- Apr 11 Sat 2009 15:39
-
[Notes]筆記另類觀點：筆記成功術
- Apr 09 Thu 2009 11:25
-
[Notes]在自己的文章中引用程式碼
在css中 加入下面code
用這個方法來做programming的筆記真的很方便
ps. 還有另一個不錯的方法syntaxhighlighter，我目前使用上還會有程式碼斷行的問題，等問題解決了在放上來
code{
display: block; /* fixes a strange ie margin bug */
font-family: Courier New;
font-size: 11pt;
overflow:auto;
background: #f0f0f0 url(http://chenkaie.googlepages.com/Code_BG.gif) left top repeat-y;
border: 1px dashed #333;
padding: 5px 5px 5px 20px;
max-height:200px;
line-height: 1.2em;
margin: 5px 5px 0 15px;
}
用這個方法來做programming的筆記真的很方便
ps. 還有另一個不錯的方法syntaxhighlighter，我目前使用上還會有程式碼斷行的問題，等問題解決了在放上來
code{
display: block; /* fixes a strange ie margin bug */
font-family: Courier New;
font-size: 11pt;
overflow:auto;
background: #f0f0f0 url(http://chenkaie.googlepages.com/Code_BG.gif) left top repeat-y;
border: 1px dashed #333;
padding: 5px 5px 5px 20px;
max-height:200px;
line-height: 1.2em;
margin: 5px 5px 0 15px;
}
- Apr 06 Mon 2009 18:25
-
[article]轉錄‧困境與抉擇
這是一篇很棒的文章，可以勉勵許多人!
有時候，在做某些選擇，太執意強求會換來患得患失的失落，也許文章中所提到的"生命是一種長期而持續的累積過程，絕不會因為單一的事件而毀了一個人的一生，也不會因為單一的事件而救了一個人的一生。"，而"一條看起來很遠的路，說不定是最短的路"，這句話也不斷的在我腦中不斷的響起...
有時候，在做某些選擇，太執意強求會換來患得患失的失落，也許文章中所提到的"生命是一種長期而持續的累積過程，絕不會因為單一的事件而毀了一個人的一生，也不會因為單一的事件而救了一個人的一生。"，而"一條看起來很遠的路，說不定是最短的路"，這句話也不斷的在我腦中不斷的響起...
- Apr 01 Wed 2009 18:39
-
[article][CUDA]轉錄‧CUDA程式設計 I -- 簡介
最近在研究nVidia CUDA剛好在PTT VideoCard板上，看到有人很專業的的分享程式設計心得
本文已徵求作者同意將轉錄在自己的blog做成筆記
//本文轉錄自 VideoCard 看板
//作者：a5000ml
[引言]
從軟體設計角度來看，CUDA 寫作並不困難，困難之處在於如何最佳化，它需要了解不少硬體的細節。一般而言未最佳化的程式，在 GPGPU 上面執行，可以比傳統的 CPU 快上 5~10 倍，而最佳化過的程式，往往還能再增速 5~10 倍，達到 25~100 倍的效能。
CUDA 和傳統 C++ 最大的差異在於「平行化的程式設計」vs.「序列化的程式設計」，例如傳統上透過迴圈執行數千次的程式碼，在 CUDA 上就可以將它拆解成數百個同時執行的執行緒，每個執行緒只執行十幾次而己，因此產生數十到數百倍的效能。
CUDA 是在傳統 C++ 的基礎上，加入一些延伸語法，以及輔助的函式庫，而形成的一種程式語言，一般而言只要熟悉 C 或 C++，就很容易上手 (可能比 Java 容易上手，因為傳統 C/C++ 的函式庫都可以用)，編譯好的程式碼，也可以跟其它語言做不錯的聯結。
在接下來幾篇文章中，我們從簡單的軟體設計角度開始，介紹 CUDA 如何寫作，逐漸帶出硬體細節，然後介紹最佳化的一些技術。
本文已徵求作者同意將轉錄在自己的blog做成筆記
//本文轉錄自 VideoCard 看板
//作者：a5000ml
[引言]
從軟體設計角度來看，CUDA 寫作並不困難，困難之處在於如何最佳化，它需要了解不少硬體的細節。一般而言未最佳化的程式，在 GPGPU 上面執行，可以比傳統的 CPU 快上 5~10 倍，而最佳化過的程式，往往還能再增速 5~10 倍，達到 25~100 倍的效能。
CUDA 和傳統 C++ 最大的差異在於「平行化的程式設計」vs.「序列化的程式設計」，例如傳統上透過迴圈執行數千次的程式碼，在 CUDA 上就可以將它拆解成數百個同時執行的執行緒，每個執行緒只執行十幾次而己，因此產生數十到數百倍的效能。
CUDA 是在傳統 C++ 的基礎上，加入一些延伸語法，以及輔助的函式庫，而形成的一種程式語言，一般而言只要熟悉 C 或 C++，就很容易上手 (可能比 Java 容易上手，因為傳統 C/C++ 的函式庫都可以用)，編譯好的程式碼，也可以跟其它語言做不錯的聯結。
在接下來幾篇文章中，我們從簡單的軟體設計角度開始，介紹 CUDA 如何寫作，逐漸帶出硬體細節，然後介紹最佳化的一些技術。
- Mar 03 Tue 2009 20:43
-
[music]縱貫線SUPERBAND-亡命之徒
出發啦 不要問那路在哪? 迎風向前 是唯一的方法
出發啦 不想問那路在哪? 運命哎呀 什麼關卡?
