The video below was made using the latest version of Open Shot, a video editor for Linux. As I am no programmer, nor a professional editor, Open Shot is a great tool to have in your Linux software repertoire, as it offers easy video editing, with video transitions, compositing etc., as well as audio editing. Open Shot 2.0 UI is very intuitive as you can see from the screenshots below. Open Shot 2.0 is relatively lightweight, and should run on lesser-specced Linux boxes as well.
The flying part: A Bombardier CRJ-700 enroute from Toronto to New York, leveling off at 10,000ft doing 330 knots, or 380 mph.
Arch Linux. That is my baby with which I begin my day, and end it. So, how does a guy like me, get a shot with a girl like you, I mean, MacBook Pro? Walk to an Apple Store, drop $2k (minimum) and suck up on the Apple juice? Well, yes.
After a week with the ‘Pro, I can understand (mostly) where the 2,000 of my hard-earned-dollars went. It is a glorious machine indeed, I must admit. Screen, superb. Keyboard, on par, if not better than the T-lineup from Lenovo. Weight? The same as my ’09 Alienware M11X. Yes, 12 inches of Alien-goodness from the previous decade weighs the same as 15+ inches of today. And the ‘Pro has less flex, better battery etc. Thank Zeus for engineers and tech progress!
I do have some gripes though.
1) The ‘+’ (zoom?) button on windows doesn’t really maximize the window, but just makes it slightly bigger. You can go all-in full screen, but there’s no easy double-clicking the taskbar for maximizing window option. Really. Maximizing windows is not a Mac thing, or am I missing something here?
2) Window manager in general, in addition to the ‘+’ window button function. Aero-snap, or whatever you want to call it, a pretty basic function of modern window management, yet is missing. Windows has it. Linux, with its plethora of tiling window managers, in addition to the basic desktop environments, obviously has it. OS X? Not so much. Download “iSnap”? Sure. Works like a version 0.9-beta. Not exactly on par with Apple’s “it just works” experience. Maybe I’m missing something, again.
3) Chrome. Not really Apple’s fault I guess, but Chrome (and especially Chrome Canary), doesn’t feel very good on the MBPr. System heats up, occasionally gets all jerky, and sucks up on battery life. I know, all this is a known bug with Chrome’s Mac arm, but the Apple answer is to use Safari. Really. Really? FFS, that’s equivalent to using IE on Windows. Never. Again. Ever. Never. Ever.
4) Dual screen setup. I have a decent ASUS 30″ 1920×1200 IPS display that works very well on Arch + Gnome. Good resolution, painless setup with my Nvidia card. I know my new MBPr doesn’t have a dedicated graphics card (I’m cheap, $2,500 for a laptop? Nein, danke). While setting up an external/dual screen with any Apple laptop is a breeze, literally a plug-and-play, the 30″ display looks like poop with the MBP driving it. It could very well be a matter of the Intel Graphics having to do all the heavy lifting, and while usable for casual web browsing, I’m not going to do anything with a this kind of dual screen setup. The Mac settings do offer various options for scaling etc but in the end, it all looked liked like crap on the 30″ ASUS. I guess I can’t really blame the MBP for it, since the more expensive one with a dedicated graphics card is known for driving 2, if not 3 additional displays. But still…
5) App Store and Software/Package Management. While better than on Windows, by far one of the biggest issues I have with Macs and the Apple ecosystem is finding, downloading and installing software. When it comes to package management, OS X is a sort of a hybrid between Windows, with no real centralized location to find packages and applications, and Linux, where probably 90% of the packages, or in Arch Linux’s case, 100%, can be downloaded through repositories with a simple terminal download & install command. With my MBP, I can find some applications (most of them paid) in the App Store, which offers not much better user experience than the god-awful iTunes, or direct downloads from a developer’s website. Many of the cross-platform applications I use, e.g. VLC, GIMP, FileZilla etc., are not found in the App Store. The actual installation process of non-App-Store packages is not very intuitive either. Even less intuitive can be uninstalling said software, and while the “Apple approved” apps can be simply tossed in the trash (which is actually very nice way of uninstalling), some of the non-App-Store software seem like they can’t be uninstalled at all, until you read the “Read Me” file(s) (which is one way of saying RTFM, I guess).
