I've used Synergy too. It's an extraordinarily useful program, but it's NOT a way to extend computer #1's desktop onto computer #2 via RDP... it's a way to have computers 1 and 2 running independently side by side, but use a single mouse and keyboard with both (treating them both AS IF they were actually one computer with its desktop extended onto a second monitor).
I wish Microsoft would give us the ability to treat connected RDP clients as additional monitors that the current desktop can be extended onto. For example:
1. Boot into Windows with laptop.
2. Boot into Windows with large-screen dual-OS tablet like a Chuwi Hi12
3. Launch RemoteDesktop on the tablet, and set up a new connection... but check the new, extra checkbox that says, "use this as an external monitor and extend the Windows desktop onto it"
4. Connect Remote Desktop on the tablet to Windows on the Laptop. Voila, instant second monitor.
Sadly, AFAIK, there's STILL no good way to achieve this. I know there are a couple of third-party Android apps that try to accomplish this (by running a host app under Windows and using VNC at the Android end)... and they all suck miserably. RDP is unique, because it hooks directly into the Windows rendering pipeline and has extraordinarily high performance. I'm pretty sure that (in theory, at least) you can even run DirectX at nearly full performance over RDP (as long as the GPU at the client end is powerful, since IT'S the one that ultimately gets used). I know that in the past (when I still had a normal desktop PC), I used to routinely run programs on my desktop PC, then use them over the LAN via RDP on a lower-powered laptop elsewhere in the house because they ran faster than if I tried running them directly on the lower-powered laptop itself.
I also wish that Linux had something with performance remotely close to that of Windows via RDP. VNC is dog-ass slow, and remote X11 over a network is (surprisingly) even slower. I'd originally had high hopes for Wayland, until I read that approximately four years ago, its architects declared point-blank that high-performance remote rendering over a network wasn't even on the table or open for negotiation as a design goal. Period, full-stop, end of story.
College students who want a minimally-compromised gaming system they can easily lug to their parents house for weekend visits and holidays?
Kids in high school who want a computer they can easily haul over to a friend's house?
People whose jobs require "travel" in the sense of "living in some city besides the one where you officially live, for days, weeks or months at a time... but not LITERALLY living out of a suitcase and running through airport terminals every day"?
People whose jobs involve "real" content-creation work that a mere laptop is inadequate for, but still want to have some degree of luggability so they can do demos outside of their office without the risk of having it not work properly on a lesser computer (or the need to waste hours ensuring that the demo DOES work properly on a lesser, more portable computer, making sure they give themselves enough time to fix whatever doesn't work if it DOESN'T work properly on the other computer)?
From what I understand, panel availability is the main problem. I believe that as of this moment, there's exactly one company that manufactures 17.3" panels with higher resolution than 1920x1080. The panels are 2560x1440 (not 3840x2160), are limited to 60hz, and apparently are known for having poor contrast & saturation.
I think part of the problem is also that nobody wants to go to the trouble of making a display capable of 2160p120, but making one capable of 1080p120 and 2160p30 without requiring yet another major overhaul of HDMI 2.x would require having some kind of agreed-upon standard for frame-packing (whereby you send 3840x2160@30fps as four 1920x1080 frames in a row at 120fps, and display 1080p120 by doubling each pixel to 2x2) that didn't exist back when that currently-available 17.3" panel was designed and went to manufacturing. I think the mess between FreeSync and Gsync compounded it, since historically Nvidia wouldn't allow licensees to support both standards, and refused to support FreeSync with their own cards in the hope that it could make Gsync the One Official Standard everyone else was forced to license from them. Now that VESA has apparently given Nvidia "the finger" by making FreeSync the officially-blessed standard for TVs, and Nvidia itself is apparently going to start supporting Freesync output, that might change in the near future.
Another issue with 2560x1440 panels: they suck for displaying natively 1080p content because every pixel HAS to be partially-scaled. With 3840x2160, 1080p can neatly scale 2:1 without losing sharpness (unless you WANT to do smoothing), and 720p has enough oversampling when scaled up to 3840x2160 to make Nyquist happy and avoid the really bad higher-order artifacts seen when scaling 1920x1080 up to 2560x1440.
Personally, I wish someone would just take the next step up and make a laptop that uses a 22" 3840x2160 display. AFAIK, "22" panels that support Gsync and 120+Hz framerates already exist... it would be too big to USE on an airplane, but then again, I'm pretty sure even 17.3" is too big to fit on a tray table without encroaching into the space of whomever is sitting next to you... and I think 22" panels are *just* small enough to fit into an enclosure that fits within the airline-defined maximum dimensions for carry-on baggage. It obviously wouldn't be a "laptop", so much as an "easily-luggable mobile workstation".
God forbid, if they went THAT big, they could probably even design its enclosure to BE the size of an airline-approved carry-on suitcase & have it support ATX-type components. I remember seeing something like that ~15 years ago, but unfortunately the company that manufactured it refused to sell the cases + monitors as separate components, and only sold them as complete (and HORRIFYINGLY expensive) systems (something like $8,000 for the cheapest one). The market for "AirTX"-formfactor cases might be small, but as long as it supported normal-formfactor motherboards, cards, etc, it wouldn't really MATTER... people would just buy the case-monitor combo, add everything else themselves, and evolve the system in a Ship of Theseus manner over the next 10 years by keeping the case, monitor, snap-on keyboard, etc, and replacing the other components as they became obsolete or wore out.
