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)

When the Heavy test is run on an empty drive, the large SLC cache of the Crucial P1 enables it to deliver an average data rate that is competitive with most high-end NVMe SSDs. When the drive is full and the SLC cache's size is greatly reduced, the performance drops to well below that of a typical mainstream SATA SSD. This behavior is essentially the same as that shown by the Intel 660p.

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

The empty-drive run of the Heavy test doesn't push the latency of the Crucial P1 up any higher than is typical for high-end NVMe drives. Things get more interesting on the full-drive test run, where the average latency from the P1 increases by a factor of 12 and the 99th percentile latency increases by a factor of 58. The average latency from the P1 is only slightly worse than the Intel 660p, but the 99th percentile score is three times that of the 660p.

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

The average read latency of the Crucial P1 is significantly affected by whether the test is run on a full or empty drive, but even the worse of the two scores is still clearly better than what the Crucial MX500 manages. On the write side of thing, filling the drive has an almost catastrophic effect on latency, driving the average up by a factor of 17, an even more severe impact than the Intel 660p shows.

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

The 99th percentile read latency from the Crucial P1 is typical for high-end NVMe SSDs for both the full and empty drive test runs. The high overall 99th percentile latency is due entirely to the write portion, where filling the drive increases 99th percentile write latency by almost two orders of magnitude.

ATSB - Heavy (Power)

The total energy consumption of the Crucial P1 during the Heavy test is significantly higher than for the Intel 660p, and the difference between the empty and full drive test runs is larger for the P1 than for any other drive.

AnandTech Storage Bench - The Destroyer AnandTech Storage Bench - Light
POST A COMMENT

68 Comments

View All Comments

  • Billy Tallis - Thursday, November 8, 2018 - link

    A.) Samsung didn't send me a 970 PRO. B.) The 970 PRO is pretty far outside the range of what could be considered competition for an entry-level NVMe SSD. It's a drive you buy for bragging rights, not for real-world performance benefits. The Optane SSD is in that same category, and I don't think the graphs for this kind of review need to be cluttered up with too many of those. Reply
  • PeachNCream - Thursday, November 8, 2018 - link

    Not to be obtuse, but by price the 970 PRO is well within the range of competition for the P1 given that the 1TB 970 retails for $228 on Amazon right now and the MSRP for the 1TB P1 $220. Buyers looking for a product will most certainly consider the $8 difference and factor that into their decision to move up from an entry-level product to a "bragging rights" option given the insignificant difference in cost. Your first point is valid. I would have stopped there since its reasonable to say, "Physically impossible, don't have one there pal." Reply
  • Billy Tallis - Thursday, November 8, 2018 - link

    I think you were looking at the price for the 1TB 970 EVO. The 1TB 970 PRO is currently $392.99 on Amazon, closer to twice the price of the Crucial P1. I think it is occasionally reasonable to get something like the 970 EVO for a high-end system. Going past that to a 970 PRO isn't reasonable. Reply
  • PeachNCream - Thursday, November 8, 2018 - link

    Whoops, you're correct! Please accept my apologies for that one. Reply
  • DigitalFreak - Friday, November 9, 2018 - link

    Even then, the 970 EVO wipes the floor with the P1. Reply
  • Death666Angel - Thursday, November 8, 2018 - link

    Who wants QLC NVME drives as the first widely available consumer QLC tech? Not me I tell you! :D
    I am fine with 3D TLC for my performance needs both from a performance and price point at the moment. 500GB is enough for many casual enthusiasts and 1TB isn't too expensive either. I'd really like 2.5" SATA and M.2 SATA QLC for my casual media storage needs.
    Reply
  • Lolimaster - Friday, November 9, 2018 - link

    QLC is such useless product except for manufactures, they give you a WORST product for basically the same price or more than a TLC.

    MX500/860 EVO 1TB for $160-180.
    Reply
  • Lolimaster - Friday, November 9, 2018 - link

    *Edit 155-160. Reply
  • piroroadkill - Friday, November 9, 2018 - link

    QLC doesn't seem to make any sense in an M.2 PCIe NVMe format - it's just really slow compared to even a good SATA 6Gbps SSD. QLC seems to make sense in a 2.5" SATA format, with an enormous capacity. 1TB makes no sense for this shitty performance level. It needs to be there to replace larger drives. Actually, even that makes no sense for a home user - where long term retention is more important, and a hard disk is therefore more useful. QLC drives will probably come into their own at the ~4TB mark in Enterprise storage arrays as a mid-tier storage solution, with hard disks under, and MLC NAND above. Reply
  • npz - Friday, November 9, 2018 - link

    Why are these QLC drives with all their downsides still expensive?

    On Amazon
    Samsung 970 EVO 1TB = $227
    Crucial P1 1TB = $219

    Are you kidding me? Why would I ever choose a P1?
    Reply

Log in

Don't have an account? Sign up now