New OWC Express 4M2 enclosure performance results

Mine fine is working even if I just plug in the power cable and it is not turning off even after long time.
Does your fan work also without any Thunderbolt cable plugged in?

If no cable is plugged in, the fan is not on, no.

Let me ask a product manager for the M2 and see if there is anything I should know about it that could explain what you are seeing.

Let me ask a product manager for the M2 and see if there is anything I should know about it that could explain what you are seeing.

That would be very helpful. I must admit I really like the fact how supportive this forum is. Glad to buy OWC bundled with SoftRAID.

You are welcome!

We tried to make his forum is for all kinds of questions, and the information provided seems to be helpful to new users also.

Let me ask a product manager for the M2 and see if there is anything I should know about it that could explain what you are seeing.

I have confirmed with OWC that there is a problem with the PCb board to cause the fan to be on all the time.

I recommend you replace it under warranty. It does not appear there is a "user serviceable" option.

 

Thank you for your support with this case. I needed to send it back under warranty and I hope to get a quick replace with the new enclosure. I miss it already!

Mac:
iMac (Retine 5K, 27-inch, 2017)
3.5 GHz Intel Core i5
40 GB 2400 MHz DDR4

Drives:
4 x Samsung SSD 970 Evo Plus 2TB, FW 2B2QEXM7

Good results. We note that some SSD Blades such as the OWC, have somewhat faster write performance (up to 2GB/s), but these numbers are great.

Did you have to update your firmware so the Plus models would work?

Just purchased and setup my new 4M2 enclosure on my Mac Pro but I am disappointed with the results.

I have four WD BLACK SN750 250GB NVMe's(over 3000 MB/s read/write capable) set to RAID 0 but am not getting the throughput I would expect. My results are:

Read: 1260 MB/s
Write: 1250 MB/s

Granted I'm using thunderbolt 2 with the TB3 to TB2 adaptor and Apple's TB2 cable and don't expect to get the maximum performance since the 4M2 is built for TB3 but the TB2 interface should handle 2500 MB/s. It was mentioned in a previous post that there is overhead to account for but losing half of the throughput from overhead seems a bit high.

To test it further I moved the 4M2 to my Mac mini 2018 using it's TB3 port and OWC's supplied TB3 cable and am getting only slightly better performance. Is this to be expected when TB3 can handle 5000 MB/s? Performance via TB3 is:

Read: 1565 MB/s
Write: 1572 MB/s

Sorry, here are my results with the requested test configuration:

I doubt the WD can do 3000 MB/s.
Try creating one volume on one blade and see what you get.

Thunderbolt 2 is limited to about 1400MB/ in practice.
Thunderbolt 3, closer to 2800MB/s.

If the WD can do 3000 each, you should see that with one blade in the M2/4 enclosure on the Mac Mini 2018

Remember manufacturers always over hype theoretical throughput on buses. 50% of the bandwidth of Thunderbolt is video. With Thunderbolt 3, only a short cable can get you maximum throughput. If you have a 1 meter cable for instance, it cannot get maximum throughput.

The M.2/4 is a simple pass-through box. It will not limit performance.

I doubt the WD can do 3000 MB/s.
Try creating one volume on one blade and see what you get.

Thunderbolt 2 is limited to about 1400MB/ in practice.
Thunderbolt 3, closer to 2800MB/s.

If the WD can do 3000 each, you should see that with one blade in the M2/4 enclosure on the Mac Mini 2018

Remember manufacturers always over hype theoretical throughput on buses. 50% of the bandwidth of Thunderbolt is video. With Thunderbolt 3, only a short cable can get you maximum throughput. If you have a 1 meter cable for instance, it cannot get maximum throughput.

The M.2/4 is a simple pass-through box. It will not limit performance.

From my understanding, the way the 4M2 is setup, you cannot get over 700 MB/s per blade and would need all four to achieve the 4M2's maximum throughout. So your suggestion should show only the WD blade achieving 700 MB/s for either TB2 or TB3 regardless of whether it really can achieve 3000 MB/s?

Thunderbolt 2 is capable of 20GB/s and while its not likely to achieve that, shouldn't the 4M2 be closer to 2500 MB/s?

I really enjoy having the 4M2 but am just disappointed I can't get more out of it when it seems that I should be able to.

Badly worded on my part. Do you see the WD blades maxing out the bus, though? (700MB/s?)

I have only tested with OWC Blades, or Samsung blades. But never saw AJA above 1500MB/s over TB2. remember half of bandwidth is reserved for video. Also, TB3 only operates at full speed with the .5 meter cabling.

I originally had the WD 250 GB blades for my testing but now have WD 1 tb versions. i tested them individually and they all came back with the same results:

Write: 670 MB/s
Read: 770 MB/s

As a striped set they came with the same results as my previous striped set:

Write: 1270 MB/s
Read: 1300 MB/s

If half of the bandwidth is reserved for video then it makes sense why I'm getting these results on the TB2 port. Though I still don't understand why I would get no better than 1500 MB/s on my Mac Mini with TB3. But I haven't tested that with the new blades and will tonight but I expect the same results as before.

I did some reading up on Thunderbolt technology and have a better understanding of the technical side. Now I just have to find a way to take advantage of what I have on this Mac Pro until I get the new one. ;)

If I were to set up a RAID 1+0 with a second 4M2 with the same blades using a different TB chip on my Mac Pro, can I expect to get double the performance I'm getting now?

Also, here is some info on the Western Digital blades I'm using and their performance:
https://www.pcworld.com/article/3336866/wd-black-sn750-nvme-ssd-review.html

So I just tested the 4M2 enclosure on my Mac mini with the TB3 port and got the same result(1500 MB/s)

Then I switched out the OWC cable that was about 3 feet long and used the Apple cable that was a foot long and boy did it make a difference. So this verified the shorter cable recommendation.

3 foot cable gave me about 1500 MB/s for read and write
1 foot cable pushed 2300 MB/s for read and write

Its unfortunate that such a difference in cable length makes that big of a difference as its not practical for me to use a foot cable with my setup. But the home for this 4M2 drive won't be on the mini but rather the Mac Pro using TB2. I tested it via TB2 using a shorter cable hoping I could gain more performance(like TB3 did) but alas, I was disappointed. :(

But I'm happy with the improvement I have and look forward to getting a second 4M2 drive and utilizing Raid 1+0. Hoping it will double my bandwidth on my Mac Pro. ;)