My personal fantasy: AirTX with three monitors, arranged like shutters over a window... a ~22" 3840x2160 main landscape-orientation display, flanked by a pair of portrait-orientation displays that were somewhere between 1440x2160 and 1600x2160 (attached with hinges so that when closed, the screens of the flanking displays faced the main screen). Add a range of keyboard options that includes model-M buckling-spring, Cherry blue/green/clear/black/red/brown switches, and Thinkpad-like scissor switches (all with two pointer sticks... one between the GHB keys, and one below the spacebar for thumb use, configurable so that one or both can be used for mouse-control, with one optionally used as a fau
Honestly, for modern computers, a full-size mid-tower case is gross overkill, even for gamers. Nowadays, if a graphics card (or cards) generates lots of heat, it just sucks in air from the outside and blows it directly outside without involving the case's interior ambient airflow. Ditto for CPUs. In the past, case airflow in high-performance systems was mostly due to 10kRPM and 15kRPM hard drive arrays, which are now basically moot -- we use SSDs for things that used to use 10k/15k hard drives, and use slow drives for bulk storage that spend most of their time spun down anyway (usually, with their 50-200 gigabytes of most recently-used blocks cached to SSD for both performance and to let them remain spun-down).
Form-factor wise, we're still stuck with "shoe box" form factor and microATX size if you want to be able to use normal cards and external drives without worrying about half-height cards, riser cards, and laptop-formfactor optical drives, but there's a world of difference size-wise between a compact microATX case the size of two stacked shoe boxes and a traditional mid-tower case that was common 20 years ago.
IMHO, the biggest reason why we've never been able to evolve much smaller is because Intel is no longer sufficiently dominant to unilaterally dictate standards anymore, and getting a CONSORTIUM of companies with opposing interests to agree about ANYTHING has historically proven to be a nearly lost cause... at best, you end up with standards that either require implementing "the kitchen sink" to accommodate 5 original conflicting standards, or you end up with a standard nobody actually uses.
Back in the late 90s, Intel dictated several new form factors... only ATX caught on. We now have a few semi-informal conventions for CONNECTORS inside a laptop (like mPCIe, which is used by discrete laptop graphics cards and internal WLAN cards, and the various standards for internal SSD cards), but no OFFICIAL vendor-agnostic standards for form factor, bolt placement, heat-removal, physical access to the outside world, etc. Laptop optical drives seem to have settled down into one of two form factors, including the placement of USB3.0 and mPCIe/mSATA ports for them to plug into, but getting a true thirdparty drive to work in a given laptop still involves a bit of bezel-related surgery (even in cases where the drive is designed to be semi-easily removable and swappable).
The main problem with graphics cards is that there's no official standard governing the interface + physical form factor + thermal profile, so there's no official standard for card manufacturers to advertise compatibility with, nor an official standard for mobile workstation vendors to advertise compatibility with. At best, you're stuck buying parts online with your fingers crossed, knowing there's a good chance you'll end up having to sell it at a loss on eBay when it ultimately doesn't fit or work.
Thinkpads have the best keyboards & pointer sticks, but Lenovo never uses anything but Quadro graphics cards (hence, my other comment in this article re Dell happily using non-Quadros). For 97% of likely users, Quadro cards suck... mega-expensive, and crippled performance.
Then the solution is for Congress to make it illegal for content providers to force cable companies into all-or-nothing contracts as a natural application of antitrust law.
Truth be told, it wouldn't even affect ESPN or Disney's profits much. If ESPN went from being a $15 bundle of channels everyone had to pay for to a premium channel bundle that cost an extra $45/month, people who really care about televised sports will grudgingly pay it anyway. For Comcast, and even ESPN, it would be largely revenue-neutral, and merely represent a massive shifting of costs from "everyone" to "the subset of customers who really care about that particularly-expensive bundle of channels".
Damn it, someone finally came out with a laptop that has good specs, a G-sync 144hz display, and a RTX videocard that ISN'T a Quadro... then ruined it by using a keyboard that doesn't have a pointer stick.:-(
It doesn't have to. All cable channels are encrypted now and selectively unlocked box-by-box anyway. If the customer doesn't pay the "local channel" fee, don't show the local channels from the cable company. If the customer lies about having an antenna, they'll just have no local channels at all.
If companies like Comcast were smart, they'd do two things:
1. Allow customers to use their own antenna for OTA and get a $12-15/month discount. In other words, for each customer who puts up their own antenna, Comcast would save having to pay the local carriage fees for that customer, so it would be revenue-neutral for Comcast & would allow customers who are likely to bolt to rationalize staying around a little longer. This is EXACTLY what DirecTV used to do about 15 years ago... you connected your OTA antenna to the STB/DVR, and it seamlessly inserted the local channels into the lineup and treated them exactly as if they were from the satellite.
2. Allow customers who don't care about Disney and/or sports to trade one or both for channels like HBO, Showtime, and/or Starz.
We don't need literal "a-la-carte" pricing. Most of the channels on the mid-tier lineup only cost a few cents per month anyway. What we NEED is the ability to prune away the most expensive low-hanging fruit. ESPN, the RSNs, and Disney cost more per month per customer than HBO and Showtime. SlingTV has already proven that there's a market for people who'll happily sacrifice sports & Disney for other channels (Sling "Orange" allows one stream & includes ESPN and Disney, Sling "Blue" allows multiple streams, has more channels, and DOESN'T include ESPN or Disney).
> That, and if you're in the city, get a simple unidirectional antenna; if you're far from the city get one that you can point toward the city.
And $150-300 for an OTA DVR. And unless you're seriously masochistic, don't even THINK about buying a $22-50 Mediasonic "DVR" -- they're a pain to use, have a UI straight out of the early 2000s, and the one I bought literally died for no apparent reason 5 days after the warranty ended. On the other hand, if you have an old PC with Windows 7 and a HD-Homerun, Windows Media Center is STILL awesome as a DVR.
