LG 5K display must be kept at least 2 meters away from Wi-Fi routers | Hacker News Reader

LG 5K display must be kept at least 2 meters away from Wi-Fi routers

arstechnica.com

103 pointsminikites80 comments

This monitor was Apple's chance to ease the USB-C transition and IMO they blew it.

The previous gen Thunderbolt display effectively brought all the useful ports[1] back to the MacBooks with one cable, and made even the Air viable for a professional. It certainly was a big factor for me back in 2011.

Unfortunately with the current 5k display, the three included ports aren't even USB 3.1 gen 2. Like many other pros, I prefer to have dedicated keyboard/mouse, speakers, and Ethernet, and would have been happy to run a second cable if it brought those things back.

Right now it's looking like our best 'pro 5k' option is duct taping a hub[2] to the back of someone else's display. And it's not clear which hubs support the 15" MBP 87W charging requirement.

Feels like such a missed opportunity.

[1] https://www.amazon.com/Apple-MC914LL-27-inch-Thunderbolt-Dis...

[2] https://eshop.macsales.com/preorder/owc-thunderbolt-3-dock/

> Like many professionals, I prefer to have dedicated keyboard/mouse, speakers, and Ethernet, and would have been happy to run a second cable if it brought those things back.

Each port provides 5V/1.5A; enough for multiple USB-A peripherals on a single port with a USB-C to multi-USB-A. (I have a LG 5K and I use a keyboard, mouse, and microphone on one port; it works great) And USB-C speed (5 Gbps) will not bottleneck data transfer in that case.

Glad that's working nicely for your needs. Are you happy with the display overall?

I agree for the low bandwidth devices you've mentioned it's fine to hang everything off the one line. There is of course a speed hit for higher bandwidth needs[1] e.g. RAID arrays but that was true for the previous gen too.[2]

Nevertheless I still find it a shame we lost the variety of integrated ports from the previous gen. If we have to buy extra dongles anyway I'm equally ready to check out competitor panels.

AFAIK the LG 5k is the only panel offering 87W power output, right?

[1] http://appleinsider.com/articles/17/01/07/review-lg-ultrafin...

[2] http://www.macworld.com/article/1163773/storage/thunderbolt-...

I'm pretty happy with it overall, especially with the current discount. (I do not have the connection issues mentioned in the article)

Worst case for super-high-bandwidth devices you can connect directly to the laptop. (I do not have a external drive bottlenecked by the 5 Gbps limit yet, but may get an external SSD soon / eGPU once that is viable)

> AFAIK the LG 5k is the only panel offering 87W power output, right?

Currently, that is the case. The LG 4k only has 60W output.

I have this 5k monitor. It sits around 30cm from my ASUS 2.4 and 5ghz router. It's connected to my Macbook Pro 2016 15". I connect one USB hub via a USB A to USB C adapter. This hub powers a MIDI keyboard, Native Instruments Maschine, FocusRite Scarlett audio interface, SD cards when I put them in, USB hard drives when I plug them in. The hub hides behind the monitor. I personally love sitting down to work with this monitor each day and I don't understand the arguments against it. There is nothing better on the market today for the spec you get.

> "There is nothing better on the market today for the spec you get."

Aside from the small detail of the monitor getting blown away by a router nearby - a router nearby is not exactly an uncommon occurrence in a computing environment. Not to mention the other issues like possible kernel reboots and flickering caused by unknown factors.

In context, it is a small detail (anyone serious enough buy this monitor and able to afford it will likely be able to relocate their router), and there's still no other 5k option on the market.

Yeah but mine doesn't, so I'm quids in ;)

Nice to hear you have an audio interface working there! Sounds like you do professional audio -- curious what kind of bandwidth needs you're expecting to have?

It's a minor nit, but one reason I prefer Thunderbolt hubs is because they hang directly off the PCI bus.

Historically speaking, with USB the host CPU intermediates all bus traffic. The 'fire' in FireWire (vs e.g. USB2) was how devices could access host memory directly.

It's probably a negligible amount of overhead these days, but my take is, if it's possible to use Thunderbolt and free up CPU cycles for whatever you're doing, why not?

Love to hear opinions from someone who has more recent knowledge on this.

If I'm totally honest, I have no idea what you're talking about :)

I do a lot of audio work for video and composition through Logic using lots of plugins and bits and it just handles it nicely. That's really all I'm bothered about ;)

Does this mean LG is violating the FCC requirement that a device must accept interference?

