v******l
发帖数: 512 | 1 I don't quite get what you mean by "can not alias it". Your sample code is a
typical aliasing and as long as there is no memory issue (overflow or
something) there is nothing wrong with it.
Look at it this way: You need a buffer by using malloc you get a big chunk
of memory pointed by void *p0. There is nothing wrong you first cast it to
int* and fill it with a bunch of integers, then cast to float* and reads
them out as floats. It is merely a re-interpretation of data and it behaves
just like a... 阅读全帖 |
|
t****t
发帖数: 6806 | 2 What you describe is exactly undefined by standard, as I showed in my
example. it is as undefined as "a[i]=i++", or "j=(i++)*(i++)". read standard
3.10, and gcc manual "-fstrict-aliasing".
you may use union to achieve the desired result with defined behaviour.
a
behaves |
|
v******l
发帖数: 512 | 3 Ok I think I see your point. Still, there is a difference between strict-
aliasing rule and "a[i]=i++" situation, the later is semantically ambiguous
by itself, while the alias issue is actually a BREAK of otherwise perfect
logic as a price paid for aggressive optimization.
standard |
|
t****t
发帖数: 6806 | 4 you may have a point, but "otherwise perfect logic" is defined by whom?
AFAIK, strict-aliasing is enforced as early as C89. maybe some compilers
allow it, but that doesn't make it legal.
ambiguous |
|
v******l
发帖数: 512 | 5 I said "perfect logic" because there is no underline logic flaw as it is.
Problem only occurs when the strict-aliasing rule assumes "pointers of
different types cannot point to the same memory address". This assumption
adds complexity to the language and is anti-intuitive, thus make the
learning curve a little bit steeper, but sure it is well justified by
performance improvement and other benefits. |
|
r*****3
发帖数: 143 | 6 中文名: 新C++标准：C++0x 概述
原名: Overview of the New C++ : C++0x
作者: Scott Meyers
图书分类: 软件
资源格式: PDF
版本: 英文文字版
出版社: Artima
书号: 978-0-596-80197-6
发行时间: 2011年
地区: 美国
语言: 英文
简介:
内容介绍：
Specification of the next version of C++ (“C++0x”) is nearing completion,
and many compilers (e.g., Visual C++ and Gnu C++) already offer several
features from the revised language. And such features! auto-declared
variables reduce typing drudgery and syntactic noise; Unicode and threading
support address important functional... 阅读全帖 |
|
t****t
发帖数: 6806 | 7 it's the requirement of C, so it's the problem of the program. |
|
|
t****t
发帖数: 6806 | 9 这是不相干的. 以前有人举的例子很好, 说有人在14街闯红灯总是不被抓, 在15街闯红
灯就被抓, 所以就得出结论说15街闯红灯就是安全的?
但是从根本上就不应该闯红灯. |
|
s********k
发帖数: 6180 | 10 你说的这个没错，但是是否证明这个是一个bad design，根本就不应该有这个限制？ |
|
g**c
发帖数: 144 | 11 我过去是用 matlab，R 之类的作图，贴在 powerpoint 里，用 powerpoint 的文本
框，箭头啥的标注。
现在要用 latex 写东西，被迫把这些加工过的图都存成 .png 格式，然后用 pdfl
atex。可惜了些。难道非要在 matlab 里加所有标注再存成 .eps 么？可以直接把
office 里的东西转成 .eps 图么（除了print成ps再转eps的那种恐怖方法）？位
图又大又傻还容易 aliasing。
还有，要是单纯画示意图，有啥软件呢？ |
|
c*****t
发帖数: 1879 | 12 I don't think aliases work in shell scripts. |
|
w*****n
发帖数: 94 | 13 check whether rm is aliased to 'rm -i' somewhere. |
|
m******g
发帖数: 91 | 14 看gentoo网上提供的说明，需要对/etc/make.conf进行配置，
让gcc针对特定的CPU进行编译优化。
好的。我的机器是pentium 233MHz MMX,
因此我改动/etc/make.conf为：
CHOST="i586-pc-linux-gnu"
CFLAGS="-march=pentium-mmx -O3 -pipe -fomit-frame-pointer"
CXXFLAGS="${CFLAGS}"
可是在执行kernel编译时，出现的却是：
gcc -D__KERNEL__ -I/usr/src/linux-2.4.20-gentoo-r5/include -Wall
