Conclusion

The Crucial P1 SSD doesn't leave us particularly excited about the near-term prospects for QLC NAND in the consumer SSD market. The P1 is a decent entry-level NVMe SSD for less intensive workloads, but the wide gap between its best and its worst performance means the P1 comes with more caveats than most of its competition.

The Crucial P1 relies very heavily on its SLC cache to provide high performance, and that cache shrinks as the drive fills up. When the drive is full, a very write-heavy workload can overflow the cache and severely impact write speeds and even read performance to some extent. However when the drive isn't close to full, it is nearly impossible to fill the SLC cache with a realistic workload, and the Crucial P1 performs at least as well as any other entry-level NVMe drive, and sometimes rivals high-end NVMe drives. Compared to the Intel 660p, the SLC cache on the Crucial P1 seems to be more write-oriented and is not as good at accelerating read operations. It seems like the P1 may be a bit quicker to evict data from the cache and compact it into QLC blocks.

Overall the Crucial P1 is primarily aimed at consumer machines, and that definitely seems like the segment it's best suited for. A typical consumer use case would involve most of the large data on the drive coming from things like movies and video games that are rarely modified, as opposed to workstation workloads that generate massive files that constantly change. This is helpful to the P1 because it reduces the actual amount of writing the drive needs to do, though it does mean that the drive's variable-size SLC cache could end up quite small. On balance, even that small cache should be adequate given the limited amount of data that does change with most consumer workloads; though to be sure, overflowing the SLC cache is something that would be far more noticeable on the P1 than most TLC-based SSDs. But it is still not something that will happen to most consumers often enough to worry about.

That leaves the Crucial P1 as usually being very fast, and definitely faster overall than any SATA SSD. The use of QLC NAND doesn't cripple the drive, and is a detail that most consumers don't have to care about. Even at its worst, the P1 is still faster and more efficient than a mechanical hard drive. NVMe SSDs should aspire to more than that, but this will probably be true even of QLC SATA drives as long as they also avoid the low capacity points where high performance is impossible.

NVMe SSD Price Comparison
  240-280GB 480-512GB 960GB-1TB 2TB
Crucial P1   $109.99 (22¢/GB) $219.99 (22¢/GB) Coming Soon
Intel 660p   $99.99 (20¢/GB) $189.99 (19¢/GB) $349.99 (17¢/GB)
MyDigitalSSD SBX $54.99 (21¢/GB) $94.99 (19¢/GB) $219.99 (21¢/GB)  
Kingston A1000 $56.99 (24¢/GB) $97.99 (20¢/GB) $219.99 (23¢/GB)  
MyDigitalSSD BPX Pro $74.99 (31¢/GB) $129.99 (27¢/GB) $259.99 (27¢/GB) $519.99 (27¢/GB)
ADATA XPG SX8200 $62.99 (26¢/GB) $107.99 (22¢/GB) $214.99 (22¢/GB)  
HP EX920 $73.99 (29¢/GB) $119.99 (23¢/GB) $199.99 (20¢/GB)  
WD Black (2018) $85.99 (34¢/GB) $138.46 (28¢/GB) $259.75 (26¢/GB)  
Samsung 970 EVO $87.90 (35¢/GB) $147.99 (30¢/GB) $227.99 (23¢/GB) $577.95 (29¢/GB)
SATA Drives:        
Crucial MX500 $52.99 (21¢/GB) $84.95 (17¢/GB) $154.99 (15¢/GB) $328.99 (16¢/GB)
Samsung 860 EVO $57.99 (23¢/GB) $82.99 (17¢/GB) $162.99 (16¢/GB) $347.99 (17¢/GB)

The downsides of QLC NAND are pretty easy to accept if they come with a significant price cut, but that is not yet the case for the Crucial P1 or the Intel SSD 660p. The Crucial P1 is 22 cents per GB and the Intel 660p is 19 cents per GB, so Micron obviously needs to drop their prices at least a little bit to be at all competitive. Meanwhile, mainstream SATA SSDs are about 16–17 cents per GB for 512GB and larger capacities, and there are some high-performance TLC-based NVMe SSDs in the 20-22 cents per GB range.

