The Pontiac GTO, like the Chevrolet Chevelle, was a classic American muscle car that debuted in 1964. Initially the GTO was merely a performance package for Pontiac’s Tempest, but became an actual separate model in 1966.
The GTO’s design was tweaked every single year of the decade it was in existence. To see how it’s evolved, we’ll rely on the excellent and consistent photography of RK Motors Charlotte, which also records 360 turntable video of the models they restore:
In the wake of the recent spate of church shootings, especially last weekend’s shooting in Sutherland Springs, TX, are you carrying in church more than you were previously? Has this changed your church’s approach to security, at all?
Have you started carrying in church, more?
I personally have been carrying in church ever since I got my CHL, and I know that where I attend we have quite a few concealed carriers in an informal network. We are starting to formalize all of this quite a bit more in the wake of last weekend’s shooting, though.
If you’re a member of a church that has a security team, sound off in the comments and let us know what’s working for you and what isn’t.
And if you’re getting more serious about your Sunday carry, let us hear from you what you’re doing now that you weren’t previously.
At some point, we’ve all been contacted by a Nigerian prince, long-lost uncle, or some guy that just can’t manage to get a bank account with the promise that if we just make one small wire transfer we’ll have millions in our account by morning.
Admittedly, I’ve responded to a few for fun just to see where the conversation can go. As The Verge reports, now one security firm has created a chatbot that allows you to respond to all those annoying scam messages and waste their time like they did yours by sending the email in the first place.
Video: ReScam
To use it, you just have to forward the first offending message to me@rescam.org. When you do, a proxy email address will start replying to the scammer’s emails for you. And the responses look pretty real.
Advertisement
Rescam has a few listed on their site. For instance, this was the chatbot’s first response to a request for a wire transfer.
It’s easy to use, and says just enough to keep the conversation going without letting on that you’re actually using a bot to do the talking.
Designed for camping in extreme weather conditions including cold, high winds, and rain, this heated mat connects to a portable external furnace, which can run for up to 24 hours using butane or propane gas. Weighing 6.4 lb total, it’s designed to pack up small for easy carrying.
If the Maxcraft is unavailable, we also like the Husky 8-in-1 Precision Slotted and Phillips Screwdriver Set. Ergonomically, we liked it more than our main pick, but the design of the storage system increases the chances of losing a bit, which is why we gave the edge to the Maxcraft.
The handle of the Husky is lightly padded, and the curved design makes it a nice tool to hold and turn. We also like that the nose of the driver is long and narrow, so in tight spots it will be a bit easier to maneuver than the Maxcraft.
The Husky comes with four double-sided bits: Phillips #000, #00, #0, and #1, and slotted 1/16, 5/64, 3/32, and ⅛. Of those, the Maxcraft doesn’t have the Phillips #1. But because that size is commonly found on regular multi-bit screwdrivers, it isn’t an essential part of a precision set.
Where the Husky falters is in the bit storage system. The compartment in the tool body can hold only three of the four bits, meaning that you always need to keep one set in the nose of the tool. The bits click into the nose with a small spring-loaded ball bearing, but the connection is not very strong, so a bit could fall out while you’re retrieving the tool from a pocket or tossing it into a toolbox. And because the bits are double-sided, they’ll be harder to replace if you ever lose one.
The Husky also lacks a pocket clip, something we liked about the Maxcraft.
Although the Husky has a spinning rear cap, a nice feature, it doesn’t rotate smoothly at all. While you’re using the tool, it’s effectively useless. The Maxcraft doesn’t have a spinning cap, though, so in comparing the two models, the ineffectiveness of the Husky’s cap doesn’t really play a role.
The Husky and the Maxcraft are similarly priced, and each presents some trade-offs. We prefer the Maxcraft due to its more secure storage system, but the Husky is a nicer tool to handle. The Husky is also widely available at Home Depot, so it may be more convenient to purchase as well.
Flying across America provides a bird’s eye view of how we humans have decided to cut the land up. In regions where we’ve completely mastered the topography, you’ll see completely unnatural-looking, perfect grids like this, punctuated with perfect circles:
So what the heck are those, and how/why are they created? Here’s the 30-second answer:
That’s a snippet from this five-minute-plus video on why certain parts of America look very different from the air, depending on which colonial master first took control of which region. It’s worth a watch if you’ve got the time.
