Just recently my trusty old Pentium 4 trading computer died and I was faced with a dilemma what computer system am I going to buy. Really there are only two options Mac or PC. I have been using both windows and Mac computers for well over ten years and I have to say that I prefer Mac and OS X over Windows any day of the week. The problem is there is very little trading software written that supports Mac. I asked myself some simple questions:
What do I need from a computer
- Large Screen.
- Needs to run all trading applications.
- Needs to be fast. (Not ultra fast).
- Ability to develop for multiple devices.
- Decent amount of RAM.
- Large hard drive.
- Good Value for money.
With all these requirements in mind I decided to go with the new 27 Inch iMac model. You can read about the full specifications of this machine 27” iMac, below is a quick rundown:
- 3.06 Intel Core 2 Duo
- 4 gig of RAM.
- 1 terabyte hard drive
- 27 inch backlight LED screen (love the resolution of this machine)
Running my Trading Software
Here is where things get a little more complicated if you want to run most trading software you will need to run Windows. Essentially you have three options for running Windows based software:
- You can use a virtual machine.
- You can use Crossover Office or Darwine (note: .NET not supported).
- Use boot camp and dual boot with OS X.
Each option has it’s strengths and weaknesses, however after weighing all the options I have decided to go with Parallels Desktop version 5. Parallels has excellent integration within OS X and coherence mode allows me to run my Windows applications and OS X applications side by side. Below is a screenshot:
So far in my limited testing I have not come across an issue running my applications in a virtualized environment. The applications run fast enough for my needs and I can export my virtual machine and run my trading system from multiple locations. It will be interesting to see how this virtualized environment performs with some statistical arbitrage systems. I am also looking forward to getting 64 bit version of Matlab up and running under OS X.