Being the obvious fan boy of Google that I am, there is a wondering if I could ever have a need for a 10 million document database. Forget Live search of your own machine… I can’t find anything useful using live.com but Google Desktop helps me find my personal data with a ctrl ctrl easily.

I do shudder the day when I have enough space to hold 10 million documents.

I had this problem too, being told that something was wrong with SSL or TLS. My solution? Rapid click on the “Ok” and then back on your unactivated device in Itunes. Within 3 minutes my pre 2.0 iPhone was activated, with my previous “wait a few minutes” was fruitless after over 15 minutes of waiting.

I am assured that rapid clicking helped. The chances of getting that free server slot is just much higher if you click fast.

When your motherboard fails, who is faster at replacing/repairing it? Intel? MSI? The supposed crap master Foxconn?

When I contacted Intel about a failed board, it took over 37 hours for the leader to get back to me about replacing the board. When the board was sent away, the replacement took 4 days to return.

MSI was asked to replace a board, taking a full 24 hours to generate an RMA # for return. Good so far, right?
Then, when the board was sent in, it took 15 days to get an email stating that a new board was in bound.

Now, on to the company some love, swear by, and others say is the worst board/company out there.
I emailed Foxconn to get an RMA number, and in 15 minutes I had an email requesting more information. I replied, and in 10 minutes I had an RMA form to fax in to a Bob Wang. After faxing Bob Wang (hold the laugh please), I had an RMA number within hours. I sent the Winfast board in and in 3 days, a new board with no problems reported since.

Anyone care to add some advice on why Foxconn is so fast? I’ve determined that since they OEM so many boards for companies out there, it’s streamlined. I’ve heard rumbles that Dell/HP/Compaq and others all buy Foxconn boards. I won’t deny that almost all plastic connectors on motherboards come from Foxconn, but that doesn’t mean it’s a Foxconn board. I do have the knowledge that Foxconn was a Connector company first.

I don’t want to hear anything about performance of alternative motherboard manufactors. Some may be clunky, due to old Winchip, K5 procs and the like from the early 90’s. It’s obvious from benchmarks that cheap motherboards are just that, not clunkers.

Never before in the history of computing has computers been so fast, for so cheap.

Early last year, Intel announced that they were coming out with a new product in the start of 2007, Q1 that would be called Clovertown. Clovertown would change a lot of things, primarily, the number of processors that you could have on one chip, then double the amount that one board could handle. It’s more important than just that however, since Clovertown is based on older core technology, and as a Xeon style product it has a lot of distribution for datacenters: it’s cheaper.

Clovertown wasn’t the only chip on the block however, as many of you know, the QX6700 was released by Intel as a Core 2 Extreme, but is now coming out with a Core 2 Quad.

Running a QX6700 will no doubt provide better performance than a single Clovertown such as the 5310, but when you start to escalate up to two 5310’s, there should be a difference that makes more sense to go with the dual Clovertowns. The Core 2 Quad however will consume even less power than the QX6700, with the Clovertown getting the same treatment as well.
771 was to be the socket of choice for Intel to release the Clovertown, and the QX6700/QX6800 is a socket 775 chip, so the following features must be kept in mind when building a system with either processor:

1. You are dealing with a single cpu that has two dual core chips mashed together
(The Q in the Intel names are not to be mistaken for the X6700 which is a dual)
Two Woodcrests=1 Clovertown
Two Clovertowns=1 OctaCore System
Two Kentfields=1 Qx6700
Two Conroes=1 Qx6800
2. Current 775 socket northbridges do not allow for a dual cpu architecture, you must go with 771

3. Current 775 quad core chips are limited, with the production runs on the 65nm process seeming to be sparse and producting really, really expensive chips

4. Not until the next chipset, Clarksboro will you see the data contention of the chipset be alleviated, and even then it’s a new chip as well, the Tigerton.

