Lightbulb

看到的笑話:How Many Gizmodo Writers Does It Take to Change a Light Bulb?

我看到之後第一個想到的是 的 FAQ:How many FreeBSD hackers does it take to change a lightbulb? (有中文版的翻譯,即 要幾個 FreeBSD hacker 才能換掉一個電燈泡?)

不過 的比較有趣 :p (尤其是更熟悉 的權力運作後再來看這個笑話)

Intel Core Duo

亮相二十天後,CPU 的部分就已經找到 34 個問題:34 Design Flaws in 20 Days of Intel Core Duo

有些問題可以在軟體層 (在作業系統或是應用程式裡處理) 避開,有些問題只是會影響到效率 (像是該 cache 的部份沒有做),有些則是 debug 的時候會產生異常情況 (平常跑的時候不會),不過仍然有不少問題可能會導致莫名其妙當機 XD

呃,比較糟糕的是 似乎沒打算修... (反正當了就重開?)

eMule 的 Encrypted Connection

那邊看到 不打算加入 Encrypted Connection 的功能:emule 發展者對加密傳輸的政策

在 eMule 的 forum 上,Frequently Asked Features, read this before posting a new Request 這篇文章裡:

Features you won't see in eMule:
Encrypted transfer - This only wastes resources and offers you no advantage. It doesn't offer you more privacy as the receiver of the data always has to decrypt it and the receiver could be any untrusted person. Encryption would only help against a Man-In-The-Middle attack which never happens for P2P-Networks as there are easier ways (see the sentence above).

不過似乎沒抓到重點,Encrypted Connection 的目的只是要閃開 P2P Filtering 的設備...

牌型的強度

www.gg.caltech.edu/~jeff/knr.cgi 可以幫你算。

6-4:
K&R (AQTxxx Kxxx xx x) = 13.65
K&R (AQ9xxx Kxxx xx x) = 13.10
K&R (AQxxxx KTxx xx x) = 13.00
K&R (AQxxxx Kxxx xx x) = 12.80
K&R (AJTxxx Kxxx xx x) = 12.30
K&R (ATxxxx Axxx xx x) = 12.30
K&R (Axxxxx ATxx xx x) = 12.20
K&R (Axxxxx Axxx xx x) = 12.00

5-5:
K&R (AJxxx Axxxx xx x) = 12.75
K&R (ATxxx Axxxx xx x) = 12.25
K&R (AJTxx Kxxxx xx x) = 12.00
K&R (Axxxx Axxxx xx x) = 12.00

5-4:
K&R (ATxxx AJTx xx xx) = 12.15
K&R (AJTxx Axxx xx xx) = 12.10
K&R (AJTxx KJTx xx xx) = 12.00
K&R (AJxxx AJxx xx xx) = 12.00

所以這些牌都應該正常開叫...