Tested: I love the exceptional long writes of the team group C47 SSD

Best prices today: TeamGroup C47 NVME SSD

PCIE 4.0 HMB NVME SSDS recently flooded our test portals. And although I would like to say that the T-Create C47 team group is one of the best of them, it has been tested under Par in synthetic benchmarks and writing our 450 GB.

That said, it continues long writes exceptionally good, and also finished second with PCIE 4.0 HMB discs in our 48 GB transfer tests. For the right price (cheaper than it was at the time of this writing), it is worth your short list.

Read on for more information, see our Roundup of the best SSDs for comparison.

What are the functions of the T-Create C47 team group?

The T-Create C47 is a 2280 (22 mm wide, 80 mm long) M.2, PCIE 4.0 X4 NVME SSD with a Maxio MAP1602A-F3C controller and 232-layer TLC Nand. It is a host memory buffer design that uses part of the memory of your PC as a primary cache.

Secondary caching, as with all NVME SSDs (Dram or HMB), is done by writing a certain part of the Nand as SLC (Single Layer/1-Bit) and then transfers the time in the background that allows time.

Writing a single bit per cell transfers the need for extensive error control to see if a multi-bit cell is written on the right tension. It is the voltage level in a cell with multiple levels that determine the numeric value. SLC is charged or not – ie on or off.

The T-CREATE C47 has a variable, but still fairly generous TBW (terabytes that can be written before only read mode) rating of 500 TBW for the 512 GB model, 1,000 TBW for the 1TB capacity, 1,500 TBW for the 2TB version and 3,000 TB for the 4TB version. The standard for TLC is only 600 TBW per capacity. TeamGroup also guarantees the ride for five years.

How much is the C47 team group?

At the time of this writing, the 1 TB T-Create was $ 75 on Amazon, and the 2TB version I tested was $ 120. The capacities of 512 MB and 4 TB were not available at the time of this writing.

In comparison with his colleagues, $ 75 is a bit high for a 1 TB, but $ 120 is perfect for a 2TB PCIE 4.0 HMB SSD – there is currently no amazing price difference in this class in this class between brands between brands.

How fast is the T-Create C47 team group?

The C47 will do the job, perhaps not as fast as other PCIE 4.0 host memory buffer SSDs. In general it is fine, even in the real world, but there are a few eye-opening dips in the synthetic results.

The most enigmatic result for the C47 was the random writing score of 32-queue. HMB SSDs are generally great at this. Both the competing WD SN5100 and the Lexar Play 2280 are PCIE 4.0 HMB designs and were considerably faster in this test. On the other hand, reading was about average. Go figure.

Although it is not an aces in our synthetic benchmarktests, the C47 did pretty well in our real-world transfer tests. Note that Fastcopy shows the real potential of the SSD. I still include Windows Explorer results, because those are what users will encounter the most, but I strongly advise you to download that more than worthy utility for large file transfers. It is free for private use.

Again, the C47 was not at the pace to write our test file of 450 GB. Not terrible, but noticeable. But that news is not nearly as bad as it seems …

… because, although the above performance may not indicate that way, the C47 is actually one of the best budget -SDs that I have tested for writing large quantities of continuous data. How so?

The 2TB drive that I tested did not delay noticeable (2.45 Gbps to 195 Gbps) during a writing of 950 GB – do most discs in this class.

It even took an extra writing of 450 GB before the writing speed fell to 500 Mbps to 800 Mbps around the 160 GB marking. It remained there for the rest of writing. In the past I have seen much worse native write rates (writing on the full bit depth of the Nand). This is on the same footing with a SATA SSD where once TLC/QLC -Nativeive writes was slower than a hard drive. Ah, progress!

Okay, so the C47 did not take the blue ribbon home in a test, except for a non-head of results. Not fully refueling during Lange writes, bears a lot of weight in our evaluation. Where the C47 is slowly reading and writes with queues – does not matter as much as not much software, including Windows, does not use queues. Unfortunately.

Do you have to buy the T-Create C47 team group?

Although the C47 is one of the slower PCIE 4.0 -host memory buffer drives in synthetic benchmarks that I have tested, I was certainly impressed by the persistent native writing speed.

Anyway, the prize is the Prime-Determinator in late model PCIE 4.0 HMB SSDS. If it is cheaper than the competition, go for it. You will never notice the difference. If it is not cheaper, go with something else like the WD Black SN7100/Blue 5100 or Lexar Play 2280.

Note: There is no relationship with the famous C-47 cargo plane that I am conscious except transporting the cargo.

How we test

Drive tests are currently using Windows 11 24H2, 64-bit that is performed from a PCIE 4.0 Samsung 990 Pro in a Asus Z890 maker WiFi (PCIE 4.0/5.0) Mother’s sign. The CPU is a core Ultra i5 225 food/fed by two Kingston Fury 32 GB DDR5 4800MHZ modules (64 GB memory total). Both 20Gbps USB and Thunderbolt 5 are integrated and Intel CPU/GPU images are used. PCIE 5.0 SSDs involved in testing are mounted in an ASUS ROG PCIE 5.0 M.2 Adapter card and a HighPoint 7604A 16x PCIE 5.0 4-Poort M.2 Adapter card.

We voeren de Crystaldiskmark 8 (en 9) uit, zoals SSD 2, en Atto 4 synthetische benchmarks (om de artikellengte laag te houden, we rapporteren er slechts één) om de potentiële prestaties van het opslagapparaat te vinden, vervolgens een reeks van 48 GB en 450 GB transfers tests met behulp van Windows Explorer Drag and Drop om te laten zien wat u ziet onder venster, As well as the distant xcopy and fastcopy to show what is possible.

A 25 Gbps two-SSD RAID 0 array on the aforementioned HighPoint 7604A is used as the second disk in our transfer tests. In the past, the 48 GB tests were performed with a ram disk.

Each test is performed on an NTFS-formatting and newly trimmed drive, so that the results are optimal. Note that with normal use, if a drive fills, performance can decrease due to less nand for secondary caching, as well as other factors. This issue has somewhat decreased with the current harvest of SSDs with more mature controllers and much faster nand in the late generation.

Caveat: The performance numbers shown only apply to the disk that we have been sent and to the tested capacity. SSD performance can and will vary per capacity as a result of more or fewer chips for shotgun reads/writes about and the amount of nand available for secondary caching. Sellers also exchange components occasionally. If you ever notice a great discrepancy between the performance you experience and what Wone Report it with all means, let us know.