Dash cams range from tiny units without screens that hide behind the rearview mirror to models with relatively large displays and a multitude of features. Here are the key features and specs to consider when you’re shopping.
Resolution and image sensor: The image quality of dash cams keeps improving. When we first published this guide in 2014, the best models had a maximum resolution of 1080p, with many models reaching only 720p. This time around, all of the models we tested could record at 1080p, with several able to capture a notably sharper 1296p and one capable of 1440p. In the higher-resolution models, we could see a clear difference in detail, and we were better able to read license plates and street signs, even when in motion. (You can set a cam’s resolution lower to take up less space on the storage card if you prefer.) The best cams also have image sensors and processors that provide a wide dynamic range that retains detail in both lighter and darker areas, as well as good color and contrast and low video noise across a variety of lighting and weather conditions.
Field of view (FOV): The models we tested have a field of view that ranges from a relatively narrow 120 degrees to a wide 180 degrees. The wider the field of view, the more of the scene that a camera can capture, which is handy for scenic vistas, sunsets, and the like. The trade-off is that wider FOVs make cars in front look a little farther away, although in our tests the sharp resolution of the best cams still made it easy to read license plates and see other details. The 160-degree FOV in our top pick is a good compromise.
Mounting: Most cameras mount to a car’s windshield in one of two ways: suction or adhesive. Some models give you a choice, while others come with only one or the other. Adhesive mounts often attach directly to the camera, which reduces the unit’s overall size. This design typically permits more-discreet placement, so the camera doesn’t intrude as much into your view, and the camera isn’t as noticeable from the outside. The ones we tested adhered very well to glass. But adhesive mounts can be difficult to remove, sometimes requiring a razor blade or a sharp paint scraper along with some patience, so you can’t easily move the camera from one car to another. Adhesive mounts also require precise placement when you install them, since you can’t adjust the placement without a lot of hassle or buying new adhesive material.
We favor a suction mount, even though the cam hangs down lower from one, because it has two key advantages: You can aim and adjust it easily, and you can remove it quickly if you want to take it down for any reason. (For instance, perhaps you have to park overnight on a street or in a lot, and you’re concerned about theft. Or you want to be able to move the camera between cars, such as for use in a rental car.)
Whatever the mount type, you want the camera to have good build quality and to remain firmly in place once positioned, so that it’s not easily jostled if you or your passengers accidentally nudge it or if your car hits a rough spot in the road.
Regardless of the mount, all models come with a long power cord that you can route to a power source. To keep it out of sight, you can often hide this cord behind the windshield molding and interior trim, but doing so can be the most challenging part of installing a dash cam. Alternatively, you can buy cord clips for a few bucks to route it out of the way, or Best Buy’s Geek Squad will install a dash cam for about $50.
Size: A dash cam and its mount should be somewhat discreet so they don’t take up too much real estate on your windshield and don’t distract you while you’re driving. The best units we looked at provide at least a small display while not blocking too much of the driver’s view.
Display: Most of the dash cams we tested have a display that’s 1.5 to 2.7 inches, measured diagonally; one model doesn’t have a display at all. We prefer dash cams with at least a 2-inch display, which makes it easier to adjust settings on the fly and preview footage. A lack of a display means that you have to use a computer, before getting into the car, to adjust the settings. The absence of a display also makes it hard to know if the camera angle is right without reviewing the footage on a separate device. And it’s impossible to know if the camera is really recording.
Second cam: Some dash cams—called dual-channel models—are able to record from two cameras simultaneously, which is helpful for recording both what’s in front of the car as well as what’s behind it or inside the vehicle. Models that record behind the car have two separate camera enclosures, one for mounting on the windshield and one for the rear window. The rear camera can also serve as a backup camera; you can display the footage from either camera on the screen or use the picture-in-picture function.
Some models have two cameras integrated into one housing, one for recording in front and the other pointing rearward toward the car’s cabin and passengers. This design can be helpful for taxi, Uber, Lyft, and other ride-for-hire drivers who have to deal with the occasional unruly customer, although the footage can be too dark to clearly see rear-seat passengers at night.