But these complaints all pale compared to the main one. And no, I’m not revealing anything earth-shattering, this is all well-known by the Linux- and Apple-communities alike; where Linux-neckbeards will tell me “We told you so” and Apple-Starbuckees will scoff “Just deal with it, Steve will always be right”, here’s my main issue: while with Linux, especially with the Arch-kind, I create my own customized “happy place”, Apple expects, nay, forces, one to adjust to the one and only. It really doesn’t matter if there’s a better way of doing things, this is the Apple way, and we all shall conform.
So why would a Linux-guy like me, drop 2k on a Mac? Because I’m a tech-omnivore. Because I absolutely love the MBPr design, hardware, screen, the keyboard, the lot. Battery life, even with Chrome/Firefox with their bugs, is probably unrivaled by anything running Linux (maybe some Thinkpads with extended batteries?). And while I feel like an unwanted guest at an affluent psychopath’s home, I still like all things that involve precision-engineering, research, and in Apple’s case, the pursuit of perfection, even if it comes with the cost of choice. So, I shall make my MBPr work like a mofo, deliver where it needs, and when I need to get a fix of Linux, I can always retreat to the bliss of my Arch-workstation.
With the exponentially ever-increasing network difficulty, my Butterfly Labs Little Single bitcoin miner is barely keeping up. When I ordered the miner in early March of 2013, the estimated delivery time had been changed multiple times by Butterfly Labs. I was late in the game, I admit, but I was still shocked how long it actually took BFL to deliver the miner (mid-October). Now, instead of making a profit through mining bitcoin, I’ve had to adjust my expectation to just not taking a loss and trying to convince that even if I never make back the purchase price, I’ll still be supporting the bitcoin network.
The miner is running on Raspberry Pi with “Minepeon”, a modified (ARM-based) Arch Linux setup. Minepeon is a command-line-interface-only distro, purposely configured for bitcoin mining. It has cgminer as well as bfgminer software installed; once the OS is loaded onto Raspberry Pi’s SD-card, all one has to do is plug in the miner via USB, and let the system automagically take care of the rest. The mining software will automatically recognize the miner as a USB device, and once you have finished setting up your mining pool information, the miner will start hacking away — in my case at a steady 30 gigahashes per second, resulting currently at 0.15 BTC’s every few days
I admit, there’s probably no substitute for Adobe’s Dreamweaver, and yes yes, we should all be coding web pages with knuckles-raw vi or notepad++, and anything more than those two is just bloated, inefficient, yucky code.
In my world though Dreamweaver has been a really useful tool. I like it’s clean UI, well integrated FTP site loading, as well as the “split view” and “live code” editors. But what to do when you don’t want to pay up, and/or you’d like to be free of expensive software and evil Microsoftesque corporations, like open-source free? Disclaimer: I must admit, my quest to remove the Microsoft shackles is, well, an ongoing quest as I still run Win 8 under VMWare’s Workstation (yes, it *does* perform better than VirtualBox, and yes, it is stable, at least on vanilla Kubuntu 12.10 linux kernel 3.5 I’m on right now).
So far I’ve tried Aptana Studio, Bluefish, Kompozer and most recently Eclipse, along with various souped-up text editors like gvim, cream, kate, gedit etc. Aptana Studio has been sort of my go-to so far, and performed ok on (K)ubuntu, but in the end it hasn’t provided me much more than Bluefish or Eclipse. Enter Bluegriffon, based on Mozilla’s Gecko, built with html5 and css3 in mind, a wysiwyg html editor. As a bare bones editor, it has a nice clean UI, very similar to what you find on Mozilla’s other products. Since I want some bling and pizzazz, I opted to buy, nay, *support*, the open source movement by shelling out ~$20 for “Live view” and “css Pro” add-ons (Mozilla’s FTP add-on is free), which turn out to be useful, but, alas, not transforming Bluegriffon into a Dreamweaver killer.