Seriously, losing your DVR is the absolute worst part of cord-cutting. Even when streaming services offer "cloud DVR", it totally sucks compared to the experience of having a real DVR that doesn't refuse to let you skip commercials, or take 10 seconds to recover when you fast-forward by 20 seconds.
Are "LS" the ones widely sold by Radio Shack back in the early 80s, or were those L/C/{other}-series? I just know that HC/HCT chips seem to be capable of surviving things that used to EASILY destroy chips ~35 years ago. I remember destroying chips all the time back then, but never seem to have that problem anymore despite occasionally connecting voltages backwards, hitting chips with unregulated 6v instead of carefully-regulated 5v, static electricity, sourcing/sinking too much current, etc. It seems like with HC/HCT-family chips, making a mistake causes malfunctioning or "erratic" behavior... older chips literally went up in smoke if you did stuff like that.
Do yourself a favor... the next time you have the house to yourself for a few days, take the TV off the wall or cabinet, and move it to a spot that's 5-6 feet in front of wherever you're sitting. If it feels like it's too low, get a $2 concrete block from Home Depot. Ideally, your natural & comfortable gaze should fall slightly below the top of the screen.
I guarantee that you'll never again be satisfied with having the TV far away or hanging high after experiencing it in that position.
The truth is, 90% of people have their TVs in totally wrong places for viewing comfort. Forty years ago, 25" console TVs put the picture tube a few inches above the floor... because it was the most natural, comfortable place to put it. Then, people started putting smaller 19" TVs in stands to raise them up so their top was approximately where the top of a 25" console TV's display would have been. Sometime in the 90s, console TVs went away, TVs gained another two inches, and... stands remained the same height, because stores didn't feel like stocking two different families of stands. And because stores were able to convince people to spend hundreds of dollars on something that's basically four sheets of laminated particleboard with a pressboard back and doors on the front, they weren't about to let people walk out the door without doing their best to sell them the highest-profit item in the entire store.
Fast forward a few more years. CRT TVs grew to 32", and kept growing towards 40"... and the stands remained the same size. By this point, TVs were uncomfortably high... but because "everyone's" TV was that high, people assumed it was somehow desirable. Perversely, this is ALSO when wall units became really popular... many of which jacked the TV up even higher so they could keep the doors symmetrical with the bottom doors in the adjacent units, themselves tall enough to accommodate two rows of books or VHS tapes. In other words, by this point, viewing ergonomics had TOTALLY gone out the door for the sake of storage space and furniture aesthetics.
Then HDTV and DLP arrived... and stores pushed even HARDER to sell stands that were STILL absurdly tall to people buying them, even though most DLP TVs already had 4-10" of height below the screen occupied by the speakers, light engine, etc. People bought plasma TVs, and blindly put them at the same height as TVs that were on stands. Or worse, stuck them in even worse places, like high on a wall above the fireplace (attempting to get some use out of a wall that builders inevitably present as the focal point of a room that REALLY gets 99% used for "viewing television and playing videogames").
The point is, for the past few decades, people have been putting their TVs in the wrong place, for all the wrong reasons. Try putting the TV back where it belongs -- near the floor -- and you'll see firsthand how much nicer it is to watch in that position.
You're assuming everyone has a relatively small (under 50") TV that's hanging on a wall on the opposite side of a large room.
If you have a 65" TV sitting 5-6 feet away, the difference between 1080p and 2160p is ABSOLUTELY visible.
There's another reason to favor 4k content, EVEN IF your display is "only" 1920x1080 -- overcompression vs oversampling.
If your content provider overcompresses 1080p video, it's going to end up with less hard detail than your 1920x1080 display is physically capable of reproducing. If the same content provider overcompresses 2160p video, it'll probably STILL end up with significantly more real detail than the official 1080p version. So if you ignore the 1080p stream you're "supposed" to use, and instead grab the higher-bitrate 2160p stream and locally resample it down to 1920x1080, it's like applying oversampling.
The thing about parallel processing is that not everything CAN be neatly decomposed into stateless parallel processes.
It's kind of like the situation with human workers. If you're excavating a big hole and have an army of slaves, adding workers/slaves to dig, fill buckets, and carry them away will generally increase your net output... until the point when they start getting in each other's way. As the complexity of the task increases, their ability to work efficiently in parallel decreases rapidly.
Computers are ultimately no different. If you're trying to perform millions of stateless computations that don't depend upon the results of other computations or their state, you can efficiently do a lot of work in parallel. The moment they have to start sharing RAM and coordinate their execution to avoid things like race conditions, your ability to compute in parallel falls off dramatically. Sometimes, a million metaphorical army ants will do the job. Other times, you metaphorically need Superman (possibly with backup from one or two other superheroes), and a thousand mere mortals will just get in the way.
Parallel programming is hard, and actually makes Djikstra's assertions about programs requiring rigorous mathematical proof of correctness start to look sane & reasonable. Traditional procedural programs can be validated experimentally. Parallel programs have to be validated primarily based on theory, because it's fundamentally IMPOSSIBLE to experimentally test all of their various runtime scenarios. And that's a really, really, huge problem, because it goes against just about every norm of real-world software development from the past half-century.
It's a problem whose scope makes rigorously validating the code used to launch a Saturn V rocket and get it to the moon look almost trivial by comparison. With a Saturn V, you basically had a single "happy path", and a few well-defined deviations whose goals could all be summarized as, "get it back ON that happy path". With parallel programming, most of time you don't even HAVE a single well-defined "happy path", and when you do, it's nearly impossible to know whether you're on or off of it until it's too late to do anything about it. Humans deal poorly with ambiguity, and computers are even WORSE at dealing with it.
> Speed of processing is definitely not the issue right now.
Clearly written by somebody who isn't actively involved with things like virtual/augmented/mixed-reality, realtime image-recognition, low-latency high-framerate photorealistic rendering, or realtime ray tracing.