Storage cards: About half of the dash cams we tested come with an 8 GB microSD memory card, although some include a more generous 32 GB. Some models don’t include any card, which increases the cam’s effective price. Most of the cams accept a larger card of 64 or 128 GB, which is an important consideration.
A larger memory card lets you capture more footage before the cam needs to overwrite older video. At its maximum 1296p resolution, the Papago GoSafe 535 (our top pick) captures less than an hour of video on its 8 GB card (or a little over three hours when you set it at the lower, 720p resolution). That might work if all you want to do is record an accident or an occasional interesting scene, or if you expect to offload the footage frequently. But if you don’t want your video overwritten so quickly, we recommend getting a larger card; a quality 64 GB microSD card will set you back about $20 currently.
Battery versus capacitor: All of the models we tested get power when you plug their 12 V adapter into your car’s 12-volt accessory outlet (the “cigarette lighter”), although some can be hardwired if you desire. Most also have an internal lithium-ion battery that lets them continue to record for a period of time if the cord is unplugged. But in extreme temperature conditions, where objects left baking in a car can reach 180 to 200 °F, these batteries can overheat and need to be cooled to perform well. They can also suffer in very low temperatures, when they’re too cold to function right away. To help in these types of situations, some models use a capacitor, which better handles extreme temperatures and provides more reliability in those conditions than a battery. The downside is that, without a battery, a capacitor cam can’t continue working when it’s unplugged; the camera will power down after a few seconds.
Automatic incident detection: All of the dash cams we tested have built-in G-sensors that detect an impact and automatically save the footage of the incident. In most cases, the cam protects this footage from being overwritten, as well, so you have the evidence should you need it.
Parking surveillance: Models with this feature use optical motion detection or G-sensors to “wake up” and start recording if something happens to or around your vehicle while it’s parked. But the odds of the camera picking up useful information are low, since the camera faces a single direction and also requires persistent power to do this. We tested the parking mode in the models that offered it, but we were unable to trigger any of the cameras. Car owners shouldn’t consider this feature a reliable security tool.
Wi-Fi: You can view footage directly on your smartphone via an ad-hoc Wi-Fi network the camera provides. This feature is convenient if you want to share video or photos on social media, but it’s more of a frill than a necessity. Dash cams with Wi-Fi typically have a companion smartphone app, where you can look at the footage.
GPS: Models with a GPS receiver typically stamp the video with geographic coordinates, as well as the vehicle’s speed and direction of travel. Some also show the vehicle’s location on a map in the model’s compatible computer software. The GPS feature doesn’t provide turn-by-turn directions, however (for that you need a dedicated car GPS navigator). This info can be useful to verify the location of an incident or to mark a site where you captured a specific view or happening. But if you’re interested only in recording what’s in front of your car, GPS is just an extra.
Safety features: Some dash cams (and car GPS devices) provide safety alerts similar to the type that are becoming more common in newer cars. These include forward-collision warnings, which sound an audible alert if you are following another car too closely, and lane-departure warnings, which let you know if your car is crossing a lane marker on the road. Both can be helpful in some situations, but as you might expect, these features on a dash cam aren’t as reliable, consistent, or sophisticated as the more expensive safety systems in new cars. In some dash cam models, the alerts are also overly sensitive, which causes them to go off too often and can make them annoying enough that you want to turn them off. Again, this feature is not really a necessity for most people.
The LG 27UD68P-B is the best 4K monitor for most people because of its great color and grayscale measurements, thin bezels, three-way adjustable stand, and easy-to-use menus. It also has FreeSync adaptive sync, which makes it even better for gaming if you have an AMD video card. The LG doesn’t have USB 3.0 hub and its grayscale tracking isn’t as good as our runner-up pick’s, although its color accuracy is much better. At a price comparable with monitors with lower resolutions, this is a great display for anyone ready to make the jump to 4K.