While I’ll be toying with on Kubuntu, it is also available on Linux’s greatest rivals, Windows & OSX (see what I did there?).
So far so good, I like Bluegriffon. While it will probably never be Dreamweaver, it’s simplicity and bare-bones-but-functional approach (with quite a few, albeit paid add-ons), it just maybe my OSS “equivalent” to the venerable Dreamweaver, a sort of OSS’s “this is all we have, plus the kitchen sink.” I know I’m giving it close hard look.
I’m kind of a sucker for “old” computing hardware. Not old as in 1980’s old, but old like about a decade old. I’ve always wanted to get my hands on some late 1990’s/early 2000’s laptops/PC’s to create a tinkerer’s Linux box, to see if I can bring the machine back to life, and make it usable for everyday use.
So now I’ve been presented with said opportunity. I’m currently working on a “donated” ThinkPad T22 (~900MHz, 384MB RAM, 32GB HDD @5400rpm) a friend used for her MBA class at UC Irvine — apparently it was a mandatory purchase for the school. Roughly a decade ago when this piece of IBM engineering was top-of-the-line, it made your wallet a wee bit lighter, we’ll call it by a solid $3,500 (Amazon has it currently at a more meager ~$120). Yeah. For a machine that spec-wise compares to a run-of-the-mill Android phone.
This machine came with Win XP Pro installed (not the original OS I presume), a docking station, DVD ROM swappable to a 1.44MB floppy disk reader — ah yes, old skool! — and a PCMCIA wireless card. Apparently the ‘Pad (see what I did there) hadn’t been booted in the last 5 years, and was collecting dust in the garage. Other than a quick CMOS complaint, the laptop booted up fairly quickly, even in modern standards, and everything partied like it was 1999. Truly a remarkable machine! Keyboard feels solid, as does the whole frame. Besides a few very minor blemishes, the display looks crisp, and everything “just works” (see what I did there again).
So after 2 mins and 47 secs of XP fun and all, I was lounging on Arch Linux forums to figure out the what/how/when to put Arch Linux on this piece of IBM history.
Arch has changed the installation process from last time I was sweating the KISS principle. No more GUI/wizard installer, and I use the term “GUI” loosely here. If previously you could sort of “wing it” with the installer, now you must have the noob guide handy. All you’ll have is the terminal, with installation done through the terminal under chroot. Though the interface is more bare, the “flow” of installation is better, simpler and quicker than previously, at least IMO.
After a fairly straightforward 45 minutes of reading the manual and basic installation (a good 15 minutes was wasted on my part on partitioning, mainly deciding between, reading and figuring out, GPT vs cfdisk), it was time for elementary post installation setup, i.e. Xorg, choosing a DE/window manager, display manager etc. I was hesitant to put resource intensive DE’s like Gnome or KDE on the T22, so I compromised and went with the trusted Xfce. For a moment there I tinkered with Enlightenment DE/WM as it is modern, lightweight yet doesn’t give in on eye-candy, but in the end, Xfce is still my choice of lightweight DE.
So far so good. Using Slim display manager, boot takes ~45 secs. Desktop is snappy, Chromium is initially loaded up in just few seconds. ThinkPad’s trackball needed zero attention, no issues with graphics driver either. I’ve read suspend-to-RAM works pretty much OOB, as does CPU scaling. As this is all is WIP, I haven’t gotten yet to customizing the desktop, keyboard shortcuts, getting PCMCIA WiFi card running, network printing etc, so the laundry list does go on.
I frequently read about how Linux brings old hardware back to life, but only now this back-from-the-dead-with-Linux is beginning to slowly become a part of my everyday computing. I can honestly see this T22 replacing my Alienware M11X as the travel machine. As I carry my gear in an overnight rollaboard, I often fear having shit broken by rampers, or stolen when I leave the bag in crewrooms etc. Now, having a sub-$100, yet still bank-vault-solid and fully capable Linux laptop for the overnights might be exactly what the pilot ordered!
3 days and some $900+ later, I have a functioning Hackintosh. The build was not easy — not because tonymacx86’s excellent blog doesn’t give you step-by-step instructions for the build, or because parts are hard to come by, thanks to the interwebz and Amazon. No, the build was difficult because the builder, that would be moi, didn’t RTFM.