These chips come in a variety of different cache options, and your application will of course dictate how much cache you want. This is the same issues as the Celeron, with some applications doing better with less cache, and even some operating systems doing better with less cache because of locking issues with more cache (?).

The benchmarks are clear, that when you stack an 8 core system against a 4 core system, there is just too much concurrency to deny a lead for a multi threaded application. Making the Freebsd kernel, or the linux kernel are very important tasks that show the incredible difference that cores make on a build.

I won’t deny however the naysayers of their two cents that some parts of the build process are not able to be built concurrently, and you must wait for one dependent library to finish before the rest can be spawned.

In short, there is a large difference for price in which market you go with, and varying levels of performance to boot. Intel won’t deny that using 2xWoodcrest will eat up more power than the single chip Qx6800, but there is more performance for multithreaded apps.

The difference is real, with two systems built for profit (and for mistakes made):

4 Core 775 system=$2,600
8 Core 771 system=$2,400

You can really tell when there is blood in the water and the sharks want a taste.

Some consider this last weekend the time when Microsoft admitted there was problems in some cases with Sleep, Hibernate or call it what you will. In short it saves the amount of time you need to boot your computer up. The point being that there is always some problems with the way various devices get shut down and how that affects your kernel as it cleans up.

I remember reading this way back, then trying the different modes of sleep on my PC (AMD64, Nforce4). I had no such problems, in fact, I could do it even when I was playing games and the game would wake back up too.

Then I read notes like this about the state of the repair of the sleep process for Vista and take it to heart, knowing for sure that Microsoft is going to issue patches for things that might be better off fixed in vendor code.

This skidmark over here starts complaining about how horrible Vista is, but he really likes black boxes. You cannot see inside black boxes, you don’t know what’s going on inside of them and you don’t know why they crash. He’s ignorant, and doesn’t even know how to interpret the blue screen beyond the color and throwing google the text he sees. At least some person with moderate sense starts to wonder if it’s drivers and not the hardware causing the problem.

One does have to start wondering why so many people blame Microsoft and not the company who made the laptop. They don’t question the chipsets, or any circuitry that might be inside, trying to do it’s own power saving. Let’s not mention that the same people expect Microsoft to buy every notebook on the planet and make sure Vista works on it. Only one company makes sure their product works on every notebook it’s designed for: Apple.

I take heart when someone posts what their system is and how well it works with Vista sleep, and while granted it would have been better to have a Detailed entry a model number is good enough. (Hey, the chipsets make a good indicator if it’ll support ACPI)

I’ve seen a Dell and an HP so far that work just fine with Vista. But I digress, I do not have the model numbers in my head. I can tell you however that the Dell only had 512mb of ram, so you know how that is going to turn out…. or do we?
As always, you can consider my opinion to be that Microsoft is not the only one to blame here, and we are operating off of insufficient information. There is no non-biased repository of information that talks about the amount of systems sleep -does- work on.

The only conclusion I can reliably draw at this moment is that no company has any business releasing a product or feature that works less than 50% of the time. Saying they would do this only infers that the Titanic didn’t sink because of an Iceberg, it sank because the designers knew 50% of the rivets would come out when it turned left.

Isn’t it wonderful? It used to just be relegated to the slim cases of notebooks, but now it’s in nearly every new Dell that’s sold.

I’m sure you are wondering why new computers have them, old ones do not, and what is going on inside of these pipes that route through a heatsink.

It’s pretty simple really why these pipes exist, and it has to do with how fast metal can move heat through them. It makes more sense to use this type of technology than a big chunk of copper, since copper can only go so far.

Intel will usually quote when this type of technology is needed to remove heat from the top of your brand new CPU. It may however be possible to achieve good airflow which is moves enough heat, but it will be louder than this type of heat conversion.

I was told once that the material inside of these pipes was Chlorine, which didn’t make much sense at the time. So I found a guy who actually made a heatpipe of his own, and I have to admit, he used some good techniques for testing and laying out what he did.

If you are looking for more information about heat pipes then you know where to look.