The 27UD68P-B is factory calibrated, and out of the box images are color-accurate and detailed. In the color saturation test, the 27UD68P-B had an overall DeltaE value of 1.94. Above 80 percent brightness, though, the individual DeltaE values for red, green, and yellow creep up toward 3.0, and yellow hits that number at 100 percent color saturation. That means you’ll see slightly inaccurate yellows in more vivid images—a picture of a field of sunflowers will look a bit unnatural. The ColorChecker test—which measures the accuracy of over 100 colors representing skin tones, the sky, and other natural elements—had a DeltaE of 2.35, and the number for primary and secondary color points was 2.25. Grayscale tracking average was 2.31, with a slightly higher number at increased brightness levels—it’s very good below 50 percent brightness.
Contrast ratio is the measurement of a fully lit white screen versus an unlit black screen at a fixed brightness level. The contrast ratio of the 27UD68P-B, at 1186:1, was the highest of all the monitors we tested. This is an excellent number for a monitor, especially an IPS panel, and it means the bright parts of the screen will look vibrant and really pop out next to blacks (think images of fireworks in a night sky).
The LG 27UD68P-B has a wide luminance range. At its highest brightness setting, a full white field on the monitor outputs 287 cd/m² of light—slightly under LG’s claimed 300 cd/m², but still enough for a sunlit room. At its lowest, the monitor light output drops to 59 cd/m², which will limit the fatigue on your eyes while playing Overwatch in a dark room.
The 27UD68P-B is VESA-compatible and has a stand that can adjust height, tilt, and pivot. The stand has 4.7 inches of height adjustment, up to a maximum height of 21.13 inches from the top of the screen to the desktop. The monitor can tilt forward 5 degrees or back 20 degrees and can pivot clockwise into a portrait position. Unlike the stands of our Dell or Benq picks, the stand cannot swivel.
Like recent monitors from Dell and HP, the 27UD68P-B has very thin side and top bezels, at just over a quarter of an inch thick on the top and sides (8 mm), though the .9-inch (23 mm) bottom bezel is significantly larger. Unlike most monitors, which are controlled with a line of buttons along the lower right of the frame, the 27UD68P-B has a four-way joystick situated underneath the bottom bezel at the center. It took a short while to get used to, but once we did we loved the ease of use. Menu navigation is quick, and being able to keep your finger in one place to move through every menu option is convenient. It’s far better than the capacitive buttons many monitor makers use.
The LG has one DisplayPort 1.2 port and two HDMI 2.0 inputs for connecting computers or other A/V sources, like a PlayStation 4 Pro or a 4K-capable Roku or Apple TV box, but it’s missing the USB 3.0 hub and the Mini DisplayPort found on both the Dell U2718Q and BenQ PD3200U.The only additional port is a headphone-out jack for audio streams sent over DisplayPort or HDMI. The volume level can be controlled by moving the joystick toggle left and right. All of the connections are on the back of the monitor, so if you need to regularly plug and unplug cables it could be tedious.
There are a few options on the 27UD68P-B that cater to gamers. The most prominent is FreeSync, which reduces screen tearing by locking the refresh rate of the monitor to the frame rate of the video card. It’s supported only over DisplayPort, not HDMI, and must be connected to either a discrete graphics card or an integrated GPU that supports it; right now, the only CPUs and GPUs that support the technology come from AMD. (Nvidia has its own adaptive-sync technology, called G-Sync, which requires both an Nvidia graphics card and a monitor with a special chip in it.)
Another gaming-oriented menu option is called Black Stabilizer. It boosts the brightness in darker gray areas of the screen, which could help keep you alive during those long online gaming sessions, especially if the room you’re playing in has ambient light that makes it difficult to distinguish the shadows on screen. We don’t recommend using the feature for anything other than gaming, though, as it sacrifices black level accuracy for an edge in gaming combat.
The LG 27UD68P-B has a one-year parts and labor warranty. LG will either repair or replace the unit, at its discretion, but it doesn’t have a bright/dead pixel policy (see “Flaws but not dealbreakers,” below). 144HzMonitors.com named it the Best 4K IPS FreeSync Monitor, MonitorNerds named it the Best 4K Gaming Monitor, and 4K.com found it to have “superb performance quality almost entirely across the board.” RTings.com gave it a good score for mixed usage—this monitor is good at just about everything.
LG sells two monitors nearly identical to the 27UD68P-B. The 27UD68-W is the same monitor with a white case and silver-colored stand, and the 27UD68-P has a stand without height or pivot adjustments. It does have VESA mounts, though, and it’s often a little cheaper than the 27UD68P-B, so it’s a good choice if you already have a monitor arm you like.
