Tip: Auto-Login Your Windows 7 User Account

An easier way - although again much less secure - is to enable auto-logins for your Windows PC. This is possible in Windows 7, as it was in prior versions, but it takes a little finagling to do so. (And for good reason, darn it.)
Still, if you must do this, here’s how:

1. Press the Windows key + R on your keyboard to launch the “Run” dialog box.
2. Type in control userpasswords2
3. Press Enter. The User Accounts window will display.
4. Uncheck the option “Users must enter a user name and password to use this computer”
5. Click “OK”
6. You will then be prompted to enter the current password and confirm it.
7. After doing so, you will no longer be prompted to enter your password upon login.

What't the difference between "Please be advised" & "please advise"

1. Please be advised that we have received your letter. This means 'We want to tell you that we have received it'. It's a standard business expression. Personally, I dislike this as it seems to serve no purpose usually. I'd omit it, and just say 'We have received your letter.
 
2. Please advise that we have received your letter. This is wrong. It makes no sense.

Optimizing Time-Based Calculations in SSAS

SQL Server Analysis Services (SSAS) provides built-in MDX functions to facilitate time-based calculations. For example, these functions are often used for period-to-period and same-period-last-year comparisons. (If you're unfamiliar with these MDX functions and calculations, see the web-exclusive article "The Concepts Behind Time Calculations in SSAS". Although approaches to leverage the MDX functions exist, they have shortcomings. I developed an approach that overcomes the shortcomings and strives to provide end users with calculations that are flexible, simple to use, and offer good performance.
I'll begin by describing commonly used approaches to time-based calculations (which involve a time utility dimension) and the shortcomings of those approaches. I'll then walk you through the approach I developed. Although I use SQL Server 2008 in the examples, my approach also works with SQL Server 2005.
Commonly Used Approaches That Involve Time Utility Dimensions
As I discuss in my web-exclusive article "The Concepts Behind Time Calculations in SSAS," adding distinct time-based calculations for each measure (and applicable time-based dimension attribute/hierarchy) can result in bloated, unwieldy cubes. A time utility dimension can be used instead.
In 2002, I came across the concept of a time analysis utility dimension (now typically referred to as time utility dimension) while reading the first edition of George Spofford's MDX Solutions (John Wiley & Sons, 2001). In Spofford's approach, you manually add a new single-member dimension (with a static value such as [Current]), then add a calculated member to the dimension for each type of desired analysis (e.g., period-to-period, same-period-last-year, year-to-date). The result is a flexible cube that has a small set of calculated member definitions.
For example, look at the pivot table in Figure 1.

Figure 1: Pivot table from a cube in which a time utility dimension was manually added (click to enlarge)

This pivot table is tracking two economic indicators: unemployment and consumer price index (CPI). A Time dimension hierarchy is displayed along the rows of the pivot table, and a Time utility dimension is displayed along the columns. The two columns underneath the Current member display the actual measure values for a given date, while the PriorPeriod and YearAgo columns represent calculated members.
The clever part of this approach is that the Current, PriorPeriod, and YearAgo members are generic (i.e., not explicitly defined against any particular measure) and defined only once. Measure values can be switched in and out of this pivot table. When you do so, the Current, PriorPeriod, and YearAgo values will update accordingly.
Let's walk through an implementation of a time utility dimension built directly into SSAS: the Business Intelligence Wizard's "time intelligence" option. (The Business Intelligence Wizard was first introduced in SSAS 2005.)
If you'd like to follow along, you can download the files in Chessman_TimeAnalysis_CodeSamples.zip by clicking the 103355.zip hotlink near the top of the page. The underlying data, which comes from the Organisation for Economic Co-operation and Development, compares economic trends across different countries.
Start by restoring the SQL Server 2008 relational database from the DebtAnalysis.bak backup file. Next, open the OLAP_GovtData.dwproj project in Business Intelligence Development Studio (BIDS). Edit the project's target server property to match the location of your SSAS instance. Finally, after editing the DebtAnalysis.ds data source to reflect the location of the DebtAnalysis relational database, process the entire database using the Database menu's Process option.
With the OLAP_GovtData project open in BIDS, make a copy of the GovtDebtAnalysis cube. (Alternatively, you can see the output in the GovtDebtAnalysis_BIW.cube file in Chessman_TimeAnalysis_CodeSamples.zip.) Open the copied GovtDebtAnalysis cube in the BIDS designer. On the Calculations tab, select the Add Business Intelligence option from the Cube menu to start the Business Intelligence Wizard. After navigating past the Welcome window, select Define time intelligence in the Choose Enhancement window, as Figure 2 shows.

