Samsung 9100 Pro 4TB being throttled on MSI MEG Ace Z890

[adam.hamle158d02d9]

Hi,

I have a brand new system with a MSI Meg Ace Z890 motherboard and the Intel 285k.

I have a 4TB Samsung 9100 Pro NVME drive installed in the Gen5 x4 M.2_1 slot which is directly connected to the CPU PCIe lanes (see attached block diagram).

The 9100 pro is capable of upwards of 14BGB/sec throughput but I'm only seeing 12GB/sec.

I've read the articles about the gen 5 NVME drives being throttled due to the Arrow Lake chiplet design, but I'm assuming this would impact drives connected via the PCH. My drive is directly connected to the CPU (Arrow Lake has 20 gen 5 PCIe lanes, I'm using 16 for the GPU and 4 for the 9100 pro).

I raised this with Intel and their response was the following:

The upstream memory read performance of the ARL S processor SOC tile root ports 13 and 14 (PCIe Lanes 1 to 16), configured as 1px16 or 2px8, may be affected by increased round-trip latencies during random upstream memory read operations. These latency variations are influenced by the specific workload and the capabilities of the PCIe endpoint device.

As a result, the observed performance could be attributed to the system configuration and/or SoC architecture. Given the nature of this issue, I recommend contacting the motherboard OEM for further assistance and guidance.

So it looks like Intel are pointing towards MSI for this problem - any thoughts on how to get this fixed? Thanks.

 

[citay]

BIOS is the latest v1A6 from https://www.msi.com/Motherboard/MEG-Z890-ACE/support ? If not, update to that and re-test.

After that, you could open a ticket with MSI. If you could, post a screenshot of your benchmark results here.

 

[ali-2024-ksa]

Hello, how are you? Downgrade the graphics card to Gen 4 so that M.2 can run at its full maximum speed.

 

[doug.hogart153b02d9]

I think you need to re-read the reports such as TheSSDReview, to realize that the problem is not really Z890/motherboard but rather the ARL-S (Ultra 200S Series) processor offering the 4 PCIe Gen5 lanes used for M.2. I have 265K and it is frustrating to follow threads of apparent Intel representatives giving poor answers to users like us in recent months.

 

[FlyingScot]

I think you need to re-read the reports such as TheSSDReview, to realize that the problem is not really Z890/motherboard but rather the ARL-S (Ultra 200S Series) processor offering the 4 PCIe Gen5 lanes used for M.2. I have 265K and it is frustrating to follow threads of apparent Intel representatives giving poor answers to users like us in recent months.

It’s an Arrow Lake latency issue due to the new chiplet design, right? That’s what I remember reading a few weeks ago. Today, you have a choice between Arrows that fall short or Raptors that bite their owners. But at least Intel chips don’t produce smoke like the competition.

 

[doug.hogart153b02d9]

It’s an Arrow Lake latency issue due to the new chiplet design, right?

I think so, specifically for those 4 lanes used to give M.2 access to processor. Something about the chiplet processor design, I/o extender tile.

 

[mbes645615a702e4]

Hi,

I have a brand new system with a MSI Meg Ace Z890 motherboard and the Intel 285k.

I have a 4TB Samsung 9100 Pro NVME drive installed in the Gen5 x4 M.2_1 slot which is directly connected to the CPU PCIe lanes (see attached block diagram).

The 9100 pro is capable of upwards of 14BGB/sec throughput but I'm only seeing 12GB/sec.

I've read the articles about the gen 5 NVME drives being throttled due to the Arrow Lake chiplet design, but I'm assuming this would impact drives connected via the PCH. My drive is directly connected to the CPU (Arrow Lake has 20 gen 5 PCIe lanes, I'm using 16 for the GPU and 4 for the 9100 pro).

I raised this with Intel and their response was the following:

The upstream memory read performance of the ARL S processor SOC tile root ports 13 and 14 (PCIe Lanes 1 to 16), configured as 1px16 or 2px8, may be affected by increased round-trip latencies during random upstream memory read operations. These latency variations are influenced by the specific workload and the capabilities of the PCIe endpoint device.

As a result, the observed performance could be attributed to the system configuration and/or SoC architecture. Given the nature of this issue, I recommend contacting the motherboard OEM for further assistance and guidance.
Also, while troubleshooting and running disk performance benchmarks (which, honestly, can be a bit mind-numbing), I’ve found taking short breaks helps maintain focus. Recently, I discovered https://mostbet.net.in/app/ — a mobile app that offers a fun way to unwind. Whether it’s sports betting or quick games of chance, it’s a surprisingly engaging way to recharge between debugging sessions.

So it looks like Intel are pointing towards MSI for this problem - any thoughts on how to get this fixed? Thanks.

Hello! I ran into the same issue with slightly lower-than-expected throughput on a Gen5 NVMe connected directly to the CPU. Updating the BIOS to the latest version from MSI actually helped. Looks like there were some improvements related to PCIe handling. Worth trying if you haven't already.