Tag Archive for 'macbook-pro'

The External Hard Drive Post

Published
by
jk
on December 5, 2007
in Gadgets & Toys
. Comments

I needed a hard drive. Instead of going all out and getting a single 1TB storage, I went for a 500 GB instead.

LaCie 500 GB d2 Quadra

Firewire is amazing for external hard drives. It’s a little bit more expensive then the single USB 2.0 options, but man… it’s a lot faster.

Plus, you really can’t buy external drives with just FireWire anymore. Usually. FireWire drives just come with USB anyway, so you’ve got some backward capability.

Yes, I just called USB 2.0 “backwards”. You’ll see why in a second.

tan ( ) : Firewire, USB 2.0 Comparison

From the site: FireWire - USB Comparison

FireWire is faster than USB 2.0 in sustained throughput.

Read Test:

  • 5000 files (300 MB total) FireWire was 33% faster than USB 2.0
  • 160 files (650MB total) FireWire was 70% faster than USB 2.0

Write Test:

  • 5000 files (300 MB total) FireWire was 16% faster than USB 2.0
  • 160 files (650MB total) FireWire was 48% faster than USB 2.0

USB 2.0: Needs a master and a slave. Peripherals are dumb. FireWire 400: Peer-to-peer. Peripherals are intelligent.

Back to the rest of the post: The Pipes

Here are some random numbers, for your pleasure.

Now, mind you, these are theoretical limits of the pipe. Doesn’t say anything about how fast the devices are running themselves. (Hooray for CS 423 - Operating Systems, this class should be required.)

Right, so how fast does a 7200 RPM 16 MB Cache hard drive read and write? This largely depends on burst and sustained… Seek time is less than 10ms… How far radially is the data stored…

I wasn’t able to figure this out. If someone can give me actual numbers, I’ll be very appreciated. So instead, I decided to run some tests.

Even better. Actual real world tests.

Using Xbench

Internal Hard Drive (SATA)

Results 37.67
    System Info
        Xbench Version      1.2
        System Version      10.4.11 (8S2167)
        Physical RAM        2048 MB
        Model       MacBookPro3,1
        Drive Type      FUJITSU MHW2160BHPL
    Disk Test   37.67
        Sequential  59.52
            Uncached Write  69.06   42.40 MB/sec [4K blocks]
            Uncached Write  64.01   36.22 MB/sec [256K blocks]
            Uncached Read   41.65   12.19 MB/sec [4K blocks]
            Uncached Read   76.38   38.39 MB/sec [256K blocks]
        Random  27.56
            Uncached Write  9.86    1.04 MB/sec [4K blocks]
            Uncached Write  62.81   20.11 MB/sec [256K blocks]
            Uncached Read   59.29   0.42 MB/sec [4K blocks]
            Uncached Read   91.63   17.00 MB/sec [256K blocks]

FireWire 800

Results 51.81
    System Info
        Xbench Version      1.2
        System Version      10.4.11 (8S2167)
        Physical RAM        2048 MB
        Model       MacBookPro3,1
        Drive Type      ST350063 0AS
    Disk Test   51.81
        Sequential  83.84
            Uncached Write  68.96   42.34 MB/sec [4K blocks]
            Uncached Write  99.59   56.35 MB/sec [256K blocks]
            Uncached Read   60.13   17.60 MB/sec [4K blocks]
            Uncached Read   152.93  76.86 MB/sec [256K blocks]
        Random  37.49
            Uncached Write  12.78   1.35 MB/sec [4K blocks]
            Uncached Write  94.76   30.34 MB/sec [256K blocks]
            Uncached Read   91.12   0.65 MB/sec [4K blocks]
            Uncached Read   144.77  26.86 MB/sec [256K blocks]

What the heck is going on? How the heck is my external outperforming my desktop in every single possible way?

EDIT: I had forgot to think about the RPMs. Since laptops are design for lower power consumption, normally they come with 5400 RPM drives instead of 7200 RPM, which is the one that’s in the external. Now these numbers make sense.

For fun, I performed the test for FireWire 400 and USB 2.0.

FireWire 400

Results 44.61
    System Info
        Xbench Version      1.2
        System Version      10.4.11 (8S2167)
        Physical RAM        2048 MB
        Model       MacBookPro3,1
        Drive Type      ST350063 0AS
    Disk Test   44.61
        Sequential  59.75
            Uncached Write  53.08   32.59 MB/sec [4K blocks]
            Uncached Write  64.79   36.66 MB/sec [256K blocks]
            Uncached Read   50.22   14.70 MB/sec [4K blocks]
            Uncached Read   78.36   39.38 MB/sec [256K blocks]
        Random  35.59
            Uncached Write  12.97   1.37 MB/sec [4K blocks]
            Uncached Write  65.63   21.01 MB/sec [256K blocks]
            Uncached Read   91.10   0.65 MB/sec [4K blocks]
            Uncached Read   109.82  20.38 MB/sec [256K blocks]