Figure 2: Selecting the Define time intelligence option (click to enlarge)

Next, choose the [H-Year] hierarchy and select the following check boxes in the Available time calculations section:

• Year Over Year Growth
• Quarter Over Quarter Growth
• Month Over Month Growth

Finally, in the Define Scope of Calculations window, select the applicable measures to be included in the calculations. In my example, I selected all the measures in the wizard. (Note that in the cube's calculations script, I manually removed the measures for the Fact OECDGDP measure group from the quarterly and monthly growth calculations because these measures are defined at an annual grain—i.e., detail at the annual level.) Click Next to review the changes the wizard will implement, then click Finish to implement them. The major changes implemented by the Business Intelligence Wizard include the following:

• A new named calculation (H-Year DimDate Calculations 1) is added to the dsv_DebtAnalysis data source view with a static value of [Current DimDate]. Likewise, an attribute (also named H-Year DimDate Calculations) is added to the DimDate dimension.
• A calculated member is created for each time calculation you selected in the wizard. Each member is initially defined with a value of NA.
• Each calculated member is associated to specific measures and cube cells (specifically, along the attributes in the DimDate dimension) through assignments.

Figure 3 shows a pivot table from the resulting cube. The time utility dimension and unemployment rates are displayed in the columns. The H-Year hierarchy is displayed in the rows.

Figure 3: Pivot table from a cube in which a time utility dimension was added using BIDS (click to enlarge)

The Shortcomings
In versions earlier than SSAS 2005, time dimensions were relatively easy to work with. SSAS 2005 and its introduction of the Unified Dimensional Model (UDM) changed everything. Today, dimension attributes, multiple dimension hierarchies, and role-playing dimensions make it difficult to implement flexible and comprehensive time-based calculations.
Most of the time-based calculation examples I've encountered ignore these issues. Examples in which time utility dimensions are manually added are typically limited to a single dimension hierarchy. The built-in Business Intelligence Wizard doesn't effectively address these challenges either because it's limited to building calculations for either a single-attribute or single-dimension hierarchy. Furthermore, the Business Intelligence Wizard has relatively weak error-handling capabilities to account for missing data along a time dimension.
A New Approach
My approach to time-based calculations involves manually creating a time utility dimension. It's based on concepts from David Shroyer's white paper "A Different Approach to Implementing Time Calculations in SSAS", ideas from Mosha Pasumansky's blog "Time Calculations in UDM: Parallel Period", and the results of my own research. At a high level, I used Shroyer's framework for creating a time utility dimension. I integrated performance optimizations from Pasumansky's blog and used an explicit attribute-based method of scoping the calculations across the time dimension (and any role playing dimensions based on the time dimension).
To walk through this new approach with me, make a copy of the GovtDebtAnalysis cube in the OLAP_GovtData project. (Alternatively, the GovtDebtAnalysis_TUD.cube file in Chessman_TimeAnalysis_CodeSamples.zip contains a completed implementation.) Then follow these steps:

1. Add a new dimension attribute to the existing time dimension.
2. Create a new dimension.
3. Add the new dimension to the cube.
4. Add the time calculations to the cube.
5. Test the cube.

Step 1. Add a New Dimension Attribute
The first task in adding a new dimension attribute to the existing time dimension is to add a new named calculation to the DimDate table in the dsv_DebtAnalysis data source view. Name it TimeCalcs and assign it the static expression value of N'Current'. Next, add TimeCalcs as an attribute to the DimDate dimension, setting the AttributeHierarchyVisible property to False. This attribute isn't used in the DimDate dimension, but it's needed for configuring a referenced dimension relationship in Step 3.
Step 2. Create a New Dimension
Now you need to create a new shared dimension. Right-click Dimensions in Solution Explorer, and select the New Dimension menu item to start the Dimension Wizard. As you work through the Dimension Wizard, choose the Use an existing table option, select DimDate as the main table, and select the named calculation (i.e., TimeCalcs) as both the Key column and Name column. Name your dimension DimTimeCalcs.
The DimTimeCalcs dimension will be used for calculations, not aggregations. Therefore, we don't want an All level in the dimension. With your DimTimeCalcs dimension open in the editor, select the TimeCalcs attribute. Change the IsAggregatable property to False and change the DefaultMember to [Current], as Figure 4 shows. Save the changes and close the dimension designer.

