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, but also the ability for the manufacturer to optimize USB speed, audio quality (based on audio codec), POST time and latency. This can come down to the manufacturing process and prowess, so these are tested.

For B550, we are running using Windows 10 64-bit with the 1909 update.

Power Consumption

Power consumption was tested on the system while in a single ASUS GTX 980 GPU configuration with a wall meter connected to the Thermaltake 1200W power supply. This power supply has ~75% efficiency > 50W, and 90%+ efficiency at 250W, suitable for both idle and multi-GPU loading. This method of power reading allows us to compare the power management of the UEFI and the board to supply components with power under load, and includes typical PSU losses due to efficiency. These are the real-world values that consumers may expect from a typical system (minus the monitor) using this motherboard.

While this method for power measurement may not be ideal, and you feel these numbers are not representative due to the high wattage power supply being used (we use the same PSU to remain consistent over a series of reviews, and the fact that some boards on our testbed get tested with three or four high powered GPUs), the important point to take away is the relationship between the numbers. These boards are all under the same conditions, and thus the differences between them should be easy to spot.

Power: Long Idle (w/ GTX 980)Power: OS Idle (w/ GTX 980)Power: Prime95 Blend (w/ GTX 980)

The B550D4-4L performs very well in our power consumption testing, which can be attributed to the lack of controllers onboard in comparison to other B550 models tested. Despite being a commercial-focused motherboard, it includes support for Precision Boost Overdrive on Ryzen processors, which from our test at full-load is clearly visible. 

Non-UEFI POST Time

Different motherboards have different POST sequences before an operating system is initialized. A lot of this is dependent on the board itself, and POST boot time is determined by the controllers on board (and the sequence of how those extras are organized). As part of our testing, we look at the POST Boot Time using a stopwatch. 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

In our non-UEFI POST time testing, the ASRock did well, considering professional boards tend to take much longer to POST into Windows. The first time we POSTed up the board, it did take considerably longer due to the BMC, but even at the default setting and when stripped of non-essential controllers, there wasn't much difference in POST times.

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 test DPC latency with default settings straight out of the box, and as our graph shows, the ASRock isn't close to being competitive with other models we've tested. The B550D4-4L also doesn't include integrated audio, so users looking to build an audio workstation will need to rely on external audio controllers.

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

73 Comments

View All Comments

  • fmyhr - Thursday, May 20, 2021 - link

    While this is true, AMD and motherboard manufacturers are distressingly cagey about whether ECC and ECC error reporting actually work. If you care about this, you need to do your own searches. There have been cases of ECC support being added or removed on successive motherboard BIOS revisions. The different mainstream mfgs have different attitudes regarding ECC RAM: MSI pretty much ignores it, Gigabyte says they support ECC on _some_ boards, Asus seems somewhat better, and ASRock appears to be the best bet. If only Supermicro would give us a non-Threadripper Ryzen board...
  • AntonErtl - Friday, May 21, 2021 - link

    AFAIK all ASUS (as well as all Asrock) boards support ECC. We have several servers with (working) ECC with Ryzen CPUs (without Pro): 1600X, 1800X, 3700X, 3900X, 5800X. If AMD sold the Pro models in retail and guaranteed ECC functionality, we would be willing to pay a little extra for that. As for the Pro models, I once compared the specification of one with the corresponding non-Pro model, and wrt ECC they were the same. Can anyone name a Pro model where AMD guarantees more ECC functionality?
  • mode_13h - Friday, May 21, 2021 - link

    The difference between ECC support of Pro and non-Pro CPUs is supposedly that AMD only tests and guarantees it on the Pros. For the non-Pro CPUs, it's up to the motherboard vendor to test and support.

    As for APUs, AMD disables ECC support on the non-Pro APUs. I guess that's because the main customers for APUs with ECC are corporations, and so it's like a favor to big OEMs, giving them a lock on the corporate market (since the Pro versions seem to be EOM-only).
  • mode_13h - Friday, May 21, 2021 - link

    > EOM-only

    typo: should be OEM-only.
  • Slash3 - Friday, May 21, 2021 - link

    Correct. The board does indeed support ECC in that way. Gavin misinterpreted the specifications; no idea why, as it is quite clear.

    https://www.asrockrack.com/general/productdetail.a...

    "DDR4 288-pin ECC*/non-ECC UDIMM
    * For AMD Ryzen Desktop Processors with Radeon Graphics, ECC support is only with Processors with PRO technologies."

    Non-Pro APUs have always been the exception, and do not support ECC on any platform.
  • Jorgp2 - Friday, May 21, 2021 - link

    >Uhh.. ECC memory works just fine with bog standard (aka "not Pro") Ryzen CPU's and has LITERALLY since their launch in 2017.

    I don't think you understand what kind of board this is.

    If the data sheet says it only supports ECC for select SKUs, then it only supports ECC for select SKUs.
    There is no halfway for the target market.
  • leexgx - Friday, June 18, 2021 - link

    Ecc functionality still works even with the non-pro CPUs (just official stance is it doesn't work even thought it does, not like Intel where if its an i5 or higher ecc automatically doesn't work) ddr5 is going to change this problem with Intel as ecc is baked into ddr5 and can't be disabled and sold "as a enterprise" feature
  • MeJ - Friday, May 21, 2021 - link

    "The B550D4-4L also doesn't include integrated audio, so users looking to build an audio workstation will need to rely on external audio controllers."

    With respect, this comment is illogical. I have never heard of any DAWorkstation using on-board audio, and don't ever expect to. NOT having on-board audio to disable is a major advantage for a DAW. Also, the DPC issues here are perhaps characteristic of early drivers. There is no inherent reason I can see for this board to have worse performance than others with the same chipset... Is there? I agree that 10G would be preferred for a DAW.
  • mode_13h - Friday, May 21, 2021 - link

    Lol. Yeah, integrated audio on server boards that even have it tends to be a minimal implementation, with lots of crosstalk and interference.
  • mode_13h - Friday, May 21, 2021 - link

    ASPEED BMCs are such garbage. This has the same ARM11 core as a first gen Raspberry Pi. Just imagine how slow software rendering is on such a core, and that's the graphics performance you get on these things.

    I have an ASRock board with one of these BMCs, and 2D graphics even feels slow at 1024x768 (which is the resolution that the EDID of my analog KVM seems to advertise, even though the monitor is higher).

Log in

Don't have an account? Sign up now