-Wstrict-prototypes -Wno-trigraphs -O2 -fno-strict-aliasing
-fno-common -fomit-frame-pointer -pipe -mpreferred-stack-boundary=2
-march=i686 -DKBUILD_BASENAME=main -c -o init/main.o init/main.c
是针对i |
|
W******r
发帖数: 789 | 15
我直接贴吧。
Deadly Diabetes Deception
Introduction
If you are an American diabetic, your physician will never tell you that
most cases of diabetes are curable. In fact, if you even mention the "cure"
word around him, he will likely become upset and irrational. His medical
school training only allows him to respond to the word "treatment". For him,
the "cure" word does not exist. Diabetes, in its modern epidemic form, is a
curable disease and has been for at least 40 years. In 2001, the most
recent y... 阅读全帖 |
|
O******e
发帖数: 734 | 16 Using ** to deal with 2D arrays normally means that the array is not
stored contiguously in memory, and for nonsparse algorithms manipulating
the entire array at once the fragmentation can inhibit code optimization
since the compiler will generally have difficulty resolving aliasing and
flow
dependency issues.
FFTW and other references suggest allocating the entire array using *,
then using ** to point to the beginning of each row if you want the
convenience of using [][] indexing.
memory
[]
th |
|
h**********c
发帖数: 4120 | 17 The figure above is 3.14 iis a pstex file
3.15 is a p png file.
The .tex is built to dvi. From the dvi we print screen as the above.
From further experiments, it is best to convert pstex to eps file. Then
include the .eps file in the .tex file. While building pdf file ((fedora
kdvi export). Kdvi seems to be able do the anti-aliasing work.
Although in the built pdf file, the eps file looks poor, but the printed
quality is better than the the png file.
So, if you try to do a presentation with your |
|
a****l
发帖数: 8211 | 18 depends on your application. If your application is noise-sensitive, you
MUST have anti-aliasing filter, otherwise, you can do away with it.
带宽
力 |
|
h********t
发帖数: 555 | 19 The original question is why quantization noise is band limited to fs/2.what
you claimed is that the spectrum of the sampled data is band limited to fs/
2. From design aspect,as an approximation, we may assume that is true, but
strictly speaking, neither of them is really band limited to fs/2.
with ideal impulse sampling, for the sampled data, the high frequency
spectrum of the original signal will be aliased back to lower frequency,
but the lower frequency components will get high frequency |
|
n****t
发帖数: 170 | 20 采样频率得严格大于带宽的两倍，你的例子刚好是等于的情况，你可以考虑频域上，有
ALIASING产生 |
|
n****t
发帖数: 170 | 21 这个问题涉及到傅立叶级数的收敛，有L1和L2收敛的分别，较好的理解需要建立在分析
测度的概念上，这样才可以从严格意义上处理广义上的函数。我想在gallager的书
Principles of Digital Communication中（第四章）有很好的介绍，我比较喜欢。
ALIASING |
|
x****g
发帖数: 2000 | 22 离散时域信号的频谱是你连续时域信号的频谱在-fs/2到fs/2之间的部分
如果采样率过低会有aliasing
时域信
如何说？谢谢 |
|
s**g
发帖数: 66 | 23 其实他做这个已经很久啦，不知道有什么新进展。他做这个也是有渊源的，八十年代初
他在continuous time filter 方面发了不少文章，这方面的研究到上个世纪末也就差
不多了。那就要找一个相关的，于是这个ctdsp来了。可是好像跟进的人不多。
Power saving is a big advantage for similar ckt; but I feel SNR penalty is
high without anti aliasing.
I also thought it might be interesting to integrate a log amp front end into
the system to improve dynamic range and furthermore the DSP in log domain
might be really power saving - multiply becomes sum.
