The blog & portfolio of Matthew J. Rogers

Posts tagged ‘Mac’

Living in Harmony with Plex

January 11, 2009

Harmony 500

Plex, the fantastic media center software I’ve written about before, added some icing on the cake recently: out-of-the-box support for Logitech Harmony universal remotes.

Logitech Harmony remotes are great because Logitech maintains a massive database of all the codes for pretty much every consumer electronics device ever made (the Harmony 550 was my pick for the best gadget gift under $100). You (usually) don’t have to sit there “training” the Harmony with your original remotes — you just hook it up to your computer, type in the model numbers of your TV, receiver, game console, etc, and you’re off. It vastly improves (and simplifies) control of your home entertainment center.

However, while Logitech did have remote codes for the Mac Mini (which is what I’m using to run Plex), the six buttons of the Apple remote aren’t enough to control a full-featured software suite like Plex. So I wound up buying a piece of software called Remote Buddy that interprets signals from remote controls and turns them into keystrokes. Basically, I spent 2 hours having my Harmony learn IR codes from 7 “virtual” Apple remotes (7 distinct sets of 6 IR codes) to give me 42 effective buttons, but then I had to create keystrokes for each of those 42 IR codes in Remote Buddy so that the proper commands would be sent to Plex (as keystrokes). It was extremely tedious, and I wound up having to do most of it over again because one of the learned IR commands was corrupted (resulting in double-presses).
Read the rest of this entry »

Plex - Band of Brothers episode list

Plex - Band of Brothers episode list

A couple months ago I wrote my initial impressions of Plex, a Mac-centric fork of the venerable XBMC software that was originally written for hacked 1st-generation Xboxes. In early September I bought a Mac Mini to run Plex, and have been using it exclusively as my living-room media streamer since then. The original post got a number of comments, and I’ve received a lot of emails as well so I wanted to follow up now that I’m more familiar with the software.

My original endorsement of Plex stands: it’s a beautiful piece of software, and is being actively developed so improvements are constantly forthcoming. It has taken our interaction with the terabytes of video I keep on hand to a whole new level. It’s so slick and easy to use, as a matter of fact, that sometimes I’ll pick something to watch through Plex even though I have the DVD right there on the shelf next to the TV. I hate physical media, I find it to be a nuisance, and Plex delivers a much richer way of experiencing your video library.
Read the rest of this entry »

Hackintosh at 6 months

October 16, 2008

Toward the end of last March, I wrote about how I had ditched my iMac for a self-built Hackintosh. Given the huge cost savings and flexibility you get from a Hackintosh, I field a lot of questions about how well it works and if I’m still glad I did it. I thought after six months, it was time for a follow-up.

Am I glad I did it? Mostly yes, but a few niggles remain. On the whole, the machine runs brilliantly. It’s never kernel panicked or crashed, and with the Core 2 Duo at 2.8 GHz and 4GB of RAM, I can keep all the apps I regularly use open all the time with very little slowdown. I can pack a whole bunch of hard drives into the case, and have the connections necessary for up to two 30″ monitors should I want to connect them — all things that you can normally only do on the $2500+ Mac Pro. For $600, I can’t complain too much.

But I can complain a little. For one thing, the card reader in my Dell monitor has never worked right. This is solved with a separate USB card reader, but it’s another thing to have to hook up. The Line In/Microphone don’t work…which would be a significant problem if I didn’t also have a MacBook Pro for the times I need to Skype or something. And finally, system updates can be a pain. The small updates aren’t a problem, but the big ones (10.5.4, 10.5.5, next up is 10.5.6, etc) will hose your Hackintosh if you just run them straight from Software Update. They require varying degrees of handholding and third-party packages provided by the Hackintosh community, along with some dedication and patience.

So I am glad I tried this project, and it’s working out great for the most part. Will I do it again? Probably not. Ultimately, I’d like to buy a Mac Pro. Another option I’m considering is seeing if I can consolidate my life down to one machine — a MacBook Pro — for simplicity’s sake, which I would just dock at my desk when I needed a larger monitor. The home-based massive file store (2.5+ TB) would return to either a basic Linux server or a Drobo. I could use the home theater Mac Mini to run any secondary tasks (like long downloads) that I can’t leave my MacBook Pro behind for.

That’s very pie-in-the-sky stuff at the moment though, because I’ve never been without a desktop computer to fall back on. And while the simplicity of a single computer appeals to me, insofar as I would no longer have to be constantly moving files around and maintaining two primary machines, the lack of redundancy scares me a bit. So for now, the Hackintosh stays and continues to chug along. I’ve definitely gotten my money’s worth, and if you’re a bit of a computer tinkerer a Hackintosh might work well for you too.

