Once your car gets about 10 years old, one of the most annoying things is that headlights dim and yellow. This is due to a number of reasons – but primarily due to the degradation of the UV coating. You can buff it off, but it often quickly returns and you’re stuck with an annoying chore almost every year.
Another option is to buy replacement headlights. In the old days, you simply unscrewed the old bulbs and put in the new ones. Now you need to remove the assemblies – which often involves removing the bumper and surrounding shrouds – as is the case with mid 2010 Subarus.
The Crosstrek/Impreza’s in the 2015 era were actually not that bad to replace. TRQ does a great job showing you how to do the job yourself – including how to re-aim the headlights. It’s a great video.
Engineers at Intel released an open-source tool that tries to quantify the issues from increasing amounts of upscalers, frame generators, and AI rendering techniques. Ironically, the tool itself is an AI trained on large datasets. Their paper about the methodology is located here.
CGVQM is a video quality metric that predicts perceptual differences between pairs of videos. Like PSNR and SSIM, it compares a ground-truth reference to a distorted version (e.g. blurry, noisy, aliased).
What sets CGVQM apart is that it is the first metric calibrated for distortions from advanced rendering techniques, accounting for both spatial and temporal artifacts.
I got my start in programming with type-in BASIC programs. Back in the 80’s, almost every computer had BASIC built-in, but almost no kid could afford games. Or even get them – the nearest store that sold software from me was over 30 miles away. Mail order took 2-3 weeks. On top of that – kids are notoriously broke. What I did have was a library, and plenty of time.
Enter Compute! magazine. After ravenously devouring all the programming books our small Carnegie library had, I branched into magazines. BYTE was too news oriented and didn’t have type-in programs; though reading about the technology was fun. When I found Compute! – I was hooked. I eventually checked out just about every single magazine they had a dozen times over. I remember digging in the downstairs old issue stacks in search of any I might not have seen. I spent whole weekend afternoons typing the programs in – and then even more hours debugging each line to figure out where I’d gone wrong.
Nate Anderson recently wrote an article about those early days of type-in programs. Even more fun is the comments section full of people sharing their similar experiences.
With the internet and instantly available content and content development tools – it makes me wonder how the next generation’s engineers will develop. How will the instantly available world of free software and tools shape them compared to our generation of type-in programmers?
Thankfully, all these wonderful magazine scans have been saved in the Compute! Magazine Archive on the Internet Archive. I even sat down and typed one in (well – heavily utilized OCR as well!). What a blast.
When is quantum cracking going to happen? Much sooner than bitcoin owners would like
Q-Day is the day when classical computational cryptography we use today is slated to be obsolete, because quantum computers will finally be powerful enough to crack them. It is likely to have a similar effect as the Y2K crisis in that many digital security systems are not using quantum cryptography safe algorithms. The good news is that people can, and are, starting to fix things now. The bad news is that when it happens, the effects will be very immediate and catastrophic. Even that, however, is only half the story. We knew exactly when Y2k would happen (January 1, 2000 at 12:01am), but we don’t know when Q-day will hits us – until it’s already happening.
Konstanitinos Karagiannis provides one of the best, fullest discussion of the upcoming crisis. He gives a much clearer idea of when quantum computers will be able to break just about all existing cryptography – including all the encryption underlying Bitcoin and other online digital currencies. And it’s much sooner than people were thinking even 2 or 3 years ago. Like fusion power, it was always thought Q-day was 10-20 years away. It’s certainly what bitcoin promoters will tell you.
The summary?
The NIST says that all systems should have switched to quantum computing safe security algorithms by 2035 – but Konstanitinos says it’s MUCH more likely that we’ll see real quantum cracking happen sometime at early as 2027 based on the recent rapid developments in quantum computing and algorithm improvements. He points out its likely to start from government backed security agencies or very powerful, well funded organized crime groups.
What does this mean? It means any companies not updated to quantum secure cryptography will have computing systems almost completely vulnerable to having financial accounts emptied, customer data stolen, system take-overs or destruction, and ransom attacks. Secure emails and chat communications will be perfectly readable and usable for blackmail or extortion. Secure government and military communications will become vulnerable to infiltration. Infrastructure systems from airline traffic control, public transit, water systems, government computing services, to power systems become vulnerable to ransom attacks, havoc, and destruction.
