Question *STABLE* NVMe - USB Adapter?

Page 39 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

cyberjedi

Junior Member
Apr 4, 2010
23
13
81
I'm looking to buy a *RELIABLE* NVME-USB external adapter, that doesn't disconnect randomly, can keep up the close to 1GB/s speed.
So far all adapter's reviews contain frighteningly high % of reports of malfunction, random disconnects, sub-par speeds, falling to even under USB 2.0 speeds, etc.
There are 3 main chipsets in the market: JMicron JMS583, Asmedia ASM2362, and Realtek RTL9210 (no product has surfaced with the latter yet, so no REAL reviews, experiences yet UPDATE: it is on the market already).
Also ASUS has a new product, ROG Strix Arion, but there isn't enough feedback about it yet, and I mean real life feedback not just being able to run a few benchmarks, but real stress test to see if the device (adapter) can keep it up, without disconnecting or slowing down hard.
I would really appreciate any feedback from those who have GOOD experiences during stress-tests, like tons of random 4k writes with a stable speed, cloning 100's of GB-s at a stable high speed close to 1GB/s.
I know SSD's very well, so please stick to the point and don't start to educate me on how their speed falls after some writes, etc. I'm aware and this is NOT the point.
You can help if you can point me to a STABLE and RELIABLE NVMe-USB-C adapter that WORKS as expected.
Thank you in advance!
 
Last edited:

Ananas

Junior Member
Jan 16, 2022
24
6
41
the Realtek chip has to be connected through a USB hub chip. Maybe that interaction is causing some data corruption at high speeds.
That is a good theory indeed. Also in my case the I/O errors occurred with an USB hub inbetween (MacBook Pro → Thunderbolt hub → USB hub → RTL9210B → NVMe SSD).
 

LuckMan212

Member
Jan 21, 2005
30
0
66
Anyone got any news of a more recent firmware for JMicron JMS583 chips? I'm using v02.01.02 with a Mac M1, occasionally the thing will just go brain-dead (slow, freeze on access, or just fail to mount) until I unplug it for a few seconds and then plug it back in.
 

ApplesOfEpicness

Junior Member
Nov 16, 2022
13
9
36
That is a good theory indeed. Also in my case the I/O errors occurred with an USB hub inbetween (MacBook Pro → Thunderbolt hub → USB hub → RTL9210B → NVMe SSD).
Well I ended up getting this SSD Hub: https://www.amazon.com/gp/product/B09WHXVXZK/ref=ppx_yo_dt_b_asin_title_o01_s00?ie=UTF8&psc=1

Testing it with the same SSD, the same file, and even the same cable has yielded nothing. There are no issues and everything transfers just fine. ACASIS better have a good explanation for this because if they don't a nice bomb of a review is coming soon.
 
  • Like
Reactions: Ananas

fantom3l

Junior Member
Mar 22, 2020
15
32
91
Good day. Today i came across a new firmware for RTL9210B - v1.30.21.070622 so i have made a new package with all in wonder, custom configs, expalined ones and also the last firmware for RTL9210A. What's new, i have no ideea.You will find also updated MPtool. You can download it from next link.
https://www.sendbig.com/view-files/?Id=78770afe-a795-36e7-d2eb-4ad3d8dc9d6e
 
Last edited:

admanny

Junior Member
Nov 22, 2022
3
0
11

R81Z3N1

Member
Jul 15, 2017
77
24
81
I just installed an Orico M2PV-C3, like the enclosure will keep an eye on it. It was used to tame a unruly Samsung Evo 256G drive that gives Linux error logs, due to bad firmware.

Every time I would boot, it would increase the error count by one, will not touch another Samsung drive. Was up to 546 error count logs even if it out classes the Orico as far as I am concerned it deserves the downgrade.

Anyone know when a good USB4 enclosure will come along, yes I know about the Thunderbolt 3 devices but since USB4 is newer, and competes with Thunderbolt 4 I figured that would be the newness folks are after.

R81Z3N1
 

Ananas

Junior Member
Jan 16, 2022
24
6
41
Anyone know when a good USB4 enclosure will come along, yes I know about the Thunderbolt 3 devices but since USB4 is newer, and competes with Thunderbolt 4 I figured that would be the newness folks are after.
Thunderbolt is multiplexed and serialized PCIe, and NVMe SSDs are PCIe, too, so I guess Thunderbolt — no matter whether version 3 or 4 — should always be a better (more direct = simpler = less error prone) way to connect them than USB.
 
  • Like
Reactions: bsly1314

saddy

Member
Oct 11, 1999
146
0
76
I just installed an Orico M2PV-C3, like the enclosure will keep an eye on it. It was used to tame a unruly Samsung Evo 256G drive that gives Linux error logs, due to bad firmware.