Mac Mini as a media center

September 12, 2008

Plex on Mac Mini

I have a lot of video media stored on my server (almost 2 terabytes), and I need a way to stream it to my television so we can watch the stuff from a couch instead of a desk chair. Years ago I started with a basic PC running Windows, then moved to a hacked original Xbox with Xbox Media Center, then to a hacked Apple TV. Today, I replaced my hacked Apple TV with a Mac Mini running the excellent Plex media center software.

The Plex interfacePlex is a fork of the original XBMC software that I used way back with my original Xbox. It has been updated and beautified, and definitely sports one of the sexiest user interfaces you’ll ever see gracing your screen. One of its greatest strengths is gathering information about your videos. You point it to your media, and as long as the files are named in a sane way it will go out and fetch all the metadata from online movie and TV databases — it downloads the episode/movie names, descriptions, and even cover art. This makes for a fantastic experience as you’re browsing through your media. You can see in the image here an example of this as I look at the movie “300″.

All in all, it’s fantastic using the Mini has a media center. I have full control over my huge media library using just my Logitech Harmony remote, but I can surf the web or do anything else I want with the computer just by pulling out the bluetooth keyboard and mouse from under the coffee table. Plex is a great piece of software already, and lots of improvements are planned. For digital media junkies, I think this is about as good as it gets right now.

Mac Mini

Hackintosh setup

Hello what?

As detailed in many places around the web (and to which an entire community is dedicated), a “hackintosh” is a computer built of off-the-shelf PC components that can, with just a little bit of tweaking, run the Mac OS X operating system. Ever since Apple made the switch to Intel processors a couple of years ago (and thus to a platform very similar architecturally to what every other computer vendor uses), this has been a growing community. Fed up with the iMac’s glossy screen and 4GB memory limitation (and partly because, as my wife repeatedly points out, I just always “have to have a project”), I finally jumped in myself.

I built myself a hackintosh, bought a Dell 24″ screen to be the primary display, installed the latest OS X Leopard (10.5.2), and now about 36 hours later I’m not missing the iMac one bit. I’ve been working on my web sites, editing RAW photos from my D40, watching movies, and basically doing everything I can possibly think of to stress-test it, and it’s been absolutely fantastic. For me, this fills a hole in Apple’s desktop lineup. The Mac mini is just too weak and can only run one monitor. The latest iMacs come with crappy-quality displays, at least for graphics work (they’re actually worse than the previous generation — something I didn’t know when I purchased the newer one because I just, you know, assumed that they wouldn’t actually downgrade something on a new model), and they’re limited to 4GB of RAM (I’m going to stick 8GB in my hackintosh eventually). Then from the iMac you jump up to the Mac Pro, which is very expensive and in reality is too much power for me. I don’t need a lot of CPU or GPU power — I just need lots of RAM and big, good-quality monitors. Basically, I’d be happy with a small Mac tower that fell somewhere in between the Mini and the Mac Pro.

For about $500-$600, this is what I bought (which, save for two cores, is very near a $2500 Mac Pro for most people’s purposes):

  • Gigabyte GA-P35-DS3R motherboard (if you don’t need as many SATA ports, get the GA-P35-DS3L for ~$25 less)
  • Intel Core 2 Duo E6550 2.33 GHz (these suckers are hugely overclockable — mine’s currently up to 2.8 GHz and hasn’t even gotten anywhere near hot); add about $150 and you can have yourself a quad-core 2.66Ghz
  • 4GB of RAM
  • 500GB SATA hard drive
  • Silent (fanless) Nvidia 7600GS PCI-E video card — bump up to an 8600GT for ~$40 if you care about gaming at all
  • Antec Sonata III whisper-quiet case and efficient 500W power supply
  • SATA DVD-RW drive

The install process for this particular hardware set, detailed in this thread, just requires a few extra package installations after the initial boot, and now everything works 100% just as on a real Mac. I’m really quite pleased. It was a fun project, and now I have a more powerful and more flexible machine than my iMac — and a bigger, far better monitor to boot.

I still love my MacBook Pro to death, and money being no object, yes a Mac Pro would be awesome. But I like getting my hands a little dirty, and this was very easy to do. Now of course as mentioned in every other post about building a hackintosh, it’s legally questionable — but I know many who have done this (including myself) buy a copy of OS X Leopard so then at least it’s somewhat morally justifiable (don’t steal OS X — it’s such a great piece of software, make sure Apple keeps developing it!).