The gun gurus at USA Today attempted to explain, for some reason, not only the features of the Ruger AR-15 used by the demented coward responsible for the Texas church massacre, but the “possible modifications” one can make to an AR as well. Including, naturally, the ever-popular chainsaw bayonet.
Wait. What? Yup. This actually happened. See USA Today’s video below, followed by a small sampling of the internet’s hilarious, meme-laden response…
To be fair, chainsaw bayonets do actually exist. I mean, they’re an $800 gag gift movie prop but, much to everyone’s surprise, somebody did really design one and does [attempt to] sell them.
In this blog post, I’ll look at MySQL and Linux context switches and what is the normal number per second for a database environment.
You might have heard many times about the importance of looking at the number of context switches to indicate if MySQL is suffering from the internal contention issues. I often get the question of what is a “normal” or “acceptable” number, and at what point should you worry about the number of context switches per second?
First, let’s talk about what context switches are in Linux. This StackOverflow Thread provides a good discussion, with a lot of details, but basically it works like this:
The process (or thread in MySQL’s case) is running its computations. Sooner or later, it has to do some blocking operation: disk IO, network IO, block waiting on a mutex or yield. The execution switches to the other process, and this is called voluntary context switch.On the other hand, the process/thread may need to be preempted by the scheduler because it used an allotted amount of CPU time (and now other tasks need to run) or because it is required to run high priority task. This is called involuntary context switches. When all the process in the system are added together and totaled, this is the system-wide number of context switches reported (using, for example, vmstat):
root@nuc2:~# vmstat 10
procs -----------memory---------- ---swap-- -----io---- -system-- ------cpu-----
r b swpd free buff cache si so bi bo in cs us sy id wa st
17 0 0 12935036 326152 2387388 0 0 0 5 0 1 9 0 91 0 0
20 0 0 12933936 326152 2387384 0 0 0 3 32228 124791 77 22 1 0 0
17 0 0 12933348 326152 2387364 0 0 0 11 33212 124575 78 22 1 0 0
16 0 0 12933380 326152 2387364 0 0 0 78 32470 126100 78 22 1 0 0
This is a global number. In many cases, however, it is better to look at it as context switches per CPU logical core. This is because cores execute tasks independently. As such, they have mostly independent causes for context switches. If you have a large number of cores, there can be quite a difference:
The number of context switches per second on this system looks high (at more than 1,000,000). Considering it has 56 logical cores, however, it is only about 30,000 per second per logical core (which is not too bad).
So how do we judge if the number of context switches is too high in your system? One answer is that it is too high if you’re wasting too much CPU on context switches. This brings up the question: how many context switches can the system handle if it is only doing context switches?
It is easy to find this out!
Sysbench has a “threads” test designed specifically to measure this. For example:
sysbench --thread-locks=128 --time=7200 --threads=1024 threads run
Check the vmstat output or the Context Switches PMM graph:
We can see this system can handle up to 35 million context switches per second in total (or some 500K per logical CPU core on average).
I don’t recommend using more than 10% of CPU resources on context switching, so I would try to keep the number of the context switches at no more than 50K per logical CPU core.
Now let’s think about context switches from the other side: how many context switches do we expect to have at the very minimum for given load? Even if all the stars align and your query to MySQL doesn’t need any disk IO or context switches due to waiting for mutexes, you should expect at least two context switches: one to the client thread which processes the query and one for the query response sent to the client.
Using this logic, if we have 100,000 queries/sec we should expect 200,000 context switches at the very minimum.
In the real world, though, I would not worry about contention being a big issue if you have less than ten context switches per query.
It is worth noting that in MySQL not every contention results in a context switch. InnoDB implements its own mutexes and RW-locks, which often try to “spin” to wait for a resource to become available. This wastes CPU time directly rather than doing a context switch.
Summary:
Look at the number of context switches per logical core rather than the total for easier-to-compare numbers
Find out how many context switches your system can handle per second, and don’t get too concerned if your context switches are no more than 10% of that number
Think about the number of context switches per query: the minimum possible is two, and values less than 10 make contention an unlikely issue
Not every MySQL contention results in a high number of context switches