Trust me, there are PLENTY of things left capable of soaking up enormous amounts of computing power.
The "realtime" part, in particular, is a nasty bitch. There are quite a few things that don't necessarily require SUSTAINED high-performance... but when they need performance, they need it INSTANTLY (example: recognizing road hazards & deciding how to handle them while driving a car).
Moore's Law isn't dead, only the "cheaper and cheaper, for less and less power" part that was a common consumer ASSUMPTION, but was never actually included by Moore himself. In the early 2000s, we hit a point when we hit computing power that was (kind of) "good enough", so vendors focused almost entirely on reducing cost and power, even while still increasing transistor counts (but at a lower rate). AI and VR/AR/MR are the next round of applications that are going to put us back into "everything you buy today will be hopelessly obsolete and unusably slow 2-3 years from now" mode.
Going with the observation that things need "bursty" high power, expect the NEXT major round of high-performance computing to come from stepping back from multiple cores back to multiple physical CPUs (or at least, multiple cores that are thermally-separate by a fair amount of space, possibly bathed in some closed-loop non-conductive coolant). Why? A CPU like the i7 can "burst" in single-core mode at speeds significantly higher than they can run with multiple cores, but we've increasingly hit a brick wall insofar as thermal management of ultra-dense CPU cores. So, if an i7 can burst (briefly) in single-core mode to 4GHz, but can only SUSTAIN 2-3GHz, the way to SUSTAIN 4GHz performance is to physically turn it back into an array of multiple CPUs, each of which can run continuously at slightly less than 4GHz, and burst for a few milliseconds at a time up to 4.5-5GHz(*). When you need REALLY high performance, you treat them like an orchestra of virtuoso soloists, each taking turns to step up and bear the full burst-load until they're about to melt before throwing the metaphorical hot potato to the next CPU in line.
---
(*) or incorporate some kind of closed-loop liquid cooling with some non-conductive liquid, so you can take the intense heat from tiny point sources and spread it around to something that can be viably air-cooled without melting itself.
Be careful about the "more money" part. Someday, when China has per-capita income & GNP comparable to the US, it'll be able to EFFORTLESSLY raise 2-4x the tax revenue to fund its space program, just by virtue of having 3-4x the population of the US (and 3-4x as many taxpayers). The US will HAVE to cooperate with Europe, just to maintain comparable levels of funding. And Russia has even LESS choice in the matter... it has space infrastructure & history, but approximately half the population of the US.
Personally, I think excluding China from programs like the ISS is a horrible long-term mistake. Do we *really* want to end up with US+Europe+Russia+(Japan, Canada, etc) vs China (with China feeling like an outsider who owes nothing to anybody & has a chip on its shoulder), as opposed to making China a nominal partner whose actions can at least slightly be kept in check & influenced by others without constantly inflaming that sense of "outsider" status?
And how would west Boston suburbanites get to the airport? I remember what getting to Logan airport USED to be like pre-Big-Dig. To call it a "total clusterfuck" would be a compliment.
Sometimes, you just need to do the job right, costs be damned. TBD is an example of "doing it awesomely right".
The Big Dig was expensive, but that's partly because it involved a lot of untested new technologies developed & used for the first time (ex: freezing muddy land under railroad tracks and using hydraulic jacks to ram tunnels through the soggy mud below... or building an underground freeway big enough to play multiple football games in, side by side, while the elevated freeway above remained in use). Some things worked, some things didn't. But SOMEBODY had to be the first to try. Some of those technologies ended up being useful for other projects, but the r&d cost is still counted entirely against TBD.
1) 4 daily rush hours instead of one: to work, home (empty) to park for free, to work (empty) to pick up, and to home.
2) people using city streets as free parking lots by having their car do laps around the block (especially after driving empty from home to work, while waiting for passenger to arrive.
Some people will pimp out their cars with Uber, but most people won't want their BMW or Mercedes to get destroyed like a public bus & will just send it home empty.
Before the "car and oil companies destroyed trolleys" tripe gets rolled out yet again, no. They were destroyed at the behest of suburbanites who hated trolleys because they caused accidents & clogged roads.
Getting rid of trolleys (and curbside parking) is what enabled us to have divided 4 & 6-lane roads with proper left turn lanes in built-up areas without having to demolish every building on one side to make the room.
The entire reason Miami voters keep wanting Metrorail (elevated heavy rail) expansion is the hope it'll get OTHER drivers off the road without getting in the way of cars. Light rail is nicer for riders than buses, but totally FUCKS UP traffic wherever LRT and cars share pavement.
The single worst transit project in Miami's history was the South Dade Busway. Instead of building Metrorail south from Dadeland to Cutler Ridge (mostly at-grade, ducking under cross streets every mile or so like the Washington DC Metro's Yellow Line through Arlington), they built a shitty BRT line that literally DOUBLED the time it took to drive from Dadeland to Cutler Ridge along US-1. Pre-Busway, Dade County spent YEARS fine-tuning the traffic light timing along that road. The Busway shot it all straight to hell by invalidating years of optimizations & tweaks in a single shot. It even fucked up cross-traffic by putting right turns on (usually, red) traffic lights as well.
Transit NEEDS to be grade-separated, or it sucks for everyone (ESPECIALLY the people whose taxes pay for it). At least, wherever it crosses a divided 4/6/8-lane road with traffic lights (2-lane residential roads that only have stop signs to begin with have a bit more room for negotiation)
Slashdot Mirror
I've used Synergy too. It's an extraordinarily useful program, but it's NOT a way to extend computer #1's desktop onto computer #2 via RDP... it's a way to have computers 1 and 2 running independently side by side, but use a single mouse and keyboard with both (treating them both AS IF they were actually one computer with its desktop extended onto a second monitor).
