Question My "Samsung 980 PRO with heatsink" quandary

[mikeymikec]

When I needed to do a SSD upgrade on my old PC (Haswell), I chose the 980 PRO + integrated heatsink because the motherboard didn't include the typical 'heatsink' for M.2 SSDs that boards do these days, and I chose a gen4 drive because I anticipated a system upgrade that I would use the drive in (in the near future).

In hindsight I wonder whether getting one with a heatsink was really worth it since it didn't appear to have any great effect on the drive temp (especially considering that system was going to push it to a max of 800MB/sec), but hey ho.

Now for the quandary with is two-fold. New system (AM5) obviously has M.2 heatsinks, but when spec'ing the system I thought that the drive already has a heatsink and I'd simply not use the heatsink that the board came with. I figured I could screw in the M.2 drive like I've done many times before for older systems, however since boards with M.2 heatsinks have become commonplace, the securing point (any idea what it's called?) for the drive/heatsink is designed with the heatsink in mind:



I tried to put a screw in it but it just wasn't happening, so I used the plastic securing bit (that I assume is really just meant to keep the drive in position before the heatsink + screw is used). I've been running it like this since late November 2023 without issues so far, but I'm wondering if it's safe in the long term.

Quandary 2 - the drive passes quick performance tests (e.g. Samsung Drive Magician, or say the Disks utility in Linux Mint), hitting read and write speeds of somewhere in the neighbourhood of 7000MB/sec, but the usual ATTO benchmark that I do on SSDs to make sure everything is basically performing within specs consistently shows signs of throttling (like 90% throttling) halfway through. I mentioned Linux Mint but ATTO is being run from within Win11. I'm using the default NVMe driver, but I've used plenty of gen4 PRO drives in other systems I've built and haven't seen that sort of performance drop-off during an ATTO run.

Surely Samsung's own heatsink is capable of cooling the drive sufficiently under load? I would report drive temps but there are two things that are odd in this respect: One is how variable the 980 PRO temp is, especially despite the fact that Linux doesn't need to read the drive at all (I'll sometimes see say 32C, right now I'm seeing 48/48/54C in psensor). I can make notes of the drive temp during an ATTO run but I don't recall it getting crazy high before the test was completed.

I guess overall I'm wondering whether I should detach the heatsink from that drive and start using the board heatsink, though I am mindful of a guy on imgur who sliced open their hand when trying to take the heatsink off

 

[In2Photos]

Unless you're constantly doing large file transfers I don't see the drive heating up that much anyway. But I'm using the heat sinks from the motherboard on my drives. They rarely reach 50C even during long gaming sessions where the entire system heats up. From what I've seen removing the heat sink from most hard drives doesn't require tools. I've seen most installers just use their hands. But if you do need a tool I suggest a plastic one.

 

[Tech Junky]

didn't appear to have any great effect

what it's called?

a latch

Samsung Drive Magician

is useless other than FW upgrades

90% throttling

probably cache exhaustion on the drive controller.

heatsink is capable of cooling the drive sufficiently under load?

they're typically more of a gimmick being a sticker and all. the real issue is whether or not things are making contact with the controller on the M2 or not.

constantly doing large file transfers

this is where things tend to start cooking

50C even during long gaming sessions

that's not enough load to really heat things up. the intense transfers are when launching a game and pushing the data to the GPU. once you're up and running the disk activity drops to minimal.

--------------------
In general we look for things to nit pick when digging into system stats. Reality is het sinks are a joke unless you have really bad air flow to begin with or mount GPUs above all of your drives and push the GPUs under load and create a heat soak due to bad ventilation. Now, if you have the M2 heat sink that's an actual metal passive cooler attached to it then it could help if it's a single sided drive vs having controllers split on each side of the PCB. The MOBO HSs are substantial enough and work in two ways.... contact with the controller and shielding the drive from passive heat from inside the case.

Now, if you want to step up your storage game look into U.2/U.3 drives which pack NVME speed into a 2.5" enclosure. I picked up a Kioxia that hits the same speeds ~6.5GB/s but stays cooler ~40C. Using an M2 adapter + 50cm cable to mount it somewhere other than in the middle of the heater core on the MOBO where they tend to stick the M2 sockets.

