n*n
发帖数: 202 | 1 ☆─────────────────────────────────────☆
NNN (两耳不闻窗外事 一心只读圣贤书) 于 (Wed Dec 13 20:47:38 2006) 提到:
点击 https://odr.bbb.org/odrweb/public/GetStarted.aspx
点 Next ==> 分别选择第一个点击next ==> 输入zip code点击next
再search URL中输入 www.compusa.com点击Next
选择第一个，达拉斯那个，这个是总部
出现
BBB of Metropolitan Dallas
(Dallas, TX)
1601 Elm, Suite 3838
Dallas, TX 75201
Phone: (214)220-2000
Fax: (214)740-0321
Email: i**[email protected]
Web: http://www.dallas.bbb.org
点击next
Step 1 Verify the nature of your Complaint
全部选No， Next
Step |
|
|
L****c
发帖数: 1466 | 3 好用就行，收费无所谓。
发现买家都不留feedback, 搞的现在ODR很高。 |
|
|
|
c****u
发帖数: 3277 | 6 yes, that's possible. but for MP, I think it is still in favor
of passing. Even when 3NT is makable, 2D x -2 is usually good match point.
Because when it's marginal, many players may not bid 3NT.
Anyway, structure in major suits usually means higher ODR, hehe.
Passing pd's negative double with 3-3-3-4 is Mashall Mile's idea.
I am not that extreme. I would pass it only with strong holding in D.
Perhaps I can send Marshall Mile an email message to ask him about this hand.
I guess he'd not doubl |
|
c****u
发帖数: 3277 | 7 I like none of them...
I think the best treatment for 3H is to show a constructive raise hand
with 4 trumps. something like this:
Sxxx HAxxxx DKxx Cxx or
Sxxx HKxxx DAxxxx Cx.
It shows about good 6 to 9 HCP, high ODR hands, about 8.5 - 8 losers.
I don't like preemptive raise because the most hands you gain from jumps
are those hands that your side have 17-23 HCP,
if your side has lower than 17 HCP, you usually can't shut out opps.
So it's rather bad that you disclose your information of trump le |
|
w****b
发帖数: 623 | 8 That looks like an interesting project. Count me in. |
|
w****b
发帖数: 623 | 9 Not sure if I posted this before or not. But we can take some ideas from here |
|
w****b
发帖数: 623 | 10 BTW with this, the hand A QJ AKQJTxxxx x evaluates to 23.95pt. Now you judge
if you wish to open 2C or 1D.
here |
|
f*****x
发帖数: 545 | 11 sounds interesting, how can i join? i really know nothing about programming.【
在 cozofu (但为君故) 的大作中提到: 】 |
|
c****u
发帖数: 3277 | 12 you are also welcome providing some comments and suggestions.
Actually, I think we need a lot of discussions before we reach the
programming stage. |
|
w****b
发帖数: 623 | 13 Sigh. I posted a long message this morning and then the server went down and I
lost it all.
Anyway to start with, I was proposing the ratio of winners when defense over
the losers at offense as a starting point. The initial consideration can be
restricts to only the two main suits (oppo's and our's).
Assume both 9 card fit without a loser, then the expected winner on defense is
about 1.4, and the expected loser at offense is definite. So in a 3-1, 2-2,
1-3 case, the initial ratio will be 1.4/1, |
|
w****b
发帖数: 623 | 14 Well Soloway was playing with Seymon Deutsch and I trust Soloway more :-)
This hand has such a high ODR value that I think double is out of the question
once pd made a voluntary 4S bid, which either showed 6S, or some offensive
values, and a non minimum hand, as we are not at a forcing pass situation yet
after a mere limited raise (or better).
Passing should be forcing here, so I think my choice would be pass, and
convert double to 5S, raise 5S to 6S. Pd needs only as little as AJTxxx xxx Jx
Ax |
|
w****b
发帖数: 623 | 15
It's double fold. First a doubleton Qx usually is worthless in offense, while
in defense, sometimes it offers a guesswork for oppo. For Kx, you have the
equal 50% chance of winning a trick; and for Ax, you always will win a trick.