Figure 4: Changing the IsAggregatable and DefaultMember properties (click to enlarge)

Step 3. Add the New Dimension to the Cube
The next step is to add the new DimTimeCalcs dimension to the cube. In Solution Explorer, double-click the cube to open it. Right-click anywhere on the Dimension Usage tab and select Add Cube Dimension. For each measure group, set the relationship type to Referenced, as seen in Figure 5.

Figure 5: Configuring the relationship type of each measure group (click to enlarge)

After you're done configuring the measure groups' relationship type, highlight the new dimension in the Dimensions Usage tab and set the HierarchyUniqueNameStyle property to ExcludeDimensionName. Because this dimension doesn't contain user-defined hierarchies, you can simplify any MDX reference to include just the attribute name.
Step 4. Add the Time Calculations to the Cube
You're now ready to add the time calculations to the cube. Select the Calculations tab and switch to the Script view. First, you need to add and define the new calculations. For example, the following code adds the PriorPeriod and YearAgo calculations to the [TimeCalcs] dimension attribute of the DimTimeCalcs dimension, and defines their default value as Null:

CREATE MEMBER CURRENTCUBE.
  [TimeCalcs].[PriorPeriod] AS Null;
CREATE MEMBER CURRENTCUBE.
  [TimeCalcs].[YearAgo] AS Null;

(Although these lines wrap here, you'd enter each CREATE MEMBER statement on one line.)
Next, you need to limit, or scope, the calculations to each desired dimension attribute in the DimDate dimension, as Listing 1 shows.

Listing 1: Code That Scopes the PriorPeriod Calculation

Notice that this code defines only a PriorPeriod calculation and not a YearAgo calculation for the Century dimension attribute. Including a YearAgo calculation doesn't make sense given that this dimension attribute's context is century. However, for many of the other attributes, it does make sense, so you need to define both the PriorPeriod and YearAgo calculations, as shown in Listing 2.

Listing 2: Code That Scopes the PriorPeriod and YearAgo Calculations

To scope the YearAgo calculation for the Calendar Qtr dimension attribute, you can simply use MDX's Lag function with an index of 4, as callout A in Listing 2 shows. I prefer using the Lag function rather than MDX's ParallelPeriod function because ParallelPeriod requires a hierarchy level as an input.
In case you're unfamiliar with the Lag function, it returns the member that's the specified number of positions before the specified member. In this case, the specified member is the current quarter ([DimDate].[Calendar Qtr].CurrentMember) and the index is 4, so the lag function returns the quarter that was four quarters ago (i.e., a year ago). Alternatively, you could use the Lag function with the Month dimension attribute in code such as

[DimDate].[Month].CurrentMember.Lag(12)

to return a year-ago value. However, you shouldn't use the Lag function with the Date dimension attribute to obtain a year-ago value. Code such as

[DimDate].3/17/2010 11:42:31 AM.CurrentMember.Lag(365)

typically doesn't work due to leap years. In this instance, you'd need to use the ParallelPeriod function. None of the measure groups in my cube have a daily grain, so I didn't use the ParallelPeriod function.
You can find all the code needed for step 4 in the MDX script in the GovtDebtAnalysis.cube file in Chessman_TimeAnalysis_CodeSamples.zip. After you've finished scoping the time calculations for each desired attribute, you need to process the cube.
Note: There's a shortcut you can take to scope all the attributes participating in a dimension hierarchy. I included an example in the comments of the MDX script in the GovtDebtAnalysis_TUD cube, and you can see the same approach in the MDX code generated by the Business Intelligence Wizard. I decided to define calculations one attribute at a time because, for large cubes, it can be difficult to keep track of which attributes do and do not participate in a hierarchy.
Also, for some types of time-based calculations (e.g., PeriodsToDate), you need to explicitly define and scope multiple calculations for an attribute participating in multiple hierarchies. For example, a PriorPeriod or YearAgo calculation should be the same regardless of hierarchy, but a year-to-date calculation will be different when querying a Calendar versus Fiscal hierarchy.
Step 5. Test the Cube
Before you let end users query the cube, you should test the cube by using the browser in BIDS to simulate an end user or by creating a connection to the cube from an application such as Microsoft Excel. If you use BIDS, you might encounter an "unknown error" on certain navigation paths because of a bug in the BIDS browser.
These errors don't show up in Excel. If you're using Excel, make sure you go into the PivotTable options and select Show calculated members from OLAP server on the Display tab.
Figure 6 contains a sample pivot table that shows historical unemployment rates in the United States.