 

[mikeymikec]

I ran the ATTO benchmark again today and I thought it was a bit odd that I was seeing different figures in Task Manager and HWMonitor compared to ATTO.

I decided to update my version of ATTO (which dated back to 2009), and now I'm getting much more reasonable figures (~6.3GB read, ~4.2GB write, temp went up to 59C max) and a normal looking throughput curve. Odd though as I'm sure I've run that version of ATTO on 980 PROs before and seen reasonable figures. Perhaps it's a quirk of AMD7000 + old ATTO?

Any thoughts on using the latch on its own in the long term?

 

[Tech Junky]

Any thoughts on using the latch on its own in the long term?

As long as you're not fiddling with it constantly it should be fine. The only concern though since it's plastic would be extended periods of high heat might make them become brittle over time. Drive should till stay seated though w/o it. It's just there to hold things in place if you're putting on a HS or digging around near the drive. It's definitely not grounding it or anything though like a screw might due to it's makeup.

quirk of AMD7000

There are a lot of quirks to AMD you don't see on Intel. It's a combo of things but mainly from what I've seen it's mostly obfuscated info due to AMD not being open source with their stuff. This will be changing though when they update to using OpenSIL instead of AGESA for the UEFI components.

With Intel had a few things showing up with generic info but AMD has a ton of them in comparison. I tried all sorts of things to get them to populate in the output commands on Linux but it's still scattered about.

