The Supermicro C9Z390-PGW Motherboard Review: The Z390 Board With PLX and 10GbE
by Gavin Bonshor on February 1, 2019 9:00 AM EST- Posted in
- Motherboards
- Intel
- Broadcom
- Supermicro
- ATX
- PLX
- PLX8747
- Supero
- Z390
- C9Z390-PGW
- PEX8747
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 manufacturing process and prowess, so these are tested.
For Z390 we are running an updated version of our test suite, including OS and CPU cooler. This has some effect on our results.
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 test bed 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.
For a flagship motherboard, the Supermicro C9Z390-PGW displays quite efficent power draw when at full load. It's consistent with other models at long idle and in idle states when tested with our power hungry i7-8700K test bed chip.
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.)
The Supermicro C9Z390-PGW represents one of the slowest boards recorded in our POST time test. Supermicro doesn't have a good record in this regard as three out of four LGA 1151 boards that we've tested sit at the bottom.
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.
None of the LGA 1151 motherboards tested so far has been optimized for DPC latency. The C9Z390-PGW sits middle of the road with a result of 127 ns; it performs just better than the MSI MEG Z390 ACE and sits below the ROG STRIX Z370-F Gaming.
Facebook
Twitter
RSS
42 Comments
View All Comments
Flunk - Sunday, February 3, 2019 - link
The 10 series cards don't have enough SLI connectors to run in more than 2x SLI without severely downgraded bandwidth. It takes both connnectors for a 2 card HB SLI bridge. For the 20 series the RTX 2080 is the cheapest card that supports SLI. Nvidia is definitely limiting SLI to the most expensive cards.FastCarsLike - Friday, February 1, 2019 - link
Pretty much abandonware. The z390 Master from Gigabyte is significantly better. Audio improvements, 10gb internet, 3x NVME slots (although capped to 3 gb/s), and significantly better support than Asus. In addition, supermicro is way too cautious in the bios department than other manufacturers. (Bought a brand new Skylake Xeon motherboard from them, and had to select "experimental" NVME x4 to run a 760p from Intel!) Not worthy of mainstream.JoeyJoJo123 - Friday, February 1, 2019 - link
I've had similar issues with a SuperO board with PLX chip in the past, probably 2 years. Very common 16GB DDR4 set of Corsair Memory (DDR4-3000, CL15), wouldn't post with XMP enabled. I had to submit a support ticket to see what they could do about getting the memory kit supported. They had to order the memory in and find some stable memory timings, and sent me over the list of timings I should change. (Yeah, quite a few of the memory subtimings beyond the main 5 or so needed configuring... One of those subtimings went from a default 8 cycles to 22 cycles or something.)I wouldn't say they provide terrible value or terrible boards, but uhhh... Sometimes you just want stuff to work out of the box, and right now they're still a bit behind the usual motherboard vendors in regards to ease of use to work with.
Rocket321 - Sunday, February 3, 2019 - link
It does, however, sound like they provided you with an amazing level of support. Above and beyond what I would have expected the response to be.StrangerGuy - Saturday, February 2, 2019 - link
Reviewing a mobo with 10GbE/PLX but not do any testing on both features seems to miss the whole point completely.The_Assimilator - Saturday, February 2, 2019 - link
Another swing and miss from Supermicro. SLI is going the way of the dodo and quad-SLI especially is already extinct, so the PLX chip is well-nigh useless. Then we have not one, but two U.2 ports which are also useless.I dunno, man. Every time I read a review of a SM board it's like they just grabbed a bunch of arbitrary features and threw them together without any thought as to whether they even make sense together.
"There are two USB 2.0 headers, two USB 3.1 G1 headers and one USB 3.1 G2 Type-C header."
It's pretty interesting how you supposedly did a visual inspection yet failed to notice there are not two, but one USB 2.0 header; ditto for the USB 3.0 header.
jabber - Sunday, February 3, 2019 - link
Yep I would just prefer 1x16/4x4Aenra - Saturday, February 2, 2019 - link
Lurker that couldn't help but (finally) comment..- Testing this board with a 8700K is not exactly ideal; one would have expected a 9900K, which you do own. A very bad start from the get go. No one can know what the board's really capable of, so there goes the entire review. Well done.
- You mention an LLC of 6, but without specifying what it entails; there are board manufacturers that use high numbers for lowest, others for highest. A serious omission here. Were you even aware?
- To add to the confusion (one thinks lack of understanding's the issue here), you criticise vdroop twice and that's before you even mention the LLC..
- You mention the PLX chip time and time again, but fail to even convince us you grasp of its downsides. A testing of any PCIe interface card other than GPUs would have been a good start towards that. Further testing performance with just a GPU to compare would've been even better.
- We've zero interest in the thermal throttling of your specific CPU, nor any knowledge as to whether it might be a 'dud', or not. Using its thermal throttling so as to 'comment' on the board's OC capabilities is.. ridiculous, sorry.
- Automated RAM timings are what's usually the make or break in terms of frequencies.. and you don't even mention how they're handled when on a manual RAM overclocking.
- Personally? I'm not sure you should be doing board reviews in the first place.
Not the first Anandtech review i see that is bad, but.. do please improve.
Korguz - Sunday, February 3, 2019 - link
Aenra... if you think you can do better.. then please do...Personally ? im not sure you would be able to do a better review...
yobbo7 - Saturday, February 2, 2019 - link
So, TU-62/NP175 are both mid-Tg generic FR4 class materials with a dielectric constant of over 4.This is standard stuff, it is not special at all and I would be surprised if anyone in the industry used anything worse on this class of motherboard.
There are still better materials available from the likes of Isola before you get into the RF materials like Rogers.