For SAR ADC I see more sampling/standby/sampling scheme are implem... 阅读全帖 |
|
k****e
发帖数: 126 | 24 hosted环境和freestanding环境区别还是蛮大的，lz这种因该没那么容易，尤其是没源
代码的情况。
多大的字符集，CJK还是GB18030？什么字体？点阵还是要TrueType（系统支持TrueType
不），要不要支持anti-aliasing？都关系到要花多少钱买，免费汉化一些桌面软件小
打小闹可以，做产品当心人家和你打官司哈。
不给源代码的情况下，俺觉得比较可行的办法是加一套单独的显示系统上去，转换原来
系统里的显示信息，在对应的去渲染每个汉字或者整个string到显示屏上。如果想在原
系统上只加入字库信息进行汉化，不给源代码恐怕很难实现啊。 |
|
z*****n
发帖数: 7639 | 25 你说的是quantization noise。
他说的有alias存在的情况，不考虑quantization noise，
如果sampling frequency is Fs，对于某一频率f的所有的
高频aliases，包括Fs-f, Fs+f, 2*Fs-f, 2*Fs+f...，都
会被Fs表现为分量f。 |
|
d********i
发帖数: 91 | 26 is this for AD -> DA or DA -> AD?
Are the folded aliases (back to f) regarded as the noise to degrade the SNR? |
|
z*****n
发帖数: 7639 | 27 分清quantization noise和aliasing。 |
|
c**l
发帖数: 159 | 28 | | 比较难搞。换成 x^2 行不行。 可以看一下这片。
K. G. Gard, H. M. Gutierrez, and M. B. Steer, “Characterization of spectral
regrowth in microwave amplifiers based on the nonlinear transformation of a
complex Gaussian process,” Microwave Theory and Techniques, IEEE
Transactions on, vol. 47, no. 7, pp. 1059–1069, 1999.
你的是real的，应该有些经典结果。你这个还有 aliasing的问题，非线性产生的频谱
扩展又alias back，导致频谱平了。 |
|
c**l
发帖数: 159 | 29 | | 比较难搞。换成 x^2 行不行。 可以看一下这片。
K. G. Gard, H. M. Gutierrez, and M. B. Steer, “Characterization of spectral
regrowth in microwave amplifiers based on the nonlinear transformation of a
complex Gaussian process,” Microwave Theory and Techniques, IEEE
Transactions on, vol. 47, no. 7, pp. 1059–1069, 1999.
你的是real的，应该有些经典结果。你这个还有 aliasing的问题，非线性产生的频谱
扩展又alias back，导致频谱平了。 |
|
h**e
发帖数: 9290 | 30 如果sub-sampling,数字信号同样有噪声Aliasing的问题
其实数字信号也可以同样先混频，滤波 |
|
s***d
发帖数: 15421 | 31 正解 所以采样的时候 采样前都要有anti aliasing filter
★ 发自iPhone App: ChineseWeb 7.8 |
|
w**a
发帖数: 1024 | 32 有重叠部分，就会叠加起来。这也是ALIASING 的原因啊。
结果 |
|
b**g
发帖数: 335 | 33 C has unrestricted pointer aliasing, but Fortran does not,
so C compiler cannot generate highly efficient code.
http://en.wikipedia.org/wiki/Restrict
C99 has "restrict" keyword to address this issue.
