Stop Taking Lossy JPG Screen Shots for Text in OS X. Use PNG.

If you take a lot of screen shots in OS X using the built-in screen capture functionality (e.g., cmd-shift-4) and present them to other people, you should really consider changing one of the default OS X settings. This is especially true if the content contains a lot of text. To be clear about what I mean, consider these two captures of a random gDoc form:

Lossy capture using JPG:

Lossless capture using PNG:

You can probably see the difference quite clearly. The jpg looks blotchy and the color of red has been distorted. This is because the JPG format will degrade image quality to produce a smaller file. Given that we’re talking 9 vs. 13 KB in this case, the savings don’t really matter. What does matter is that the image looks terrible and makes you look oh-so-non-tech-savvy.

The solution is quite simple if you’re using OS X. Simply change the default format for screen captures. Launch the Terminal app and issue the following commands:

$ defaults write com.apple.screencapture type png
$ killall SystemUIServer

That’s it! All of your screen captures will now be in the much cleaner PNG format. If you decide you’d rather have JPG again, just replace “png” with “jpg” in that first command.

5TB LVM Volume with an LSI 9265-8i RAID Controller

This article outlines how to get a 5TB LVM volume created with an LSI 9265-8i RAID controller.

Background

RAID Array

I’ve been running software RAID for a while. Specifically, I’ve got an ASUS P6T Deluxe V2 motherboard with 6 SATA ports. Up until now, I’ve had 1 SATA connected to a single 1 TB drive with the Fedora OS on it, one to a SATA DVD/Blu-ray drive, and the other 4 to a 4x1TB software RAID 5. This has worked great. When I started to max that out, I had a decision to make. It seems I could either:

  • Continue with the small array and just continue to increase the disk size.  This is easiest, but given that 4 disks in RAID 5 give you a 25% loss of storage space (i.e., 3 used, 1 for parity), you have to buy bigger disks and the biggest ones usually cost the most.
  • Make the 1-time investment to get an 8-port RAID card and grow the array with disks that are large, but not necessarily the largest out there.

I decided latter made more sense for me and went with the LSI 9265-8i based on various reviews.  My plan was to build a 6x1TB SATA array (5TB storage) with 2 available ports on which I could add 2 additional drives when/if needed.

Continue reading

Converting VHS to Digital Video (DV) With Canopus (Grass Valley) ADVC-300

Canopus / Grass Valley ADVC-300My parents have mountains of VHS tapes that are slowly degrading in various boxes, cabinets and shelves throughout their house.  For Christmas, I got them a Canopus (now Grass Valley) ADVC-300.  The results were pretty awesome.  Here’s what we did:

  1. Purchased the ADVC-300 online from Electronica Direct via Amazon.
  2. Went to Wal-Mart and got the cheapest VHS player they had.  I think this is important.  If you’ve got a ton of videos to convert, make sure you have a player that’s ready for the load.  If you dig out the old player from 1987, you might get poor results.
  3. Went to Best Buy and got a Firewire 400 to 800 converter cable.  The ADVC-300 comes with a 400 to mini-400 cable, but my parents have a newer iMac that has a Firewire 800 port so the extra cable was necessary.
  4. Plugged everything in: VHS player to ADVC-300 to iMac
  5. Opened iMovie, clicked import, pressed play on the VHS player

The results were very nice.  Obviously quality is determined by the tapes, but the process was generally hassle free.  The only pain point was that iMovie stops importing every time it reaches empty tape.  So if you’ve got several things on one tape and a few seconds between each video set, then iMovie will stop importing at the end of each and you’ll have to manually restart the import for the next set.  I’m guessing that if you used a “pro” application like Final Cut, this could probably be avoided.

If you’ve got a stockpile, now’s the time.  I found tapes with mold in the cassette and one tape broke during playback due to brittle plastic.  I’m glad we converted to DV because I’m not sure those tapes would last much longer and it would be a shame to lose 20+ years of video records.

Bottom line: the ADVC-300 is a solid purchase for anyone looking to convert VHS tapes to digital video.

Power Book Broadcom Wireless In Fedora Linux

Man do I hate Linux sometimes.

So I’ve got an old PowerBook G4 867Mhz that is just not keeping up with the latest Mac software. Rather than scrap it, I thought I’d see if I could give it a second life. I decided to try Fedora 8 as I’m very familiar with that OS. My wife only really uses this computer to surf the net and check email, so I figured a base install would be perfect. All went well until I got to the wireless card. It didn’t work out of the box so I started Googling only to find tons of incorrect information.

The most common was the application of a program called bcm43xx-fwcutter which is used to extract data from a proprietary driver and create useful drivers for Linux. Well, 99% of the information surrounding this kept leading me to the same error:

# /sbin/modprobe bcm43xx
FATAL: Module bcm43xx not found.

# sudo ifconfig wlan0 up
SIOCSIFFLAGS: No such file or directory

After literally 2 days of trying variations of the bad info, I had a sudden flash of common sense: check dmesg to see what’s failing. I ran the following:

dmesg | tail -n 30

Buried in the output was a message telling me to go to linuxwireless and get new drivers. I followed the simple instructions there and bam, it works like a champ. Spread the good word!

Linux Server Load With Top

I always find it interesting when I believe something to be true for years and then one day someone says, “that’s incorrect.” Today was one of those days.

In Unix, Linux, OS X, or most any good operating system, you’ll find a standard command called top. It’s a simple programs that, according to the manual, “provides a dynamic real-time view of a running system.” At the top of the top (heh) output is a header that looks like:

top - 15:43:10 up 14 days, 15:25,  4 users,  load average: 0.27, 0.12, 0.10
Tasks: 106 total,   2 running, 104 sleeping,   0 stopped,   0 zombie

Note the top right load average numbers. I had always thought that those numbers represented the percent load on the processor averaged over the last minute, 5 minutes, and 15 minutes respectively. Well, I was right about the time intervals, but wrong about the information. The load actually reflects system load such that a value of 1 means the processor on average had 1 process waiting to run, i.e., was loaded 100%. So, in the values above, the CPU was only 27% busy over the last minute, 12% in the last five, and 10% in the last 15. Anything over 1 means that there is an overload and a queue of pending processes.

This would explain why I saw values of 40 when OmniNerd was last Slashdotted. I thought the server was doing well with a 40% load. In reality it had a 4000% load and was dropping requests left and right.

Oh well. Live and learn.