Figure 6: Pivot table from a cube in which a time utility dimension was added using the new approach

The TimeCalcs dimension is displayed in the columns. The PriorPeriod and YearAgo calculations are properly scoped at yearly, quarterly, and monthly levels. If you have Excel 2007, you can open the pivot table spreadsheet (Excel_PivotTables.xlsx) and explore the various measures, date attributes, and date hierarchies.
Note that the GovtDebtAnalysis cube you just created doesn't contain role-playing dimensions. If you have time-based role-playing dimensions (e.g., OrderDate, ShipDate) in a cube, you can copy and paste all the scoped calculation definitions, then replace the dimension name with the role-playing dimension name. Because my approach uses a referenced time utility dimension rather than using the [TimeCalcs] attribute directly from the date dimension, the calculations will work across all the role-playing dimensions.
Benefits of the New Approach
So what can your end users gain from all this effort? Consider the original goal of providing end users with flexibility, simplicity, and good performance:

• Flexibility. Regardless of the measures, date attributes, or date hierarchies being queried, the [Current], [PriorPeriod], and [YearAgo] calculations are automatically computed and appropriately displayed. Other approaches to time utility dimensions don't do this.
• Simplicity. Only two calculated members were added to provide time-based analysis across the entire cube, avoiding the problem of calculated measure explosion inherit in traditional approaches. (For an example of calculated measure explosion, see "The Concepts Behind Time Calculations in SSAS".)
• Good performance. The MDX script contains no runtime-dynamic statements (e.g., CASE, IIF). Instead, the script uses static scoping to avoid potential performance issues. It also efficiently eliminates erroneous calculations at the start and tail end of the date dimension. For example, in Figure 6, notice that years without unemployment rates (1790-1989) don't appear in the pivot table and the PriorPeriod and YearAgo results for the first year containing data (1990) correctly show a null value.

Give It a Try
I encourage you to try this new time-calculation approach in your existing cubes and any new cubes you need to create. Your end users will be more productive and independent. Although the implementation involves a bit of up-front effort, it should reduce the amount of overall time spent on custom calculations.

Selecting Speaker Cable for Whole House Audio

There are many types of speaker cable on the market. Most are not suitable for whole house audio applications. We sell speaker cable designed specifically for in-wall wiring.

• Rated Cables
• Gauge Selection
• Power Loss Budget Table
• Speaker Cable Comparisons
• Purchasing Information

Rated Cables When speaker cable is installed inside the walls, it is important to use cable that is rated for in-wall use. This is often referred to as premise wiring rated. Most speaker cables sold for consumer use are not rated for premise wiring. There are two strong reasons for insisting on premise rated wiring. First, it complies with the National Electric Code. Second, it assures you the materials and the design of the cable meet well established standards for fire safety. Gauge Selection Because the impedance of loudspeakers is quite low, (usually 4 or 8 ohms) the resistance of the cable feeding the speakers becomes quite relevant in determining how much power actually reaches the speaker. As an example, a hundred foot run of 16 AWG cable will have a round trip resistance of 0.8 ohms. If this is used to feed a 4 ohm speaker, approximately 17 percent of the power will be lost to the cable, and 83 percent will reach the speaker. Larger wire will reduce the power loss. From a practical standpoint, the wire sizes normally used for whole house audio speaker wiring are either 16 AWG or 14 AWG. These sizes provide a good compromise between line loss, cost, and ease of installation. Most of the equipment available for whole house audio systems have connection devices designed for these wire sizes. Going to a non-standard wire size can immensely increase the difficulty of installation. Power Loss Budget Table The following speaker wire power loss budget table provides a guideline for selecting the right size speaker wire. Speaker Wire Power Loss Budget Table Select Speaker Impedance and Maximum Acceptable Loss, then look up maximum cable run distance. Speaker Ohms db Loss Power Loss 16 AWG Run in feet 14 AWG Run in feet 4 0.5 11% 60 100 4 1 21% 130 210 4 2 37% 290 460 4 3 50% 500 790 8 0.5 11% 120 190 8 1 21% 260 410 8 2 37% 580 930 8 3 50% 990 1580 A 3db loss may seem like a lot, but since the ear has tremendous dynamic range, you probably won't notice that size loss and with the power of the amplifiers typically used, it is easy to overcome a 3db loss. There is another effect that you should consider. Some high end amplifiers have special speaker compensation circuits. Over the full range of frequencies handled by a speaker, it will perform better at some frequencies than others. Some high end amplifiers sense the response characteristics of the speaker and automatically compensate for non-linear responses. This is sometimes called servoing. When the resistance of the speaker wire is a significant part of the speaker impedance seen by the amplifier, it can impact the effectiveness of an amplifiers compensation capabilities. For most amplifiers intended for whole house use, this is of little relevance.

