AnandTech Storage Bench - Heavy

Our Heavy storage benchmark is proportionally more write-heavy than The Destroyer, but much shorter overall. The total writes in the Heavy test aren't enough to fill the drive, so performance never drops down to steady state. This test is far more representative of a power user's day to day usage, and is heavily influenced by the drive's peak performance. The Heavy workload test details can be found here. This test is run twice, once on a freshly erased drive and once after filling the drive with sequential writes.

ATSB - Heavy (Data Rate)

The Crucial MX500 turns in the best average data rate score on the Heavy test that we've seen from a 1TB-class SATA drive. When the test is run on a full drive, the MX500's performance falls below that of the Samsung 850 PRO and 850 EVO.

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

The average and 99th percentile scores of the Crucial MX500 are typical for a good SATA SSD. The MX500 doesn't set any records here, but at least the latency doesn't climb out of control when the test is run on a full drive. This is a notable improvement over the MX300.

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

The average read latency of the Crucial MX300 is fast by SATA standards, but it doesn't quite match the Samsung 850 PRO. When the test is run on a full drive, the average read latency suffers and the MX500's score is merely average for a mainstream 1TB drive. The average write latency is slightly below average in both cases, but not to a degree worthy of concern.

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

The 99th percentile read latency of the MX500 falls in the middle of the pack, though running the test on a full drive has a bit more of an impact than for most drives. The 99th percentile write latency is reasonably low whether or not the test is run on a full drive.

ATSB - Heavy (Power)

The power consumption of the Crucial MX500 on the Heavy test is significantly higher than the record-setting MX300, but the MX500 certainly doesn't qualify as power-hungry compared to the broader field of competitors. The Samsung 850 PRO and EVO drives require much more power on this test than the MX500.

Introduction AnandTech Storage Bench - Light
Comments Locked

90 Comments

View All Comments

  • StrangerGuy - Tuesday, December 19, 2017 - link

    Hey let's pay >3x/GB for SLC so we can have extra drive endurance where TLC already doesn't matter for 99.999% of the target market, of course he's the smart one and rest of us are teh dumbs.

    10TB written over 3.5 years on my M550 1TB, yup can't wait for the drive to finally die in 300 years if it was *only* 1000 writes/bit TLC so I can buy a new one. Because TLC endurance sux amirite?
  • PeachNCream - Tuesday, December 19, 2017 - link

    TLC is pretty good these days compared to planar MLC from 2012. You probably wouldn't have anything to worry about since a drive like the MX500 would be obsolete before endurance becomes a problem. But if you're worried, there's the BX300 that's still for sale like MajGenRelativity suggested. Mushkin was also selling a 3D MLC drive, the Reactor Armor3D that was released in January-ish of this year. They have a 1TB model available and it uses the same Silicon Motion SM2258 controller so you'd probably see similar performance.

    https://www.anandtech.com/show/11035/mushkin-launc...
  • ddrіver - Tuesday, December 19, 2017 - link

    BS. This study proved SLC and MLC aren't actually meaningfully different. By inference TLC is good enough for almost anything. If you're the kind of person who can't do with TLC it's either overstating your usage scenarios or you're in the NAND destruction business.

    https://www.usenix.org/conference/fast16/technical...
  • mode_13h - Tuesday, December 19, 2017 - link

    I wouldn't get so hung up on # of bits per cell. What matters is performance, write endurance, and (for some use cases) power-off data retention.
  • mode_13h - Tuesday, December 19, 2017 - link

    I'm really impressed with the QD1 random read performance of this drive. I didn't think you could achieve so many QD1 IOPS with SATA, or else Samsung probably would've done it.
  • DoveOfTheSouth - Wednesday, December 20, 2017 - link

    The QD1 results are impressive in the tables but the charts seem to show lower figures (QD1 read: 44.7 v c. 35; write 164.2 v c. 146). Which is correct?
  • StrangerGuy - Tuesday, December 19, 2017 - link

    The only thing that really matters is the 4K QD1 benchmark and that was extremely impressive for the asking price.
  • shatteredx - Thursday, January 11, 2018 - link

    Agreed. MX500 hard to beat for the price now.
  • LordConrad - Wednesday, December 20, 2017 - link

    "With the MX500 arriving at $259.99 for the same capacity but with a longer warranty..."

    The Samsung 850 EVO and Crucial MX500 both have a 5-year warranty.
  • diceman2037 - Thursday, September 2, 2021 - link

    Unfortuantely, what this review couldn't account for was these SSDs have a serious defect where uptimes correlate to significant increases of write amplification in comparison to their older mx300 siblings and not fit for use as an OS drive.

    https://forums.tomshardware.com/threads/crucial-mx...

Log in

Don't have an account? Sign up now