Guy Builds Wind Tunnel-Cooled Computer to Help Cure Cancer

I’ve had problems with computers and video game systems overheating over the years, so I’m all for good cooling systems. And while some computer modders have solved the problem by adding more cooling fans and heat sinks, I think this build might just be a bit more than most of us need.

wind tunnel computer 1

What you’re looking at here is the Wind Tunnel Computer. It was built by Mike Schropp over at Total Geekdom, and it’s nothing short of awesome, if you ask me. Those giant ducts you see on the left and right are air intakes and outputs for the 6-foot-wide system, which is mounted in a contraction chamber in the middle designed to take full advantage of the increased air speed as it flows through.

wind tunnel computer spec

The beefy Intel Ivy Bridge 3770k powered system is overclocked to a whopping 4.5GHz, along with overclocked Radeon 7970 GPUs running on Sapphire Dual-X cards. But it doesn’t just run fast, look cool and stay cool – Mike built it for a good cause. It’s connected to the World Community Grid project, and its CPU and GPUs spend their days and night crunching numbers to help cancer researchers find a cure. And thanks to the massive throughput of the system, it’s able to push through about 20x the workload of an average 4-core system. While that’s all awesome, I do wish Mike had actually built a small wind test chamber inside of it for other science experiments. Maybe for version 2.0.

wind tunnel computer 2

He built the system using large MDF wood panels for the vent sections, and there’s a big box fan installed in one end of it to generate the wind. Mike assembled the main computer section using custom-cut Lexan so you can see through to it, and aluminum pieces to finish the look. It’s even got a control panel on the front with LED temperature gauges and key switches for turning the fans on and off.

wind tunnel computer 3

Using a wind-speed tester, Mike was able to determine that the inlet takes in air at about 0.6MPH, and the exhaust port pushes out air at 1.4MPH. However, it’s the center contraction section where the computer is that really benefits from the design, where the wind reaches 12MPH – or 9MPH with the computer in place. He actually was able to push wind speeds as high as 26-30MPH with the fan set on a higher setting. He even used smoke testing to ensure optimal airflow over the components.

wind tunnel computer 4

Check out the complete build log, along with many more photos over on Total Geekdom.

Facebook Poke and Snapchat bug saves “deleted” videos

A security flaw that quietly saves private videos in Snapchat and Facebook Poke for future recovery, when the sender expects them to be deleted after a few seconds, has been identified in the new “sexting” apps. Both apps promise a simple way to share photos and video temporarily, with the promise that they’ll be deleted shortly after viewing; however, BuzzFeed discovered, a simple third-party file browser can pull the video files from a cached store on your iPhone or iPad, allowing the clips to be archived and shared.

facebook_poke_extract_videos

Snapchat and Poke – Facebook’s recently released copy of the original app – each cater to the temporary messaging market. Senders of messages, photos, and video can choose how long they wish the content to be available, for up to ten seconds, after which time the promise is that they are deleted from the app and the company’s own servers.

Although a wide variety of uses are suggested, the most commonly-cited purpose for Snapchat and Poke is “sexting” where the ephemeral nature of the media leaves users more confident that DIY pornography won’t hang around to haunt them later. While screenshots can be saved on the device, both apps warn the sender that a shot has been taken.

snapchat_save_videos

This new hack, however, gives no such warning. Users simply need to load – but not open – any received clip, then plug their iOS device into their computer and use a third-party file browser such as iFunBox to navigate through to the Snapchat/Poke temporary file folder (either Snapchat/tmp or library/caches/fbstore/mediacard). The video clips are stored there; in Facebook’s case, they are deleted after being watched, whereas Snapchat appears to save them forever.

The hack only works with videos – photos don’t appear to be cached – but serves as another warning that even the most casual digital content could be extracted if users have the right tools. Despite the relative simplicity of the cache access, Snapchat doesn’t appear to be concerned. “The people who most enjoy using Snapchat are those who embrace the spirit and intent of the service” founder Evan Spiegel told BuzzFeed. “There will always be ways to reverse engineer technology products — but that spoils the fun!”


