Comments on: How I Found A Bug In iPhone Software 2.0 http://mazedlx.net/2008/07/16/how-i-found-a-bug-in-iphone-software-20/ what the f prompt Fri, 09 Jan 2009 22:59:57 +0000 http://wordpress.org/?v=2.7 hourly 1 By: maze http://mazedlx.net/2008/07/16/how-i-found-a-bug-in-iphone-software-20/comment-page-1/#comment-2038 maze Thu, 18 Sep 2008 19:00:01 +0000 http://mazedlx.net/?p=2824#comment-2038 I tried to reproduce the same bugs on my iPod touch and you're right. When using the calc that comes with MacOS X i get 2,688117141816134e+43 for the first equation, the iPod shows me 268811,714... which is in fact off by 1e39. I can also confirm the second bug. So whatever you do, planning a skyscraper or a nuclear bomb, do not use the iPod touch / iPhone for calculations! I tried to reproduce the same bugs on my iPod touch and you’re right. When using the calc that comes with MacOS X i get 2,688117141816134e+43 for the first equation, the iPod shows me 268811,714… which is in fact off by 1e39.

I can also confirm the second bug. So whatever you do, planning a skyscraper or a nuclear bomb, do not use the iPod touch / iPhone for calculations!

]]> By: Marc http://mazedlx.net/2008/07/16/how-i-found-a-bug-in-iphone-software-20/comment-page-1/#comment-2037 Marc Thu, 18 Sep 2008 18:53:45 +0000 http://mazedlx.net/?p=2824#comment-2037 Here is a real calculator bug that I have submitted to Apple via their feedback form: Flip to the scientific (landscape) mode and enter 100 [displays 100] +/- [displays -100] e^x [displays 3.72...e-44 which is correct] 1/x [displays 268,811.714... which is off by a factor of 1e39] Here's a probably related bug where there is a loss of precision: 1 EE 32 +/- = [displays 1e-32 the '=' is necessary to work around another bug. if you skip the '=' you will see another bug] 1/x [displays 1.111...e+32] Here is a real calculator bug that I have submitted to Apple via their feedback form:

Flip to the scientific (landscape) mode and enter
100 [displays 100]
+/- [displays -100]
e^x [displays 3.72...e-44 which is correct]
1/x [displays 268,811.714... which is off by a factor of 1e39]

Here’s a probably related bug where there is a loss of precision:
1 EE 32 +/- = [displays 1e-32 the '=' is necessary to work around another bug. if you skip the '=' you will see another bug]
1/x [displays 1.111...e+32]

]]>