It's the reverse. The FCC is about jamming the signal not being jammed.

Elaborating, my understanding is that the 'accept interference' bit is more about placing such devices low on the legal totem pole: The TV owner or manufacturer doesn't have a legal right to _not_ be interfered with; it _must accept_ interference. Whether it continues to work or fails after 'accepting' the interference is its own business. With that interpretation, they would be in violation of that clause if they sued the router manufacturer over this.

It's both:

Title 47 - Telecommunication. § 15.5 (b) [...] subject to the conditions that no harmful interference is caused and that interference must be accepted [...]

https://www.gpo.gov/fdsys/pkg/CFR-2015-title47-vol1/xml/CFR-...

Not true.

When the FCC says you must accept interference, that means you have to live with it if the interfering device meets emissions standards. It doesn't mean the device has to be functional in the face of interference.

Are you sure? Seems to be pointless if it just means that it will not be irrecovably damaged.

It doesn't even mean it won't be irrecoverably damaged, it just means the manufacturer of the damaged hardware can't sue the manufacturer of the interfering hardware.

They must accept the interference in a philosophical manner more than a technical one.

Let me try an example: I have an amateur radio rig that transmits only on the allowed frequencies and it meets FCC standards. You are my neighbor using an antenna to get your TV. You allege that my radio transmissions interfere with your TV viewing. Legally, I'm allowed to tell you to pound sand and you have no recourse. Because your TV has to accept radio interference that is otherwise legal, whether your reception is affected or not.

(Of course, because I'm not a complete dick, I would assist my neighbor in properly shielding his TV or otherwise mitigating the situation, rather than tell her to pound sand.

No, but it raises the question of whether the lack of shielding causing this issue also causes the monitor to radiate excessive RF noise that may interfere with other devices. If it's sensitive to 2.4Ghz signals, there may be signals generated inside it in the same frequency band which can likely escape just as easily as the signal from a router can intrude.

USB 3.0 has been known to interfere with WiFi for quite a while. I think Intel mentioned it when they announced USB 3.

Of course, this is a slightly different signalling mechanism (Thunderbolt 3), but basically, you have to put the spurious emissions somewhere that doesn't interfere with licensed radio services, which basically means 900MHz, 2.4GHz or 5GHz, and they picked 2.4GHz.

http://www.intel.com/content/www/us/en/io/universal-serial-b...

Sounds like it's going above and beyond the regulations by not just accepting the interference, but it's inviting it with open arms.

The clause about accepting interference means that the device must not fail when recieving interference in a way that would cause it to radiate on another frequency. Product functional failure otherwise is not regulated in the US. The FCC doesn't care if your product bricks itself as long as it doesn't jam the spectrum.

I've been through this testing before. They blast your device with many different frequecies and measure what comes out.

The test was harder to pass in the old days when there many discrete mixers inside radios and poor isolation between the RX and TX paths through the duplexer.

I'd be curious if an enterprising soul with some foil and a ground could identify the specific vulnerable hardware and put out some notes on how to, err, patch it.

I have one of these displays at the office (as well as the 4K version) and they both very much feel like an Apple product sans the aluminum casing.

So this makes me suspect that the metal case was really supposed to be there when the electronics were put together.

I disagree. I used one at an Apple store and it was terribly slow. It looked great, of course, but as soon as you tried to use Expose or Mission Control (whatever it's called now) it turned into a slideshow and almost made me think the computer froze up.

The computer being unable to render it's UI fast enough has exactly what to do with the display...?

I'm not a fan of Apple, but I think it's unfair to blame Apple in this case. If it's an LG monitor, it's LG's problem.

Apple designated this monitor as the successor to the discontinued Thunderbolt Display, and they sell it alongside Macs in the Apple Store.

The kernel panics and lockups are not LG's fault.

I somewhat agree with you, but the manner in which Apple is promoting this display combined with the utter lack of ports on the MBP (hooking up my previous two displays would have been a dongle nightmare) make me less willing to excuse them.

Yeah seriously. The display crashed multiple El Capitan running MBPs at my office.. so they had to upgrade to Sierra. You think it would work flawlessly instead.

Well I'm glad I didn't invest in this monitor yet! My desk in less than 2 meters from my wifi and I can't move either easily... (wifi mounted to ceiling, nowhere else for the desk)

I have an ASUS VG248QE (144hz) monitor, and when i put my phone under it, the screen goes black

Bet that USB-C cable is the issue. I've come across plenty that had horrible RF shielding.

