Slow computer? Two free tools can help

This post was updated 7.24.10

*****************************

Is your computer getting a bit long in the tooth? Does it seem especially slow with visual applications like games and Photoshop? Perhaps it's also choking on Office2007 or Office 2010 graphics. The problem may lie with your graphics processing unit (GPU), otherwise known as a video card.

TheRaven found a free and easy tool you can download to benchmark GPU performance. This tool works equally well on desktops and laptops, although generally restrictive laptop designs greatly limit GPU upgrade options. Conversely, you have so many upgrade options in the desktop world it may be difficult to choose one. A slow desktop computer can be remedied for as little as $50 with the right GPU upgrade.

The first free tool is called FurMark and it's available here. Downloading and running FurMark takes no more than 5 minutes. It benchmarks your GPU's performance and provides a link to the most recent 400 benchmarks posted by other FurMark users. FurMark uses a (graphic representation of a) furry donut to test your GPU, which explains the FurMark logo:

The beauty of FurMark is that it boils down all factors affecting GPU performance to a single number.  FurMark data provides a basis for comparison of GPU upgrade options if you aggregate scores by GPU and compare results. Ozone3d.net provides a great tool but unfortunately you're on your own with analysis of FurMark data. That is, if you don't use a free data analysis tool developed by TheRaven.

This tool must be used with Excel2007 or Excel2010 and it can be downloaded (free) from Docstoc, here. It stores up to 5,000 benchmark scores and compares any three GPU's you select. Comparison selection is easy. The tool is currently loaded with 2,569 benchmark scores (obtained during July 2010) that correspond with 193 GPUs. The tool automatically generates a report of all GPUs, two sets of performance charts, Top-20 assessment and a comparative look at any three, user-selected GPUs. Here's an example of the latter:

 Click image to see full size

The analysis tool also contains instructions, explanations and user controls not shown in output. It's designed for use by people with minimal Excel skills. If you can copy data from ozone3d.net, follow a few simple instructions and operate a mouse, the analysis tool will help you identify a GPU solution. It automatically updates charts & report when new data is copied in. Do not attempt use with Excel 2003, because your computer will explode and your house will burn down (OK, not really, but the tool won't work with Excel2003).

Here's a short case study in how the data analysis tool can help. Have a look at this Top-20 list:

Click image to see full size

The ATI Radeon HD 5870 (4th from top) has an average FurMark score of 6,269. This average is based on 157 user submissions. The analysis tool applied data filters to remove outliers from extreme ends of reported GPU performance. Therefore, the average benchmark is a fair representation of Radeon HD 5870 performance, which is at the high-end of all reported GPUs.

Same for the NVIDIA GeForce GTS 250, which appears in the last line. It's average benchmark, which is one-half of the Radeon HD 5870, is based on 121 submissions. The Radeon HD 5870 is the primary component of video cards priced above $400 while the GeForce GTS 250 is found in cards priced between $100-$150, with many priced around $100.

This means that the ATI Radeon HD 5870 offers twice the performance of the GeForce GTS 250 but costs roughly 4-times more. Depending on budget and need, either choice, or one of many others, might be appropriate. The analysis tool helps cut through clutter in a crowded marketplace by converting data into information.You are equipped with fair, simple comparisons to use in evaluating product performance and value.


There are two things you must know before ordering a new GPU:

1. The type of motherboard slot the GPU must plug into and how many available slots, of each type, are in your computer.

2. Output (in watts) from your computer's power supply.

New, high-end GPU's typically impose a 350 or 400 watt power requirement but many desktop computers purchased 2+ years ago only have 300 watt power supplies. You'll need to do some research to find GPUs that work with 300 watt power supply and provide an adequate processing improvement. First, visit the website of your computer manufacturer to obtain its specs. Second, don't order any GPU until you've visited the manufacturer's web site and confirmed power requirements. Video card retailers are generally consistent in disclosing interface requirements but they do a poor job with power requirements.

It costs roughly $50-$100 to upgrade a computer power supply, assuming you'll provide the labor.

Send flowers

The Bureau of Labor Statistics (BLS) publishes occupation pay and employment data in May of each year. The data is "as-of" May of the preceding year, since the BLS needs a year to vet data, which covers pay range and employment for 800 occupations in 400 metro areas and 291 industries. Occupations are defined by the BLS while industries are defined using the North America Industry Classification System (NAICS). This post covers changes in employment by industry from 2005 to 2009.