當車聲隆隆 夢開始陣痛 它卷起了風 重新雕塑每個面孔
夜霧那麼濃 開闊也洶湧 有一種預感 路的終點是迷宮....
這段副歌的旋律一直在我耳邊不停的重複
我也好想去現場聽呀!!這是多麼熱血~
出發啦 不想問那路在哪? 運命哎呀 什麼關卡?
當車聲隆隆 夢開始陣痛 它卷起了風 重新雕塑每個面孔
夜霧那麼濃 開闊也洶湧 有一種預感 路的終點是迷宮....
這段副歌的旋律一直在我耳邊不停的重複
我也好想去現場聽呀!!這是多麼熱血~
- Feb 19 Thu 2009 21:36
-
[閒聊]我與Gina的讀書計畫
2009年讀書計畫
- Feb 13 Fri 2009 13:04
-
[article]轉錄‧CUDA
最近可能要開始玩CUDA，先來搞懂CUDA是甚麼!!
CUDA(Compute Unified Device Architecture, 計算統一設備架構)是NVIDIA所推出的技術，是NVIDIA的GPGPU的正式名稱。透過這個技術，用家可利用NVIDIA的GeForce 8以後的GPU和較新的Quadro GPU進行計算。亦是首次可以利用GPU作為C-編譯器的開發環境。NVIDIA行銷的時候，往往將編譯器與架構混合推廣，做成混亂。實際上，CUDA架構可以相容OpenCL或者自家的C-編譯器。無論是CUDA C-語言或是OpenCL，指令最終都會被驅動程式轉換成PTX代碼，交由顯示核心計算。
CUDA(Compute Unified Device Architecture, 計算統一設備架構)是NVIDIA所推出的技術，是NVIDIA的GPGPU的正式名稱。透過這個技術，用家可利用NVIDIA的GeForce 8以後的GPU和較新的Quadro GPU進行計算。亦是首次可以利用GPU作為C-編譯器的開發環境。NVIDIA行銷的時候，往往將編譯器與架構混合推廣，做成混亂。實際上，CUDA架構可以相容OpenCL或者自家的C-編譯器。無論是CUDA C-語言或是OpenCL，指令最終都會被驅動程式轉換成PTX代碼，交由顯示核心計算。
- Feb 12 Thu 2009 19:39
-
[article]轉錄‧C++ Gossip: C++爸爸（Bjarne Stroustrup）給C++初學者的信
- Jan 09 Fri 2009 16:50
-
[閒聊]懷念下午有咖啡喝的日子
離開Sinica已經有9天了!
懷念下午喝咖啡及chatting的日子
近況，在老爸的公司當雜工中
懷念下午喝咖啡及chatting的日子
近況，在老爸的公司當雜工中
- Dec 24 Wed 2008 16:41
-
[閒聊]再一次的半馬拉松挑戰(2008Taipei ING Marathon)
今年的路跑活動就在12月21日的2008Taipei ING Marathon畫下句點，感謝一路支持我的Academia Sinica Team(Ay、羿喬還有實驗室的同事)、Gina、Dumpling、James、Paul等等。
從未跑過超過10km的我一路從8月份的Nike 10K+開始、11月的太魯閣馬拉松的21km，在到現在的ING馬拉松21km，這些都是不可思議的故事!!我想我一輩子都難以忘懷我通過終點的那份喜悅與榮耀!!
