Archive for December, 2008

Testing eboostr on an older machine

I decided to try out the eboostr 3.0 beta to see if it can speed up my mom’s computer (an ancient eMachine T1221 with a Celeron 1.3 Ghz). Normally, adding more memory is best, but the machine is maxed out at 512Mb. I figured that adding a disk cache to an usb stick may be able to add additional performance. Apparently, I was wrong.

The machine is setup to use a 2Gb stick of Cruzer Micro (readyboost ready). HDtach returned about 27 Mb/s read speed with an access time of 0.6ms. The machine has an USB 2.0 card, so it should be fast enough.

I ran several test like startup, shutdown, opening different applications, opening web pages. All of the test were no faster than before eboostr. I examined the cache and noted that the files being opened are in the cache, it’s just that they didn’t return quickly enough. If it worked, it should be noticeable.

I decided to do some investigating and ran the eboostr speed test. It came back with a ratio of only speed ratio of 1.13 and 100% cache hit. A 13% speed bump is probably too slow to make a difference and this is with a reasonably fast stick. I installed eboostr and used memory as a cache, but the ratio only went up to 1.59. This is a lot better, but surprisingly low for a memory cache. Keep in mind that while it’s faster to find a file on the USB drive, the cpu cost for getting it is higher. HDtach indicated that it only use 5% CPU to read from the IDE drive and 19% CPU to read from the USB drive.

I also tried eboostr on an AMD 2600+ desktop with 2Gb of memory and allocated 1Gb of memory to cache. There was no noticable speed difference before and after eboostr. The speed test indicated that the speed ratio is only 1.89, which is better than the eMachine, but not good enough in my opinion.

eboostr did not work for the two older machines that we tried it on.

Advertisements

December 31, 2008 at 2:56 pm 2 comments

How I selected my Antivirus software for Windows

For a long time, I used AVG free on my machines. Recently, I notice that the memory footprint for AVG was getting large and I wanted to know if there was a better free Anti-virus software out there. There are essentially 3 main free Anti-virus software: Avast!, Avira, and AVG. If you do a google search, you’ll notice tons of post comparing the merits of one over the others. The following links is an actual benchmark that compare the Anti-virus products and were used as reference.
http://www.av-comparatives.org/seiten/ergebnisse_2008_11.php
http://www.passmark.com/ftp/antivirus_09-performance-testing-ed1.pdf
http://www.maximumpc.com/print/4230

Memory Footprint Ranking
If you have plenty of memory, you can probably careless about this category. However, a lot of older machine or netbook may have limited memory, so every bit of memory saved counts. Keep in mind I am more interested in idle memory footprint than when it’s scanning.

Memory footprint of the anti-virus product at idle can be ranked in the following order according to the passmark article. I also try installing each product on my Vista machine and verified the ranking.

Product Memory Usage in Mb
Avast! Antivirus Home 18.05
Avira Antivir free 22.3
AVG Free 56.06

While Avast! And Avira were similar to each other in memory footprint, AVG appears to be using twice the memory of the other two product.

Virus Detection
How well does the three product detect virus. It appears that Antivir is somewhat better than the other. I am not convinced that Antivir is actually that much better, but various post online indicates that its detection rates were better than average.

Product Detection Rate (from av comparative)
Avira Antivir free 71%
AVG Free 43%
Avast! Antivirus Home 40%

Background Performance
How much performance loss do you suffer running the anti-virus software. Judging from the test results from passmark, the three products are very similar in performance. Differences show up in File Open where AVG and Avira is nearly twice as fast as Avast and file copy where AVG is noticeably slower than the other two products.

The Maximum PC article seems to complain that the AVG had the biggest negative performance.

Scanning Speed Ranking
Of the three products, the scan speed varies by quite a lot according to passmark.

Product Time (sec)
Avira Antivir free 68.8
Avast! Antivirus Home 116.87
AVG Free 364.2

The Maximum PC article indicated that the Avast is very slow but didn’t provide numbers on how it compare with the other two product.

User Interface
User interface is very subjective, but in my opinion, AVG has the best and clearest interface. Avast is the worse. The interface looks like a media player, which doesn’t quite fit the paradigm of a virus scanner at all. Avira is somewhere in between the two. It’s not great, but at least it works.

Registration
How annoying is the registration process for the Virus. AVG is probably the least intrusive. You are prompted to register, but you do not have to. Avira is similar but throws up a pop-up every times the program updates the virus definition. The most annoying is Avast, which requires you to registered for a license key or the program will expire in 60 days. The key last only a year, so you must registered again next year.

Features
All of the product offers email and spyware protection. Avast also has IM scanning, but does not have scheduled scanning. Unlike the other two product, AVG does not have rootkit protection.

Recommendation
In my opinion, the best free anti-virus is currently Avira Antivir. It has the fastest scan, best detection rate, and second best memory footprint. It’s annoying to get a pop-up on every update, but that is a small price to pay for a free product. I also want the anti-rootkit feature and scheduled scanning since I want to do the scanning while I am not using the machine.

December 26, 2008 at 11:19 am 4 comments


Calendar

December 2008
M T W T F S S
« Nov   Mar »
1234567
891011121314
15161718192021
22232425262728
293031  

Posts by Month

Posts by Category