TheRaven has summarized occupation-by-industry data into broadly defined industry groups. Groupings attempt to vertically align industries according to basic economic themes. To be clear, the data is authoritative and covers the entire U.S. workforce, except for farming; domestic help and freelancers. Bearing in mind that industries overlap and that no grouping system is perfect, this post uses summarized BLS data to throw light on a few aspects of the US job market not found in typical reportage.  

Click any chart to see it full-size.

Reported: bankruptcies of GM and Chrysler, loss of Big-3 jobs.
Missed: the Big-3 workforce is a very small piece of total auto industry employment. Due to Big-3 outsourcing and inroads by Toyota, Honda, Subaru, Hyundai, et al, component manufacturers employ the bulk of the auto manufacturing workforce and they took the brunt of layoffs. Big-3 employees got the ink but lesser-known, lower-paid people felt the pain.

Confused by the chart? 

Left side - employment by industry, by year.
Center - color codes (chart legend)

  Important: color code sequence matches left & right, reading from top to bottom.

Right side - percentage distribution of employment, by industry, by year.

Reported: individual retailers struggled during the recession, retail layoffs, Las Vegas implosion.
Missed:
1) Florists have fared worse than the gambling industry, with a (25%) employment drop vs. (23%) in gambling.
2) "General merchandise stores" employment is up 29% over 2005, with 2009 gaining 3%.

Reported: oil prices, gas prices, "pain at the pump", "drill baby, drill", coal mine disasters, West Virginia's eternal dependence on coal, rising atmospheric carbon, climate change, oil spills.

Missed: employment in core industries of the carbon-based economy has been slowly falling, even as oil & gas exploration ramped up, and these industries account for less than 2% of the U.S. workforce.

Reported: "bailouts", tea bagger rage at the Federal government, flourishing conspiracy theories and extremist groups, "bloated Federal bureaucracy", slow dispersal of stimulus money.

Missed:
1) State & local governments employ four-times more people than the Federal government.
2) 87% of new government jobs created by the 2009 hiring surge were State & local.
3) all government jobs in 2009 accounted for only a slightly larger share of the U.S. workforce as compared to 2005.

Reported: anything branded "Apple", every Intel press release, lawsuits, video game releases, all "smart" objects, including the stupid ones.

Missed:
1) U.S. electronic manufacturing in long-term decline.
2) Digital economy jobs are up almost 10% over 2005, even after a 2% dip in 2009.
3) Systems design is flourishing.

Reported: newspaper financial trouble, Leno vs. Conan, Lost, 24.

Missed:
1) Old-media industries are in an across-the-board, long-term decline (it's not just newspapers).
2) Industry group share of U.S. workforce fell from 2.2% to 1.9%.
3) Internet-driven creative destruction is hitting in-force (aided by journalist incompetence).

Reported: organic vs. non-organic, Michael Pollan, franchise restaurant financial trouble, farming efficiencies have reduced farm employment to a tiny fraction of the U.S. workforce.

Missed:
1) Industry employment ratios from 3-4 generations ago have reversed. Where once a vast number of farmers produced food that moved through thinly-staffed distribution, a comparatively tiny farming workforce now plugs into a vast distribution network.
2) Where 3-4 generations ago very few people ate out, America now has a huge, sophisticated set of restaurant industries that have significantly displaced meals at home.
3) Almost one-in-eight U.S. workers is connected to food and this industry grouping shows a 3.8% net gain over 2005 employment, after shedding 177,980 jobs in 2009.

Reported: so-called Obama-care, so-called town-hall meetings, health-care share of GDP, hospitals feeling effects of uninsured population

Missed:
1) Hospitals & outpatient centers account for more than one-third of health care employment.
2) While hospital employment grew 7% between 2005 and 2009, ambulatory/outpatient industries grew a (combined) 18%.
3) Home care had the fastest employment growth, increasing by one-quarter between 2005 and 2009.

Reported: collapsing real estate markets, over-building, death of the suburbs, Case-Schiller, home buyer credits, foreclosures, short sales, home builder financial issues, mass layoffs.

Missed:
1) Similar to the auto industry, collapse of home building rippled through many supplier industries.
2) One-in-eight jobs related to home construction was lost with most of the decline in 2009.
3) This industry group dropped from 8.0% to 6.9% of the U.S. workforce.

Reported: "off-shoring", jobs lost to overseas competition, "no job safe".

Missed:
1) Domestic out-sourcing industries are an immense group, accounting for 1-in-9 U.S. jobs.
2) In 2009, this industry group gave up almost all the 1+ million jobs it gained over 2005-2008.
3) Employment services were crushed, losing 1 million jobs between 2007 and 2009.

****************************************

A subsequent post may look at more winners & losers among individual industries and occupations. This was the introductory, top-level view.

(All analysis by TheRaven)