Every time I would boot, it would increase the error count by one, will not touch another Samsung drive. Was up to 546 error count logs even if it out classes the Orico as far as I am concerned it deserves the downgrade.

Anyone know when a good USB4 enclosure will come along, yes I know about the Thunderbolt 3 devices but since USB4 is newer, and competes with Thunderbolt 4 I figured that would be the newness folks are after.

R81Z3N1

Not sure if it would help, but this firmware might be worth checking out.
 

forart.it

Junior Member
Jul 2, 2022
5
3
41
The funny thing about the ICY BOX IB-1817M-C31 is that while it does indeed have a JMS583 chip on the PCB (I own one and just had a look) it identifies via USB as 0x152d:0x0562, i.e. it claims to be a JMS562. This might be the reason for the untypical firmware.


Where did you get this from?
Hi there, we just buyed a Kingston SNV2S/2000G as external storage drive (dedicated to ATEM Mini Extreme ISO recordings) and here's our speed tests' results:

PCI Express x16 slot (5.0)ICY BOX IB-1817M-C31 @USB 3.2 Gen 2x2 port
NV2.pngNV2_usbC.png
RIG: Intel i5-12600 @ Gigabyte Z690 UD DDR4 (rev. 1.x) / latest BIOSRAM: 4 x Ripjaws V F4-4000C18-8GVK @ 3600/16-20-20-36 CR1
Last but not least, here's the latest official IB-1817M-C31 firmware update screen:
ICY.png
...can someone tell if there's a suitable fw upgrade that can improves compatibility/performances ?
 
Last edited:

Ananas

Junior Member
Jan 16, 2022
24
6
41
...can someone tell if there's a suitable fw upgrade that can improves compatibility/performances ?
If it works so well, why bother? The enclosure only supports Gen 2 @ 10Gbit/s (and the ATEM Mini Extreme ISO apparently only Gen 1 @ 5Gbit/s according to the specs page you provided the link to), so the transfer speeds you're seeing are as good as it gets.
 

White7561

Junior Member
Apr 17, 2020
20
2
51
I've used Asus Strix Arion enclosure (ASM2362) for years now and it's been nice and stable for me (used in old MBA 2014 with Intel and MBP 2021 with M1 Pro chip). Now I'm getting a new enclosure and it says that it supports both NVMe and SATA which could only mean RTL9210 right? It's still on the way so I can't physically check. Anyways, if it is RTL9210, does anyone know the latest safe firmware version to flash? Since I've heard a few people bricking their enclosure flashing RTL9210 firmwares so just wanna be careful. Thank you!

I would probably upgrade just to make sure it has bugs fixed and to disable the sleep timer.
 

STSC

Junior Member
Jan 17, 2023
1
0
6
Is there also a stable M.2 SATA case on the market, which can be used 24h for a NAS. I only found a case with a JMicron JMS579 (Icy Box IB-183M2), but no cases with e.g. Asmedia ASM1351 or ASM225CM. Most time for M.2 SATA cases the RTL9210 is used, which supports both M.2 SATA and M.2 NVMe SSDs.

Can the RTL9210 also be recommended for M.2 SATA SSDs?
 

peroko

Junior Member
Jun 8, 2021
15
5
41
Is there also a stable M.2 SATA case on the market, which can be used 24h for a NAS. I only found a case with a JMicron JMS579 (Icy Box IB-183M2), but no cases with e.g. Asmedia ASM1351 or ASM225CM. Most time for M.2 SATA cases the RTL9210 is used, which supports both M.2 SATA and M.2 NVMe SSDs.

Can the RTL9210 also be recommended for M.2 SATA SSDs?
You need RTL9210B variant for that.
 
Jan 4, 2022
30
6
51
I've used Asus Strix Arion enclosure (ASM2362) for years now and it's been nice and stable for me (used in old MBA 2014 with Intel and MBP 2021 with M1 Pro chip). Now I'm getting a new enclosure and it says that it supports both NVMe and SATA which could only mean RTL9210 right? It's still on the way so I can't physically check. Anyways, if it is RTL9210, does anyone know the latest safe firmware version to flash? Since I've heard a few people bricking their enclosure flashing RTL9210 firmwares so just wanna be careful. Thank you!

I would probably upgrade just to make sure it has bugs fixed and to disable the sleep timer.

To be fair, if you don't encounter issues, a firmware update may not be necessary.
 
Jan 4, 2022
30
6
51
2 years longtime no see this thread, I have RTL9210 all working stable. no touch firmware update. no headache.
I still recommended Realtek chipset. :grin: also flashdrive style enclosure no mess up with cable quality.

View attachment 73407

Surprised to see the Orico holding up even after 2 years. Which PCB revision do you have?