(F90 has pointers, and by design they are all restrict) |
|
m*****h
发帖数: 2292 | 34 你如果要和别人说为什么用数字来取代模拟把采样定理拿出来说事则是没有说中要点
多了采样只能说明数字比模拟要多做一步，系统收发要多两个模块，但并不能说明数字
比模拟好
你说我讲句话本来就是模拟信号，你偏要先把它转成数字的，然后调制成模拟的，再采
样量化成数字的，再转成模拟的信号来听，多折腾啊，你用这个怎么来说服别人数字比
模拟好？（如果说信息丢失很少只能说明数字的性能不比模拟差多少。。和好没关系）
所以奠定了数字时代的基础的Shannon 1937, Shannon 1948两论文里没采样定理什么事
，采样只是数字化的手段，不是原因
~~~~~~~~~~~~~~~~~~~~~~~~~~~~no，当然可以直接发送数字信号
真正的调制是很复杂的，不是什么只有调频，调幅，或调相之分
首先是在baseband做两级调制，然后再在passband做调制
而且调制中恰恰又要利用aliasing而不是避免它 |
|
m*****h
发帖数: 2292 | 35 你如果要和别人说为什么用数字来取代模拟把采样定理拿出来说事则是没有说中要点
多了采样只能说明数字比模拟要多做一步，系统收发要多两个模块，但并不能说明数字
比模拟好
你说我讲句话本来就是模拟信号，你偏要先把它转成数字的，然后调制成模拟的，再采
样量化成数字的，再转成模拟的信号来听，多折腾啊，你用这个怎么来说服别人数字比
模拟好？（如果说信息丢失很少只能说明数字的性能不比模拟差多少。。和好没关系）
所以奠定了数字时代的基础的Shannon 1937, Shannon 1948两论文里没采样定理什么事
，采样只是数字化的手段，不是原因
~~~~~~~~~~~~~~~~~~~~~~~~~~~~no，当然可以直接发送数字信号
真正的调制是很复杂的，不是什么只有调频，调幅，或调相之分
首先是在baseband做两级调制，然后再在passband做调制
而且调制中恰恰又要利用aliasing而不是避免它 |
|
J****r
发帖数: 682 | 36 有一问题请教，在申请回事考试的指纹卡上的"Residence of person fingerprints 和
"signaniture of person fingprinted" 都填申请人的名字吗？还有一个问题是“
Aliases"
是别名的意思吗？ 如果没有别名（ 曾用名），应该怎样填写？
谢谢回答！ |
|
c*******1
发帖数: 240 | 37 y = x
is called aliasing. x and y refer to the same list in memory.
you can type t
id(x)
id(y)
to see it.
x.remove
is a method of list, it modifies the list x points to
x = [1,2]
creates a new list and changes the address x is referring to
it is no different from
x = 'asdf'
or
x = 1.201
equal sign is an operation on the name x, not the object x points to.
just remember python has no variable. every so-called variable is a pointer |
|
r****y
发帖数: 26819 | 38 这个？
http://rss.acs.unt.edu/Rdoc/library/micEcon/R-ex/maxNR.R
### Name: maxNR
### Title: Newton-Raphson maximisation
### Aliases: maxNR
### Keywords: optimize
### ** Examples
## ML estimation of exponential duration model:
t <- rexp(100, 2)
loglik <- function(theta) sum(log(theta) - theta*t)
## Note the log-likelihood and gradient are summed over observations
gradlik <- function(theta) sum(1/theta - t)
hesslik <- function(theta) -100/theta^2
## Estimate with numeric gradient and Hessian
a <- maxN |
|
u*******r
发帖数: 2855 | 39 主要是怎么建模的问题
首先你要考虑如何定义时间和空间的类别，比如
1. 空间是fixed effect
则要考虑它是continuous variable or categorical variable
如果是categorical variable，你可以采用上面某位建议的，建立一组dummy variable,
A, B,..., X
y(space=A)=K1*time+A+ε1
y(space=B)=K1*time+B+ε2
...
y(space=X)=K1*time+X+ε3
三种方式处理dummy variable:
1)sum(A+B+...+X)=0;
2)A=0
3)不加限制，但是只有dummy variable之间的相对大小有意义
当然你也可以加入二次方，比如y=K1*time+K2*time^2
如果你把空间定义为continuous variable，那么
你可以像上面建议的那样只考虑线性:
y=K1*time+K2*space+ε
也可以考虑interaction
y=K1*time+K2*space+K12*time*space+ε
可以考虑更高次方
y=... 阅读全帖 |
|
T****i
发帖数: 15191 | 40 以下摘抄在水版有代表性的回帖 （详情请看水版原帖） ：
发信人: majia88888 (马佳佳), 信区: WaterWorld
标 题: Re: x5 先生你又开始滥用职权了。 (转载)
发信站: BBS 未名空间站 (Sun Oct 28 22:40:31 2012, 美东)
她的话题狠好阿。 只是不合你们几个bully的口味而已
————————————————————————————————————
发信人: egon (schiele), 信区: WaterWorld
标 题: Re: x5 先生你又开始滥用职权了。 (转载)
发信站: BBS 未名空间站 (Mon Oct 29 11:48:13 2012, 美东)
LZ被和谐而后又被封的是这篇帖子. 咱们凭良心说话, 这个帖子怎么就troll了?