This is the old HomeTech Solutions website.
Please visit our new website here

Speaker Cable Comparisons

Two speaker cables we recommend are the Monster Cable in-wall cable and new speaker cable from Belden in their New Generation® series.

Speaker Cable Comparisons
Manufacturer	Type	Characteristics
Monster Cable	S162 (16AWG, 2 Cond.)	Rated UL CL3 for premise wiring. Very fine strands (14AWG consists of 100 strands). Quite flexible. White outer jacket. East to strip and install. Monster Cable is a highly respected manufacturer of premium audio cables.
	S164 (16AWG, 4 Cond.)
	S142 (14AWG, 2 Cond.)
	S144 (14AWG, 4Cond.)
Belden	5102UE (14AWG, 4Cond)	Rated UL Type CL3R for premise and riser wiring. Fine strands (14 AWG consists of 27 strands.) Easier to connect to terminals than Monster Cable because of larger strands. Grey outer jacket. This New Generation® Speaker Cable by Belden is new on the market. Belden is a highly respected manufacturer of general use and speciality wires.

Getting Wired: How to Choose the Right Speaker Wire for Your Home Audio System

With so many different brands and grades of speaker wire on the market, it can be tough to narrow down your choices. Believe it or not, your final decision doesn’t need to be based on a famous name or hefty price tag – on the contrary, it’s more important to look at specifications and characteristics. Not sure what to look for? Just read on: CableOrganizer.com shares tips on how to find the perfect speaker wire for your audio system.

How much distance does the speaker wire need to cover?
The distance of your speaker-wire run directly corresponds to the gauge (or thickness) of wire you’ll need to use. All speaker wire is based on the AmericanWire Gauge (AWG), a wire-sizing system that assigns numbers (which range from 0 to 40) to standardized wire thicknesses. In the AWG system, the higher the number is, the thinner the wire. Speaker wire typically falls within the 12 to 18-gauge range.
The farther an audio signal travels down a wire, the more resistance – and power loss – it encounters. Because thicker, lower-gauge wire offers the least resistance, it has the best capacity for carrying audio signals long distances – generally speaking, a 12 or 14-gauge speaker cable should give you great results. But don’t go on assumption alone – the best way to find the right wire gauge for your speakers is to know the exact length of the cable runs, then compare them to the chart below.

Distance Between Amplifier/Receiver and Speaker	Wire Gauge Needed (AWG)
Up to 79 feet (24 meters or less)	16
80 - 200 feet (24 - 61 meters)	14
Over 200 feet (longer than 61 meters)	12

Will the wire need to go through a wall?
Because fire safety considerations come into play, choosing the right speaker wire for in-wall installation takes a little more consideration than is necessary when you’re picking out standard speaker cable. Start out by consulting local fire and building codes for exact specifications. If you’re going to be running speaker wire in-wall, you’ll need to select a type that is UL-rated and labeled either CL2 or CL3 (Class 2 or Class 3, respectively). These designations ensure that the wire has been tested extensively for current-based heat generation, flammability, and susceptibility to damage, and that Underwriters Laboratories has given it their stamp of approval for safe consumer use and in-wall installation. We recommend Monster XP® CI Compact Speaker Cable.
On the other hand, if your audio setup doesn’t require in-wall wire runs, any type of speaker wire will work, provided that you’re satisfied with its quality. One great all-around speaker wire to try is Monster XP® Speaker Wire by Monster Cable®.

What level of sound quality are you aiming for?