Facebook Poke and Snapchat bug saves “deleted” videos is written by Chris Davies & originally posted on SlashGear.
© 2005 – 2012, SlashGear. All right reserved.

Verizon FIOS Hacked (Rumor)

On Saturday evening, there was a story making its rounds that a hacker claimed to have nicked information from approximately 3 million Verizon Wireless customers which was in plain-text, and this hacker allegedly posted the information for 300,000 customers online from Saturday onwards. Calling himself “TibitXimer”, the hacker mentioned over Twitter (prior to the account being removed) some clarifications, stating that the names and information were taken from a list of Verizon FIOS customers, where content of such information include sensitive data such as serial numbers, names, addresses, starting date with Verizon, password to one’s account, and phone numbers among others.

What about the folks over at Verizon? How are they taking it? The carrier has stepped forward with a statement of their own, where Verizon spokesman Alberto Canal claimed that “The ZDNet story is inaccurate. The ZDNet story is inaccurate. We reported this incident to the authorities when we first learned of it months ago and an investigation was launched. No Verizon systems were breached, no root access was gained, and this incident impacted a fraction of the number of individuals being reported.”

I guess everyone else can sleep easy this Christmas, but I won’t be surprised of TibitXimer comes up and throws a spanner in the works along the way.

By Ubergizmo. Related articles: Google Releases Happy Holiday Cards, Wishes Everyone Happy Holidays, Windows 8 Start Menu Replacement Application ‘Pokki’ Registers 500k Users,

Hacker allegedly leaks Verizon FiOS customer data, Big Red points the finger at marketing firm

Hacker allegedly leaks Verizon Fios customer data, big red points the finger at marketing firm

Verizon customers saw a flash of excitement this weekend when an alleged hacker claimed to have pilfered personal data for some three million of its wireless customers. Twitter user TibitXimer shared 300,000 names from the file, claiming to have collected them as early as July 12th. According to Verizon, the would-be hacker’s claims are bunk — the leaked data has been available for months, and it’s populated by Verizon FiOS customers. More importantly, Verizon says that its servers weren’t hacked at all.

“There was no hack, and no access gained,” it said in a statement to The Next Web “A third party marketing firm made a mistake and information was copied.” Verizon says the leak was reported to authorities months ago, and insists that recent claims are inaccurate and exaggerated. Sure enough, security researcher Adam Caudill recalls seeing the file back in August, guessing this is probably a file leaked from a telemarketing agency. Either way TibitXimer’s account has vanished from the social network, demonstrating, if nothing else, that Twitter is serious about its Trust & Safety policies.

Comments

Source: The Next Web

DIY Google Project Glass: I Am Locutus of Nerd!

Even though people haven’t been that hyped up about Google’s Project Glass of late, some technophiles still couldn’t wait and decided to hack together their own augmented reality eyewear. This particular design makes you look like some sort of space pirate.

space cyborg pirate google glass

Gregory McRoberts designed his augmented reality specs to look like an angry eyepatch. Apart from serving as a cool disguise for Halloween, it was designed as a aid for the visually impaired. Instead of an LCD screen to show information, the patch uses a simple set of glowing LEDs to translate temperature and distance information to the wearer, helping provide depth and safety cues to those without 3D vision.

space cyborg pirate google glass naked

A flashing green LED will blink faster or slower depending on whether an object is closer or farther away. Blue LEDs will shift to red if an object or a room is above 80 degrees Fahrenheit.

Gregory also suggests that the device could be fitted with vibration motors to provide similar feedback to the totally blind.