"
我选择4／3和m4/3的心路历程
Tianzi (tt) 于 (Sun Oct 28 10:19:53 2012, 美东) 提到:
以前对相机什么的不感兴趣，基本就用傻瓜，拍拍风景，拍拍LD，拍拍虫子，觉得挺有
乐趣的。几年前的一天，我知道了MITBBS，知道了Ph... 阅读全帖 |
|
T****i
发帖数: 15191 | 41 以下摘抄在水版有代表性的回帖 （详情请看水版原帖） ：
发信人: majia88888 (马佳佳), 信区: WaterWorld
标 题: Re: x5 先生你又开始滥用职权了。 (转载)
发信站: BBS 未名空间站 (Sun Oct 28 22:40:31 2012, 美东)
她的话题狠好阿。 只是不合你们几个bully的口味而已
————————————————————————————————————
发信人: egon (schiele), 信区: WaterWorld
标 题: Re: x5 先生你又开始滥用职权了。 (转载)
发信站: BBS 未名空间站 (Mon Oct 29 11:48:13 2012, 美东)
LZ被和谐而后又被封的是这篇帖子. 咱们凭良心说话, 这个帖子怎么就troll了?
"
我选择4／3和m4/3的心路历程
Tianzi (tt) 于 (Sun Oct 28 10:19:53 2012, 美东) 提到:
以前对相机什么的不感兴趣，基本就用傻瓜，拍拍风景，拍拍LD，拍拍虫子，觉得挺有
乐趣的。几年前的一天，我知道了MITBBS，知道了Ph... 阅读全帖 |
|
e**n
发帖数: 5876 | 42 x5一贯是滥用职权双重标准封人的, 只不过没封到你或者你的熟人的时候你没注意罢了.
几天前封Tianzi的"我选择4／3和m4/3的心路历程". 虽然Tianzi平时说话比较夸张,
但是管理执法要对事不对人. 是不是?
Tianzi这篇"我选择4／3和m4/3的心路历程"凭什么就被删贴封ID了? 哪儿违规了? x5至
今拒绝给解释.
下面是被封的原帖. 咱们凭良心说话, 这个帖子怎么就违规了?
"
我选择4／3和m4/3的心路历程
Tianzi (tt) 于 (Sun Oct 28 10:19:53 2012, 美东) 提到:
以前对相机什么的不感兴趣，基本就用傻瓜，拍拍风景，拍拍LD，拍拍虫子，觉得挺有
乐趣的。几年前的一天，我知道了MITBBS，知道了Photogear，看到这里的大师们华山
论剑，我看了之后就产生了羡慕嫉妒恨，于是就开始置办单反。
当时对单反一无所知，于是从B&H的产品目录开始学习。LD有一个尼康的N80，所以我开
始是倾向于入尼的。不过看了B&H的产品目录，决定还是入个小众化的便宜的。于是就
入了4/3的套机E-510。那时候纳伊夫啊，没考虑到4/3的高档... 阅读全帖 |
|
C******g
发帖数: 2930 | 43 We are excited to announce the God of War Collection for this Holiday Season
. Now fans and newcomers to the series can experience the epic journey of
God of War and God of War II in 720p HD form. Both titles have been
remastered with anti-aliased graphics, running at 60 frames per second for a
smooth gameplay experience, only on the PS3 system. Additionally, the Blu-
ray Disc compilation will bring the PlayStation Network (PSN) trophy support
to the franchise for the very first time.
This is yo |
|
C******g
发帖数: 2930 | 44 God of War Collection was announced earlier this week for the Playstation 3.
The remastered port will contain the original God of War and God of War II
with anti-aliasing, high-definition resolutions, 60 frames per second and
added trophy support.
Apparently, the announcement raised some eyebrows over at Capcom as the
publisher is gauging interests from fans on possible remastered Playstation
2 ports of Capcom games for the Playstation 3. Christian Svensson, the
Corporate Officer and Vice-presid |
|
w*******y
发帖数: 60932 | 45 is there a way to filter out the multi poster (selma and aliases) I am
getting tired of all the non deals they are posting... it is really getting
old, and I am sure it is losing traffic for this site...
|
|
|