If you’re choosing speaker cable based on the sound quality you want to achieve, you’ll need to take both conductor characteristics and shielding into consideration. When it comes to conductor material, look for speaker wire that’s based on high-purity copper, which is recognized as one of Earth’s finest conductive metals.
Next, think about gauge. If you have a high end, audiophile-quality system that’s geared toward finely nuanced sound reproduction, opt for a thicker gauge speaker wire, even if cable runs between the amp/receiver and speakers are on the short side. A thicker conductor (12 or 14-gauge) provides a wider, clearer path for audio signals to travel along, and allows every little detail to come through.
Shielding – a tightly woven metal braid that surrounds a cable’s conductor – can also improve overall sound quality by blocking interference from nearby power cords and fluorescent lighting. Shielded speaker cable is an especially smart choice for in-wall installations, which often run in close proximity to electrical wiring.
For a smaller scale, everyday audio system, high-grade speaker cable isn’t a necessity. In cases like this, basic 16-gauge copper speaker wire does the job just fine, and will help to keep your budget under control.

最佳科技公司雇主 Facebook力压谷歌苹果

BusinessInsider评选出25家最适宜工作的科技公司，Facebook、谷歌、Adobe、苹果名列前茅，微软、英特尔、诺基亚、思科等科技巨头尽管也榜上有名，却跌落十名开外。

Facebook：逼退谷歌

公司评级：4.4 排名：No.1
CEO认同率：马克·扎克伯格(Mark Zuckerberg)，92%

Facebook夺取了第一位的宝座，成为科技界最佳雇主。员工表示，该公司拥有美味的食品、轻松的工作环境、视频游戏、充满活力的工作氛围，以及科技员工都渴望的没有办公室政治。

谷歌：不只是华而不实的五星级酒店

公司评级：3.9排名：No.2
CEO认同率：埃里克·施密特(Eric Schmidt)，97%

人人都知道谷歌提供的美食，但那不是领先科技公司的唯一原因。谷歌雇佣天才员工，信任他们去开发产品、推陈出新。但是关于谷歌的消极评价都指向了中层管理者，他们在早期发展阶段被雇佣，当时主要是处于便捷而少考虑资质，今天这些人已经成为职业发展的障碍。

Adobe：死盯华尔街

公司评级：3.9排名：No.3
CEO认同率：山塔努·纳拉延(Shantanu Narayen)，78%

评估结果显示，Adobe产品和合作氛围是受爱戴的原因。然而该公司裁员不断，高管则更加关注Adobe在华尔街的表现，而非员工士气。

苹果：“薪”情不错，但太强调顾客

公司评级：3.8 排名：No.5
CEO认同率：史蒂夫·乔布斯(Steve Jobs)，98%

有的顾客认为，“顾客永远是对的”这一古老谚语在苹果这里发挥到极致。他们认为消费者调查太经常影响雇佣和解雇决策了。一旦员工熟谙了应对顾客之道，就会被付以高薪，特殊折扣及赠品，还有偶尔的优惠——但是“偶尔”的间隔时间太长了。另请注意乔布斯的赞同率。

高通：员工过劳

公司评级：3.7 排名：No.8
CEO认同率：保罗·雅各布(Paul Jacobs,)，90%

美丽的圣地亚哥风光、健身房、田径场……这些都无法阻止高通员工对沉重的工作压力的关切。

微软：工资高，院子大!  

公司评级：3.6 排名：No.12
CEO认同率：史蒂夫·鲍尔默(Steve Ballmer)，52%

微软的好处和缺点都因为太大了。尽管薪水好、上升空间广阔，但是这型号看起来太笨拙了，谁想在这里“脱颖而出”?难!

英特尔：人聪明，产品棒，股价……低

公司评级：3.5 排名：No.18
CEO认同率：保罗·欧德宁(Paul Otellini)，82%

即使是最差评语也感谢英特尔的大名和产品，以及高智商人品好的同事，可是那些握有公司期权的员工着实为公司股票在华尔街的糟糕表现感到沮丧。

诺基亚：需要聚焦(众多问题之一)  

公司评级：3.6 排名：No.21
CEO认同率：康培凯(Olli-Pekka Kallasvuo)，65%

目前的共识是，为了摆脱对手竞争，保留全球市场上的宝座，诺基亚不断进行战略转移。为了高报酬、好福利、饭碗稳，员工也都在这样变换的战略中漂移。

思科：成也大公司，败也大公司

公司评级：3.4 排名：No.25
CEO认同率：约翰·钱伯斯(John Chambers)，80%

思科的规模对员工来说有着巨大影响。积极评价都认为，这有助于员工在不同部门中流动，消极评价则认为这造成了一种反应迟钝的环境。

"System won't boot" and "no video output" checklist

This checklist is a compilation of troubleshooting ideas from many forum members. It's very important to actually perform every step in the checklist if you want to effectively troubleshoot your problem.

