Retina Mac External display problem

https://github.com/oyiptong/retinadisplayproblem

by sayhello • 13 years ago

96 39 13 years ago

github.com

octo_t• 13 years ago

I can't seem to reproduce, but I've got two monitors (VGA + HDMI)

safwanc• 13 years ago

Same here. MacBookPro10,1 with two monitors connected (VGA + HDMI) and I can't reproduce.

cmsj2• 13 years ago

I find it most amusing that you'd buy the laptop with the best screen available and then connect it to a monitor with the worst available connection method (VGA) ;D

singular2• 13 years ago

Not experiencing this with a mid-2012 Mac Book Pro Retina hooked up to an external display via an HDMI/DVI adapter.

I wonder if this is some strange prank designed to get people to look at a screenshot of https://optin.stopwatching.us/, esp. given the '1px reproduction' of this is entitled 'We are watching you!' If so, I really, really don't get it :-S

sayhello5• 13 years ago

Not a troll. Actually, let me host another version of the page.

The reason it's got the "We are watching you!" is because i noticed the problem while viewing https://optin.stopwatching.us/

Edit: hosted here: https://people.mozilla.com/~oyiptong/retinadisplayproblem/

singular2• 13 years ago

Ah fair enough, that made me suspicious in particular as that title seemed unrelated to the bug :-)

I wonder whether my HDMI -> DVI adapter is making a difference somehow?

zoop2• 13 years ago

This works for me with my first gen 15" Retina. This also happened to me at https://optin.stopwatching.us/, and I assumed it was exploiting some kind of bug with fullscreen for an intentional effect. I am using an external 20" older generation Apple display.

sayhello5• 13 years ago

It wasn't intentional. I'm not on the team working on stopwatching.us, but I had the issue and wanted to get to the bottom of this.

If you've got the bandwidth, could you please post your monitor model and/or snap a video?

zoop2• 13 years ago

Apple Cinema Display ‪23-inch DVI‬ (1920 x 1200) connected through a DVI/DP converter. It doesn't happen when I open the image in photoshop.

kunai• 13 years ago

This must be exclusive to the Retina MacBook Pro, then, because my MacBook Air doesn't show the same symptoms.

sayhello5• 13 years ago

So far, I've only been able to reproduce it on first-gen 15" Retina MacbookPro's

jlgreco• 13 years ago

For what little it is worth, it doesn't repro on chromebook pixels.

marc• 13 years ago

Can't reproduce. mid-2012 MBP Retina + Cinema Display 24"

freshyill• 13 years ago

I can't reproduce it using the latest update to the Retina 15" MBP and an iMac in target display mode, connected via Thunderbolt.

moraljto2• 13 years ago

off-topic: how does he/she get those vertical tabs in Firefox in that screenshot?

https://raw.github.com/oyiptong/retinadisplayproblem/master/...

(warning: that is the picture that apparently breaks certain displays)

sayhello5• 13 years ago

That's Tree Style Tab, great for tab hoarders like me https://addons.mozilla.org/en-US/firefox/addon/tree-style-ta...

moraljto2• 13 years ago

awesome, thanks!

faramarz• 13 years ago

No problems for me.

  Retina, 13-inch, Late 2012 HDMI Out

kogir• 13 years ago

Can't reproduce with MacBookPro10,1 + 2x Mini-DP->Dual-Link DVI + 2x Dell 3007WFP on 10.8.4

mbell3• 13 years ago

Just a guess:

The patterns in the examples seem to be horizontal, meaning the raw (pre 'cable encoding', usually 8b/10b) bit pattern of the display output would be repetitive . It wouldn't be out of the question that a slightly improperly electrically balanced or terminated output could cause signal integrity issues, including ringing (voltage exceeds spec on rising / falling edges) which could trigger safeguard circuitry in a display (shut it off).

sayhello5• 13 years ago

I don't have any electrical knowledge. care to elaborate?

Is that a problem with the cable, terminals on the devices, the mini-displayport dongle or either?

The problem also only seems to occur on Retina Macbook Pro's

mbell3• 13 years ago

If my guess is correct, it would be a problem in the design of the output transceiver circuit on the RMP's motherboard, assuming the cable and receiver in the display are built to spec.

Could be an IC issue, could be the impedance control of the connector, could be the impedance control of the traces on the circuit board, could be poorly chosen passives, could be interference from nearby circuits, etc.

I wish I could give you a better answer but high speed signal integrity is one of the most obtuse and most black art aspects of electronics design, there's a fair amount of non-intuitive physics going on, e.g. signals going down a wire don't just go, they 'ricochet' off impedance differences. It's far too large a topic to treat in a HN comment and its whole system encompassing so 'what it could be' is a large set.