I wish Microsoft would give us the ability to treat connected RDP clients as additional monitors that the current desktop can be extended onto. For example:
1. Boot into Windows with laptop.
2. Boot into Windows with large-screen dual-OS tablet like a Chuwi Hi12
3. Launch RemoteDesktop on the tablet, and set up a new connection... but check the new, extra checkbox that says, "use this as an external monitor and extend the Windows desktop onto it"
4. Connect Remote Desktop on the tablet to Windows on the Laptop. Voila, instant second monitor.
Sadly, AFAIK, there's STILL no good way to achieve this. I know there are a couple of third-party Android apps that try to accomplish this (by running a host app under Windows and using VNC at the Android end)... and they all suck miserably. RDP is unique, because it hooks directly into the Windows rendering pipeline and has extraordinarily high performance. I'm pretty sure that (in theory, at least) you can even run DirectX at nearly full performance over RDP (as long as the GPU at the client end is powerful, since IT'S the one that ultimately gets used). I know that in the past (when I still had a normal desktop PC), I used to routinely run programs on my desktop PC, then use them over the LAN via RDP on a lower-powered laptop elsewhere in the house because they ran faster than if I tried running them directly on the lower-powered laptop itself.
I also wish that Linux had something with performance remotely close to that of Windows via RDP. VNC is dog-ass slow, and remote X11 over a network is (surprisingly) even slower. I'd originally had high hopes for Wayland, until I read that approximately four years ago, its architects declared point-blank that high-performance remote rendering over a network wasn't even on the table or open for negotiation as a design goal. Period, full-stop, end of story.
College students who want a minimally-compromised gaming system they can easily lug to their parents house for weekend visits and holidays?
Kids in high school who want a computer they can easily haul over to a friend's house?
People whose jobs require "travel" in the sense of "living in some city besides the one where you officially live, for days, weeks or months at a time... but not LITERALLY living out of a suitcase and running through airport terminals every day"?
People whose jobs involve "real" content-creation work that a mere laptop is inadequate for, but still want to have some degree of luggability so they can do demos outside of their office without the risk of having it not work properly on a lesser computer (or the need to waste hours ensuring that the demo DOES work properly on a lesser, more portable computer, making sure they give themselves enough time to fix whatever doesn't work if it DOESN'T work properly on the other computer)?
> Screen resolution of 1920 x 1080 pixels.
From what I understand, panel availability is the main problem. I believe that as of this moment, there's exactly one company that manufactures 17.3" panels with higher resolution than 1920x1080. The panels are 2560x1440 (not 3840x2160), are limited to 60hz, and apparently are known for having poor contrast & saturation.
I think part of the problem is also that nobody wants to go to the trouble of making a display capable of 2160p120, but making one capable of 1080p120 and 2160p30 without requiring yet another major overhaul of HDMI 2.x would require having some kind of agreed-upon standard for frame-packing (whereby you send 3840x2160@30fps as four 1920x1080 frames in a row at 120fps, and display 1080p120 by doubling each pixel to 2x2) that didn't exist back when that currently-available 17.3" panel was designed and went to manufacturing. I think the mess between FreeSync and Gsync compounded it, since historically Nvidia wouldn't allow licensees to support both standards, and refused to support FreeSync with their own cards in the hope that it could make Gsync the One Official Standard everyone else was forced to license from them. Now that VESA has apparently given Nvidia "the finger" by making FreeSync the officially-blessed standard for TVs, and Nvidia itself is apparently going to start supporting Freesync output, that might change in the near future.
Another issue with 2560x1440 panels: they suck for displaying natively 1080p content because every pixel HAS to be partially-scaled. With 3840x2160, 1080p can neatly scale 2:1 without losing sharpness (unless you WANT to do smoothing), and 720p has enough oversampling when scaled up to 3840x2160 to make Nyquist happy and avoid the really bad higher-order artifacts seen when scaling 1920x1080 up to 2560x1440.
Personally, I wish someone would just take the next step up and make a laptop that uses a 22" 3840x2160 display. AFAIK, "22" panels that support Gsync and 120+Hz framerates already exist... it would be too big to USE on an airplane, but then again, I'm pretty sure even 17.3" is too big to fit on a tray table without encroaching into the space of whomever is sitting next to you... and I think 22" panels are *just* small enough to fit into an enclosure that fits within the airline-defined maximum dimensions for carry-on baggage. It obviously wouldn't be a "laptop", so much as an "easily-luggable mobile workstation".
God forbid, if they went THAT big, they could probably even design its enclosure to BE the size of an airline-approved carry-on suitcase & have it support ATX-type components. I remember seeing something like that ~15 years ago, but unfortunately the company that manufactured it refused to sell the cases + monitors as separate components, and only sold them as complete (and HORRIFYINGLY expensive) systems (something like $8,000 for the cheapest one). The market for "AirTX"-formfactor cases might be small, but as long as it supported normal-formfactor motherboards, cards, etc, it wouldn't really MATTER... people would just buy the case-monitor combo, add everything else themselves, and evolve the system in a Ship of Theseus manner over the next 10 years by keeping the case, monitor, snap-on keyboard, etc, and replacing the other components as they became obsolete or wore out.
