What the 4K: High-def DisplayPort vid meets reversible USB Type C The new Type C USB connector is causing a lot of excitement, thanks in part to its reversibility (you can plug it in either way up) and high rates of data and power transfer. But there's now another reason to buy into in: DisplayPort support. The Video Electronics Standards Association (VESA) on Monday finalized DisplayPort …
Firewire was better? It exposed the entire contents of your RAM directly to any connected device ffs! It couldn't have been designed any better by the NSA themselves to defeat the security of any machine. If you thought viruses and trojans spread through USB thumb drives were bad, at least they needed the operating system to actually auto-run them, hence a properly configured OS was immune. Firewire thumb drives and peripherals would be able to compromise any system, no matter what the OS and in fact the OS would be powerless to prevent it.
> It exposed the entire contents of your RAM directly to any connected device
I think if the NSA has nobbled your screen to leak data, your CPU and RAM are probably not so safe anyway.
I believe TB is the same - it gets DMA access, as do PCIe cards. You should be afraid!
For PCIe you at least need access to the inside of the machine, which isn't the case for firewire or thunderbolt. It would be a problem for an external express card slot on a laptop.
I wonder why four people down-voted, there's nothing inaccurate about saying the DMA of firewire is a huge security risk. It's hardware level access which cannot be disabled by software, and cannot even be disabled by the BIOS on some motherboards. Apple fans maybe?
>I wonder why four people down-voted, there's nothing inaccurate about saying the DMA of firewire is a huge security risk.
The attacker will need physical access to your machine to plug in a rogue FW device. In the environments in whih FireWire is used, any thief will just make off with your external HDDs or expensive video camera before they start faffing around with a FireWire-based attack.
"The attacker will need physical access to your machine to plug in a rogue FW device."
Yes, a an ordinary thief might, if they only wanted the data on a hard drive and not ongoing covert surveillance, passwords and encryption keys.
A compromised FW device shipped from China (other countries may apply) would be plugged in by a user without them being any the wiser. If you think that's far fetched then just google for the story about Chinese irons shipped to Russia with wifi snooping circuitry, or the surveillance/spy stores online which sell all sorts of innocent looking peripherals designed to be given to the intended target as gifts.
Most corporate environments forbid external drives. If someone wants your data, the chances are that data is in a hard drive in your main computer. If I were to steal data from someone I'd pop a usb drive in and copy their my docs folder onto it and bugger off. Imagine if I could do that silently though, by slipping a device in and having it automatically do that while the computer is locked...
Maybe you didn't read about how Stuxnet got propogated through social engineering?
They just left some "infected" 16 Gb thumb drives laying in the parking lot, same thing could be done with ANY other technolgy, even cables or cameras or harddrives.
YOU will do the "physical access" connection part for the criminals or spys.
"I wonder why four people down-voted, there's nothing inaccurate about saying the DMA of firewire is a huge security risk."
I suspect the downvotes were due to the implication that because of the security risk, Firewire was a stupid or worthless system with no useful application.
Given that Firewire was offering 400-3200Mbps at a time when USB 1/1.1/2 was offering 1.2-480Mbps, the use case for demanding video and photographic applications was overwhelming. And piping video back from your GPU to get to the USB controller is indeed a stupid and wasteful thing to do.
In a closed environment where you used Firewire exclusively for importing material from your camera, etc it was the choice between getting on with it or leaving it overnight, and realistically caused no significant security risk.
The gap has closed now, with USB 3/3.1 offering 5-10Gbps, which in most applications far outstrips the I/O of connected hard drives, cameras or memory devices, rendering the speed advantage of Thunderbolt to be less than compelling for all bar the most demanding users, who will potentially find a means of hooking up their Red Rocket direct via a PCIe breakout port (which is what TB is effectively) very useful on laptops or mobile workstations.
That's hardware which would - in a conventional tower or workstation be connected via PCIe anyway. Therefore providing a means to connect it to devices without conventional PCIe slots (like laptops) has value, and doesn't put your system at any more risk than it would be anyway - provided you treat TB like your genitalia and take care what you do with it and who you give access to...
Best buy should be delighted. More cables and adapters for them to sell - no sooner had I bought a displayport adapter for my DVI KVM switch than I got a laptop with a mini-DP socket, requiring another adapter. Coupled with the HDMI adapter I already had, I can now wait to buy a USB C adapter. Should be good business.
However much I may hate BB though, I have tried cheap and less cheap USB cables, and the super-thin, super cheap versions are much less reliable than the nice thick hefty ones.
If it's unreliable, take it back to the retailer, or at least let the retailer know that you're unhappy.
Whatever the price, products must be fit for purpose. Cheap products are not a licence to rip off the customer. Retailers who want repeat custom are likely to be sensitive to complaints about products that are more trouble than they're worth.
PoE can do 30W over Cat5 cable, with only a 5 degree C raise in temperature.
Don't see what's so impossible about 100W over a USB cable which, in my experience, can be as much as 2-3 times thicker than some of the Cat5 that I manually crimp and patch.
If PoE isn't dangerous, then 100W over this USB connector probably isn't either. And I assume that someone has sat down and worked out the numbers before releasing millions of devices to the public that might get hot.
To be honest, I bet your cable heats up more just from sheer conductivity from that huge TV...
The USB Power Delivery spec accounts for this - there is a presence detection & handshake process that makes sure the connected device and cable can support the required voltages and currents before the power is actually switched on. If the handshake doesn't happen, you just get the standard 5 volts/500mA. This handshake process also allows for the power to flow "backwards" (i.e. from target to host), so that when you plug your laptop into a USBPD-enabled monitor, the monitor will be able to provide power to the laptop.
