Question *STABLE* NVMe - USB Adapter?

[cyberjedi]

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!

 

[mervincm]

I have since purchased one additional cable. With this new cable (1meter Cable Matters C<>C USB3.1G2 10Gbps CERTIFIED) and the Jeyi 2.08 firmware, my JMS583 based unit is still solid on the rear port, and still flakey on the case port. My fractal designs case offers aq front USBC 3.1G2 port, but it seems that extra cable from the motherboard to my case port makes the difference, and causes issues with my JMS583 devices.

 

[apoelistasaudi]

I have a miniitx pc that came with usb 2.0 front header. That header was later made available by manufacturer to support usb 3.0 or 3.1 (Silverstone). However one should also buy the newer cable too which was thicker and connects header on one side and motherboard on the other. So yes there is also a cable from motherboard to front usb header that should also have same specs as the usb header otherwise you are in trouble. You may need to upgrade usb header too??? Just a thought.

 

[bricoleur]

Thanks everyone who has contributed time and energy to this thread.

Interesting update: Plugable has quietly updated the chipset in its enclosure from the JMS583 to the Realtek RTL9210, though they don't mention it on the Amazon product page.

Though, at $50, it almost feels worth it to experiment with the cheap (~$100) Thunderbolt 3 drives that have started to appear.

 

[papapapapapapapababy]

i have a black FIDECO M204CP NVME with a JMS583 (dont know what version) with the latest firmware, no disconnection problems. but it doesn't like the second gen 2 port of my asus b450 prime also have a sata version M204CS with a via chipset, i get I/O device errors unless a short cable is used and its perfectly stable. also fast.

 

[BrallyO]

Interesting update: Plugable has quietly updated the chipset in its enclosure from the JMS583 to the Realtek RTL9210, though they don't mention it on the Amazon product page.

Though, at $50, it almost feels worth it to experiment with the cheap (~$100) Thunderbolt 3 drives that have started to appear.

I posted that info weeks ago! But yes, they did.

 

[BrallyO]

my JMS583 based unit is still solid on the rear port, and still flakey on the case port

That mirrors my experience with jms583 adapters. They need a lot of power, and putting any kind of USB extension cable in between is enough resistance that they will start acting flakey. I observed this behavior with both external USB 3 extension cables and front-side chassis ports which were extended from the motherboard.

 

[papapapapapapapababy]

fideco gave me an update tool with newer firmware, 2.09 but im unable to load the file ....

 

[Andalu]

fideco gave me an update tool with newer firmware, 2.09 but im unable to load the file ....

If you have the possibility to upload the file somewhere I can try it.

Same malfunction for the JMS583 chipset here on USB 3.1 Gen.2 port n.2 of my asus prime Q370

 

[papapapapapapapababy]

here you go https://we.tl/t-y7zf02DyEb

 

[Andalu]

Firmware successfully updated, thanks
Here the shots:


On the first attempt I got the "open firmware file error" message. I decided to rename the file .bin included in the zip archive to "JMS583_0209" (as visualized in the image) and using the "JMMassProd.ini" from the firmware v2.0.8 where I renamed the "FwFileName=JMS583_0208.bin" to "FwFileName=JMS583_0209.bin".
The "JMMassProd.ini" included in the zip package seems to be referred to another ProductID (0562).

The update was done in Win10 x64 with only the enclosure connected to the usb port (without mounting the nvme disk) and at the end of the operation it must be disconnected and reconnected to view the firmware updated.

Now I need some time to test the updated enclosure.

 

[papapapapapapapababy]

great, let me know how it goes, i migth give it a go, but removing the nvm is a pain * pad applied edit. my poduct id is that
USB\VID_152D&PID_0562&REV_0208
USB\VID_152D&PID_0562

 

[Andalu]

first good news: now the nvme finally goes into stand-by mode after the 5 minutes set even after using Crystaldisk Mark and CrystalDisk Info. Also the resuming from the stand-by mode is OK.

The first two consecutive performance tests (enclosure connected into the USB port n.1 to avoid problems ) are good as expected for an nvme disk even if the temperature reached 62 degrees already at the end of the second benchmark, but this is not new.

 

[Andalu]

great, let me know how it goes, i migth give it a go, but removing the nvm is a pain * pad applied edit. my poduct id is that
USB\VID_152D&PID_0562&REV_0208
USB\VID_152D&PID_0562

Is this about the Hardware ID for the sata enclosure?

