Performance Consistency

Starting things off, our performance consistency test saturates the drive with 4kB random writes for a full hour, with a queue depth of 32, the maximum supported by the AHCI protocol used by SATA and most PCIe drives. This puts the drive's controller under maximum stress and writes enough data to exhaust all free space and spare area on the drive. This is an unrealistic workload for any client use, but it provides a worst-case scenario for long-term heavy use, and it sheds light on how different SSD controllers behave and if their performance will hold up as they fill up.

The average of the last 400 seconds of the test gives us a steady-state IOPS rating that is usually very different from what the manufacturer specifies for a new, empty drive. We also quantify the consistency of the drive's random write performance, and provide plots of the performance over the course of the test.

Steady-State 4KB Random Write Performance

The BX200 is off to a poor start, with very low steady-state IOPS where the BX100 managed to place closer to the middle of the pack.

Steady-State 4KB Random Write Consistency

The BX200's consistency is also at the bottom of the chart, indicating that there's wide variability between its best and worst performance even after entering steady-state.

For a brief moment, the BX200 performs almost as advertised, and then for a few minutes it performs well for a budget drive, but when it runs out of cache and spare area, performance hits the floor.

Save for the periodic but infrequent excursions to 9k IOPS and 20-25k IOPS, the BX200's steady-state hovers between 200 and 700 IOPS: better than a hard drive, but not what we want to see from a SSD.

Introduction, The Drive & The Test AnandTech Storage Bench - The Destroyer
POST A COMMENT

86 Comments

View All Comments

  • Mugur - Wednesday, November 4, 2015 - link

    I must buy another BX100 until the stocks will dry... Reply
  • redzo - Wednesday, November 4, 2015 - link

    It's a disaster. I think that Crucial was better of not releasing the damn thing. What were they thinking? The performance gap is huge compared to samsung's evo. This product just shows that samsung rules TLC. All educated buyers will think twice before purchasing it even at 15/20$ lower than this. Reply
  • KAlmquist - Wednesday, November 4, 2015 - link

    What makes it a disaster is that Crucial is replacing a successful product (the BX100) with this thing. Keep the BX100 around, give the new drive a designation like AX100, market it as a budget alternative to the BX100, and the drive might be a modest success.

    When the BX100 disappears, that will leave the Samsung 850 EVO as the paradigm of the SATA SSD sweet spot, with good balance of performance, quality, and attractive pricing.
    Reply
  • paulgj - Monday, November 9, 2015 - link

    They should name it the BX50 :-) Reply
  • tabascosauz - Wednesday, November 4, 2015 - link

    The BX200 should be compared to the SP550 from ADATA. Both are budget drives, both SM2256 (which has left a much crappier first impression than the actually promising SM2246EN), both TLC, yet the SP550 does better as a budget drive. Reply
  • KAlmquist - Friday, November 6, 2015 - link

    Yup. I would be excited about the performance of the BX200 if I worked for ADATA. Reply
  • Dritman - Wednesday, November 4, 2015 - link

    This is actually embarrassing, to put this product into today's market, at any price. You cannot step backwards THIS far.

    A crap review is like one of the cheap OCZ drives, except they're understandably bad, because they're so cheap and kinda in line with the price.

    With this thing though, it's much worse than anyone could have imagined.
    Reply
  • bennyg - Tuesday, November 10, 2015 - link

    80mb/s sequential write is not that bad in a SSD. If it were made in 2009. Reply
  • TonyCL6 - Friday, November 6, 2015 - link

    The comparison chart should add Samsung 840EVO (planer 19nm TLC) and ADATA SP550 (planer Hynix 16nm TLC) to be more meaningful. It helps users understand the sustain Seq. write (after pSLC cache ran out) performance differences between planer TLC of Samsung 19nm, Toshiba 19nm, Hynix 16nm and Micron 16nm. Reply
  • Kutark - Sunday, November 15, 2015 - link

    Can someone answer a question for me. It says these are capable of roughly 500MB/s read and write, however SATA3 obviously saturates around 200. Are those figures for m.2 PCIE versions of the drive or? Cus I see them advertised for the 2.5" SATA drives and im a little confused.

    I was about to build a skylake setup as a Christmas gift to myself, however I was looking at some of the NVMe drives etc, to get better perf than SATA3 can offer, however, they're significantly more expensive per GB. So, if I can get something with a sustained 500 read/write like this, (or maybe a "pro" version, i.e. non NVME) I'd rather save the money, as realistically in a gamer situation the difference between 500mb/s and 1000 is gonna be negligible.
    Reply

Log in

Don't have an account? Sign up now