AnandTech Storage Bench - Light

Our Light storage test has relatively more sequential accesses and lower queue depths than The Destroyer or the Heavy test, and it's by far the shortest test overall. It's based largely on applications that aren't highly dependent on storage performance, so this is a test more of application launch times and file load times. This test can be seen as the sum of all the little delays in daily usage, but with the idle times trimmed to 25ms it takes less than half an hour to run. Details of the Light test can be found here. As with the ATSB Heavy test, this test is run with the drive both freshly erased and empty, and after filling the drive with sequential writes.

ATSB - Light (Data Rate)

The Crucial SSDs occupy the bottom half of the average data rate rankings for the Light test, as the other 3D NAND SSDs in this bunch are able to deliver higher peak performance. The BX300 is slower than the MX300 when the test is run on an empty drive, but for a full drive the BX300 is the fastest Crucial SSD and also faster than the ADATA SU800.

ATSB - Light (Average Latency)ATSB - Light (99th Percentile Latency)

The average and 99th percentile latency scores for the BX300 are worse than the other 3D NAND SSDs in this comparison, but there's enough of a gap for it to matter.

ATSB - Light (Average Read Latency)ATSB - Light (Average Write Latency)

The average read latency of the Crucial BX300 on the Light test is better than any other Crucial drive, but is unimpressive compared to the 3D NAND SSDs from other brands. The average write latency is significantly higher than most of the other SSDs (excepting the BX200), but is not enough to cause real problems for light workloads.

ATSB - Light (99th Percentile Read Latency)ATSB - Light (99th Percentile Write Latency)

The 99th percentile read and write latencies tell pretty much the same story as the averages for the BX300: it performs fine for read operations, but is a bit slower for writes.

ATSB - Light (Power)

The Crucial BX300 turns in another second-place score for power efficiency, behind the MX300. The Light test doesn't put too much stress on the MX300's SLC caching, so it keeps its first-place efficiency even when the test is run on a full drive.

AnandTech Storage Bench - Heavy Random Performance
Comments Locked

90 Comments

View All Comments

  • BrokenCrayons - Wednesday, August 30, 2017 - link

    You're right, I don't know every aspect of every engineering problem that exists. You don't either and, on a much smaller scale, you probably didn't read my comment closely enough to understand every aspect of it before you mistakenly assumed I'd adopted a particular viewpoint. If you read closely, you'll see we're attempting to make the same point.
  • Alexvrb - Wednesday, August 30, 2017 - link

    If you agree that you'll never burn out a 3D TLC equipped drive, then they are both effectively equal on that front. If the MLC-equipped drives are worse performing, you've paid the same for a slower product. I think it might be more reasonable to ask why anyone would bother with a product which is saddled with an inferior controller just because it has a sticker than says "MLC!!!!oneone1eleven".
  • BrokenCrayons - Wednesday, August 30, 2017 - link

    Well, I hate to sink your ship, but they're not worse performing. :)
  • plopke - Tuesday, August 29, 2017 - link

    thats the thing, it is priced cheap enough, if you are a enthusiast , you might have a M.2 PCIe4 drive , but I could see myself adding a BX300 to expand storage. It just looks like great bang for buck not?
  • MajGenRelativity - Tuesday, August 29, 2017 - link

    It does seem like it is priced well
  • Alexvrb - Wednesday, August 30, 2017 - link

    Except at 1TB-2TB range. Then the ADATA SU900 and MX300 are worth a look. Especially if you're using it for secondary storage only.
  • Lolimaster - Tuesday, August 29, 2017 - link

    NVME basically offers you higher max transfer which helps when working/and or moving huge files.

    What most people actually complain about TLC is sustained transfers when the SLC cache depletes.
  • bug77 - Wednesday, August 30, 2017 - link

    Motherboards have a lot more SATA connectors than they have NVMe. That may have something to do with it.
  • doylecc - Friday, November 3, 2017 - link

    Ever heard of RAID??? It is frequently less expensive to use two or three inexpensive SATA SSDs in RAID 0 to achieve transfer rates comparable to the very expensive NVME drives. Most motherboards only have one M.2 slot, so you can't RAID the NVME M.2s. Older motherboards that lack an M.2 slot can still support SATA RAID setups.

    On an older AMD motherboard, the SATA controller maxed out with 3 SSDs in RAID 0 (over 1200 MB/S). When I added a fourth SSD, performance actually declined in some tests, so I figure the controller was saturated.
  • ImSpartacus - Tuesday, August 29, 2017 - link

    Did I miss the performance consistency section? I always like that.

Log in

Don't have an account? Sign up now