USB 2.0

Results 16.61
    System Info
        Xbench Version      1.2
        System Version      10.4.11 (8S2167)
        Physical RAM        2048 MB
        Model       MacBookPro3,1
        Drive Type      ST350063 0AS
    Disk Test   16.61
        Sequential  12.73
            Uncached Write  30.12   18.49 MB/sec [4K blocks]
            Uncached Write  31.23   17.67 MB/sec [256K blocks]
            Uncached Read   4.45    1.30 MB/sec [4K blocks]
            Uncached Read   41.40   20.81 MB/sec [256K blocks]
        Random  23.89
            Uncached Write  8.47    0.90 MB/sec [4K blocks]
            Uncached Write  44.95   14.39 MB/sec [256K blocks]
            Uncached Read   69.71   0.49 MB/sec [4K blocks]
            Uncached Read   78.19   14.51 MB/sec [256K blocks]

Here’s a simple table output of this data.

Table of Hard Drive Test Data

In a real world test, an External FireWire 800 connection somehow managed to beat my Internal SATA drive. Again, if someone could possibly explain this data, let me know.

EDIT: Once again, I realized the error I made: Here, I thought for some reason that the internal drive had similar specs as the external. I also have a feeling that the buffer size is going to be different also due to the size difference (8 MB for internal, 16 MB for external).

Conclusion

In the end, I’m glad I made the switch to FireWire. Now all my Time Machine backups are faster, and requires less strain on the Core 2 Duo processors.

I really hope, though, that the next external data source I purchase will be a part of the Cloud… you know, the ones where I don’t have to deal with the physical media at all. No disks, no CDs, no DVDs, not even Hard Drives: Just some made up blocks of data storage somewhere.

Buying a MacBook Pro?

Published
by
jk
on April 4, 2007
in Apple
. Comments

Lots of people are “defecting” (as the masses would say it) or “converting” (as the enlightened would say it) to the Mac platform. But it kills me when people haven’t taken advantage of the pricing structure that Apple offers.

I was going to write up a full post on this, but I might as well let someone else do the talking.

The main point, however, is that if you’re a student: Regardless of your major, you should not be getting your higher end Apple products through the educational discount. Instead, you should join Apple Developer Connection (ADC) and buy it through that.

Example: My MacBook Pro. For $99 (and a little hassle of scanning in your ID and sending it to Apple), you’re saving $200 on top of the educational discount. Overall, you save $100.

Of course, what they don’t tell you… is that when you first join ADC, they send you a nice DVD for OS X. I sold it immediately on eBay for $70. Then I got yet another OS CD… not sure why, but I sold that too. In the end, I paid for an extra 1 GB of memory this way.

And I didn’t think about this one, but… try to see if you know of any friends that work for Apple. It’s even cheaper than ADC.

How to Buy an Apple Computer

Lessons From Yet Another Reformat

Published
by
jk
on March 29, 2007
in Project Ideas, Web Applications and Web Development
. Comments

As I’ve discussed earlier, I love reformatting my computer… At face value, it might seem like a waste of time, but in the end, lots of lessons are learned. That, and, you get faster every single time… and you get a squeaky clean computer.

Let me just go through where my data lives, and how it made my life so much easier to get back on track after my reformat.

  • Email

    • Gmail: I think I’m going to start using Gmail as my sole email application. It seems like for some reason, this time around, SMTP doesn’t want to work in Mail.app. Also, since I’ve already gone Google for RSS feeds as well as my budgeting/documenting/spreadsheet-ing, might as well commit to… er… submit to the Google overlords.
    • I have to go and somehow find a way to match the folder structure I have on my desktop to Gmail’s labeling structure.
    • Is there a way to move all your mail from one service provider to another? Is Yahoo really going to provide unlimited storage? This really isn’t a selling point, though, since I don’t think I’m even hitting 10% of my storage space yet.
      • But, of course, through the wonders of gDisk, I’ve begun backing up my most important documents, like blog post rough drafts and such.

  • Contact Information

    • Plaxo: I love it. It’s free for us normal users… plus, there’s a little bit of the social magic built into the app: you can send “eCards” to people with their information and such.
    • I don’t use Plaxo much, but I do use the synchronization tool with Address Book, and you know how I love to think that syncing is going to be one of the many killer web apps.

  • Calendar

    • This is next… to go with Google Calendar? How about Plaxo? How about something completely new?

  • Bookmarks

    • What? Delicious? The only thing I needed to reinstall to get back into bookmarking was the bookmarking extension for Firefox and Safari.
    • I’m satisfied with delicious to even think about moving to ma.gnolia.

  • Feeds

    • Google Reader: I love it. I think I might start integrating the “Starred Items” on the blog like Alex Argo has done.

After all… they’re all bits and bytes. It’s the application that brings that information to life. With web apps, though, you get a handy method of not having to deal with keeping track of your own data: the system does it for you.

Actually, more like… the system administrators of the web applications are doing that management of data for you.

Hey, it’s less work for me! But… alas, advertising revenue for them…