Monday, November 29, 2010

Matrices in MATLAB

Let X be some matrix. Then, X(:,i) will give you the i-th column with all the rows, and X(i,:) will give you the i-th row with all the columns.

Monday, August 16, 2010

Flash Drives

If you're having problems figuring out what your flash drive is called in Arch Linux, less /var/log/messages.log. Life saver.

Wednesday, April 28, 2010

How to win CapSim

If you want to do well in the CapSim part of your strategic management class, you have to be willing to put in the hours. Some of our classmate competitors did not realize how much effort my teammate and I put in; we individually logged in more than many other teams did collectively. We viewed the CapSim as an enjoyable game; when other Georgia Tech students played WoW or LoL, we played CapSim.


Do the situation analysis online, but don't forget that your demand analysis should be based on what is in the Round 0 edition of the CapStone Courier. Moreover, don't forget that each segment has a different growth rate. I made the previous mistake TWICE, which cost my team lost profit. Luckily, we never turned a loss, and we never made any mistakes after Round 3.


If you start a price war, don't raise your price! Following this piece of advice could've saved us $18,000,000 in an emergency loan as well as given us twenty additional points for the final balanced scorecard.


Read the CapStone Courier cover to cover. It contains all the competitive intelligence you need to defeat your competitors. Be sure to pay attention to the production analysis page; there is a lot of information on that single page alone. In particular, take note of whether any of your competitors is launching a new product.


Read the manual cover to cover. There is a plethora of information regarding pricing, budgeting for sales and promotion, human resources, finance, and so forth. Moreover, if your professor has activated advanced modules, then read the accompanying descriptions online word for word. The additional information for TQM explicitly stated that $1,500,000 was the magic number: you got the most bang for your buck with that investment. Yet, in what I can only describe as evidence of the failure of the "divide and conquer" strategy, one of our competitors put in a total of $8,000 the first round TQM was on. That team already lost money in that round; their TQM investment gave them no return and served only to put them $8,000 more in the red.


Invest in automation and TQM. The manual describes the pitfalls of automation, but if TQM is activated, you can invest in the initiatives that reduce R&D cycle time, negating the major drawback of automation (ie. increased R&D cycle time). Automation reduces your labor costs as well as the needed complement. We invested in automation so thoroughly that during the second labor negotiation, we caused all but one of our competitors to strike for the maximum length. We also invested in TQM earlier on so we didn't have to do it later on. Toward the end, automation and TQM had reduced our costs to the point where we sold more products than the competition at lower costs.


Invest in human resources. A higher productivity index means you can get the same output with fewer workers or more output with the same amount. You can taper off your expenditures in later rounds without hurting your productivity index, though your turnover rate may never be low enough to get you full points in the balanced scorecard. Many of our competitors made no investment in HR when the module was first activated. Consequently, they were never able to catch up to our productivity index increase.


Pay attention to advanced marketing, if it is activated. It makes getting full awareness and full accessibility MUCH harder, if not downright impossible. Again, many of our competitors did not pay any attention to advanced marketing and saw precipitous declines in their awareness and accessibility. Ours declined as well but not by nearly as much.

Saturday, January 16, 2010

Multi-monitor update

Due to winter break, I had not updated my Archlinux install for about a month. During that time, there appeared to have been several major changes, including a kernel update as well as an Xorg update.


When I rebooted, I found that my multi-monitor setup was borked. It was doing a very bad job of mirroring the displays when I wanted them to be extended.


It turns out that the Xorg update changed the reference names for the monitors. VGA and LVDS are now VGA1 and LVDS1, respectively. A bit of playing around with xrandr in the command line solved that nicely, though I admit I was in a panic and ready to ask the forums for advice.