Here are the specs:

• 12 or 24 High-Intensity LEDs
• Offered in 1 and 2 meter in cables
• On/Off and 3-way brightness control
• Narrow cable fits in tight paces.

When we opened the shipping box, we were immediately surprised by the remarkably small package.  Had we seen this on a shelf, we would not have guessed that the small black coil of wire was a meter long.  It was obvious before we opened the package that the design is remarkably compact.

The clear packaging also reveals that everything needed to install the LED strip is included.    So far, this product has given us a great first impression.

The contents of the package are shown below:

The contents include the LED strip, a PCI packet for controlling the power and brightness, a screw installing the bracket, a power cable, and mounting clips for the LEDs.

The LED strip is shown below.  There are 12 LEDs on the meter-long wire, spaced roughly 3.25 inches apart.  The first LED is 3.25 inches from the start of the cable.  Since the cable must plug into the PCI bracket (more on that below), this puts constraints on where the LED strip starts and ends.

The kit includes 6 clips for mounting the LED strip.  The clips mount to any flat surface with double-sided tape.  The wire itself is slightly flattened, which allows the clips to keep the LEDs properly oriented.

The LEDs get their power from a PCI bracket.  The PCI bracket holds a smaller circuit with the necessary resistors.  It provides an On/Off switch on the back panel, and a brightness control with 3 settings.  The power comes from a small cable that plugs into the edge of the board.

Below, the bracket is connected to the LED strip and the power cable.  The power cable seems a bit short.

Due to the short power cable, we had to route a spare cable from the PSU down to the PCI bracket.  This isn’t ideal for cable management (the PSU cable is quite large), but we managed to tuck it out of sight behind the graphics card.

After reinstalling the graphics card and securing the power cable, we couldn’t wait to try out the LEDs.  We were amazed at the brightness.  The photo below hardly does justice to their intensity.

Finally, we used the included mounting clips to run the lights around the outside of the case.  The clips  held the wire firmly, and kept the LEDs pointed in the right direction.  The LED cable was long enough to run all the way across the bottom and up one side of a Thermaltake Armor case.

The end result is shown below, with the side panel reattached.  The case is illuminated brilliantly, even with LEDs along only two sides.  The two-meter version might wrap all the way around the case, but LEDs on two sides provide plenty of lighting.

This case was previously lit by dual six-inch cold cathodes.  The cold-cathodes are a bit old, and no longer at their peak intensity.  However, the illumination from the LEDs greatly exceeds the cold cathodes at their best.  For comparison, the case is shown below, lit with only the cold cathodes on the left, and only the LEDs on the right.  Both pictures were taken with identical manual exposure settings under the same ambient lighting.  The dynamic range of our camera was not sufficient to truly capture the intensity difference.   While larger or newer cold cathodes would be brighter, the LEDs are more than sufficient.

Overall, this is a great product.

Pros:

• Very bright.
• Cheap ($9.99 directly from NZXT)
• More compact than cold cathodes.
• No inverter required.
• Multiple brightness levels.

Cons:

• Short power cable.
• Short lead before the first LED.

The short lead on the LEDs was only a minor issue.  Moreover, using a different PCI slot would allow for some flexibility.   The short power cable dropped our rating a little, since it required routing a new cable from the PSU.  We give this product 4.5 stars.

Comments? view forum post

“Today, AMD launched its Radeon HD 6800 series graphics processors, with all its AIB partners bringing out their designs. If you haven’t read our review of the Radeon HD 6870 already, please do so, to understand what the card is worth in its single GPU form, it may also clear some lingering misunderstandings about the product and its naming scheme, which you may have.

The Radeon HD 6870 supports 2-way AMD CrossFire multi-GPU technology, which allows you to upscale performance by pairing two HD 6870 cards of any other make, you can also pair an HD 6870 with a HD 6850, which may not yield the maximum performance increase but provides an additional upgrade path.

In this review we’re going to show you what to expect from Radeon HD 6870 CrossFire, a roughly $480 solution at present. Since AMD got off on the premise of promising us Radeon HD 5800 series kind of performance at very affordable price points, it will be interesting to see if HD 6870 CrossFire can surprise us.”