[via Pinterest via Adafruit]

25 GPU Cluster Guesses Windows Passwords In Under 6 Hours

The amount of raw computing power found in a 25-GPU cluster is more than enough to take a crack at 6.6 quadrillion password combinations. Just in case your mind is boggled by the amount, let’s just put it simply. That amount is more than enough to check just about every possible eight-character password, ranging from upper/lower-case letters to digits and symbols in a matter of 5.5 hours. Microsoft’s NTLM cryptographic algorithm that has seen action since Windows Server 2003, is relatively weak and particularly insecure in some enterprise settings by today’s standards. Let’s just say that access to a hash of a workstation password, this machine will most probably take less than a day to crack it.

Of course, we are talking about this machine guessing passwords of up to eight characters within a reasonable time period, and throwing in an additional character (to nine characters in total) would take approximately 500 hours to crack. When you have 10 characters in the equation, then such a machine would consume 5.4 years of cracking time. Just how many characters does your password contain for most of your online accounts?

By Ubergizmo. Related articles: Google Discounting Chromebooks For Schools To $99 Until 12/21, Intel Readies 13W CPUs, Dives Deeper in Low-Power Territory,

Tumblr blogs hacked to display racist, offensive posts

If you maintain a blog on Tumblr, it might be a good idea to stay away from the site for the time being. Apparently, hacker group GNAA has taken advantage of a major security hole to publish racist and offensive posts to Tumblr blogs, and visiting infected blogs while logged into Tumblr helps these posts spread. Buzzfeed originally reported on the attack, posting images of the inflammatory posts being published (just a warning: they’re pretty offensive to anyone other than an Internet troll).


Buzzfeed suggests that if you go to Tumblr, only go to your dashboard. By avoiding other Tumblr blogs, you can prevent this exploit from spreading and posting things you probably don’t want to your blog. It seems that this exploit first targeted, of all things, Tumblr’s Brony tag, with the company making a statement on the exploit. Here it is in full:

There is a viral post circulating on Tumblr which begins “Dearest ‘Tumblr’ users”. If you have viewed this post, please log out of all browsers that may be using Tumblr immediately. Our engineers are working to resolve the issue as swiftly as possible. Thank you.

No word on when a fix is scheduled to arrive, but hopefully it’s delivered at some point today. The post themselves are shooting for shock value in a big way, so make no mistake, this definitely isn’t anything you want your visitors to see. It’s possible that this exploit is spread by running a script in the video embed field, so this is a big security problem indeed.

Of course, breaches like this are nothing new, with hackers doing their best to constantly keep us on our toes. Even though there isn’t any evidence that GNAA is accessing accounts to put these posts up, it’s probably still a good idea to change your password. Better safe than sorry, after all. Keep it tuned here at SlashGear and we’ll keep you posted any new developments with this breach.


Tumblr blogs hacked to display racist, offensive posts is written by Eric Abent & originally posted on SlashGear.
© 2005 – 2012, SlashGear. All right reserved.

Instagram exploit could see accounts stolen

A security flaw in popular photo filter and sharing app Instagram could allow hackers to steal user accounts, after it was discovered that certain cookies are sent in plain-text rather than secured. The exploit, discovered by Carlos Reventlov and apparently flagged to the Instagram team in early November, relies on the fact that while log-in and profile editing is done over encrypted links, Computerworld reports, a plain-text cookie is sent to the Instagram servers when the app is loaded. If the hacker intercepts that cookie – among other criteria – they can subsequently gain access to the account and lock out the legitimate user.

“Once the attacker gets the cookie” Reventlov says, “he is able to craft special HTTP requests for getting data and deleting photos.” The flaw was identified and repeated using the latest version of the Instagram app for iOS, v3.1.2, on the iPhone 4, and relies on the fact that the company does not use HTTPS for API requests, Reventlov points out.

For the exploit to work, however, both the hacker and the legitimate Instagram user must be connected on the same LAN. That’s obviously not going to be the case if the Instagramer is using their mobile data connection, but if they’re on the same WiFi hotspot then it leaves them potentially susceptible.