First off, since I didn’t RTFM, I ended up buying a retail DVD for OS X 10.6 Snow Leopard. So right off the bat I was going the wrong way; you see SL works great on Intel i3/5/7 Sandy Bridge Hackintoshes, but not on my newly-purchased i5 Ivy Bridge CPU. Oh yes, there’s a clear warning on tonymacx86’s web site, but did I read it carefully? Of course not. Ideally I was to first build the Hack with Snow Leopard, then upgrade to OS X 10.7 Lion. This was not to be. At least the way I envisioned it.
In order to have the bootloader to “trick” Mac OS into thinking it is in fact running on Apple-ordained machinery, one must be able to create a thumb drive or a CD to be run during the OS X installation process. You can create the slightly older iBoot CD, or, if you already have a Mac, a MultiBeast thumb drive. iBoot works well with Snow Leopard, but if you want to play with Lion, you have to have Multibeast. My problem? I don’t have a Mac so I couldn’t create the MultiBeast thumb drive. Of course I could always burn a copy of iBoot, if I was to eventually boot to Snow Leopard, which of course I couldn’t, since Snow Leopard doesn’t support Ivy Bridge CPU’s. What to do?
Enter the fantastic world of virtualization. I’ve played with Oracle’s VirtualBox quite a bit in the past, and while otherwise a great virtualization tool, getting USB devices recognized has been a hit and miss. With my official copy of Snow Leopard? More of a miss. I moved on to VMware, another commonly used application for virtual computing, and while not the easiest of setups, I was able to get Snow Leopard running and eventually create the MultiBeast boot USB. Finally I was making progress.
A new day began with me trying to figure out how to duplicate my success from virtual world to real world hardware. Like I mentioned, tonymacx86 et al have done a fantastic job creating detailed instructions how to go about getting OS X on PC hardware. My installation was not to be without hiccups though; for some of the problems I ended up googling stuff, since not every error log entry from a kernel panic or boot-loop can be addressed on a single website. Two key changes for my installation process were 1) not to use USB 3.0 port for the thumb drive boot, and 2) disabling “Limit CPUID Maximum” on the motherboard BIOS. Once Snow Leopard was up and running, a quick installation of basic software bundle from Multibeast (kexts, drivers etc). Then on to upgrade to Lion. So far so good.
Final day had its own set of issues. Sleep mode was not working, and during boot I’d have to enter “Graphics Mode”=”1920x1080x60” to get full resolution out of the i5-3750k’s HD4000 graphics. Alas, HD4000 hardware acceleration for graphics is not available until Apple releases Ivy Bridge machinery. I was very hesitant to include extra kext files from MultiBeast to resolve the sleep and resolution issues. With regards to sleep, it is a matter of disabling Apple’s own CPU power management, and using a patched kext file instead. But since I am complete noob to the art of Hackintoshing, I didn’t want to risk my otherwise stable setup for a slew of kernel panics. The same went for the resolution issue. Now I know adding the appropriate kexts from MultBeast would’ve had solved both of my issues.
I had been waiting for over 2 weeks for my graphics card to arrive, to no avail. I sh!tcanned the Amazon purchase order, and went to Fry’s and got myself EVGA’s GTX 550 Ti, one of the more popular and tested graphics cards for Hackingtoshing. After a quick realization that my monitor does not have a DVI-D “in”, and the GTX 550’s other “out” option is mini-HDMI for which I didn’t own a cable (a smarter, more methodical buyer would’ve naturally done this before leaving the house for said box store…), I was down another $30 for a new monitor cable. Oh well.