[url=http://www.techpowerup.com/reviews/ATI/Radeon_HD_6870_CrossFire/1.html]Read the full review on Techpowerup.com[/url]

Comments? view forum post

The site was bought from Smoked1 by a small team that had banded together on VW – Voided Warranty * A4T in her old guise to the un-aware *

The team brought me in as a Co-Owner and I’ve stayed ever since! It’s been a long bumpy road and we still have some work to do, But I am confident with our current team of old and new staff alike, Appetite For Tech has a future on the World Wide Web!

Working in the background, I’ve become known as *The Fixer* And I’m ususally one of the team that keeps the forums running smoothly.

So what are you waiting for? Got an appetite for tech? Register and join in with our growing community!

Paul

Corsair memory has  a reputation among PC enthusiasts for producing top-notch RAM.  Founded in 2001, they quickly became known for both impeccable quality and high performance.  We recently got our hands on a triple channel Dominator kit, designed for Core i7 systems.

This kit is massive–a whopping 12 GB–and fast, featuring 6 DDR3 1600 modules.  Today A4T’s high-performance computing specialist will subject these modules to the mother of all stress tests, to see if they are worth their weight in silicon.

First, here are the specs:

• Model: CMD12GX3M6A1600C8
• Capacity: 12 GB (6 x 2 GB)
• Speed: DDR3 1600 (PC3-12800)
• Timings: 8-8-8-24
• Voltage: 1.65V
• Intel XMP (Extreme Memory Profile) Support
• Dominator heat sinks
• Cooling fan included

Installation

The Dominator heatsinks make these chips considerably larger than most DIMMs.  We had no trouble installing them in our water cooled testbed.  However, due to their height, they may get in the way of a large CPU cooler.

The DIMMs aren’t much wider than a typical DDR3 module equipped with a heat-spreader.  We had no trouble fitting them in adjacent DIMM slots.

The modules installed in our testbed:

Just in case the ginormous heat sinks weren’t enough, the Dominator kit also includes an active RAM cooler, equipped with two 60mm fans:

Some water cooling enthusiasts may balk at the thought of adding not one but two noise-producing fans to their systems.   We viewed the fan as a necessary evil, given the magnitude of the flogging we planned for this memory.  We didn’t find the fan noise terribly bothersome–it barely made an audible difference with the side panel closed.

The fan is, of course, gigantic.  It covers all 6 modules.  This may be a problem for large CPU coolers.

Installing the fan was easy.  It attaches to the clips on the DIMM slots:

Testing

We tested the memory on an EVGA X58 SLI motherboard with a Core i7 920 processor ocerclocked to 2.88 GHz/5.8 GT/s QPI.  We set the memory timings at 8-8-8-19 1T.  The system was running Windows 7Professional 64.  The screenshots from CPU-Z below show the detailed specs (click to see a larger image).

We benchmarked the memory bandwidth and memory latency in SiSoftware Sandra 2010.  The results are below.  As a testament to the power of this memory, we found that running the memory-intensive benchmarks did not result in any noticeable system-wide slowdown.

Sandra memory bandwidth benchmark. The top bar in each graph is our system. Larger values are better.

Sandra memory latency bencmark. The top bar in each graph is our system. Smaller values are better.

The Sandra benchmarks compare our test system to synthetic benchmarks for other configurations.  The synthetic benchmarks are only estimates.  However, our test system beat all but of the synthetic benchmarks available in Sandra 2010.  That synthetic benchmark (which we tied) was for the same memory modules that we are testing!

Benchmarks are not always good indicators of real-world performance.   We were therefore faced with the dilemma of finding a “real world” application that would stress 12 gigabytes of RAM.  We turned to an algorithm from a machine learning project.  The algorithm itself is a complex procedure used in a variety of applications.  For the more mathematically inclined, we computed the singular value decomposition of  a 50,000 x 50,000 matrix.   In layman’s terms, we extracted meaningful patterns from nearly 10 gigabytes of data.  The screenshot below shows the Task Manger as the algorithm was running.

This much memory consumption alone would be enough to make a typical PC die a thrashing death, and yet our test system continued to function without a hiccup.  However, A4T review points must be earned, so we devised a plan to push the RAM even further.   We re-ran the test above while playing Crysis.

The results were impressive.  Even at peak memory usage, Crysis ran at 40 to 60 FPS on a GeForce GTX 285 with all the display settings on the highest quality.  There was no noticeable lag, despite the thorough flogging that our memory was receiving in background.

The screenshot below shows Crysis running in windowed mode, with the task manager (showing the memory usage) in the background.  Click to see a larger image.

This is impressive to say the least.

In summary:

Pros:

• It’s fast.
• 12 GB is a lot of RAM
• Lifetime Warranty
• The heat sinks make nice addition to a windowed case.
• You can factorize large matrices in the background while playing Crysis.

Cons:

• May get in the way of some CPU coolers.
• It’s not cheap (but you get what you pay for).

Overall, this RAM is a great buy.  We’re proud to give it A4T’s top rating.

Comments? view forum post

Created by Blake Foster, Rui Wang, and Erik Learned-Miller

The human eye is amazingly adept at tracking moving objects. The process is so natural to humans that it happens without any conscious effort. While this remarkable ability depends in part on the human brain’s immense processing power, the fast response of the extraocular muscles and the eyeball’s light weight are also vital. Even a small point and shoot camera mounted on a servo is typically too heavy and slow to move with the agility of the human eye. How, then, can we give a computer the ability to track movement quickly and responsively?

Thanks to recent progress in camera miniaturization, small, easily manipulable cameras are now readily available. In this project, we use a first person view (FPV) camera intended for use on model airplanes. The camera is mounted on servo motors which can aim the camera with two degrees of freedom. The entire assembly weighs only 32 grams, only slightly more than a typical human eyeball. Coupled with a GPU-based tracking algorithm, the FPV camera allows the computer to robustly track a wide array of patterns and objects with excellent speed and stability.

The video clip above shows a short demonstration.   The camera remains locked on the target [me] even as he jumps, squats, paces about, and turns 360 degrees.  We show both the view captured by the tracking camera (the smaller video), and the view from a different camera that shows the movement of the tracking camera (the larger video).

Parts List (links to parts included)

• Pan-Tilt Camera
• Video Cable
• Portable 12V Power Supply
• Video Transmitter (optional)
• Video Receiver (optional)
• Antenna (optional)
• RF Link Transmitter (optional)
• RF Link Receiver (optional)
• USB Frame Grabber
• Arduino Diecimila
• Voltage Regulator
• Heat Sink (optional)
• Capacitors (0.1 μF)

Note: the software (downloadable below) requires a PC with CUDA-capable graphics hardware (GPU).

The Camera
We used a 420-line pan-tilt camera manufactured by Fat Shark. The camera is mounted on two servo motors, which allow for about 170° of rotation on the yaw axis and 90° of rotation on the pitch axis. The camera produces composite video in PAL format. An NTSC version of the camera is available as well, but it was out of stock when we ordered our parts.

Power
Because the video transmitter requires the 12 volts, we power the camera with a 12V rechargeable Lithium battery. We use a voltage regulator to provide 5 volts for the camera and the servos. We added capacitors before and after the regulator to eliminate any voltage fluctuations. Both the video and servo cables from the camera connect to headers on the voltage regulator circuit, which provide the regulated 5V power supply. The servo control signals and the video output are passed through to a second set of headers, which connect to the frame grabber and the Arduino, respectively. The video output header additionally provides a 12V power supply for the video transmitter.

Digitizing the Video

We used a USB frame-grabber manufactured by StarTech to read the video into the host PC. The frame grabber supports both NTSC and PAL composite video, so the NTSC camera could be used without any hardware changes. We used a video cable sold by Digital Products Company to connect the frame grabber to the video output header on the voltage regulator circuit. The cable also has a power jack, which provides 12 volts to the video transmitter.

The frame-grabber provides 640 x 480 interlaced video at 25 FPS. For efficiency, we downsample the video to half resolution for tracking. Our downsampling filter discards the even lines to eliminate errors due to combing artifacts. We display the video at full resolution, after eliminating combing artifacts with a standard deinterlacing filter.

Controlling the Servos

We used an Arduino Diecimila to generate the control signal for the servos. The Arduino receives the desired pulse widths for the servos over its serial port. Each pulse width is encoded as a 16-bit integer, with 1 bit reserved to select one of the two servos. We use the servo library included with the Arduino software to generate the PWM signals.