My personal fantasy: AirTX with three monitors, arranged like shutters over a window... a ~22" 3840x2160 main landscape-orientation display, flanked by a pair of portrait-orientation displays that were somewhere between 1440x2160 and 1600x2160 (attached with hinges so that when closed, the screens of the flanking displays faced the main screen). Add a range of keyboard options that includes model-M buckling-spring, Cherry blue/green/clear/black/red/brown switches, and Thinkpad-like scissor switches (all with two pointer sticks... one between the GHB keys, and one below the spacebar for thumb use, configurable so that one or both can be used for mouse-control, with one optionally used as a fau
> Now I'm puzzled what ATX cases are for
Honestly, for modern computers, a full-size mid-tower case is gross overkill, even for gamers. Nowadays, if a graphics card (or cards) generates lots of heat, it just sucks in air from the outside and blows it directly outside without involving the case's interior ambient airflow. Ditto for CPUs. In the past, case airflow in high-performance systems was mostly due to 10kRPM and 15kRPM hard drive arrays, which are now basically moot -- we use SSDs for things that used to use 10k/15k hard drives, and use slow drives for bulk storage that spend most of their time spun down anyway (usually, with their 50-200 gigabytes of most recently-used blocks cached to SSD for both performance and to let them remain spun-down).
Form-factor wise, we're still stuck with "shoe box" form factor and microATX size if you want to be able to use normal cards and external drives without worrying about half-height cards, riser cards, and laptop-formfactor optical drives, but there's a world of difference size-wise between a compact microATX case the size of two stacked shoe boxes and a traditional mid-tower case that was common 20 years ago.
IMHO, the biggest reason why we've never been able to evolve much smaller is because Intel is no longer sufficiently dominant to unilaterally dictate standards anymore, and getting a CONSORTIUM of companies with opposing interests to agree about ANYTHING has historically proven to be a nearly lost cause... at best, you end up with standards that either require implementing "the kitchen sink" to accommodate 5 original conflicting standards, or you end up with a standard nobody actually uses.
Back in the late 90s, Intel dictated several new form factors... only ATX caught on. We now have a few semi-informal conventions for CONNECTORS inside a laptop (like mPCIe, which is used by discrete laptop graphics cards and internal WLAN cards, and the various standards for internal SSD cards), but no OFFICIAL vendor-agnostic standards for form factor, bolt placement, heat-removal, physical access to the outside world, etc. Laptop optical drives seem to have settled down into one of two form factors, including the placement of USB3.0 and mPCIe/mSATA ports for them to plug into, but getting a true thirdparty drive to work in a given laptop still involves a bit of bezel-related surgery (even in cases where the drive is designed to be semi-easily removable and swappable).
The main problem with graphics cards is that there's no official standard governing the interface + physical form factor + thermal profile, so there's no official standard for card manufacturers to advertise compatibility with, nor an official standard for mobile workstation vendors to advertise compatibility with. At best, you're stuck buying parts online with your fingers crossed, knowing there's a good chance you'll end up having to sell it at a loss on eBay when it ultimately doesn't fit or work.
Thinkpads have the best keyboards & pointer sticks, but Lenovo never uses anything but Quadro graphics cards (hence, my other comment in this article re Dell happily using non-Quadros). For 97% of likely users, Quadro cards suck... mega-expensive, and crippled performance.
Then the solution is for Congress to make it illegal for content providers to force cable companies into all-or-nothing contracts as a natural application of antitrust law.
Truth be told, it wouldn't even affect ESPN or Disney's profits much. If ESPN went from being a $15 bundle of channels everyone had to pay for to a premium channel bundle that cost an extra $45/month, people who really care about televised sports will grudgingly pay it anyway. For Comcast, and even ESPN, it would be largely revenue-neutral, and merely represent a massive shifting of costs from "everyone" to "the subset of customers who really care about that particularly-expensive bundle of channels".
Damn it, someone finally came out with a laptop that has good specs, a G-sync 144hz display, and a RTX videocard that ISN'T a Quadro... then ruined it by using a keyboard that doesn't have a pointer stick. :-(
It doesn't have to. All cable channels are encrypted now and selectively unlocked box-by-box anyway. If the customer doesn't pay the "local channel" fee, don't show the local channels from the cable company. If the customer lies about having an antenna, they'll just have no local channels at all.
If companies like Comcast were smart, they'd do two things:
1. Allow customers to use their own antenna for OTA and get a $12-15/month discount. In other words, for each customer who puts up their own antenna, Comcast would save having to pay the local carriage fees for that customer, so it would be revenue-neutral for Comcast & would allow customers who are likely to bolt to rationalize staying around a little longer. This is EXACTLY what DirecTV used to do about 15 years ago... you connected your OTA antenna to the STB/DVR, and it seamlessly inserted the local channels into the lineup and treated them exactly as if they were from the satellite.
2. Allow customers who don't care about Disney and/or sports to trade one or both for channels like HBO, Showtime, and/or Starz.
We don't need literal "a-la-carte" pricing. Most of the channels on the mid-tier lineup only cost a few cents per month anyway. What we NEED is the ability to prune away the most expensive low-hanging fruit. ESPN, the RSNs, and Disney cost more per month per customer than HBO and Showtime. SlingTV has already proven that there's a market for people who'll happily sacrifice sports & Disney for other channels (Sling "Orange" allows one stream & includes ESPN and Disney, Sling "Blue" allows multiple streams, has more channels, and DOESN'T include ESPN or Disney).
> That, and if you're in the city, get a simple unidirectional antenna; if you're far from the city get one that you can point toward the city.
And $150-300 for an OTA DVR. And unless you're seriously masochistic, don't even THINK about buying a $22-50 Mediasonic "DVR" -- they're a pain to use, have a UI straight out of the early 2000s, and the one I bought literally died for no apparent reason 5 days after the warranty ended. On the other hand, if you have an old PC with Windows 7 and a HD-Homerun, Windows Media Center is STILL awesome as a DVR.
Seriously, losing your DVR is the absolute worst part of cord-cutting. Even when streaming services offer "cloud DVR", it totally sucks compared to the experience of having a real DVR that doesn't refuse to let you skip commercials, or take 10 seconds to recover when you fast-forward by 20 seconds.
