Interfacing A GPU With A CPU

December 26, 2012 by Mike Szczys 11 Comments

interfacing-a-gpu-with-a-cpu

[Quinn Dunki] pulled together many months worth of work by interfacing her GPU with the CPU. This is one of the major points in her Veronica project which aims to build a computer from the ground up.

We’ve seen quite a number of posts from her regarding the AVR-powered GPU. So far the development of that component has been happening separately from the 6502 centered CPU. But putting them together is anything but trivial. The timing issues that were so important to consider when developing the GPU get even hairier when it comes writing to the VRAM from an external component. Her first thought was to share a portion of the external RAM between the CPU and GPU as a way to push rendering commands from one to the other. This proved troublesome both in timing and in the number of pins available on the AVR chip. She ended up using something of a virtual register on the AVR chip that can receive commands from the CPU asynchronously. Timing dictates that these commands be written only during vertical blanking so this virtual register also acts as a status register to let the CPU know when it can send the next command.

Her post is packed with the theory behind the design, timing tests on the oscilloscope, and a rather intimidating schematic. But the most important part is the video showing her success in the end.

25 GPUs Brute Force 348 Billion Hashes Per Second To Crack Your Passwords

December 6, 2012 by Mike Szczys 66 Comments

It’s our understanding that the video game industry has long been a driving force in new and better graphics processing hardware. But they’re not the only benefactors to these advances. As we’ve heard before, a graphics processing unit is uniquely qualified to process encryption hashes quickly (we’ve seen this with bitcoin mining). This project strings together 25 GPU cards in 5 servers to form a super fast brute force attack. It’s so fast that the actual specs are beyond our comprehension. How can one understand 348 billion hashes per second?

The testing was used on a collection of password hashes using LM and NTLM protocols. The NTLM is a bit stronger and fared better than the LM, but that’s not actually saying much. An eight character NTLM password will fall in 5.5 hours, while a 14 character LM hash makes it only about six minutes before the solution is discovered. Of course this type of hardware is only good if you have a copy of the password hashes themselves. Login protocols will lock out after a certain number of attempts and have measures in place to slow down automated systems like this one.

[via Boing Boing]

PC Temperature Monitoring System Lights Up When Things Get Hot

October 4, 2011 by Mike Nathan 23 Comments

gpu_overheating_warning_system

[Taylor] popped a new graphics card into his computer, but before he could settle in for a round of gaming, his card started to overheat. He eventually tracked the problem down to an undersized power supply, but the prospect of cooking his new GPU to death made him think twice about how he was monitoring his system’s health.

To continually keep tabs on his video card’s temperature going forward, he put together a small circuit that will alert him if things start to get too hot. He mounted a small temperature sensor on his graphics card near the GPU, wiring it to an Arduino. The Arduino monitors his video card, lighting an RGB LED blue if conditions are alright. If the temperature rises above 50C, the LED changes to red, signaling a problem.

We’re aware that there are all sorts of software applications that can monitor component temperatures for you, but the appeal of [Taylor’s] system is that it can be easily seen from across the room rather than via the desktop. That said, we think that his system could take advantage of his PC’s case fan lighting for a more visible warning, and it wouldn’t hurt to wire in an auto-shutdown feature in case his computer overheats while he’s away.

Paddle Controller For GPU Overclocking

August 25, 2011 by Mike Szczys 10 Comments

[Fred] likes to squeeze every cycle possible out of his graphics card. But sometimes pushing the clock speed too high causes corruption. He figured out a way to turn a knob to adjust the clock speed while your applications are still running.

The actuator seen above is a Griffin Powermate 3.0. It’s a USB peripheral which is meant to be used for anything you can imagine. [Fred] uses an AutoHotKey script that he wrote to capture the input from the spinner, process that information, then adjust GPU clock speed in the background. Since the clock on his ATi Radeon 5800 can be adjusted using the AMD GPU clock tool, it’s an easy choice for this application. Now better graphics are at the tips of his fingers. See for yourself in the video after the break.

Of course if you don’t want to shell out for the fancy hardware you could always build your own paddle controller.

Continue reading “Paddle Controller For GPU Overclocking” →

Converting PWM To DC Signaling For More Precise Fan Control

August 15, 2011 by Mike Nathan 24 Comments

pwm_to_dc_fan_control

[hedgehoginventions] wrote in to share a little modification he made to his video card in order to keep it from overheating during strenuous 3D tasks. Having swapped out the stock cooler on his Nvidia 9600GT graphics card, he found that it did not need to utilize the fan while doing mundane things like checking email, but that it still required extra air flow while playing games.

He figured he get the fan to shut off by tweaking the PWM signal, but he found that he could not get the duty cycle under 20% using software, which still caused the fan to run at all times. The circuit he built takes the PWM signal output by the card, cleaning it up before converting it to a corresponding DC voltage. The fan then runs at the same speed it would if driven directly by the PWM signal, though it can now turn off completely when not required.

It’s a nice way to do automatic fan control when you can’t otherwise get your GPU fan to shut off. Nice work!

Creating Music From GPU Noise

July 24, 2011 by Mike Szczys 27 Comments

Yep, that’s a picture of a Laptop rocking out on an electric guitar. In what can only be described as a truly bizarre hack [CNLohr] discovered that the RF noise from the computer can be used to play music through the guitar’s pickup.

Check out the clip after the break to hear an annoying, but very discernible rendition of Jingle Bells. Once [CNLohr] stumbled onto the fact that changes in what the graphic processing unit is doing was affecting the pitch detected by the pickup he started writing some code. Now he’s got a program that automatically calculates the size of the window, and produces a white square on a black background to dial in the GPU at the right frequencies.

He mentions in the notes accompanying his video that he had to turn off Vsync to get this to work right. We don’t understand why but we’d love to hear what you think in the comments.

Continue reading “Creating Music From GPU Noise” →

GPU Password Cracking Made Easy

June 1, 2011 by Mike Szczys 37 Comments

The power that a Graphics Processing Unit presents can be harnessed to do some dirty work when trying to crack passwords. [Vijay] took a look at some of the options out there for cracking passwords and found that utilizing the GPU produces the correct password in a fraction of the time. On a Windows machine he pitted the Cain password recovery tool which uses the CPU for its calculations against ighashgpu which uses ATI or Nvidia graphics cards to do the deed. Hands down ighashgpu is the fastest; with Cain taking about one year to crack an eight character password while ighashgpu can do it in under nineteen hours.

We were very interested to see how easy it is to use this package. We looked in on GPU cracking in September but didn’t focus on the software packages that are out there. Now that you know how easily your password can be unearthed perhaps you will get some use out of this article discussing the usability and security of longer passwords which we ran across over on Reddit.