Wireless Operation

We can eliminate the wired connections to the host PC with a wireless transmitter and receiver for the video, and a wireless RF link for the servos.

The RF Link

The wireless RF link transmits the servo angles digitally, with a range of up to 500 feet. Both the transmitter and receiver connect to Arduinos running the Virtual Wire library. The transmitter Arduino (connected to the host PC) broadcasts each 2-byte angle, followed by a byte of all zeros to keep the transmitter and receiver in sync. The receiver Arduino updates the servo angles when an angle is transmitted correctly (that is, all 3 bytes are received). Because Virtual Wire is incompatible with the Arduino’s servo library, we use the Software Servo library to control the servos.

Wireless Video

The wireless video transmitter broadcasts NTSC or PAL video with a range of up to 500 meters. The transmitter plugs directly into the RCA jack from the camera. The 12 volt line on the video output header powers the transmitter. The RCA jack on the receiver plugs directly into the frame grabber. The receiver can be powered by a generic 12 volt power adapter.

Software

The software for the project, including source code and build files, is available via the links below.   The tracking software is based on an algorithm developed jointly by the UMass Computer Vision and Computer Graphics labs. It runs on a CUDA-capable NVidia GPU. The GPU acceleration can process video at up to 60 frames per second, which enables the real time tracking we show in the video. Our implementation has only been tested in Windows 7, but we do not foresee major difficulties in porting it to other versions of Windows.  We compiled and ran the software under Linux (Ubuntu), but we were unable to receive video due to driver problems with the frame grabber. We do not anticipate any problems running our system on Linux with a compatible frame grabber.

More details about building the software are included in the README files.

Download Software:
Windows
Linux

view forum post

Comments? view forum post

“Xigmatek is ready with its newest PC case, the Pantheon. This ATX mid-tower comes in four variants: Pantheon Black (all-black, with an extra intake instead of a side window), Pantheon Black W (all-black, with a side window), Pantheon Silver (has two chrome lines running down the front), and Pantheon Silver W (the same, with window). The case measures 206 (W) x 543 (H) x 500 (D) mm, has a perforated metal front, including perforated drive bay covers. Interiors are all-black, too; including tool-free drive and expansion card retention knobs.

A unique feature here is the air-intake and 3.5″ drive cage design. The front panel over the drive cage area is detachable, giving you access to six hot-swappable 3.5″ drive trays. Instead of its conventional location, the air-intake fans are moved to its side, consisting of no less than two 120 mm spinners. Other fans include one 140 mm rear, and slots for one bottom intake, two top exhaust. The motherboard tray has holes that help with cable-management, and a cut-out at the CPU socket area to help manage coolers with back-plates. Front-panel includes two USB 3.0 and two USB 2.0 ports, apart from audio. The company did not give out a price.”

Comments? view forum post

It has come to my attention that there is an actor named Blake Foster.  I am older than he and I had the name first, which in my opinion makes me the real Blake Foster. The other one is just an actor.

My personal website is at www.blake-foster.com

Frequently, while at the office or out of town, I find (to my dismay) that I need to access files on my desktop computer at home. The inevitable results are special trips and wasted time. While I could leave my desktop on indefinitely and access files remotely as, the inordinate amount of electricity that it consumes literally doubles my electric bill. Wake on LAN (WOL) would be another option, but I don’t want to give anyone in the world the ability to start my computer. Moreover, if I should ever crash my system while logged in remotely, I need a way to force it to reboot. With this project, I’ve solved both problems for good, for the cost of about 3 months of running my desktop all day.

Parts:

• Arduino Duemilanove
• Arduino Ethernet Shield
• Arduino ProtoShield Kit
• Wire Wrap Pins
• 3:8 Decoders (x2, optional)
• 16 Pin IC Socket (x2, for the decoders)
• SPST 5VDC Reed Relay
• PNP General Purpose Transistor (such as the 2N3906)
• Resistors (1 4.7k and 1 10k)
• Various plugs/jacks for connecting the Arduino to my PC.
• Power Switch Extension Cab/le (If you don’t want to cut the power switch cable in your case)

The Arduino

The Ethernet shield
The power switch circuit. I used wire wrap pins to extend the pins on the ProtoShield, in order to clear the Ethernet jack. Download a schematic here.

The heart of this project is an Arduino Duemilanove microcontroller, equipped with an Ethernet adapter, and a home-made circuit that connects it to power switch on my computer. The Arduino is always on, so at any point in time (and from anywhere in the world), I can command the Arduino to start up my computer.

The simplest possible implementation would have the Arduino trigger the power switch any time it received a packet on some specific port. While this might work in practice, there is one obvious problem–it would allow anyone in the world to start up or shut down my computer (most likely accidentally). I’ve addressed this problem in the carefully crafted software running on both the Arduino, and my laptop, from which I access the Arduino remotely.

My first layer of security is a set of randomly-selected 64-bit binary strings that encode the messages to the Arduino. The Arduino component of my software will respond only to one of these exact strings. The chances of a stray packet carrying the startup signal reaching the Arduino are negligible.

As a second layer of security (which is undoubtedly overkill), I implemented some rudimentary cryptography in the communication protocol. Prior to sending any message, my laptop and the Arduino exchange a secret key with Diffie-Hellman key exchange. All messages between my laptop and the Arduino are encrypted with the key. Moreover, the key is invalidated after each message, which ensures that no message to or from the Arduino will be repeated in my lifetime.

To avoid the lose of unsaved work, I need to ensure that my software won’t accidentally shut off my computer. I’ve handled this problem with some simple status checking. Before the Arduino triggers the power switch, it sends my desktop an ICMP echo request (ping). If it receives a response, it knows that my desktop is already on, and ignores the power request. Only if it receives will trigger the power button.

Because I can’t change the Arduino software remotely, most of the code runs on my laptop. The Arduino supports only a small set of low-level functions for communication. A typical exchange works as follows:
1. My laptop randomly picks number a, and then computes g^a mod p, where p (a prime) and g are randomly selected integers, known in advance by both my laptop and the Arduino. My laptop places the resulting 64-bit number in a single UDP packet. The packet also includes 8-bit sequence number (since UDP does not guarantee in-order delivery) and an 8-bit code that indicates a key-exchange request.

2. When the Arduino receives the packet, it first checks the 8-bit code and concludes that the packet is a key-exchange request. It then proceeds to pick a random number b, and computes g^b (mod p). It places the result in a response packet contains, which it sends back to my laptop. The response contains the same sequence number (so that my laptop knows which message the Arduino is responding to), and the same 8-bit key exchange code.

3. The Arduino now knows b (which it picked itself) and g^a (mod p) (which it received in the key request packet). It then computes (g^a)^b mod p. Similarly, my laptop knows a (which it picked itself) and g^b mod p (which it received from the Arduino). It then computes (g^b)^a (mod p). Each computation produces the same result, which serves as a secret key known by only the Arduino and my laptop. This is mechanism is called Diffie-Hellman key exchange.

4. After my laptop has computed the key, it needs to determine if my desktop is already on. To this end, my laptop sends a second packet to the Arduino, carrying a pre-determined message that indicates a status query. The first byte of this packet is a code which indicates that the packet contains an encrypted command. As with the key exchange, the second byte contains a sequence number. The final 8 bytes contain the pre-determined message, encrypted with the key. 5. When the Arduino receives the packet, it once again starts by investigating the first byte, which reveals that the packet contains an encrypted command. The Arduino then proceeds to decrypt the command with the key. Finally, by comparing to the pre-determined constants, the Arduino discovers that the command is a status request.

6. To determine if my desktop is on, the Arduino sends an ICMP ping. If, after four attempts, it does not receive a response, it assumes that my desktop is off. In each case, the Arduino sends a pre-determined random binary string as its response, also encrypted with the key. The first and second bytes contain, respectively, the 8-bit message code and the sequence number used by the incoming packet. The Arduino also invalidates the key, which will cause it to ignore any incoming commands until a new key is exchanged.

7. When my laptop receives the response, it decrypts with the key (which it has not yet invalidated itself). If the key indicates that my desktop is on, my laptop does nothing more. If my desktop is off, my laptop initiates a new key exchange.

8. After the new key exchange, my laptop sends another pre-determined message encrypted with the new key, indicating that the Arduino should trigger the power button. The Arduino responds by sending the power-on signal to the switch-control circuit. The power-on signal is a 6-bit code, written in parallel to 6 of the Arduino’s digital output pins. The switch-control circuit uses two 74ALS138 3:8 decoders to select a single combination of these 6 pins as the switch-control signal. This is intended to prevent the Arduino from accidentally triggering the power switch after a software or hardware glitch.

