeping max power when gaming is, I'm not 100% sure if the adaptive setting is quick enough to almost instantly ramp up the power everytime? Seems clear it is most of the time though. However in many casesĪdaptive will be just as good with the added benefit of saving power Seeing as I am not too concerned about GPU power usage I can conclude from this that "prefer maximum performance" is the most certain way to get proper/max performance. RAPOO E9270P Black 5GHz wireless / HP backlit Logitech MX Anywhere 2 Laser wireless / Logitech M330 wireless Seasonic Focus Plus Gold 750W / Powerbrick Panasonic SA-PMX94 / Realtek onboard + B&O speaker system / Harman Kardon Go + Play / Logitech G533 MSI RTX 2070 Super Gaming X Trio / Intel HD620 ![]() Noctua NH-U12A 2 fans + Thermal Grizzly Kryonaut Extreme + 5 case fans / Fanģ2GB DDR5 Corsair Dominator Platinum RGB 6000MHz C元6 / 8GB DDR4 HyperX CL13 Gigabyte Z690 Aorus Master / HP 83A3 (U3E1) This represents mid-range up to high-end video card performance.Intel i7 12700K P-Cores 5Ghz / Intel i3 7100U We are using an NVIDIA GeForce RTX 2080 SUPER FE and an NVIDIA GeForce RTX 2060 SUPER FE. We chose two video cards to use to collect all this data. We have two games benchmarked with a manual run-through, and two games benchmarks using the game’s built-in in-game benchmark runs. Game Performance – Finally, we will see if it all adds up to changing game performance between each power mode. Will it give us more or less GPU frequency? Will the card run faster under “Prefer Maximum Performance”? That’s the question. Don't forget to check you're dealing with NVidia. ![]() When you're done hit 'F3' again and repeat the whole process for any other hits. When you're happy you've got the NVidia, set PerfLevelSrc to 2222, and the four PowerMizer values to 0. ![]() The goal here is to see if any of these power modes actually changes the potential of the GPU. If it's the Intel just hit 'F3' and it'll look for the next match in the registry. GPU Clock Frequency – This one is very important we are tracking the GPU clock frequency in real-time as we play a game. GPU Temperature – We are also collecting GPU temperature data to see if any option is more demanding on the GPU or not. With these two different power data points we’ll have a good grasp on the power demands with each option. In addition to total system Wattage, we will report the Power Consumption that GPUz finds for the graphics card. Therefore, we will look at the total system Wattage at Idle and Full-Load in a game and compare them. Idle and Full-Load Wattage – Naturally, we need to find out which options affect power usage and by how much. ![]() To test this practically, in a real-world environment we are taking several data samples that will be important to look at.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |