Thursday, November 11, 2010

Dust in heat sink

What can four years of not cleaning do to your CPU’s heat sink?

The CPU was reaching temperatures of more than 100 degree Celsius at which time it would reboot.

Upon opening the heat sink was found to be clogged with dust. Cleaning it resulted in collection of this much dust:

2010-11-11 10.42.54

2010-11-11 10.43.18


1) Clean your heat sink regularly

2) If your PC abruptly reboots, it maybe time to clean the heat sink

My PC has been running smoothly since with max temperatures going only upto 60 degrees C.


Also posted at

Monday, November 01, 2010

London Eye.


This has quickly become one of the symbols of London. At night, lit in resplendent colours, it shines brightly over the Thames river.

Friday, October 29, 2010

Pictures of the day.


Windows Live Writer Test.


This is a post just to test if I can publish to blogger from Windows Live Writer. I have heard good things about this piece of software and hope it lives up to expectations.

Friday, April 23, 2010

Modulo 3

Taken from Codercorner..
Thanks to Christer Ericson, one of the fastest way to do:

int Index1 = n; // 0, 1 or 2
int Index2 =
int Index3 =


int Index1 = n; // 0, 1 or 2
int Index2 = (1 << Index1) & 3;
int Index3 = (1 << Index2) & 3;

Monday, December 29, 2008

An overrated problem.

This was a comment in response to Over-rate blues by Michael Jeh


I think over rates are an overrated problem (sorry about that). The blame for the Mohali test is surely to be placed at the authorities for scheduling it there in spite of knowing that Mohali, as well as other test centers in North India, would be severely affected by fog and short days. Surely other centers, like Bangalore, Hyderabad or Kolkatta could have been employed.

Another point is how people say that the last Mohali test kills test cricket when there is no mention of the Ind/Aus Nagpur test. Ponting's "tactics" to speed up the overrate did more to kill off test cricket by robbing an absorbing contest. I think the spirit of any sport is that all teams would strive to win (or draw if that is not possible) at all times. When a team like Australia does not do it when there was a distinct possibility, it does a lot of harm.

Disclaimer: This comes from an Indian supporter. Even to me it was plain obvious that the tactics resulted in India escaping making the result frivolous.

Sunday, December 21, 2008


"He is literally making the ball talk" -- Arun Lal.

I would like to see that.

Thursday, December 18, 2008

Optimal first innings score in a Test Match.

In this post, the optimal score that provides the team batting in the first innings the best chance of winning will be analyzed using statistics of matches gone by.

The thought occurred during the India - Australia series in October, 2008. India won the match in Mohali when they made a score of 400+ in the first innings. In Delhi, India made a score of 600+, but the match ended in a draw. At this point, I wondered whether a score of 400 - 500 would be better for the team batting first than a score in excess of 500. Sure enough, though due to idiotic 'tactics' by Ricky Ponting, India won in Nagpur when they made a score of 400 something in the match.

Closer inspection was warranted.

Using Cricinfo's stats guru, the following data can be obtained.

Team batting first scores 0-100 runs:

The stats show
Win Percentage: 0 / 12 = 0%
Loss Percentage: 11/ 12 = 91.7%
Draw Percentage: 1 / 12 = 8.3%

Team batting first scores 100-200 runs:

The stats show
Win Percentage: 8 / 84 = 9.5%
Loss Percentage: 65 / 84 = 77.3%
Draw Percentage: 11 / 84 =13.1%

Team batting first scores 200-300 runs:

The stats show
Win Percentage: 34 / 127 = 26.8%
Loss Percentage: 75 / 127 = 59.1%
Draw Percentage: 18 / 127 =14.2%

Team batting first scores 300-400 runs:

The stats show
Win Percentage: 33 / 94 = 35.1%
Loss Percentage: 35 / 94 = 37.2%
Draw Percentage: 25 / 94 =26.6%

Team batting first scores 400-500 runs:

The stats show
Win Percentage: 45 / 78 = 57.7%
Loss Percentage: 14 / 78 = 17.9%
Draw Percentage: 19 / 78 =24.4%

Team batting first scores 500-600 runs:

The stats show
Win Percentage: 23 / 55 = 41.8%
Loss Percentage: 2 / 55 = 3.6%
Draw Percentage: 30 / 55 = 54.5%

Team batting first scores more than 600 runs:

The stats show
Win Percentage: 13 / 23 = 56.5%
Loss Percentage: 0 / 23 = 0%
Draw Percentage: 10 / 23 =43.5%

The above records show the split for the matches over the past 10 years only. Considering only the first innings scores, it is clear that teams batting first win more when they score between 400-500 runs (win percentage of 57.7%) followed closely by when they score more than 600 runs (win percentage of 56.5%). Strangely, when the score is between 500-600, the win percentage drops to 41.8%. At the same time, when the score is between 400-500, the loss percentage is 17.9% whereas when the score increases above 500, it decreases dramatically to 3.6% when the scores are between 500-600 and 0% when it is above 600.

This is probably due to several reasons. A score of above 500 could indicate that the pitch is really good for batting making draw very likely. It could also indicate that to make scores of above 500 takes up more time than that required to make a score in excess of 400.

To conclude, if winning -- and not 'not losing' -- is the main objective, teams are better off scoring between 400 and 500 runs rather than more than 500 runs.