However, a sure trick in oppo's trump may be more than that. It represents a
control and timing. For example, you might arrange a ruff with trump control.
So trading an honor from your main suit to their main suit tremendously
decrease ODR.
This is true in the formula |
|
c****u
发帖数: 3277 | 16 2S is perfectly fine with me, because the ODRs high and
the shape is reasonable.
I'd bid 3D with north's hand.
for 4S, west holds at least 5 spades, perhaps 6. if he holds 6 spades,
you'll have a hard time, because he may switch a trump and you'll
have little play unless he has 6-3-2-2 shape.
If west has 5 spades, you have better chance to make it.
A killing defence might be a diamond return after two clubs which might
have killed your entry. Then you have to play long spades with stiff diamond. |
|
a****s
发帖数: 524 | 17
Yeah, I agree with you, Holding a hand with too high ODR, It's better to
choose
preemptive open.
How high is too high? My rule is to open Namyats, one has 8-9 tricks and
roughly 80%-90% of the tricks come from trumps. With one side Ace,
this hand is qualified.
My rule of open 2C is 9+ tricks with 2+ defensive tricks OUTSIDE the
dominant suit. The original hand is obviously a 2C open.
With only one side Ace, It is a Namyats open (if you play that convention).
BTW, I hate to open 2C and try to fi |
|
w****b
发帖数: 623 | 18 strategy说到底是hand evaluation问题。
人们建议6-6-1-0抢打是因为这种牌一般ODR比较高。但这手牌，到现在为止已经显示非
常垃圾。你的两套都不好，没有控制，在同伴短红心时也未必能拿到将吃。敌人如果不
是在玩我们，那么绝对不会有9张黑心配合，所以我们现在叫的话，有可能：
1。错误的牺牲，被罚上西天。
2。把敌人拱入正确的S或D定约。
3。把敌人拱上满贯发现打不宕。
4。泄漏牌型，帮助敌人找到不可能的深飞。 |
|
w****b
发帖数: 623 | 19 In IMPs you obviously need to bid. In MP I'll risk a pass. Pass would allow
us to collect tops in many different scenarios but only bottom when it makes
.
In our hand, values are quite spread and the ODR value isn't that high
despite the shape.
Pd always can bid his 2nd suit if he's 5-4 or more shapely. So lack of it, I
'd assume he's at most 4-4-3-2. |
|
p***r
发帖数: 20570 | 20 I actually don't think it's merely a partnership thing. It is a matter of
evaluations and anticipations. For unbalanced hands, I would tend to bid 4S
with high ODR.
For example, with Axxx A xxx KQJxx, I tend to pass 4H, because this hand
contains at least about 2.5 (I offer some discount on the CKQJ because of C
length and opps' 4H overcall) defensive tricks, we are likely to beat 4H if
partner can also provide 1.5 defensive tricks; and we need partner to hold a
nice hand to make 4S.
For hands l... 阅读全帖 |
|
|
|
|
e****d
发帖数: 895 | 24 Declaration allows name lookup to find the entity.
Definition defines the entity.
A definition can also be a declaration.
You declare a static data member in a class and define it outside the class.
One definition rule doesn't allow you define a static data member
in a class becuase multiple includes will violate the ODR. |
|
g*********s
发帖数: 1782 | 25 sorry, i'm more confused.
1. the initializer should go with declaration or definition, or it
doesn't matter?
it's ok to have:
a.cpp: int x = 0; // definition.
b.cpp: extern int x; // declaration.
but is it ok to have the following?
a.cpp: int x; // definition.
b.cpp: extern int x = 0; // declaration.
2. why definition of a static variable in a class violates the ODR?
because we can define a static integral variable in a class, it seems
not a problem to allow any static variable to be defined in ... 阅读全帖 |
|
t****t
发帖数: 6806 | 26 didn't you read xentar's post? even you can write initializer to const
integral static data member, it doesn't mean you don't have to write
separate definitons.
however, in most cases, const static integral will be optimized out as
constant, and there will be no references to the symbol anymore in the link
stage. and in compilation stage, "no diagnostic required" (9.4.2 clause 5),
so in most cases you can get through without any errors. BUT, i can easily
construct a counter-example:
#include 阅读全帖 |
|
b*****e
发帖数: 474 | 27 You are right & your example is great. Thanks.
