Saturday, January 30, 2010

New Computer

After slowly watching my home-built, home-updated desktop getting slower and slower at various things (I think at least one of the hard drives is sick, and its age is denoted by the fact that its motherboard has USB 1.1 ports), I decided to just get a new one. I did look into getting the parts, motherboard, CPU, case, etc., etc., but after looking at prices of pre-built vs. doing that, it made no economic sense. This is, of course, a matter of being on the lookout for good buys.

I know that for many, HP is a popular choice, but I have been wanting to support Dell for their support of Linux (Ubuntu), not that I feel restricted to getting a Dell with Ubuntu already installed.

So a couple of weeks ago I got an email about one of their Vostro desktops, checked it out, looked at the deal ($312 off), compared that to a custom build, and bought it, with a few mods - upgrade to and E8400 Intel Core Duo, 3GB RAM, 320GB hard drive. It came with one of various versions of Windows 7 -- all 32-bit, interestingly enough, but this didn't matter so much to me (details below), so I just got the one that didn't cost me extra.

My aging desktop is a dual-boot Linux-WindowsXP machine. Dual boot machines give you some flexibility but actually are a pain in the butt. This rebooting into the other OS for something or other is aggravating. Also, because of Windows only being able to see vfat formatted drives, I would always have some vfat formatted partitions for sharing data without needing to reboot.

Now comes virtual machines, and more specifically VirtualBox, originally made by a German software company that was bought by Sun, and it's available for free. So now I could run Linux on a Windows machine or vice versa, and it's the vice versa I'm more interested in -- why? Because if I want to browse or check my email, Linux is by far safer. Everything else considered, this is the decision-maker. Also, Linux users are familiar with and expect multiple desktops, so I can leave my virtual machine up and running full size in one desktop, and flip over to another one in Linux any time.

So I received my Dell (Vostro 220s) yesterday, set it up, booted into Win7 once to make sure it worked, then shut it down and blew Win7 away to install 64-bit Fedora 12. In the past I wouldn't have had the courage to do this, but after doing some research, worst-case scenario #1 was that I would need to reinstall Win7, and it did come with a reinstall disk. Worst case #2 was that I would send it back to Dell for a "restocking fee."

So now this morning, I've installed VirtualBox, set up my Win7 virtual machine, installed Win7*, and I'm good. Well, almost there. The screenshot shows what I have now. My monitor is capable of a 1600x900 display, but Win7 could only offer 4:3 ratios, so what you see there is the biggest I could go and keep it all on the screen when I flip to full-screen mode in VB. My guess is the answer is on the drivers disk from Dell.


Update: no the answer is not on the Driver disks from Dell
Further update: The answer came from a command in VirtualBox. Since the VMs can only see what VB allows, I needed to create display parameters that the VM could see. You do this with VBoxManage:

VBoxManage setextradata WinXP_1 "CustomVideoMode1" "1600x900x32"

after which WinXP can find this display resolution. I made another one for Win7.

BTW, Fedora recognized my 1600x900 monitor and selected that all by itself. It also found my printer as soon as I hooked it up, and set it up -- no searching for a driver, no getting a driver disk. Fedora probably took longer to install, mainly because I typically install a lot of software (all free, of course), including both Gnome and KDE window managers (if you can, why not?)

So the final chapter will come when my Windows XP Pro 64-bit disk arrives in the mail. Why that "dinosaur"? One of the whole points of this purchase is that I need to be able to connect to the hospital and to my office from home (you probably thought this post had nothing to do with medicine). My hospital requires Windows XP and IE6 (yes the dreaded IE6) for connectivity. So that will be the other virtual machine I make on this computer, the one I'll probably use more often.

Soon I'll post about how I've begun using Scribus, my favorite desktop publishing software, in the hospital.


* Incidentally, I did nothing illegal. I looked at the license that came with my disk, and it says that I can install this on the hardware that I purchased with it, and furthermore, there is a specific clause that allows me to install using virtualization on the hardware I purchased.

Sunday, January 24, 2010

Dementia News

The January, 2010 issue of Archives of Neurology has some articles I find of interest.

MCI and Exercise

There were 2 articles on MCI (Mild Cognitive Impairment) and the effects of exercise. The first, from the Puget Sound VA, was a controlled trial of 33 men and women, testing aerobic exercise vs only stretching. Women were more likely to see benefit from this, especially with tests of executive function (more or less products of the frontal lobes).

The second article was a population-based study, in other words, taking a group of people with or without cognitive impairment, then seeing how they may differ, in this case in regard to exercise. Like some other studies, this one too showed that those with better cognition were more likely to carry out some form of moderate exercise on a regular basis. These kinds of studies always have questions about what is the cause of what -- are you more likely to avoid dementia by exercising, or more likely to exercise if you do not have dementia? One has to suspect a bit of both is true.

Does donepezil slow Alzheimer's progression?

Donepezil is the generic name for Aricept, widely used these days, not only in diagnosed Alzheimer's disease, but also by some for MCI, with the idea that it might forestall the development of Alzheimer's. This study used processed MRI images to follow the volume of the hippocampi (structures in the brain that have to do with memory formation) at a 2-year interval. No apparent effect of donepezil was seen, however.

Wednesday, January 20, 2010

This is an empty blog except for this notice. I decided to delete all content, and since I didn't save any of it, it is simply gone, unless Google has secretly kept some memory of it somewhere.