sudo lspci
00:00.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14d8
00:00.2 IOMMU: Advanced Micro Devices, Inc. [AMD] Device 14d9
00:01.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14da
00:01.1 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 14db
00:01.2 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 14db
00:02.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14da
00:02.1 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 14db
00:02.2 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 14db
00:03.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14da
00:04.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14da
00:08.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14da
00:08.1 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 14dd
00:08.3 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 14dd
00:14.0 SMBus: Advanced Micro Devices, Inc. [AMD] FCH SMBus Controller (rev 71)
00:14.3 ISA bridge: Advanced Micro Devices, Inc. [AMD] FCH LPC Bridge (rev 51)
00:18.0 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14e0
00:18.1 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14e1
00:18.2 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14e2
00:18.3 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14e3
00:18.4 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14e4
00:18.5 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14e5
00:18.6 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14e6
00:18.7 Host bridge: Advanced Micro Devices, Inc. [AMD] Device 14e7
01:00.0 PCI bridge: Intel Corporation Device 4fa1 (rev 01)
02:01.0 PCI bridge: Intel Corporation Device 4fa4
02:04.0 PCI bridge: Intel Corporation Device 4fa4
03:00.0 VGA compatible controller: Intel Corporation DG2 [Arc A380] (rev 05)
04:00.0 Audio device: Intel Corporation DG2 Audio Controller
05:00.0 PCI bridge: Intel Corporation Thunderbolt 4 Bridge [Maple Ridge 4C 2020] (rev 02)
06:00.0 PCI bridge: Intel Corporation Thunderbolt 4 Bridge [Maple Ridge 4C 2020] (rev 02)
06:01.0 PCI bridge: Intel Corporation Thunderbolt 4 Bridge [Maple Ridge 4C 2020] (rev 02)
06:02.0 PCI bridge: Intel Corporation Thunderbolt 4 Bridge [Maple Ridge 4C 2020] (rev 02)
06:03.0 PCI bridge: Intel Corporation Thunderbolt 4 Bridge [Maple Ridge 4C 2020] (rev 02)
07:00.0 USB controller: Intel Corporation Thunderbolt 4 NHI [Maple Ridge 4C 2020]
09:00.0 USB controller: Intel Corporation Thunderbolt 4 USB Controller [Maple Ridge 4C 2020]
46:00.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f4 (rev 01)
47:00.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
47:04.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
47:05.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
47:06.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
47:07.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
47:08.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
47:0c.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
47:0d.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
48:00.0 Non-Volatile memory controller: KIOXIA Corporation NVMe SSD Controller CD8 (rev 01)
4d:00.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f4 (rev 01)
4e:00.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
4e:01.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
4e:02.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
4e:03.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
4e:04.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
4e:08.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
4e:0a.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
4e:0b.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
4e:0c.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
4e:0d.0 PCI bridge: Advanced Micro Devices, Inc. [AMD] Device 43f5 (rev 01)
4f:00.0 PCI bridge: ASMedia Technology Inc. Device 2812 (rev 01)
50:00.0 PCI bridge: ASMedia Technology Inc. Device 2812 (rev 01)
50:02.0 PCI bridge: ASMedia Technology Inc. Device 2812 (rev 01)
50:03.0 PCI bridge: ASMedia Technology Inc. Device 2812 (rev 01)
50:08.0 PCI bridge: ASMedia Technology Inc. Device 2812 (rev 01)
50:0a.0 PCI bridge: ASMedia Technology Inc. Device 2812 (rev 01)
50:0b.0 PCI bridge: ASMedia Technology Inc. Device 2812 (rev 01)
52:00.0 Ethernet controller: Aquantia Corp. AQC111 NBase-T/IEEE 802.3bz Ethernet Controller [AQtion] (rev 02)
53:00.0 Ethernet controller: Aquantia Corp. AQC111 NBase-T/IEEE 802.3bz Ethernet Controller [AQtion] (rev 02)
55:00.0 Ethernet controller: Aquantia Corp. AQC111 NBase-T/IEEE 802.3bz Ethernet Controller [AQtion] (rev 02)
56:00.0 Ethernet controller: Aquantia Corp. AQC111 NBase-T/IEEE 802.3bz Ethernet Controller [AQtion] (rev 02)
59:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8125 2.5GbE Controller (rev 05)
5a:00.0 Non-Volatile memory controller: Sandisk Corp WD Black SN770 / PC SN740 256GB / PC SN560 (DRAM-less) NVMe SSD (rev 01)
5c:00.0 Network controller: Qualcomm Technologies, Inc Device 1107 (rev 01)
5d:00.0 PCI bridge: Pericom Semiconductor PI7C9X2G304 EL/SL PCIe2 3-Port/4-Lane Packet Switch (rev 05)
5e:01.0 PCI bridge: Pericom Semiconductor PI7C9X2G304 EL/SL PCIe2 3-Port/4-Lane Packet Switch (rev 05)
5e:02.0 PCI bridge: Pericom Semiconductor PI7C9X2G304 EL/SL PCIe2 3-Port/4-Lane Packet Switch (rev 05)
5f:00.0 Multimedia video controller: Conexant Systems, Inc. CX23887/8 PCIe Broadcast Audio and Video Decoder with 3D Comb (rev 04)
60:00.0 Multimedia video controller: Conexant Systems, Inc. CX23887/8 PCIe Broadcast Audio and Video Decoder with 3D Comb (rev 04)
61:00.0 USB controller: Advanced Micro Devices, Inc. [AMD] Device 43f7 (rev 01)
62:00.0 SATA controller: Advanced Micro Devices, Inc. [AMD] Device 43f6 (rev 01)
63:00.0 USB controller: Advanced Micro Devices, Inc. [AMD] Device 43f7 (rev 01)
64:00.0 SATA controller: Advanced Micro Devices, Inc. [AMD] Device 43f6 (rev 01)
65:00.0 Non-Volatile memory controller: Sandisk Corp WD PC SN810 / Black SN850 NVMe SSD (rev 01)
66:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Raphael (rev c2)
66:00.1 Audio device: Advanced Micro Devices, Inc. [AMD/ATI] Rembrandt Radeon High Definition Audio Controller
66:00.2 Encryption controller: Advanced Micro Devices, Inc. [AMD] Family 19h PSP/CCP
66:00.3 USB controller: Advanced Micro Devices, Inc. [AMD] Device 15b6
66:00.4 USB controller: Advanced Micro Devices, Inc. [AMD] Device 15b7
66:00.6 Audio device: Advanced Micro Devices, Inc. [AMD] Family 17h/19h HD Audio Controller
67:00.0 USB controller: Advanced Micro Devices, Inc. [AMD] Device 15b8