Ultimately if you are going to pay extra for a NVMe SSD instead of a SATA drive, at current prices there are far more compelling options than the Crucial P1 and Intel 660p. We're accustomed to seeing entry-level NVMe SSDs get undercut by more popular high-performance drives as prices in general trend downward, and the two QLC drives we have so far on the consumer market are continuing that pattern. When QLC comes to the SATA SSD market, prices will need to be at or below 13 cents per GB to avoid repeating this problem.

Power Management
Comments Locked

66 Comments

View All Comments

  • Mikewind Dale - Thursday, November 8, 2018 - link

    Sic:

    "A reduction in quantity and an increase in price will increase net revenue only if demand is elastic."

    That should be "inelastic."
  • limitedaccess - Thursday, November 8, 2018 - link

    The transition to TLC drives was also shortly followed with the transition to 3D NAND using higher process (larger) from smaller planar litho process. While smaller litho allowed more density it also came with the trade off of worse endurance/higher decay. So the transition to 3D NAND effectively offset the issues of MLC->TLC which is where we are today. What's the equivalent for TLC->QLC?

    Low litho planar TLC drives were the ones that were poorly received and performed worse then they reviewed in reality due to decay. And decay is the real issue here with QLC since no reviewer tests for it (it isn't the same as poor write endurance). Is that file I don't regularly access going to maintain the same read speeds or have massively higher latency to access due to the need for ECC to kick in?
  • 0ldman79 - Monday, November 12, 2018 - link

    I may not be correct on the exact numbers, but I think the NAND lithography has stopped at 22nm as they were having issues retaining data on 14nm, just no real benefit going to a smaller lithography.

    They may tune that in a couple of years, but the only way I can see that working with my rudimentary understanding of the system is to keep everything the same size as the 22nm (gates, gaps, fences, chains, roads, whatever, it's too late/early for me to remember the correct terms), same gaps only on a smaller process. They'd have no reduction in cost as they'd be using the same amount of each wafer, might have a reduction in power consumption.

    I'm eager to see how they address the problem but it really looks like QLC may be a dead end. Eventually we're going to hit walls where lithography can't improve and we're going to have to come at the problem (cpu speed, memory speeds, NAND speeds, etc) from an entirely different angle than what we've been doing. For what, 40 years, we've been doing major design changes every 5 years or so and just relying on lithography to improve clock speeds.

    I think that is about to cease entirely. They can probably go farther than what we're seeing but not economically.
  • Lolimaster - Friday, November 9, 2018 - link

    Youre not specting a drive limited to 500MB to be as fast as a PCI-E 4x SSD with full support for it...

    TLC vs MLC all goes to endurance and degraded performance when the drive is full or the cache is exhausted.
  • Lolimaster - Friday, November 9, 2018 - link

    Random performance seems the land of Optane and similar. Even the 16GB optane M10 absoluletely murders even the top of the line NVME Samsung MLC SSD.
  • PaoDeTech - Thursday, November 8, 2018 - link

    Yes, price is still too high. But it will come down. I think that the conclusions fail to highlight the main strength of this SSD: top performance / power. For portable devices, this is the key metric to consider. In this regard is far ahead any SATA SSD and almost all PCIe out there.
  • Lolimaster - Friday, November 9, 2018 - link

    Exactly. QLC should stick to big multiterabyte drives for avrg user or HEDT.

    Like 4TB+.
  • 0ldman79 - Monday, November 12, 2018 - link

    I think that's where they need to place QLC.

    Massive "read mostly" storage. xx layer TLC for a performance drive, QLC for massive data storage, ie; all of my Steam games installed on a 10 cent per gig "read mostly" drive while the OS and my general use is on a 22 cent per gig TLC.

    That's what they're trying to do with that SLC cache, but I think they need to push it a lot farther, throw a 500GB TLC cache on a 4 terabyte QLC drive. That might be able to have it fit into the mainstream NVME lineup.
  • Flunk - Thursday, November 8, 2018 - link

    MSRP seems a little high, I recently picked up an HP EX920 1TB for $255 and that's a much faster drive. Perhaps the street price will be lower.
  • B3an - Thursday, November 8, 2018 - link

    That latency is APPALLING and the performance is below par. If this was dirt cheap it might be worth it to some people, but at that price it's a joke.

Log in

Don't have an account? Sign up now