If you have a hackintosh, have thought about building one, or have any questions about it, just post in the comments!

Gallery: Hackintosh build

NERD ALERT: This one’s going to be pretty bad. If you aren’t interested in servers, file protocols, and the geeky ability to access your files anywhere, turn away now.

Those who know me know I run a decently big file server at home. It’s an AMD system running Ubuntu Server with about 1.6 TB of hard drive space (that’s 1,600 GB). I store most of my stuff on that server — personal documents, client projects, web site development code, TV shows, movies, pictures, music, software backups, you name it. None of my Macs actually have any documents stored on them — it’s all on the server, which I depend on having access to wherever I go.

On the road, I typically would use the somewhat klutzy method of downloading files I need via FTP, working on them on my MacBook Pro, and then re-uploading them when I was done or when I got home again, to keep the server up to date (this excludes code, for which I use the Subversion version-tracking system). At home, I have until now used the Samba protocol (the file-sharing system native to Windows) to mount the shared directories and work on them directly over my gigabit network.

I recently had an epiphany about the way I was doing things. My Samba setup was a holdover from when I had all Windows boxes. I continued to use it when I moved to Macs because it pretty much works with everything — Linux, OS X, Windows. However, while compatible with many things, it excels at nothing. It is slow, it is limited to Windows filenaming conventions (UNIX/Linux is much more flexible about file names), it has an extremely limited permissions system, and it can’t really be used over the Internet — hence why I was using FTP while away from home to access my files, because using Samba through the Internet is about as fun as pulling your fingernails out with pliers. It’s just so slow, if you can get it to work at all.

Then I discovered AFP. Despite having had it operational for less than 24 hours, I think it’s safe to say that AFP has changed my life.

As I said, an epiphany: I suddenly realized I was using a Linux server, Mac OS X clients, and a file-sharing protocol that was native to neither of them. Hence, the slow performance. Of course, even on Windows (from whence Samba, in its modern form, hath sprung) most experienced networking engineers will tell you that file-sharing performance sucks. So I started to look at other options: NFS and AFP.

AFP (AppleTalk Filing Protocol) was introduced by Apple in the mid-1990s as a way for people to access shared files and printers over a local network or the Internet. I installed it (with SSL authentication) on my Ubuntu server using this guide and had everything up and running in less than 10 minutes.

File transfer benchmarksPerformance? Freaking awesome. I connected to my home server from work and was able to browse directories and open files incredibly quickly — and all using the native OS X Finder rather than a special program for FTP. The usage is no different from being plugged into my network at home, other than the speed of course, which while not as fast as being at home is still much faster than FTP or SSH over MacFUSE or something. I don’t have any benchmarks, but I’d compare it to using a USB pen drive. Keep in mind I’m accessing the server through 15Mb down/1.5Mb up cable Internet.

It also “just works” when it comes to permissions. With Samba, since it was born of Windows, and Windows doesn’t have a very robust (ok, pretty much nonexistent) file permissions system, there were all sorts of configuration settings you had to define to tell Samba which users could read/write which files. Also, those permission settings got applied to the whole mounted share…there was no way to fine-tune permissions within the share for specific users (as you can natively in Linux). Oh, and speaking of users, with Samba you had to specifically add Samba users on top of the Linux user accounts they already had on the server. It was a real pain. With AFP, after defining where my shares were (which was stupidly simple…two parameters per line: the folder path and the share name) I was able to connect to it with my regular account on that server, and the permissions on all the files were appropriately handled. No special user creation, no special settings. It was fantastically easy.

I’m not a network engineer, and I only took one network programming course in college, but it seems to me that AFP performs so well because:

  • It was designed from the start to work over large network “distances” — i.e., complex networks like the Internet
  • It integrates into TCP/IP at a low level, reducing overhead
  • It integrates into the operating system at a low level — access to files over AFP is handled the same way as local files, without going through a lot of special protocol translations (Samba)

AFP will change the way I work when I’m away from home. If I’m in a location with a decent Internet connection, I can just work on most files directly on the server at home through AFP. Even if I have to copy files to the laptop, because they’re huge or because I might not have Internet where I’m going, it’s still much more convenient (and faster) than using FTP. I left Samba running on the server because I do occasionally need access to those shares with Windows, but that has no impact on AFP performance. If you have a file server and you’re using OS X client computers, I strongly recommend setting up AFP. It’s made things go a lot smoother for me when I’m away from home, and even the performance when I’m directly plugged into my network is noticeably improved.