But I was referring to the fact that, given:
class A {
public:
const static int a = 1;
const static char c = 'A';
};
You definitely need to define const char A::c; in your code,
but you don't need to write const int A::a; (there is danger
of not writing this, as you have pointed out).
In other words, it seems that const static int typically will be optimized
out, but const static char does not get this treatment.
ODR notwithstanding, I see the... 阅读全帖 |
|
|
d****p
发帖数: 685 | 29 I see :-) You guys are amazing - keep reading standard! |
|
e****d
发帖数: 895 | 30 Class mytest has external linkage. Both compilation units
have the same definition of mytest, which doesn't violate
the ODR. |
|
g*********s
发帖数: 1782 | 31 so if you put the following in class_x.h that is included by a couple of
.cpp, the compiler possibly treats X::f() as non-inline? Isn't that an ODR
violation?
// class_x.h
class X {
public:
void f();
}
inline
void X::f(){} |
|
q****x
发帖数: 7404 | 32 【 以下文字转载自 JobHunting 讨论区 】
发信人: quantx (X矿工), 信区: JobHunting
标 题: Re: c++ inline问题
发信站: BBS 未名空间站 (Tue Dec 6 01:53:26 2011, 美东)
下面这个例子，即使没有inline关键字，X::f()也必须inline，否则ODR违例？
这里的inline语法上必须有，但编译器还是可以处理成普通函数，还是说必须inline？
// class_X.h
#include
using namespace std;
class X {
public:
void f();
};
void X::f()
{
cout << "X::f()" << endl;
}
//class_X1.cpp
#include "class_X.h"
void f()
{
X x;
x.f();
}
// class_X2.cpp
#include "class_X.h"
int main()
{
X x;
x.f();
f();
}
g++ cla... 阅读全帖 |
|
t****t
发帖数: 6806 | 33 从语法上来说, 不管是不是在头文件里, 成员总是可以写在class外面或里面. 但是如
果不是模板, 那就一定要标inline, 不管是显式的还是隐式的, 理由说过了.
如果是模板, 那class外的可以不标inline, 并且不违犯ODR.
从优化上来说, inline从来都没有强制作用, 从来都是一个提示. 但是不依赖这个关键
字和这个关键字有没有(优化上的)作用, 还是有区别的. 现代的编译器用的策略应该是
对可做可不做的依靠提示. |
|
t****t
发帖数: 6806 | 34 除了满足ODR的需要以外, 所有的地方都不写或者所有的都写, 都是可以的. 编译器的
开关一般有更高的优先级. 但是: (1)所有的地方都不写或所有的地方都不写并非最佳
策略. (2)不管程序员用什么策略, inline关键字(隐含或显式)仍然对编译器有提示的
作用. |
|
r***e
发帖数: 1840 | 35 Not necessary. With the #ifdef guard. The header file will be included only
once in one translation unit. So only one definition.
ODR. |
|
t****t
发帖数: 6806 | 36 你这都不知道我们在说什么, 还是先搞清楚ODR是控制什么东西再说吧.
only |
|
W********u
发帖数: 58 | 37 C++必备书籍推荐
[手册类 – 所有级别]
The C++ Programming Language, 4th Edition 作者：Bjarne Stroustrup(更新到C++
11)C++之父写的经典C++书籍。第四版涵盖了C++11的所有内容，从语言内核到标准库。
C++ Standard Library Tutorial and Reference, 2nd Edition 作者：Nicolai
Josuttis (更新到C++11) 这本书是C++标准库（STL）的引导和手册。 2012年4月发行
的第二版涵盖了C++11。
[实用]
Effective C++, 3rd Edition 作者：Scott Meyers 本书以瞄准成为C++程序员必读的
第二本书籍而写，内容覆盖了50多个很容易记住的条款，每个条款深入浅出（并且有趣
）讲到了你可能没有考虑过的C++规则。
Effective STL 作者：Scott Meyers 讲解方式和Effective C++类似，但内容主要面向
于STL。
Effective Modern C++ 作者：Scott Meyers... 阅读全帖 |
|
|