Are "LS" the ones widely sold by Radio Shack back in the early 80s, or were those L/C/{other}-series? I just know that HC/HCT chips seem to be capable of surviving things that used to EASILY destroy chips ~35 years ago. I remember destroying chips all the time back then, but never seem to have that problem anymore despite occasionally connecting voltages backwards, hitting chips with unregulated 6v instead of carefully-regulated 5v, static electricity, sourcing/sinking too much current, etc. It seems like with HC/HCT-family chips, making a mistake causes malfunctioning or "erratic" behavior... older chips literally went up in smoke if you did stuff like that.
Do yourself a favor... the next time you have the house to yourself for a few days, take the TV off the wall or cabinet, and move it to a spot that's 5-6 feet in front of wherever you're sitting. If it feels like it's too low, get a $2 concrete block from Home Depot. Ideally, your natural & comfortable gaze should fall slightly below the top of the screen.
I guarantee that you'll never again be satisfied with having the TV far away or hanging high after experiencing it in that position.
The truth is, 90% of people have their TVs in totally wrong places for viewing comfort. Forty years ago, 25" console TVs put the picture tube a few inches above the floor... because it was the most natural, comfortable place to put it. Then, people started putting smaller 19" TVs in stands to raise them up so their top was approximately where the top of a 25" console TV's display would have been. Sometime in the 90s, console TVs went away, TVs gained another two inches, and... stands remained the same height, because stores didn't feel like stocking two different families of stands. And because stores were able to convince people to spend hundreds of dollars on something that's basically four sheets of laminated particleboard with a pressboard back and doors on the front, they weren't about to let people walk out the door without doing their best to sell them the highest-profit item in the entire store.
Fast forward a few more years. CRT TVs grew to 32", and kept growing towards 40"... and the stands remained the same size. By this point, TVs were uncomfortably high... but because "everyone's" TV was that high, people assumed it was somehow desirable. Perversely, this is ALSO when wall units became really popular... many of which jacked the TV up even higher so they could keep the doors symmetrical with the bottom doors in the adjacent units, themselves tall enough to accommodate two rows of books or VHS tapes. In other words, by this point, viewing ergonomics had TOTALLY gone out the door for the sake of storage space and furniture aesthetics.
Then HDTV and DLP arrived... and stores pushed even HARDER to sell stands that were STILL absurdly tall to people buying them, even though most DLP TVs already had 4-10" of height below the screen occupied by the speakers, light engine, etc. People bought plasma TVs, and blindly put them at the same height as TVs that were on stands. Or worse, stuck them in even worse places, like high on a wall above the fireplace (attempting to get some use out of a wall that builders inevitably present as the focal point of a room that REALLY gets 99% used for "viewing television and playing videogames").
The point is, for the past few decades, people have been putting their TVs in the wrong place, for all the wrong reasons. Try putting the TV back where it belongs -- near the floor -- and you'll see firsthand how much nicer it is to watch in that position.
You're assuming everyone has a relatively small (under 50") TV that's hanging on a wall on the opposite side of a large room.
If you have a 65" TV sitting 5-6 feet away, the difference between 1080p and 2160p is ABSOLUTELY visible.
There's another reason to favor 4k content, EVEN IF your display is "only" 1920x1080 -- overcompression vs oversampling.
If your content provider overcompresses 1080p video, it's going to end up with less hard detail than your 1920x1080 display is physically capable of reproducing. If the same content provider overcompresses 2160p video, it'll probably STILL end up with significantly more real detail than the official 1080p version. So if you ignore the 1080p stream you're "supposed" to use, and instead grab the higher-bitrate 2160p stream and locally resample it down to 1920x1080, it's like applying oversampling.
Or interpreting the imaging sensor on a gaming mouse. ;-)
I'd hardly call debouncing a keyboard "AI" ;-)
The thing about parallel processing is that not everything CAN be neatly decomposed into stateless parallel processes.
It's kind of like the situation with human workers. If you're excavating a big hole and have an army of slaves, adding workers/slaves to dig, fill buckets, and carry them away will generally increase your net output... until the point when they start getting in each other's way. As the complexity of the task increases, their ability to work efficiently in parallel decreases rapidly.
Computers are ultimately no different. If you're trying to perform millions of stateless computations that don't depend upon the results of other computations or their state, you can efficiently do a lot of work in parallel. The moment they have to start sharing RAM and coordinate their execution to avoid things like race conditions, your ability to compute in parallel falls off dramatically. Sometimes, a million metaphorical army ants will do the job. Other times, you metaphorically need Superman (possibly with backup from one or two other superheroes), and a thousand mere mortals will just get in the way.
Parallel programming is hard, and actually makes Djikstra's assertions about programs requiring rigorous mathematical proof of correctness start to look sane & reasonable. Traditional procedural programs can be validated experimentally. Parallel programs have to be validated primarily based on theory, because it's fundamentally IMPOSSIBLE to experimentally test all of their various runtime scenarios. And that's a really, really, huge problem, because it goes against just about every norm of real-world software development from the past half-century.
It's a problem whose scope makes rigorously validating the code used to launch a Saturn V rocket and get it to the moon look almost trivial by comparison. With a Saturn V, you basically had a single "happy path", and a few well-defined deviations whose goals could all be summarized as, "get it back ON that happy path". With parallel programming, most of time you don't even HAVE a single well-defined "happy path", and when you do, it's nearly impossible to know whether you're on or off of it until it's too late to do anything about it. Humans deal poorly with ambiguity, and computers are even WORSE at dealing with it.
> Speed of processing is definitely not the issue right now.
Clearly written by somebody who isn't actively involved with things like virtual/augmented/mixed-reality, realtime image-recognition, low-latency high-framerate photorealistic rendering, or realtime ray tracing.