If compromised, the hacked account can give up not only whatever user details have been stored, but access to the photo streams of any friends on the service. It’s also possible to change the password and lock the proper owner out, as well as delete photos that the user has taken.

According to Reventlov, the Instagram team is yet to respond to his comments on the insecurity. It’s unclear if the Android version of the app is susceptible to the same exploit.


Instagram exploit could see accounts stolen is written by Chris Davies & originally posted on SlashGear.
© 2005 – 2012, SlashGear. All right reserved.

Hotel room break-ins linked to key card lock exploit

I’d wager just about everyone has stayed in a hotel that uses the key card locks where you slide a little card into the door to release the locking mechanism. One hotel called the Hyatt in the Houston Galleria district that uses these locks had a rash of break-ins in September. The most puzzling aspect of the break-ins was that the locks weren’t picked, the doors weren’t forced open, and a key wasn’t used to unlock the doors.

During the investigation, hotel staff learned that an enterprising thief was using technology to hack the key car door locks, tricking the doors into opening with ease. The hotel staff came to this conclusion when a device capable of reading the memory on the key card lock determined that none of the hotel staff keys had accessed the lock on one of the violated rooms. In fact, the device showed no key had been used during the theft.

The thief who was behind the string of hotel room break-ins in Houston was believed to be taking advantage of a previously theoretical intrusion technique that security researchers have warned about months before. Even more troubling is that hordes of hotel rooms around the world are still vulnerable to this exact same attack. The vulnerable locks come from a company called Onity and are estimated to be in use on 4 million hotel rooms around the world.

The flaw that investigators believed to have been exploited was demonstrated by a software developer for Mozilla at the Black Hat hacker conference last July named Cody Brocious. Brocious showed that he was able to trick the lock into opening using a portable programmer device designed to be used for designating master keys and opening locks when the batteries had died. The device Brocious used was small and cost less than $50 to build. Also worth noting is that lock maker Onity isn’t paying to fix the flaw in the 4 million locks estimated to be in use, and is asking hotels to pay for the fix. That could mean many locks remain vulnerable as hotels refuse to pay.

[via Forbes]


Hotel room break-ins linked to key card lock exploit is written by Shane McGlaun & originally posted on SlashGear.
© 2005 – 2012, SlashGear. All right reserved.


US denies hacking French government

The US government has “categorically refuted” accusations that it hacked computers used by the French presidency, after a magazine claimed a US team used malware and social engineering to access confidential data. The hack, which reportedly took place during the May elections, saw a staff member of then-president Nicolas Sarkozy fooled into clicking a fake Elysée presidential office webpage, L’Express claims. That supposedly allowed the US team to harvest that employee’s login details.

With those credentials, it’s suggested, the real Elysée site was infected with a specially created worm. That – apparently similar to the Flame malware that has made headlines in recent months, and which is believed to be the handiwork of the US and Israeli governments – was used to extract “sensitive information.”

Sarkozy is said to have escaped any hacking, solely because the former president himself did not have a PC. Instead, data was taken from other high-ranking officials, including the French secretary general and chief of staff.

A team of French investigators painstakingly traced back through the malware’s route, and concluded that it was the handiwork of the US. Official lips are sealed as to the full results of that investigation, but an anonymous security forces insider suggests that the US may simply have been wanting to ensure that France was still a solid ally during a time of political upheaval.

“You can be on good terms with a “friendly country” and will, at the same time ensure its continued support, especially in a period of political transition” the insider said.

Unsurprisingly, the US has moved quickly to deny the allegations, The Verge reports, with the local embassy issuing the following statement:

“We categorically refute allegations from unidentified sources, published in L’Express, that the United States government has participated in a cyberattack against France. France is one of our best allies. Our cooperation is remarkable in the areas of intelligence, law enforcement, and cyberdefense. It has never been as good and essential to our common fight against the threat of extremism” US Embassy in Paris


US denies hacking French government is written by Chris Davies & originally posted on SlashGear.
© 2005 – 2012, SlashGear. All right reserved.