It also means bitcoin and all digital currencies based on elliptical encryption/similar algorithms are very likely to drop from their current values to zero within hours after the first confirmed cracks happen. Clever attackers will likely crack a large number of digital wallets quietly over weeks and months by simply capturing the encrypted transaction data, and then flash-liquidate as many wallets as they can before the scheme is discovered and values go to zero. It’ll likely happen in less than a day. North Korea, even without quantum computing, already is doing this to the tune of billions per year.
You’re not even safe now. It’s also highly likely governments are using record-now-crack-later strategy of recording secret communications and bank transactions now so they can uncrack them later when quantum computing is cheap and easy. It’s very likely we’ll see it used for extortion in just a few years when everyone’s communications, web traffic, and bank transactions become public knowledge. If you thought Wikileaks revealed a lot of stuff, wait until governments and organized crime groups unencrypt years worth of recorded traffic.
He also covers the good points. There are cryptographic algorithms that are secure from quantum attack – which you should be using today. He also outlines how we will detect if people are using quantum computers to crack things by describing the current cracking algorithms and their telltale signatures.
Still – quantum cryptographic cracking is likely to be like lightening from the blue. Everything will be fine until it’s discovered to be happening. It’s very possible that literally trillions of dollars could be stolen in the matter of hours or days.
Fluxfox is a floppy disk image library – written in Rust. It’s intended to serve the needs of the emulator world and supports IBM, Amiga, Macintosh, and Atari ST formats. It can even perform operations on disk images consistent with typical operations of a PC floppy disk controller, while also giving low-level access to the track bitstream for other controllers.
Back in the day, software didn’t come on encrypted, online, distributed marketplaces, they came on humble floppy disks. This made them susceptible to copying. To fight this, developers started using all kinds of interesting tricks, which hackers would try to break. Thus started a nearly decade-long war of hackers and copy protection schemes.
I particularly like his article about Copy-Lock mechanism used by Kings Quest. Copy-Lock employed several tricks such as sectors with non-standard sizes and putting purposefully incorrect CRC values on tracks to make standard copying incorrect.
In this case, Copy-Lock used a mechanism in which sector 1 on track 6 was intentionally written as only 256 bytes (instead of 512 bytes), with a 256-byte blank section to fill the gap. Additionally, the CRC was also altered to make a normal read think it was invalid. A normal INT 13h disk read would search and fail the read and CRC check.
CopyLock worked by bypassing the BIOS and talking directly to the disk controller. It would issue an INT 13h read on sector 1 track 6 that it knew would fail. This would place the head on the right track. The code would then tell the floppy controller directly to read track – and dump all 512 bytes. It was looking for the special byte 0xF7 as the final byte of that supposedly empty section of the track. The key is that it is not possible to create invalid tracks with invalid CRC’s like this using a standard IBM PC floppy controller. Copy-Lock created the special hardware that could write in this way and sold that, along with the checking code, as their solution.
His article has all the assembly code – which is really awesome.
MangoHud is a nice little Linux package that provides a Vulkan and OpenGL overlay for monitoring FPS, temperatures, CPU/GPU load, and lots of other parameters. Source is provided on github.
Grant Smith got one of the USPS delivery scam text messages. He decided to track the scammers and uncovered a Chinese-language group behind the campaign. He hacked their systems, discovered their mechanisms, and gathered victim data. He handed it to USPS, bank, and FBI investigators – as well published information about their operations online and at Defcon.
He discovered the group sold their scamming kits to set up their own operations for a $200/mo subscription. Similar scams showed up in half dozen other countries.
What’s interesting is he reported how many people fell for it. The triad sent 50,000-100,000 text messages a day. In total, US victims for just this one (albeit very large) operation entered 438,669 credit card numbers. Many people entered multiple cards.
New Wave makes some amazing, fully functioning miniature retro arcade models – or replicades. One of their more recent additions was the ever-fabulous arcade fortune telling machineZoltar.