 

[papapapapapapapababy]

ok so mangaed to update the firmware but using the same JMMassProd.in * since it is for my hid after all , removed the SSD, renamed to JMS583_0209 and redirected to C:\JMS583_0209 * left it there, the default path was a mess. i was greated with a pass pass and FAIL. but it seems it was refering to partition and since i had no SSD inside i guees it was that. disconnected, retested and now it detects on all gen 2 ports and all pcs! moved 50gb with no issues * a bit slower 750 vs 800... but i did it in the port it wasnt working before. also speed is great. so it worked! * it shows as rev 0209 now * edit lol i left partition an format, lucky me no ssd inside! cheked

 

[razel]

Thanks, also edited the ini file and updated my Pluggable USB-NVME to 2.09. Performance is down slightly less compared to the original 2.04 firmware perhaps even margin of error on my USB 3.1 GEN 1 (max 500mb/s) port. I updated prior it to Yeji 2.08 firmware from on an earlier wonderful post here and saw the same margin of error less perf.

IOW... Updating from the original 2.04 Pluggable firmware to 2.08 Yeji firmware, the performance is margin of error less. 2.04 to 2.09 firmware, it's the same margin of error less. I didn't compare 2.08 to 2.09. I seriously doubt it's anything significant.

I am glad I was able to change the name and serial number of my Pluggable back to the original. The 2.08 Yeji renamed it to Yeji with serial # ~9876543210. I doubt this thing is in warranty anymore after this. I'll chime in again if I encounter any issues.

After applying the firmware you can edit the EEPROM variables (drive name,etc) without reapplying the firmware, you can also have it try to use the HDD serial number and name, but didn't work for me.

 

[Andalu]

@papapapapapapapababy
I'm glad to know it worked
From the information you indicated in a previous post and also displayed in the image, it seems that your device even having a different Hardware ID (0562) can use the same firmware version for the ID 583. Good to know.

 

[Andalu]

@razel
I tried downgrading the firmware some time ago to check the performance drop I had noticed too updating from the version 2.0.4 to the 2.0.8 but I couldn't do it. I must admit, however, that the chance to get standby mode thanks to the latest firmware versions is better than an extra performance, considering the temperature problem of the JMS583 chipset.

 

[razel]

I probably shouldn't mention the performance drop. It's within margin of error and I re-ran the benches today this time on two different computers that I had prior 2.04 benchmarks saved and some are higher than 2.04 some are lower. I'd just flag it as the same.

However, I am limited by my USB 3.0 ports. I do not have any 1GB/s ports. All of mine are 500GB/s. I think those with 1GB/s are the ones having various issues which appear to be related to compromised or more likely cheap cables.

 

[Charlie22911]

I don’t believe the issues are power related. My dual 2.5” bus powered raid enclosure draws more power than this and doesn’t have issues.
Photo attached is power draw while doing crystal disk mark run. Cable is Apple thunderbolt 3, attached to the 10Gbps/thunderbolt 3 port on my Aero 15x.

EDIT:
Attached another image of the aforementioned dual 2.5 enclosure, also running crystal disk mark. My JMS583 enclosure will *not* run at all on this same port. Even though (as you can see from the other image) it draws much lower power (4.13w vs 7.13w).

 

[Andalu]

I decided to stress my JMS583 polycarbonate enclosure (to which I added 3 thermal pads) with the firmware 2.0.9 on the USB3.1 gen.2 port number two of my asus Q370 board and for the first time even after 4 consecutive benchmarks, I didn't recognize any performance degradation even if the temperature reached 71 degrees in the last test. It had never happened before, there was always a loss of performance within the second consecutive test, especially when connected to that specific port:

 

[apoelistasaudi]

Please confirm that standby works with firmware 2.0.9 even after running the crystaldiskmark

 

[papapapapapapapababy]

@papapapapapapapababy
I'm glad to know it worked
From the information you indicated in a previous post and also displayed in the image, it seems that your device even having a different Hardware ID (0562) can use the same firmware version for the ID 583. Good to know.

lol im confused.. its seems my Nvme case M204CP model uses a sata chipset? the JMS567 6gb instead of JMS583 ? * it claims uses ??? btw my seq read writes are 10Gbps

 

[Andalu]

Please confirm that standby works with firmware 2.0.9 even after running the crystaldiskmark

Already reported here

 

[Andalu]

@papapapapapapapababy
it could also be a wrong ID assignment, which is infrequent but that I have already seen few days ago with my PCIEx to USB 3.1 Gen.2 card with the ASM3142 chipset which is recognized in windows with the ID 2142 instead of 3142.
In your case you can get the answer from this: the disc you insert in the enclosure has an "M" key, "B" key or a "B+M" key connector?

 

[papapapapapapapababy]

@papapapapapapapababy
it could also be a wrong ID assignment, which is infrequent but that I have already seen few days ago with my PCIEx to USB 3.1 Gen.2 card with the ASM3142 chipset which is recognized in windows with the ID 2142 instead of 3142.
In your case you can get the answer from this: the disc you insert in the enclosure has an "M" key, "B" key or a "B+M" key connector?

this is the ssd im using. litteraly the one they use !!!




edit. a bit newer, mine has the nand on the other end