Trust me, there are PLENTY of things left capable of soaking up enormous amounts of computing power.
The "realtime" part, in particular, is a nasty bitch. There are quite a few things that don't necessarily require SUSTAINED high-performance... but when they need performance, they need it INSTANTLY (example: recognizing road hazards & deciding how to handle them while driving a car).
Moore's Law isn't dead, only the "cheaper and cheaper, for less and less power" part that was a common consumer ASSUMPTION, but was never actually included by Moore himself. In the early 2000s, we hit a point when we hit computing power that was (kind of) "good enough", so vendors focused almost entirely on reducing cost and power, even while still increasing transistor counts (but at a lower rate). AI and VR/AR/MR are the next round of applications that are going to put us back into "everything you buy today will be hopelessly obsolete and unusably slow 2-3 years from now" mode.
Going with the observation that things need "bursty" high power, expect the NEXT major round of high-performance computing to come from stepping back from multiple cores back to multiple physical CPUs (or at least, multiple cores that are thermally-separate by a fair amount of space, possibly bathed in some closed-loop non-conductive coolant). Why? A CPU like the i7 can "burst" in single-core mode at speeds significantly higher than they can run with multiple cores, but we've increasingly hit a brick wall insofar as thermal management of ultra-dense CPU cores. So, if an i7 can burst (briefly) in single-core mode to 4GHz, but can only SUSTAIN 2-3GHz, the way to SUSTAIN 4GHz performance is to physically turn it back into an array of multiple CPUs, each of which can run continuously at slightly less than 4GHz, and burst for a few milliseconds at a time up to 4.5-5GHz(*). When you need REALLY high performance, you treat them like an orchestra of virtuoso soloists, each taking turns to step up and bear the full burst-load until they're about to melt before throwing the metaphorical hot potato to the next CPU in line.
---
(*) or incorporate some kind of closed-loop liquid cooling with some non-conductive liquid, so you can take the intense heat from tiny point sources and spread it around to something that can be viably air-cooled without melting itself.
Do people still *use* "LS" chips? I thought HC & HCT were the default families used by most hobbyists now.
Be careful about the "more money" part. Someday, when China has per-capita income & GNP comparable to the US, it'll be able to EFFORTLESSLY raise 2-4x the tax revenue to fund its space program, just by virtue of having 3-4x the population of the US (and 3-4x as many taxpayers). The US will HAVE to cooperate with Europe, just to maintain comparable levels of funding. And Russia has even LESS choice in the matter... it has space infrastructure & history, but approximately half the population of the US.
Personally, I think excluding China from programs like the ISS is a horrible long-term mistake. Do we *really* want to end up with US+Europe+Russia+(Japan, Canada, etc) vs China (with China feeling like an outsider who owes nothing to anybody & has a chip on its shoulder), as opposed to making China a nominal partner whose actions can at least slightly be kept in check & influenced by others without constantly inflaming that sense of "outsider" status?
And how would west Boston suburbanites get to the airport? I remember what getting to Logan airport USED to be like pre-Big-Dig. To call it a "total clusterfuck" would be a compliment.
Sometimes, you just need to do the job right, costs be damned. TBD is an example of "doing it awesomely right".
The Big Dig was expensive, but that's partly because it involved a lot of untested new technologies developed & used for the first time (ex: freezing muddy land under railroad tracks and using hydraulic jacks to ram tunnels through the soggy mud below... or building an underground freeway big enough to play multiple football games in, side by side, while the elevated freeway above remained in use). Some things worked, some things didn't. But SOMEBODY had to be the first to try. Some of those technologies ended up being useful for other projects, but the r&d cost is still counted entirely against TBD.
The biggest impact of self-driving cars will be:
1) 4 daily rush hours instead of one: to work, home (empty) to park for free, to work (empty) to pick up, and to home.
2) people using city streets as free parking lots by having their car do laps around the block (especially after driving empty from home to work, while waiting for passenger to arrive.
Some people will pimp out their cars with Uber, but most people won't want their BMW or Mercedes to get destroyed like a public bus & will just send it home empty.
Before the "car and oil companies destroyed trolleys" tripe gets rolled out yet again, no. They were destroyed at the behest of suburbanites who hated trolleys because they caused accidents & clogged roads.
Getting rid of trolleys (and curbside parking) is what enabled us to have divided 4 & 6-lane roads with proper left turn lanes in built-up areas without having to demolish every building on one side to make the room.
The entire reason Miami voters keep wanting Metrorail (elevated heavy rail) expansion is the hope it'll get OTHER drivers off the road without getting in the way of cars. Light rail is nicer for riders than buses, but totally FUCKS UP traffic wherever LRT and cars share pavement.
The single worst transit project in Miami's history was the South Dade Busway. Instead of building Metrorail south from Dadeland to Cutler Ridge (mostly at-grade, ducking under cross streets every mile or so like the Washington DC Metro's Yellow Line through Arlington), they built a shitty BRT line that literally DOUBLED the time it took to drive from Dadeland to Cutler Ridge along US-1. Pre-Busway, Dade County spent YEARS fine-tuning the traffic light timing along that road. The Busway shot it all straight to hell by invalidating years of optimizations & tweaks in a single shot. It even fucked up cross-traffic by putting right turns on (usually, red) traffic lights as well.
Transit NEEDS to be grade-separated, or it sucks for everyone (ESPECIALLY the people whose taxes pay for it). At least, wherever it crosses a divided 4/6/8-lane road with traffic lights (2-lane residential roads that only have stop signs to begin with have a bit more room for negotiation)
I think the "drone" aspect was just played up as clickbait due to the recent (alleged) drone incident at the airport.