While 2009 may end up being the first year that we see widespread adoption of SSDs (Solid State Drives) in notebooks, 2008 will go down as the year that it all started happening.

My experience with a SSD on the MacBook Air was an overwhelmingly positive one. While most application usage performance didn't improve, boot and application start times were noticeably quicker. Also, amazingly enough, battery life was improved by a good 5 - 15% depending on the usage model. I was quite impressed.

The size and price of the MacBook Air's SSD option were both tough pills to swallow however. Available only as a 64GB drive and as a $999 option from Apple, the option honestly doesn't make financial sense on a notebook that has a healthy shelf life of only a year or two thanks to a non-upgradable CPU and memory.

Shortly after publishing my MacBook Air review we were contacted by DVNation with an interesting offer: to try out a 128GB SATA SSD in a MacBook Pro.

If you'll remember, one of the issues with the MacBook Air's SSD is that it's still based on older PATA controller technology. When the first SSD drives hit the market they were almost exclusively for industrial applications, where reliability not performance was the top concern and they also happened to be PATA drives. Once it became clear that there was consumer interest in SSDs, development quickly shifted to SATA drives and Flash-to-PATA controller technology lagged behind. The overall market for SSDs was small enough that it didn't make sense to commit a ton of resources to the development of these things, thus we saw SATA drives improve in performance and PATA offerings stagnate. While the Flash memory side of the Samsung SSD in the MacBook Air was fast enough, the controller became the bottleneck and thus there were some instances where the SSD option was actually slower than the already sluggish mechanical disk that ships with the Air.

I was very eager to find out what would happen if I paired the latest MacBook Pro with an even faster SSD. Built with the latest in Flash-to-SATA controller technology I expected to see both performance and battery life improve.

DVNation shipped me the drive, a Memoright MR25.1-128S (the same thing as the MR25.2-128S apparently). I didn't realize how expensive it was until I got the package. DV Nation had shipped me a 15" MacBook Pro with a sticker price of $3,074, expensive but not the shocker. The next line told me the price of the SSD: $3,819.

That's right, the Memoright 128GB SATA SSD costs almost $4K just for the drive. And you thought Apple's SSD option was pricey.

The Drive
Comments Locked


View All Comments

  • Alexstarfire - Wednesday, April 16, 2008 - link

    Efficiency != performance. You know that, right?
  • JarredWalton - Wednesday, April 16, 2008 - link

    Yeah. Using a custom IC means you don't have any extra transistors wasting space and power. An FPGA can do the same thing as a custom IC, but you usually have more gates available than you actually need.
  • nubie - Tuesday, April 15, 2008 - link

    I was under the impression that SATA used more electricity than PATA, no mention of that?

    We are going to need a dedicated low power mode for the chipsets and the SSD's

    I wouldn't be surprised if these are all switched to a Mini PCIe card when this is mainstream. The added bulk, price and inefficiency of these in a 2.5" form factor is just plain dumb, besides, let it communicate over PCIe, then you wouldn't need to have a PATA-SATA etc interface.
  • mindless1 - Friday, April 18, 2008 - link

    1) It would be very risky to design a laptop that didn't support 2.5" drives, a lot of the market isn't going to pay $4000 for their hard drive.

    2) You are obviously wrong about (significant) added bulk, that size is what allows the number of flash chips to reach capacity, the controller to improve performance, and the supporting electronics onboard. Of course there is a bit of space non-electronically *wasted* but so it goes with modular parts. When flash density increases again you could have smaller form factor with same capacity or you could have higher capacity. Also, a well optimized controller should parallelize access to the chips so more chips = better performance.

    3) SATA vs PATA power consumption is not necessarily a difference enough to be significant in total device power consumption, although in this case with a PATA SATA bridge chip there is a small (probably under 50mA) consumption by that bridge.

    4) You suggest comm over PCIe would be significant, that you wouldn't need PATA to SATA, but you would still need intermediary controller. Remember that SSDs are not developed for only one notebook, the market for a $4000 device is limited enough already so supporting the largest # of systems reasonably possible will help drive down prices.

    5) What happens when you keep packing more and more active parts into a smaller space? Power density goes up and you then either need a fan or elaborate heatsinks scattered around and a lifespan degradation of other parts from running at higher temp. Not what people want when paying $5000 plus for a laptop.

    In the end, SSD is still in it's infancy, think back to the improvements in mechanical drives made from the beginning until now. Ultimately if space savings becomes priority #1 they will just integrate the controller and flash chips onto the mainboard, forgoing the PCIe mechanical interface and parallel PCBs, that space consumption altogether.

    Personally I think we need a new memory format similar to Compact Flash, with robust pin connectors, easily slotted and removable, and native SATA300 support. Next we need a portable Windows, so switching from your laptop environment and files to your desktop is just a matter of pulling card out of slot and plugging into the other system - or any other PC in the world for that matter. The crude precursors of this are already seen in mobile apps on USB thumbdrives but to be truely portable the whole OS environment needs be fully plug-n-play not what MS calls PNP.
  • quanta - Tuesday, April 15, 2008 - link

    Last time I checked, even 8 sticks of Corsair Flash Voyager GT (16GB) flash drive cost less than $1000 total. Even with FPGAs and other extra electronics, the $3,819 price tag is a poor price excuse for extra performance. Considering that the drive doesn't even read and write at 8x the speed of Flash Voyager GT, you are better off using the extra cash to get better laptop options for other components.
  • ltcommanderdata - Tuesday, April 15, 2008 - link

    It's interesting that although your article mentions at the bottom of page 2 that and SSD would appear as just another hard drive, the DV Nation website lists the system requirements as OS X 10.5 being required. I guess they are just being conservative.

    I would like the response time of an SSD, but beyond the price, the storage capacity is still a bit constraining, even at 128GB. Personally, now that Fujitsu has announced 320GB 7200rpm 2.5" drives, I'm waiting on Hitachi to release an equivalent to replace my 160GB 5400 rpm Hitachi in my MacBook Pro. If I'm not mistaken Hitachi drives tend to be faster than Fujitsu.
  • Timothy123 - Tuesday, April 15, 2008 - link

    Honda Civics do not depreciate that much, pretty bad example actually. Honda is known for it's high resale value and continued quality through high mileage and the test of time.

    You should have picked an American car to use in that metaphor, for instance, any car made by Ford or GM would have been a good choice.

    A Ford Focus for instance, loses a lot more of it's value than a Honda Civic, hell for that matter, any American car loses it's value quicker than a Honda Civic.

    Really really really bad example. Really bad.

  • xkon - Saturday, April 19, 2008 - link

    lol. well... that kind of makes the statement true nonetheless. if the sdd depreciates in value 40% that means it depreciates a lot faster than a honda civic. if they'd chosen say... a dodge intrepid (my neighbor's '03 sold for $3000) or something like that, the depreciation value would be similar or over 40% defeating the purpose of the comment.
  • Googer - Wednesday, April 16, 2008 - link

    What the hell has that got to do with any thing in the article?
  • strikeback03 - Wednesday, April 16, 2008 - link

    I had the same thought about the lack of depreciation on Civics when I read that line. And the fact that half the comments relate to that one sentence is humorous as well.

Log in

Don't have an account? Sign up now