And judging from the Philips screw, I am guessing it's the older model? Both of mine (original and RMA unit) have a torx head.
 

somemorestuff

Junior Member
Jan 25, 2023
1
0
6
Anyone got any news of a more recent firmware for JMicron JMS583 chips? I'm using v02.01.02 with a Mac M1, occasionally the thing will just go brain-dead (slow, freeze on access, or just fail to mount) until I unplug it for a few seconds and then plug it back in.

That file actually has several ROMs, including that v208 through v212 (for apparently a JMS583 v234 variant?). I just successfully burned v212 to my JEYI A3 enclosure & seems to be working fine on Win 10 (although I can't test the full speed due to my USB 3.1 Gen 1 motherboard limitation). Temperature profile is roughly the same in CrystalDiskInfo running CrystalDiskMark on a Hynix BC501 512GB (~39°C while reading vs. ~45°C during write - and quite warm to the touch). With 1 min firmware timeout it does "cool down" slowly to <35°C if you set the auto refresh on CrystalDiskInfo to something longer than 1 minute, but it did so on v209 as well - still quite warm to the touch. For comparison, my 970 Evo Plus 2TB with a ridiculous heatsink "rests" at 29°C (yes, yes, warms up nicely to 40-45°C on writes).

NOTE: stationdrivers now has an v213, the image is actually dated 4/2022 so I suspect there's newer somewhere. It doesn't seem to be targeted to a v234 chipset variant either. Tested it out with the same or possibly slightly cooler results. No issues with connection stability, but haven't tried running it for "days" or anything.

The thing that eats me is apparently the JMS583 doesn't respond to a "Data Set Management" Command properly, so CrystalDiskInfo & Intel MAS tool can't read the parameters to report that TRIM & Volatile Memory (& lord knows what else) is actually supported with this drive. TRIMCheck (on github) does "physically" verify TRIM is working though. SMART works fine of course.

I like the compact size, all aluminum case, aluminum bottom (for long double-sided drives), custom labels (with the firmware updater), don't plan to run an OS off of it, & it was only $18 - so I think I'll keep it!
 

eddieobscurant

Junior Member
Oct 19, 2015
15
5
81
I had continuously, fast (after a couple of days) data corruption with firm. 02.01.02 which made the drive unusable. It wasn't like that a few months ago when i installed the firmware. I installed ver. 02.01.03 from station drivers, but i'm starting to believe that the jms583 controller is starting to die.
 
Jan 4, 2022
30
6
51
This is a mild annoyance, as my Realtek-based Asus occasionally gives me the 100% activity time issue. Not sure if it's sufficient grounds for a warranty claim?

Edit: The drive in question is a Toshiba 512 GB 2242 drive. Power-on time is roughly 10 days, meaning it's a relatively new drive. From what I observed so far, it only happens when the regular USB-A to USB-C cable is used. It seems fine when C-to-C cable is used.

Edit 2: After much checking and troubleshooting, it appears that the issue occurs under 2 very specific conditions.

1. Primarily when the USB-A to USB-C cables is used. Issue isn't apparent when USB-C to USB-C cable is used
2. It appears that this particular drive (Kioxia Toshiba KBG40ZNT512G) is very finicky with Realtek controllers. When paired with my faithful Strix Arion and its ASM2362 controller, the issue isn't there.
 

Attachments

  • photo_2023-01-31_22-21-07.jpg
    photo_2023-01-31_22-21-07.jpg
    62.9 KB · Views: 38
Last edited:
  • Wow
Reactions: igor_kavinski

PRSXFENG

Junior Member
Jun 25, 2022
14
3
41
This is a mild annoyance, as my Realtek-based Asus occasionally gives me the 100% activity time issue. Not sure if it's sufficient grounds for a warranty claim?

Edit: The drive in question is a Toshiba 512 GB 2242 drive. Power-on time is roughly 10 days, meaning it's a relatively new drive. From what I observed so far, it only happens when the regular USB-A to USB-C cable is used. It seems fine when C-to-C cable is used.

Edit 2: After much checking and troubleshooting, it appears that the issue occurs under 2 very specific conditions.

1. Primarily when the USB-A to USB-C cables is used. Issue isn't apparent when USB-C to USB-C cable is used
2. It appears that this particular drive (Kioxia Toshiba KBG40ZNT512G) is very finicky with Realtek controllers. When paired with my faithful Strix Arion and its ASM2362 controller, the issue isn't there.
in my experience using USBA to C cables for high speed things like these enclosures was always a bit finicky and buggy
but I have only ever used realtek ones
 
Jan 4, 2022
30
6
51
in my experience using USBA to C cables for high speed things like these enclosures was always a bit finicky and buggy
but I have only ever used realtek ones

Barely a day after these symptoms presented itself, the Kioxia drive nuked itself, rendering all my files unrecoverable.

It seems fine after formatting, but I wont be trusting that drive with anything important