Now though, I have a fully functioning HackPro. VMware Fusion runs Windows like a champ, or as “champy” as one can run good ole Windoze… By allocating 4 cores, 4GB of RAM, and with a dedicated graphics card, Windows 7 runs like it was natively installed. [A quick side note: the early plan was to dual boot Lion and Windows, both from the same SSD and subsequently both having individual HDD’s for storage. The problem is/was, as I discovered in the middle of all this, OS X uses GUID Partitioning Table for harddrives, and while it is possible to partition Windows using GPT, this must be done before any other OS has been installed, i.e. making my hard work getting OS X running useless. Needless to say, for now I am sticking with VMware Fusion and virtual Windows.] Apple TrackPad via Belkin bluetooth dongle was recognize without a hitch, and TrackPad gestures –love them! — work like on a bona fide Mac, as do sound and the rest. And my 22″ Samsung monitor displays in all of its1080p glory — ok, the one thing you can’t really have on a Hackintosh budget, is the dazzle and sleekness of Apple’s hardware design, including their fantastic displays.
But all in all, my HackPro is a beast of a machine.
While many hackintoshers are enjoying OS X Mountain Lion, it is not to be, at least so far, with my HackPro. Problem is my tried-and-true GTX 550Ti graphics card, which so effortlessly works on 10.7 Lion, but creates havoc on 10.8 Mountain Lion, to a point of being unusable. Of course I could revert back to i5 CPU’s HD4000 graphics, but I fear my desktop swappiness with Ubuntu and Windows under Fusion would greatly suffer.
While I wait for the true hackers — or Nvidia — to come up with new drivers for 550Ti, indeed for all Nvidia Fermi-cards, I decided to upgrade RAM to 16GB, and also upgraded VMWare Fusion to.5.0 with plans to eventually replace Windows 7 with 8.
1) Assemble computer with Mac OS X compatible hardware
2) Install OS X and UniBeast
3) Enjoy OS X.
1) NZXT H2 case’s front USB hub creates a short in one of the ports, resulting in a) 2 busted USB flash drives, b) 1 busted external DVD drive, c) almost burnt fingers after handling the hot, now-busted, USB flash drive.
2) In the process find out OS X 10.6 Snow Leopard is indeed not compatible with Intel Ivy Bridge CPU’s, must have Lion. In order to have Lion, you must have Snow Leopard, or shell out $69 for a USB thumb drive. USB thumb drive wouldn’t work due to trippy USB hub.
3) Should’ve bought a Mac.
In other news, I can upgrade to Windows 8 for $40 when it’s released. Oh joy.
Finally got Arch Linux running on Raspberry Pi the way it should. Initially I thought of keeping the first Pi (yes, there’s one more coming) under CLI only, running as a NAS/file server. After a few times around with Samba setup, SSH’ing from Fedora main box, I’ve given in to a more or less full DE. Enter OpenBox for window manager and desktop environment, ROX for file manager, and Midori for web browser. As you can see, name of the game is keeping things light.
As it is now, the Pi boots in ~20 sec, and with ‘startx’ I have OB running in about 4, so overall pretty snappy. I have a 22″ monitor hooked up via HDMI, and screen resolution is set from boot to 1080p. Keyboard and mouse are working fine via wireless USB dongle. But, as it has been mentioned time and again, the Pi is not for everyday use. Even Midori, as lightweight of a browser it is (I’ve yet to try ultralights like UZBL browser), loading times are a bit sluggish. Not at unacceptable/unusable levels by any means, but the Pi won’t be replacing any of the everyday machines I use. So far I’ve had the Pi run only 3-4 hrs at a time, and only thing I’ve discovered is it gets luke warm. That’s it. So box it in, leave it in the corner, it won’t complain.
And what’s also great, and the real purpose of the Pi, it’s so simple, humming (ok, so it is completely silent…) away alone, separated from the rest of the household gadgetry. Screw up something royally, and at worst, all you’ve done is maybe force yourself to another ‘sudo dd if=…’. So it really is a perfect Linux learning tool.
Plan for Pi numero deux is to run Debian Squeeze on it, with all the pre-loaded learning tools (Scratch etc.), and to have my 6+ year-old get her Girl Geek on. She’s probably a little too young for real coding yet — read: Dad will have no farking clue what to do — but maybe we’ll have a couple of productive Scratch sessions and get a simple Pac Man game going.
So all in all, the Pi is, and I’m quoting one Charlie Sheen: “Winning!”