As previewed at CES, OCZ has released an update to their budget-oriented Trion 100 SATA SSD. The Trion 150 switches from parent company Toshiba's A19nm TLC to their 15nm TLC, the densest planar NAND on the market. The transition to 15nm NAND has been slow for Toshiba and SanDisk, and the release of the Trion 150 signals that the 15nm TLC is finally ready to compete in the most cost-sensitive market segment. This is also probably the end of the road for Toshiba's planar NAND and the Trion 150 will probably be the cheapest drive from Toshiba or OCZ until their 3D NAND ships, unless they introduce a drive with a DRAM-less controller.

The specifications for the Trion 150 are otherwise unchanged from the Trion 100, but the press releases have mentioned some improvements in sustained performance. Supporting the 15nm NAND required at least some firmware tweaks and it's possible that some performance optimizations were introduced as well. It's also possible that the Trion 150 adopts more overprovisioning or larger SLC-mode caches.

OCZ Trion 150 Specifications
Capacity 120GB 240GB 480GB 960GB
Controller Toshiba TC58
NAND Toshiba 15nm TLC
Sequential Read 550MB/s 550MB/s 550MB/s 550MB/s
Sequential Write 450MB/s 520MB/s 530MB/s 530MB/s
4KB Random Read 79K IOPS 90K IOPS 90K IOPS 90K IOPS
4KB Random Write 25K IOPS 43K IOPS 54K IOPS 64K IOPS
Endurance 30TB 60TB 120TB 240TB
DevSleep Power 6mW
Idle Power 830mW
Max Power 4.8W
Warranty Three years
Price (Amazon) $45.99 $69.99 $139.99 $269.99

We initially found the Trion 100 to be a fairly poor performer compared to other modern SSDs, but its pricing of late has been very low and more recent TLC drives like Crucial's BX200 have sacrificed even more performance for the goal of affordability. The allure of the cheapest TLC SSDs has been reduced by the availability of some decent MLC drives for only slightly higher prices, such as Mushkin's Reactor. With the switch to denser NAND, the Trion 150 may be able to widen the gap and take a clear lead in affordability over MLC drives. Even if the Trion 150 turns out to be another step in the race to the bottom among value SSDs, it will still vastly outperform hard drives, and that's all that value SSDs are really aiming for at the moment. On the other hand, if it does offer significant real-world performance improvements without any price increase, it can probably be competitive against other value SSDs at some capacities.

OCZ hasn't announced pricing for the Trion 150, but Newegg is currently listing the Trion 150 at around $0.29/GB for the 240GB and larger capacities and Amazon's listing (not in stock yet) has the 120GB drive slightly cheaper.

Source: OCZ

POST A COMMENT

27 Comments

View All Comments

  • jkhoward - Friday, February 12, 2016 - link

    You do realize that the entire game wasn't updated right? If a 30 GB had to re-download itself every single time it patched itself, that would be absurd. I reckon those 7 games used less than 1 GB tops. Reply
  • ERJ - Thursday, February 4, 2016 - link

    According to http://www.anandtech.com/show/8747/samsung-ssd-850... the write endurance of the 1TB 850 Evo is 150TB so 240TB doesn't seem too bad. Reply
  • bug77 - Friday, February 5, 2016 - link

    Yet I'm having a hard time believing this planar TLC can do better than V-NAND. Reply
  • WinterCharm - Friday, February 5, 2016 - link

    Anandtech is pretty trustworthy. What's so hard to believe? Reply
  • DanNeely - Friday, February 5, 2016 - link

    Specsheet endurance numbers are almost always about market segmenting/warranty limitations, and not anywhere near the actual hardware limits. The manufacturers don't want to cannibalize their highly profitable enterprise lines with cheap consumer drives; so they set the warrantied wear limits low enough that enterprise users won't touch them because if something went wrong they'd be out of support. To minimize warranty claims from normal consumers they also set the limits well below typical physical limits; this way even a customer who got a low edge of the bell curve drive and was doing a pathological workload that resulted in abnormally high write amplification will still hit the specsheet endurance number before actually using up the available spare area on the drive and being able to claim a replacement. Reply
  • hojnikb - Friday, February 5, 2016 - link

    thats not how you get p/e.

    It would have 250 p/e if we lived in a perfect world, where write amplification was 1. But it isn't, so endurance of the actual flash is most certainly higher. I'm guessing around 400-500 p/e
    Reply
  • extide - Friday, February 5, 2016 - link

    The actual flash is probably rated at 1,000 cycles. They are being very conservative here, by probably dividing by two and then assuming avg write amp of 2.

    In all reality the flash will probably last for far more p/e cycles than 1,000. All of the long endurance tests of SSD's that I have seen have gone WAY beyond their spec'd life.
    Reply
  • MrCommunistGen - Thursday, February 4, 2016 - link

    With more of the marketshare of computers being taken by laptops (and other battery powered portable formfactors) I don't see how planar TLC SSDs can really find a lot of marketshare due to their generally poor idle and perf/W characteristics.

    Yes, I'm intentionally sidestepping discussions about endurance.
    Reply
  • MrCommunistGen - Thursday, February 4, 2016 - link

    I posted about half of my thought process. I strongly believe in SSDs for the masses, I just don't think this is the set of compromises that will get us there.

    I know so many people who use a laptop as their primary work machine, often times away from the wall. These people are also often on the go, and to me having a spinning disk drive for this usage scenario just seems silly. They're sitting around for minutes waiting for Windows to boot wasting productive time and chewing up battery. Also, since these users are highly mobile, I'm constantly terrified that one wrong jostle of the chassis could lead to disk failure and catastrophic dataloss. Sure, a backup can be a godsend to avoid losing you projects, but the amount of downtime you'd have restoring your work environment is costly in terms of lost productivity and frustration.

    The current crop of budget planar TLC SSDs, while better than a spinning disk for most of the caveats above, still have too many tradeoffs compared to the marginal price premium of an MLC or larger process-node 3D TLC.

    Performance is lower, power consumption is higher, and although long term (non-endurance related) reliability is still largely unknown due to the short time these drives have been on the market, I can't help but wonder if these "cut every possible corner to minimize BOM" drives will really stand the test of time.

    $70 for a 240GB Trion 150? I'd pay the extra $15 to get a 250GB 850 Evo every time. Don't have $15? Skip going out to for a meal once a week for the next two weeks. Don't have $85? Maybe you shouldn't be spending your money on computer parts.
    Reply
  • chlamchowder - Thursday, February 4, 2016 - link

    I disagree - the SSD industry does have to move towards lower prices (at almost any cost) to increase SSD adoption. Price/GB is the biggest barrier for SSDs, and even a BX200 will beat a 5400 RPM HDD for random accesses.

    With regards to write durability, remember that HDDs are notoriously unreliable.
    Reply

Log in

Don't have an account? Sign up now