1. Did you carefully read the motherboard owners manual?
 2. Did you plug in the 4/8-pin CPU power connector located near the CPU socket? If the motherboard has 8 pins and your PSU only has 4 pins, you can use the 4-pin connector. The 4-pin connector USUALLY goes on the 4 pins located closest to the CPU. If the motherboard has an 8-pin connector with a cover over 4 pins, you can remove the cover and use an 8-pin plug if your power supply has one. This power connector provides power to the CPU. Your system has no chance of posting without this connector plugged in! Check your motherboard owners manual for more information about the CPU power connector. The CPU power connector is usually referred to as the "12v ATX" connector in the owners manual. This is easily the most common new-builder mistake.
 


 3. Did you install the standoffs under the motherboard? Did you place them so they all align with the screw holes in the motherboard, with no extra standoffs touching the board in the wrong place? A standoff installed in the wrong place can cause a short and prevent the system from booting.
 

 4. Did you verify that the video card is fully seated? (may require more force than a new builder expects.)
 5. Did you attach all the required power connector(s) to the video card? (some need two, some need none, many need one.)
 

 6. Have you tried booting with just one stick of RAM installed? (Try each stick of RAM individually in each RAM slot.) If you can get the system to boot with a single stick of RAM, you should manually set the RAM speed, timings, and voltage to the manufacturers specs in the BIOS before attempting to boot with all sticks of RAM installed. Nearly all motherboards default to the standard RAM voltage (1.8v for DDR2 & 1.5v for DDR3). If your RAM is rated to run at a voltage other than the standard voltage, the motherboard will underclock the RAM for compatibility reasons. If you want the system to be stable and to run the RAM at its rated specs, you should manually set those values in the BIOS. Many boards don't supply the RAM with enough voltage when using "auto" settings causing stability issues.
 7. Did you verify that all memory modules are fully inserted? (may require more force than a new builder expects.) It's a good idea to install the RAM on the motherboard before it's in the case.
 8. Did you verify in the owners manual that you're using the correct RAM slots? Many i7 motherboards require RAM to be installed in the slots starting with the one further away from the CPU which is the opposite of many dual channel motherboards.
 9. Did you remove the plastic guard over the CPU socket? (this actually comes up occasionally.)
 10. Did you install the CPU correctly? There will be an arrow on the CPU that needs to line up with an arrow on the motherboard CPU socket. Be sure to pay special attention to that section of the manual!
 http://www.intel.com/cd/channel/re [...] 299985.htm
 11. If using an after market CPU cooler, did you get any thermal paste on the motherboard, CPU socket, or CPU pins? Did you use the smallest amount you could? Here's a few links that may help:
 Benchmark Reviews
 Arctic Sliver
 http://www.youtube.com/watch?v=ffK [...] re=related
 12. Is the CPU fan plugged in? Some motherboards will not boot without detecting that the CPU fan is plugged in to prevent burning up the CPU.
 13. If using a stock cooler, was the thermal material on the base of the cooler free of foreign material, and did you remove any protective covering? If the stock cooler has push-pins, did you ensure that all four pins snapped securely into place? (The easiest way to install the push-pins is outside the case sitting on a non-conductive surface like the motherboard box. Read the instructions! The push-pins have to be turned the OPPOSITE direction as the arrows for installation.) See the link in step 10.
 14. Are any loose screws laying on the motherboard, or jammed against it? Are there any wires run directly under the motherboard? You should not run wires under the motherboard since the soldered wires on the underside of the motherboard can cut into the insulation on the wires and cause a short. Some cases have space to run wires on the back side of the motherboard tray.
 15. Did you ensure you discharged all static electricity before touching any of your components? Computer components are very sensitive to static electricity. It takes much less voltage than you can see or feel to damage components. You should implement some best practices to reduce the probability of damaging components. These practices should include either wearing an anti-static wrist strap or always touching a metal part of the case with the power supply installed and plugged in, but NOT turned on. You should avoid building or working on a computer on carpet. Working on a smooth surface is the best if at all possible. You should also keep fluffy the cat, children, and fido away from computer components.
 16. Did you install the system speaker (if provided) so you can check beep-codes in the manual? A system speaker is NOT the same as normal speakers that plug into the back of the motherboard. A system speaker plugs into a header on the motherboard that's usually located near the front panel connectors. The system speaker is a critical component when trying to troubleshoot system problems. You are flying blind without a system speaker. If you case or motherboard didn't come with a system speaker you can buy one for cheap here: http://www.cwc-group.com/casp.html
 
