One of the hurdles of TLC NAND and VNAND is an inherently lower count of program / erase (P/E) cycles that the SSD can endure. This is the nature of packing more voltage levels into a cell to accommodate for the extra bits each cell can hold (yielding our higher capacity and lower cost). More voltage levels means more granularity required when attempting to read/write data, and the NAND loses its ability to accurately perform those reads / writes as it ages. Controllers have to step in to ensure longer life when using TLC NAND.

smi-2256

SSD benchmarks generally include two fundamental file I/O tests: Sequential and 4K random R/W. At a very top-level, sequential tests consist of large, individual files transfers (think: media files), which is more indicative of media consumption and large file rendering / compilation. 4K random tests employ thousands of files approximating 4KB in size each, generally producing results that are more indicative of what a user might experience in a Windows or application-heavy environment.

file-io-study

Theoretically, this would also be the test to which gamers should pay the most attention. A "pure gaming" environment (not using professional work applications) will be almost entirely exposed to small, random I/O requests generated within the host OS, games, and core applications. A particularly piratical gamer -- or just someone consuming large movie and audio files with great regularity -- would also find use in monitoring sequential I/O in benchmarks.

This article looks at a few things: What types of I/O requests do games spawn most heavily and what will make for the best gaming SSDs with this in mind? There are a few caveats here that we'll go through in a moment -- namely exactly how "noticeable" various SSDs will be in games when it comes to performance. We used tracing software to analyze input / output operations while playing five recent AAA titles and ended up with surprisingly varying results.

UPDATE: Clarified several instances of "file" vs. "I/O" usage.

This year has been full of delays in the hardware-time continuum, it seems. It feels like forever ago since Maxwell was announced, with Intel's Broadwell and HW-E / X99 platform similarly far behind us. Each of these devices will finally be shipping by the holidays, or so we're told, but that still leaves a major market segment untouched: SSDs. Other than recent innovations in Samsung's NAND lineup, the SSD market has remained relatively silent since our initial SandForce Gen3 controller analysis.

sf3700-slider2

At the Flash Memory Summit in Santa Clara this week, SandForce announced plans to meet a December, 2014 on-shelves timeline with its SF3700 SSD controller. For clarity, SandForce was acquired by LSI some years back, which was recently acquired by Avago Technologies (the same people who make most mouse sensors); Avago is presently in the process of spinning-off SandForce to Seagate, who do not yet have an in-house controller manufacturer. A confusing pattern of acquisitions, to be sure, but its impact will be deep on the controller market. That's something for a future article, though. Technically, SandForce is no longer a standalone company in the market -- it's just a technology under LSI, now. For ease of understanding with our audience, we're going to keep calling it "SandForce" rather than try to balance the various Avago / LSI / Seagate owners.

It's felt like an agonizingly slow five years, but SSDs are finally affordable for most PC builds. The 2009 consumer launch saw the arrival of Intel's X25 SSD, built atop SLC architecture and priced accordingly. I remember testing some of the first X25 SSDs and the resulting stack of $1200 paperweights that had accumulated. Thankfully, things have come a long way since then. With the advent of new NAND types that can pack multiple bits into a single cell, affordability and flexibility of use have arrived to the SSD marketplace.

ssd-arch-2

This year in particular has seen the rapid expansion of consumer-ready SSDs, particularly with a refresh of Crucial's budget-class SSDs, ADATA's forward positioning, and Corsair's updated Force lineup. And there's more, too -- Seagate, Samsung, Kingston, PNY, SanDisk, and Mushkin have continued to ship into 2014, with Samsung planning more releases in short order.

With all these choices and the beginning price-war, it's an ideal time for consumers to jump on the constant SSD sales and the rapidly collapsing price-point. This buyer's guide will introduce the best SSDs for the price in gaming and enthusiast uses, hopefully helping with tips on selecting an SSD. We're going to stay away from the high-performance / professional marketplace in this guide.

This weekend's sales roundup features an LED controller for $28, case fans at $16, an AMD CPU for $170, and a 1 TB SSD for just $400. If these deals don't whet your appetite for improving your system, first - get a better appetite, then keep posted to our Twitter and Facebook accounts for more sales and deals throughout the week. Also subscribe to our YouTube channel for build tips, interviews, and reviews.

samsung-evo

SSDs are surrounded by terminology that generally isn't understood beyond a relative level. There's this top-level concept that one type of NAND is superior to another, that synchronous is preferable to asynchronous, that endurance is tied to P/E cycles, but a lot of the knowledge halts there. We've worked closely with several SSD and controller engineers over the past year to educate ourselves on the inner workings of the storage world's biggest recent advancement; now it's time to start organizing that education in article form. Over the next weeks, we'll be releasing several "SSD Architecture" postings (so be sure to like / follow / subscribe) that focus on different aspects of solid-state drives, controllers, and NAND.

ssd-arch-2

This installment includes a video component. The video showcases a discussion with LSI's Kent Smith and spoils the basics of what we'll cover throughout this series. I highly recommend watching the video, especially for those who benefit from visual aids. We covered SSD questions pertaining to varying voltage levels on evolving NAND types (SLC, MLC, TLC), cell decay when an SSD goes unused, P/E cycles and endurance, and "what's next" after TLC for Flash types. That's a lot of stuff. Each item is complex in its own way -- hence the chronicle-like release of in-depth article components.

Today we're talking about top-level SSD anatomy and architecture, defining what "NAND Flash" actually is, evolving NAND types (MLC vs. TLC, what's after TLC), capacity calculations, and providing an "SSD primer" of other basic elements. This is what will lay the foundation for our more advanced articles.

SSDs have faced numerous challenges since their fairly recent beginnings. Initial SLC (single-level cell) SSDs were out-of-reach for most consumers, and as evolving Flash memory types allowed production of more affordable consumer SSDs (MLC, TLC), we saw serious endurance concerns. Endurance and stability concerns have subsided as controller manufacturers learn to cope with the hurdles, though we now face other obstacles -- like endurance on TLC and brushing up against the SATA bus bandwidth cap. Other features, like data integrity and redundancy, often hold greater value than pure speed.

sf3700-slider2

Speed is now a largely irrelevant metric for comparison when looking at high-end consumer SATA III SSDs; no SATA III-powered SSD can exceed the 6Gbps cap (which translates to about ~550MB/s in real-world use, accounting for overhead). Going forward, SSD controllers (this is true for Samsung, OCZ, SandForce, etc.) are not the limiting factor for speed -- it's the bus. Time to move to PCI-e.

The SandForce 3700 series controllers are now officially announced and are already in the hands of manufacturers. On a top-level, the new controller should:

  • Bring more PCI-e SSDs to the consumer market at better prices.
  • Improve overall I/O performance (especially on PCI-e, where the bandwidth isn't a bottleneck).
  • Improve endurance (lifespan) of the drive.
  •  

    With LSI Corporation's (NASDAQ: LSI) new SandForce SF3700 controller impending Q1/Q2 mass production next year, it looks like manufacturers will finally be able to bring PCI-E SSDs to mass consumption and TLC endurance concerns may fade. In this write-up, we'll look at the SF3700 SSD controller's (specifically SF3729/SF3739) specs, SHIELD feature, and PCI-e modularity - specifically as it pertains to Samsung's existing 840 Pro and XP941 controllers.

    We moderate comments on a ~24~48 hour cycle. There will be some delay after submitting a comment.

    Advertisement:

      VigLink badge