I wonder if a simple ferrite bead would fix the issue.

I had a 60in plasma TV from LG that was systematically disconnecting my DSL connection when the image was too bright. The telephone cable was going under the wooden floor under the TV. Solved it by switching to an LED TV.

What seems a complete disregard for the costumer is that the solution proposed by LG itself is to "increase the distance between the monitor and the router".

Seems that working directly with Apple made them take a hint from the "you are just holding it wrong" excuse of Steve Jobs.

Steve Jobs never said "you are just holding it wrong".

He wrote, ipsis verbis, "don't hold it that way".

https://www.theguardian.com/technology/blog/2010/jun/25/ipho...

The very source you linked says he said "Just avoid holding it in that way".

So he did not say what you said ipsis verbis, and he certainly did not say "You're holding it wrong". That was made up by reddit.

https://en.wikipedia.org/wiki/Title_47_CFR_Part_15 ?

Sounds about right. Nothing some creative use of foil wont fix. Just what I'd want to do to my new ~$1000 monitor.

Part 15 is only for transmissions. However, it sounds like it isn't handling reception very well.

I haven't had any issues with the 5K monitor and it's about 3 feet from the router. At least for me it works incredibly well.

You are lucky. I have an LG 5K, and I have experienced every symptom (excepting the "no nearby wireless" bit, as I'm not sure if there is wireless nearby) described here:

> kernel panics on the host machine, random flickering […], and USB devices not being reliably detected when plugged into the monitor.

I'll have to check where my APs are, now that I know about this association. But it's been a damn struggle; I feel like 50% of the times I connect to it, I have to disconnect & reconnect USB devices to get them to be detected on the host.

Best alternative?

I've had a couple 27" 1440p monitors since those cheap Korean (QNX and the like) came on the market a few years ago.

The natural upgrade path for me would be to ditch those and get a big-ass 5k monitor eventually, but I don't see the market really moving in that direction in an affordable way.

One 4k isn't going to cut it and I'm worried two 4k monitors is going to be to much from a field of vision perspective....maybe I'm wrong there.

Has anybody upgraded from a couple 1440p monitors to a couple 4k monitors, and what's your experience been?

Have you thought about the 40" 4K displays from Philips and Iiyama? I believe they are around $600.

I went from 2 1440p to one 32" 4K and 1 1440p.

It's been one of the biggest productivity improvements since moving from 640x480 to 1024x768.

The 1440p is now only used for minor stuff like email and not really necessary anymore.

I soon moved to a 32" 4K at home as well, because I just couldn't stand just 1440 anymore.

One of the interesting parts is that I don't use 4K fully: it's too much surface are and your head tilts too much to see everything at once. But what's great is that I can put a background process as full screen in the background and have a 3500x1800 window in front in which I do my real work. So I can observe the proceedings of the background talks without having to toggle windows all the time.

Agreed, I think 32" 4K is the sweet spot for Linux & Windows. The PPI is 140, which is "marginally retina" -- you can see pixels if you get close. Higher PPI is sharper even once you get past "retina", but it's definitely a case of diminishing returns. But the nice thing about 140 PPI is that you can run without scaling if you absolutely need to. You have to squint and you wouldn't want to do it too often, but it does work for those rare apps that don't scale well.

And the 32" size is about perfect too, IMO. The height is about the maximum you can view without head movement.

I'm starting to agree with you and Bryanlarsen that 32" 4k seems to be a sweet spot.

I run two 2560p QNX monitors. They're wonderful, would highly recommend; I think two of them will set you back around $400, a bargain compared to 4k monitors.

One 40" 4K has been perfect for me. Roughly the same DPI as my old QNIX (~110).

I got the Samsung UN40KU6290 (does 4k@60 with 4:4:4 chroma) for under $300. The PC mode is crisp and responsive. Calibrates well enough. Only complaint is slow GTG time can look blurry when scrolling text.

Seconded. I have the same monitor and it's a perfect upgrade from my old 3 27"-24" monitors. 40" seems perfect to not have turn your head too much, two would be too much for me.

I've got a UN40JU6500 (40" 4K Samsung) on my MBP (15" late 2013) but it doesn't do @60, as far as I know.

Are you driving this from a Mac or from a PC? What kind of cable are you using exactly?