詳細資料請參閱"路跑協會網站"
ING官網所提供的成績
這次的成績比太魯閣路跑成績進步了11分40幾秒，值得慶賀!!
21.0975km半馬拉松路線圖
臺北市政府前廣場→ 仁愛路 → 中山南路 → 中山北路 → 北安路 → 明水路 → 堤頂大道 → 麥帥二橋 →市民高架(折返) → 基隆路 → 信義路 → 光復南路→ 仁愛路 → 臺北市政府前廣場
值得紀念的榮耀
恭喜Ay以4小時0分52秒跑完42.195km全馬拉松
Paul、me、James and Dumpling
拖著抽筋的小腿奮力一跳...
這就是完成的成就感呀!!
預告:明年的計畫 <橫渡日月潭>
路跑花絮
- Dec 17 Wed 2008 16:53
-
[Notes]Advice for potential graduate students
We currently have room in the lab for more graduate students. Before you apply to this lab or any other, there are a few things to keep in mind. First, be realistic about graduate school. Graduate school in biology is not a sure path to success. Many students assume that they will eventually get a job just like their advisor's. However, the average professor at a research university has three students at a time for about 5 years each. So, over a career of 30 years, this professor has about 18 students. Since the total number of positions has been pretty constant, these 18 people are competing for one spot. So go to grad school assuming that you might not end up at a research university, but instead a teaching college, or a government or industry job. All of these are great jobs, but it's important to think of all this before you go to school.
Second, choose your advisor wisely. Not only does this person potentially have total control over your graduate career for five or more years, but he/she will also be writing recommendation letters for you for another 5-10 years after that. Also, your advisor will shadow you for the rest of your life. People will always think of you as so-and-so's student and assume that you two are somewhat alike. Finally, in many ways you will turn into your advisor. Advisors teach very little, but instead provide a role model. Consciously and unconsciously, you will imitate your advisor. You may find this hard to believe now, but fifteen years from now, when you find yourself lining up the tools in your lab cabinets just like your advisor did, you'll see. My student Alison once said that choosing an advisor is like choosing a spouse after one date. Find out all you can on this date.
Finally, have your fun now. Five years is a long time when you are 23 years old. By the end of graduate school, you will be older, slower, and possibly married and/or a parent. So if you always wanted to walk across Nepal, do it now. Also, do not go to a high-powered lab that you hate assuming that this will promise you long-term happiness. Deferred gratification has its limits. Do something that you have passion for, work in a lab you like, in a place you like, before life starts throwing its many curve balls. Your career will mostly take care of itself, but you can't get your youth back.
If, after reading this, you want to apply to this lab, we would love to hear from you.
source: The Johnsen Lab, Duke University, Department of Biology
Second, choose your advisor wisely. Not only does this person potentially have total control over your graduate career for five or more years, but he/she will also be writing recommendation letters for you for another 5-10 years after that. Also, your advisor will shadow you for the rest of your life. People will always think of you as so-and-so's student and assume that you two are somewhat alike. Finally, in many ways you will turn into your advisor. Advisors teach very little, but instead provide a role model. Consciously and unconsciously, you will imitate your advisor. You may find this hard to believe now, but fifteen years from now, when you find yourself lining up the tools in your lab cabinets just like your advisor did, you'll see. My student Alison once said that choosing an advisor is like choosing a spouse after one date. Find out all you can on this date.
Finally, have your fun now. Five years is a long time when you are 23 years old. By the end of graduate school, you will be older, slower, and possibly married and/or a parent. So if you always wanted to walk across Nepal, do it now. Also, do not go to a high-powered lab that you hate assuming that this will promise you long-term happiness. Deferred gratification has its limits. Do something that you have passion for, work in a lab you like, in a place you like, before life starts throwing its many curve balls. Your career will mostly take care of itself, but you can't get your youth back.
If, after reading this, you want to apply to this lab, we would love to hear from you.
source: The Johnsen Lab, Duke University, Department of Biology