9. After triggering the power switch, the Arduino responds with a pre-determined acknowledge message (encrypted with the key) and invalidates the key.

All normal exchanges between the Arduino and my laptop consist of a message sent to the Arduino containing a message code, sequence number, and message data, followed by a response from the Arduino to my laptop with the same message code and sequence number. To deal with packet loss, I use two additional pre-determined messages. First, if after a short timeout interval, my laptop has not received a response, it can send a special repeat-request packet. When the Arduino receives a repeat request packet (again indicated by the first byte) it retransmits the last packet it sent. If multiple copies of a packet are sent, my laptop uses the sequence numbers to eliminate duplicates.

If the Arduino is delayed (for example, while waiting for ping responses from my desktop), it can send a special wait message, again indicated by the first byte. If my laptop receives a wait message, it will reset its timer to zero, thus avoiding timeout.

If my desktop should ever crash while I’m logged on remotely, my laptop can also send a force shutdown message. My laptop will send this message regardless of whether my desktop is on or off. When the Arduino receives a force-shutdown message, it will press and hold the power button for 10 seconds. I do occasionally crash my computer while doing research (it’s easy to do when you’re dealing with 100,000 x 100,000 matrices), so this is a very useful feature.

The power switch control circuit uses a reed relay to short out the power switch pins on my motherboard. I installed a jack in the back of my case, which gives me access to the power pins from the outside. I tapped the wires from my case power switch, so that the front panel power button still functions.

The image below shows a screen shot of the control program running on my laptop. Note: That is not my IP address in th box, hence the multiple timeout messages.

For those interested, my software is available for download.

Arduino Software:
Arduino Sketch
Requirements:
My ICMP Ping Library
The TrueRandom Library

Control Program:
Visual Studio 2008 Project
Windows Binary (32 and 64 bit versions included)

Note: The control program will create the registry key HKEY_CURRENT_USERSoftwareRemote_Startup to store the IP address and port number. If you run the software but don’t intend to use it in the future, you can use regedit to remove this key. The software will not make any other changes to your system.

view forum post

Comments? view forum post

Performance improvements:

STALKER: Call of Pripyat Benchmark – depending on the model, screen resolution, anti-aliasing and whether you are using ATI CrossfireX™ technology, here’s an example of some performance improvements we’ve noted in our labs:
up to 20% on ATI Radeon™ HD 5870 graphics
up to 12% on ATI Radeon™ HD 5770 graphics
The Chronicles of Riddick: Assault on Dark Athena
up to 4% on ATI Radeon™ HD 5770 graphics
up to 5% on ATI Radeon™ HD 5850 graphics
up to 6% on ATI Radeon™ HD 5970 graphics
ATI Catalyst™ Application – Updated ATI CrossFireX™ Profiles

Aliens Vs. Predator – Profile update
Battlefield: Bad Company 2 – Profile update
F1 2010 – Profile update
Kane & Lynch 2 – New profile
Resolved issue highlights:

UVD clocks will now return to normal speeds after switching from High Definition to Standard Definition content

Flickering is no longer visible while running 3D games on HDMI displays with Quad CrossFireX™ enabled

Mesh corruption is no longer visible in “The Chronicles of Riddick 2: Assault on Dark Athena” when SSAO in-game settings are set to “high”

ATI CrossFireX™ now functions properly, and in-game Anti-Aliasing can now be enabled in “StarCraft II”

Adobe® Flash Player will no longer display green screen after resume from sleep

Negative ATI CrossFireX™ scaling on “Final Fantasy-XIV” no longer occurs
Intermittent performance drops no longer observed with hardware cursor enabled in “World of Warcraft” game running in extended mode
Resolved issue highlights:

Download ATI Catalyst 10.9 Now!

Comments? view forum post

“For this review I will be looking at another product from our
friends at Thermaltake. The product this time is one of their
power supply units and is the
Evo Blue Series 650watt. As usual coming from Thermaltake I expect nothing
but quality.”

Link:
XtremeComputing

Comments? view forum post