Both. For my mid-2014 15" MBP I had to get an active MiniDP-HDMI adapter. Looks like your MBP is limited to 30Hz...

https://support.apple.com/en-us/HT206587

Interesting! So it's not the monitor (TV), it's the MBP.

Oh well, will have to wait for the next MBP 15" refresh--don't see any real reason to upgrade yet to the latest TB MBP.

Plus, the @30 doesn't normally bother me, unless there's a solid large gray-ish color, which definitely shimmies.

These 40" Samsungs 4Ks really hit the sweet spot for Retina displays, at least for my old eyes. "Only" 1080P logically, but oh, so large and sharp!

Also, did you have to do any SwitchResX hackery? That's the only way I could get macOS to recognize the monitor as a Retina display.

Nope, but I run it at 100% scaling. I did notice that if the source on the TV is not labeled "PC" then the Mac uses YCbCr 4:2:0 and looks like garbage. Setting the input label to PC must send a different EDID because it seems to use RGB... On my Windows machine I can just override that in the nvidia control panel.

Oh, you run it at the full 4K (3840x2160) resolution? Must have good eyes.

Right, nothing works unless you label the relevant HDMI input "PC".

I bought this monitor the other week after hearing things like this but I couldn't get over two things:

1) It doesn't power off or on automatically. I need the remote 2) The viewing angle made seeing the corners difficult on my admittedly shallower then probably needed desk with no scaling, (40" seemed like the sweet spot for PPI without scaling).

How do you cope with (1)? I'm amazed that Intel, Nvidia and AMD haven't thrown a HDMI-CEC IP code into their GPUs and made it trivially to turn on and off displays over HDMI.

I just leave the remote on my desk and turn the eco auto-off to 8hrs so it doesn't turn off in the middle of my day. The TV controls the on/off of my receiver via CEC.

If you have 2 27" with 1440p and go to 2 27" with 4k which do you think the field of vision changes? It should be exactly the same. The potential differences would be that you can have either more screen estate (but with tiny rendering) or a much better image quality if scaling is used. I would at the moment always go for 4k due to better image quality - and not for anything else.

The trick is to put some of your monitors in portrait orientation. It's a game changer for reading text and for writing code (aka 95% of my computer use).

Especially if you can get a new main monitor that is exactly as tall as the old screen was wide. I've seen a couple people pull that off and it works really really well, better still if you do a color calibration.

This monitor is pathetic. 900 dollars for a monitor and it doesn't even support upstream.

Save your time and money and buy a DELL or something that actually works guys.

What is upstream?

According to a comment[1]" "An upstream port connects to the host device (PC) while the downstream ports are where you plug in peripheral devices (thumb drives, printers, etc.)."

Of course the 5k monitor does have three USB3 ports. (Not Thunderbolt ports, ostensibly because there isn't enough bandwidth left.)

[1] https://www.cnet.com/forums/discussions/upstream-vs-downstre...

I like how they use USB-C inputs for USB3 ports as well. Not at all false advertising and confusing for end-user.

Why? This is the promise of USB-C "Every port uses the same connector, so you only need one type of cable". Of course, with the different cables supporting different ports it is not ideal.

It's when you connect your keyboard and mouse to monitor instead of laptop/pc.

Essentially only one wire goes to your laptop/pc instead of 3.

I'm pretty sure this monitor lets you plug in keyboards and mice to have them work with the host computer.

Nope, it does not.

Really? Apple says, "And three downstream USB-C ports (5 Gbps) offer additional connectivity and power to compatible devices and accessories." What makes mice and keyboards incompatible?

Edit: according to this random internet stranger, it works fine: http://www.apple.com/shop/question/answers/product/HKN62LL/A...

I don't know where you got that impression. I have this LG 5K display and I'm typing on a keyboard connected to it right now...

Untrue. I have one and works just fine with my mouse, keyboard, sdcard reader, external hard drives, etc.

Best looking monitor I've ever owned, no problems yet.

$1000 (after tax) is a great deal compared to the long dark years where the LCDs weren't as good as the CRTs which they had totally replaced.

What irks me is the lack of an Ethernet port. Would a USB-C to Ethernet adapter work?

Yes, but there are reports of the USB hubs in these particular monitors being unreliable. Anyway why do you expect your monitor to have Ethernet?

The Apple monitors that these LGs are replacing had built-in Ethernet ports.

Apple's monitors had one and like them, LG's monitors are kinda marketed as docking stations. Missed opportunity to make an awesome product.

Reply on news.ycombinator.com