System Performance

Not all motherboards are created equal. On the face of it, they should all perform the same and differ only in the functionality they provide - however, this is not the case. The obvious pointers are power consumption, POST time and latency. This can come down to the manufacturing process and prowess, so these are tested.

For Z790 we are running using Windows 11 64-bit with the 22H2 update.

Power Consumption

In previous motherboard reviews, we focused on the overall system power, and while sometimes useful, it doesn't paint the full picture. Motherboards can vary in power, but users with more M.2 SSDs, more devices plugged into the motherboard, and additional PCIe devices such as capture cards or sound cards will experience variances in power consumption.

For our motherboard reviews going forward, we're going to be focusing on power draw explicitly from the CPU, as in real-world situations, the processor is the de-facto part of the system that will display variance in power; the graphics card will also draw power, but with so many different models of the graphics card, these values will vary massively. To capture CPU power draw from each motherboard tested, we will be using ElmorLabs PMD-USB power measurement device, which directly plugs into the 8-pin ATX EPS 12 V CPU power connectors, and lets us directly measure the power being fed into the CPU via the power delivery from the power supply.

Ideally, lower values are better, especially with Intel's 13th and 12th Gen Core processors. The other factor is, of course, performance, which goes hand in hand with power consumption depending on the workload and the amount of CPU processing being utilized on the cores. At idle, the lower value is best, but some motherboard vendors are more aggressive on voltages, and more voltage = more heat and higher power consumption.

At idle values, we are measuring the value while the system has settled down, and for load power, we're measuring the average peak value over the course of the 10-minute CB23 MT run. This is because while power-hungry at full-load, processors can sometimes spike for a second or two and as such, can skew peak results, hence why we take the average value.

CPU Package Power: OS Idle

CPU Package Power: CineBench R23 MT 10-min Run

Starting with CPU package power in an idle state, the GIGABYTE Z790 Aorus Xtreme was the most economical out of the LGA1700 boards we've tested so far with our Core i9-13900K. We measured an OS idle reading of 3.7 W, which is around half a watt less than the ASRock Z790 LiveMixer. 

Focusing on the full-load average, the Z790 Aorus Xtreme averaged 271.6 W over a 10-minute loop of the CineBench R23 MT benchmark. This means, on average, it used around 4.5% less power than the ASRock Z790 Taichi Carrara at full load.

Non-UEFI POST Time

Different motherboards have different POST sequences before an operating system is initialized. A lot of this depends on the board itself, and POST boot time is determined by the controllers on board (and the sequence of how those extras are organized). We look at the POST Boot Time using a stopwatch as part of our testing. This is the time from pressing the ON button on the computer to when Windows starts loading. (We discount Windows loading as it is highly variable given Windows-specific features.)

Non UEFI POST Time

Measuring UEFI post times into Windows 11, the GIGABYTE Z790 Aorus Xtreme took 22.6 seconds to boot into Windows from system power-up. Things that can increase booting times include additional controllers such as Thunderbolt 4, CPU and memory initialization times, and of course, storage. While the Z790 Aorus Xtreme is not a slow booter, this can be attributed to the number of premium controllers installed onto the board.

DPC Latency

Deferred Procedure Call latency is a way in which Windows handles interrupt servicing. In order to wait for a processor to acknowledge the request, the system will queue all interrupt requests by priority. Critical interrupts will be handled as soon as possible, whereas lesser priority requests such as audio will be further down the line. If the audio device requires data, it will have to wait until the request is processed before the buffer is filled.

If the device drivers of higher priority components in a system are poorly implemented, this can cause delays in request scheduling and process time. This can lead to an empty audio buffer and characteristic audible pauses, pops, and clicks. The DPC latency checker measures how much time is taken processing DPCs from driver invocation. The lower the value will result in better audio transfer at smaller buffer sizes. Results are measured in microseconds.

Deferred Procedure Call Latency

We measure DPC latency out of the box without tweaks or customizations within Windows. The GIGABYTE Z790 Aorus Xtreme managed a latency reading of 135.7 microseconds, which isn't the best we've seen but is more than capable for audio/DAW workloads and applications.

Board Features, Test Bed and Setup CPU Performance, Short Form
Comments Locked

22 Comments

View All Comments

  • Chaser - Thursday, September 7, 2023 - link

    "Great thermals"? They'd better be. As this Raptor Lake "refresh" appears to be mostly another clock speed increase like the prior edition. Back in it's day, Alder Lake was a formidable challenge to Ryzen when it launched. But it seems these "refreshes" are going to have even more demanding thermals, generate more heat, and require very effective cooling solutions to keep it in check. Great space heaters as winter is coming :)
  • duffie - Tuesday, November 21, 2023 - link

    Why only one CPU connected PCIe 5.0 x16 at that price? All other brands have two mechanical PCIe 5.0 x16 slots. Sure, they will run at 5.0 x8 speed but that can be still useful with two GPUs.

Log in

Don't have an account? Sign up now