Higher voltages allow lower resistance (heat). 12V and 24V are the standard PoweredUSB voltages, but even then you're still talking 4-8A, quite a bit for a little wire. They'd have to go up to 48V (PoE) to get it under the 2.1A that seems pretty standard on USB chargers now, and maybe that's exactly what they did.
Per the USB type C spec, the 100W is delivered by 20V @ 5A -- but note that the host and target device both need to support it. 5A does seem a lot for such a small wire and connector; however there are already USB 3.1 cables electronically marked as able to handle up to 3A or up to 5A.
The USB power delivery spec which type C supports specifies a maximum 1V drop from VBus to Gnd (round trip) at the rated current, and gives examples of using 22AWG wire for ground and 26AWG for VBus for a 1m cable rated for 3A. If I follow their example and substitute 5A for 3A, it looks like 16AWG is needed for Gnd and 22 AWG for VBus for a 1m cable rated at 5A. Not a thin cable, but the larger power conductors won't really be noticed when considering the 16 conductors needed for the other signals, plus shield.
The branding will also, hopefully, stop retailers from charging ridiculous amounts for cabling with no extra benefits: the gear should work exactly the same, whether it costs ten bucks or a thousand.
Yeah, that'll never fuckin' happen. I remember buying an Aquos a few years back, and the salesman arguing to the hilt that the $80 HDMI cable would produce better results than my $5 ones. He really wouldn't drop it, and then I found out he had three of the expensive ones at home.
No sale, mate.
Motion blur is part of how moving images work. Indeed there are some cinematic advantages in having the blur in frames rather than the eye having to reconstruct it through the persistence of vision. Bear in mind that the great majority of films used to be shot at 24fps with a two blade shutter to greatly reduce flickering.
Indeed much electronic capture is also done at 24 or 25fps to maintain a "cinematic" look.
What high (video) frame rate can help with is dealing with flicker through higher refresh rates, and there's some evidence that there's evidence that 100Hz is preferable to 50Hz in that respect. But high refresh rates has nothing to do with "motion blur". High refresh rates also don't have to be done in the source. They can be synthesised by the display device.
@Steven Jones: you're concentrating on large-area flicker whilst completely ignoring motion portrayal. The eye will track moving objects, so if you don't wish an object to go blurred as soon as it (or the camera, eg when panning) starts to move, you need a high enough refresh rate. The higher the resolution of the static image, the more noticeable will be the effect so as resolution increases, frame rate needs to increase to maintain image quality. There is evidence that several hundred frames per second can be needed for ultra-high resolution ("4K"). For further explanation, see, for instance, here.
@Steven Jones: you're missing the point - what is done currently is irrelevant - we are developing higher frame rates (as well as other improvements) to go with higher resolutions for the future. You might as well have been arguing in the 1950s that there was no point in developing colour TV because no commercial or broadcast systems captured in colour.
It is not that a 4K frame isn't better than 1080p frame. Of course it looks better paused side by side but you are at normal viewing distance* hitting some seriously diminishing returns.
It is also quite a way to misunderstand how it is that our eyes work. Our eyes track with a central high resolution focal point and periphery low resolution. If something on our periphery captures our attention, our eyes will move to focus on it and our brain magically stitches it together. You have no doubt seen the kids books in the Where's Wally series. If so you will know it can take a few minutes to find him, but if Wally had an animated hand waving at you with the rest of the image still, you would see him in under a second. Our eyes** are very good at picking movement.
In the real world, I would prefer the bandwidth used to better represent the movement in the frame, even to the point of dropping below 1080*1980 resolution briefly during that movement because that would "look better"
* er, that would be my definition of normal viewing distance; not that of my 2 year old.
** technically our brains rather than our eyes
Hint: If using proper cables, the side with the USB logo on it goes upwards. This even works in the dark by touch.
IMHO, SATA leads are the worst ruddy things to try to fit blind. Handed, sides indistinguishable and damned tricky to get in if access is poor when the right way round.
It's interesting to see Apple on board, especially for mobile. Suggests they reckon that Lightning will run out of speed and have to be replaced by USB C.
Something interesting in press release: The cable has 4 USB 3.1 lanes. Each lane can either be used for USB or re-purposed for DisplayPort. That means that you could have a 2.5kx1.4k monitor with a built-in USB 3 hub connected using 1 USB C lead. 2 lanes can drive the monitor and the other 2 provide 20gbps to the USB hub.
And I didn't know that USB C has a standard for negotiating complete changes of wire protocol. Cool. That means that a phone could contain a USB C port, but a range of reasonably-cheap adapters can expose a range of different signals. (This will be less useful when USB C takes over the world and there's no need for any other signals, but still)
... for us poor hobbyist hardware folk. USB2 is a doddle to solder, chips are plentiful and software stacks abound. Now USB C comes along with 24 pins in a tiny package. Soldering anything less than 0.6mm pitch is pretty hard, and given this is only 8.4mm across I predict I won't be building anything with USB C for a long time.
Spare a thought...
... for us poor hobbyist hardware folk. USB2 is a doddle to solder, chips are plentiful and software stacks abound. Now USB C comes along with 24 pins in a tiny package. Soldering anything less than 0.6mm pitch is pretty hard, and given this is only 8.4mm across I predict I won't be building anything with USB C for a long time.
It's not that hard, but you need to work very cleanly and make sure you have a decent temperature stabilised soldering iron to avoid pumping too much heat into the tracks. If you can handle SMD packaged ICs you ought to be fine as long as you don't drink too much alcohol beforehand :)
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