These types of things are not uncommon in hardware development, especially with high speed signaling. The interactivity of such systems can be high and hard to predict fully at times.

EDIT: Simply put - display outputs scan top to bottom, left to right. The fact that the pattern that triggers the issue appears to be horizontal means it would occur in the raw bit stream of the output in a periodic way. That means there is some low order frequency in the system caused by this pattern (maybe it repeats every 500khz or something like that). If there is something in the electronics of the output of the RMBp that resonates at that frequency (easier than you think), then it could possibly cause the described failure.

sliverstorm• 13 years ago

Sounds to me like your parent is suggesting it could be any and/or all of the above. Essentially, at very high speeds the whole link has to be very carefully balanced, so if any one piece is not built to a tight enough spec, things could fall apart.

mbell3• 13 years ago

I thought of a simpler, or at least more relatable, analogy:

Most people have been around a cell phone which was near an audio speaker and heard it make those awful snap and pop sounds.

It doesn't make sense initially that you would hear anything like that from a phone, after all they operate at very high frequency, 900 - 2400 Mhz, definitely not audible frequencies. But, the system works by breaking up radio time into chunks, several hundred chunks a second. The relative spacing of the chunks, a few hundred a second, creates noise that is easily picked up by the speaker and turned into audible sound.

In the same way with the laptop you've got this high frequency signal: HDMI / DVI but when there are particular patterns of bits being carried you can get lower frequency content, similar to the chunks being used by the cell phone. Just like the coil in the speaker gets excited by the radio chunks, you could have a problem where the output circuitry of the RMBp gets excited by these rare low frequency patterns leading to problems with the display.

testbro• 13 years ago

The readme mentions that the issue is reproducible in OSX and Windows which further supports your theory.

fearlessleader• 13 years ago

Can't Reproduce on my 15" retina with neither a dell nor a vizio tv over hdmi.

dlazerka• 13 years ago

It's funny to see how many Hacker News writers use Mac, the most closed hardware+OS, most anti-hackish.

How about renaming HN to stand for Hipster News?

lazugod• 13 years ago

A Mac is the only computer on which you can legally develop for all major operating systems, including mobile.

How is that closed?

jrockway2• 13 years ago

Apple puts artificial restrictions on what tools you can use to write iPhone apps, and Macs are that allowed tool, so OS X isn't closed?

That is literally the dumbest argument I've ever read on HN.

jawngee• 13 years ago

takes photo

Posts like these are why HN can't have nice things anymore.

stordoff• 13 years ago

Unable to reproduce this on my first-gen rMBP 15" with an LG TV connected via HDMI. Will try my external displays (Dell/HDMI; LG/VGA) tomorrow.

brisance• 13 years ago

How long before it takes effect? I don't see anything on my Retina MacBook Pro with Thunderbolt Display.

  Model Name:	MacBook Pro
  Model Identifier:	MacBookPro10,1
  Processor Name:	Intel Core i7
  Processor Speed:	2.6 GHz
  Number of Processors:	1
  Total Number of Cores:	4
  L2 Cache (per Core):	256 KB
  L3 Cache:	6 MB
  Memory:	16 GB
  Boot ROM Version:	MBP101.00EE.B02
  SMC Version (system):	2.3f35

Thunderbolt Display:

  Vendor Name:	Apple Inc.
  Device Name:	Thunderbolt Display
  Vendor ID:	0x1
  Device ID:	0x8002
  Device Revision:	0x1
  UID:	0x0001000100508120
  Route String:	3
  Firmware Version:	22.2
  Port:
  Status:	Device connected
  Link Status:	0x2
  Port Micro Firmware Version:	0.0.21
  Cable Firmware Version:	0.1.18
  Cable Serial Number:	C4M2263005QDNWFAX
  Port:
  Status:	No device connected
  Link Status:	0x7
  Port Micro Firmware Version:	0.0.21

cmsj2• 13 years ago

I also can't reproduce it with the Thunderbolt Display.

Are the original reproducers of this all using minidisplayport adapters? Or are some able to reproduce using the native HDMI port?

haxxorfreak• 13 years ago

I can't reproduce the issue on a first-gen MacBookPro10,1 2.3GHz connected to an Apple Cinema Display 27" (over mini-DP)

Apple recently swapped my LG LCD panel to a new Samsung one for ghosting but I'm in clamshell mode.

psteinweber• 13 years ago

can reproduce with all examples incl. stopwatching.us with mid 2012 rmbp and LG 23EN43 which is connected via thunderbolt to DVI adapter.

display shuts off immediately.

Was quite confused when opening stopwatching.us yesterday, thanks for pointing it our here :)

Retina Mac External display problem

Related Stories