 17. Did you read the instructions in the manual on how to properly connect the front panel plugs? (Power switch, power led, reset switch, HD activity led) Polarity does not matter with the power and reset switches. If power or drive activity LED's do not come on, reverse the connections. For troubleshooting purposes, disconnect the reset switch. If it's shorted, the machine either will not POST at all, or it will endlessly reboot.
 

 18. Did you turn on the power supply switch located on the back of the PSU? Is the power plug on a switch? If it is, is the switch turned on? Is there a GFI circuit on the plug-in? If there is, make sure it isn't tripped. You should also make sure the power cord isn't causing the problem. Try swapping it for a known good cord if you have one available.
 19. Is your CPU supported by the BIOS revision installed on your motherboard? Most motherboards will post a CPU compatibility list on their website.
 20. Have you tried resetting the CMOS? The motherboard manual will have instructions for your particular board.
 http://www.spotht.com/2010/02/rese [...] -cmos.html
 21. If you have integrated video and a video card, try the integrated video port. Resetting the bios, can make it default back to the onboard video.

I also wanted to add some suggestions that user jsc often posts. This is a direct quote from him:
 "Pull everything except the CPU and HSF. Boot. You should hear a series of long single beeps indicating memory problems. Silence here indicates, in probable order, a bad PSU, motherboard, or CPU - or a bad installation where something is shorting and shutting down the PSU.
 To eliminate the possiblility of a bad installation where something is shorting and shutting down the PSU, you will need to pull the motherboard out of the case and reassemble the components on an insulated surface. This is called "breadboarding" - from the 1920's homebrew radio days. I always breadboard a new or recycled build. It lets me test components before I go through the trouble of installing them in a case.
 If you get the long beeps, add a stick of RAM. Boot. The beep pattern should change to one long and two or three short beeps. Silence indicates that the RAM is shorting out the PSU (very rare). Long single beeps indicates that the BIOS does not recognize the presence of the RAM.
 If you get the one long and two or three short beeps, test the rest of the RAM. If good, install the video card and any needed power cables and plug in the monitor. If the video card is good, the system should successfully POST (one short beep, usually) and you will see the boot screen and messages.
 Note - an inadequate PSU will cause a failure here or any step later.
Note - you do not need drives or a keyboard to successfully POST (generally a single short beep).
 If you successfully POST, start plugging in the rest of the components, one at a time."

If you suspect the PSU is causing your problems, below are some suggestions by jsc for troubleshooting the PSU. Proceed with caution. I will not be held responsible if you get shocked or fry components.
 "The best way to check the PSU is to swap it with a known good PSU of similar capacity. Brand new, out of the box, untested does not count as a known good PSU. PSU's, like all components, can be DOA.
 Next best thing is to get (or borrow) a digital multimeter and check the PSU.
 Yellow wires should be 12 volts. Red wires: +5 volts, orange wires: +3.3 volts, blue wire : -12 volts, violet wire: 5 volts always on. Tolerances are +/- 5% except for the -12 volts which is +/- 10%.
 The gray wire is really important. It should go from 0 to +5 volts when you turn the PSU on with the case switch. CPU needs this signal to boot.
 You can turn on the PSU by completely disconnecting the PSU and using a paperclip or jumper wire to short the green wire to one of the neighboring black wires.
 http://www.youtube.com/watch?v=5FW [...] tube_gdata
 This checks the PSU under no load conditions, so it is not completely reliable. But if it can not pass this, it is dead. Then repeat the checks with the PSU plugged into the computer to put a load on the PSU. You can carefully probe the pins from the back of the main power connector."

Here's a link to jsc's breadboarding thread:
 http://www.tomshardware.co.uk/foru [...] _13_0.html

If you make it through the entire checklist without success, Proximon has put together another great thread with a few more ideas here:
 http://www.tomshardware.com/forum/ [...] g#t1934282
 Here's a couple more sites that may help:
 http://www.gamespot.com/pages/foru [...] d=26986920
http://www.playtool.com/pages/psuc [...] ctors.html