Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Displayport: A Better Video Interface

hackaday.com

Post reply on HN
Print view View original

211–220 of 426 posts

Re: Displayport: A Better Video Interface

#211

Post by babypuncher »

lucideer wrote: post #26
remix2000 wrote: post #23

With HDMI, I have both video and audio traveling over a single cable. That is _extremely_ convenient. DP _theoretically_ supports audio too, but AFAIK that's not widely implemented.

> That is _extremely_ convenient. I'm open to having my mind changed but I have not really experienced this extreme convenience in practice. The vast majority of things I plug a HDMI cable into either don't have speakers or if they do, the speakers are a bad quality afterthought. In actual fact, what has been an extreme inconvenience is the OS thinking I want HDMI audio instead of whatever much higher-quality* altern…

HDMI was designed to accommodate home theaters. That usually means a sound system with good speakers. Not every user needs this expanded feature set, but enough do that DisplayPort does not make a good alternative.

Here's where ARC/eARC really benefit me, and I don't see how this problem could be solved with DisplayPort.

I have a PC and a PS5 that support Variable Refresh Rate (VRR) over HDMI. I bought a new TV last year that also supports VRR, but I did not want to replace my perfectly adequate 3 year old reciever that does not support VRR. Even today, VRR support on recievers is questionable at best, and most users likely want their VRR devices plugged directly into their TV.

Thanks to eARC, I can plug my PS5 or PC directly into my TV and have it send the PCM 7.1 surround sound audio to the reciever over HDMI. I can still leave other devices like an Apple TV or Blu-Ray player plugged into the reciever, and everything just works.

Without eARC, I would have to fall back on TOSlink. That means extra cables, and dropping down to compressed dolby digital 5.1 if I want to keep surround sound. Using dolby digital on a game console incurs a pretty noticeable latency penalty, which is why they all default to PCM.

Re: Displayport: A Better Video Interface

#212

Post by chx »

someplaceguy wrote: post #189
chx wrote: post #168

> However, if you don’t need the full bandwidth that four lanes can give you, a DP link can consist of only one or two main lanes resulting in two This is how many USB C docks operate. The USB C connector also has four high speed lanes and there's a mode where two are assigned to carry DisplayPort data and two are assigned to be TX and RX respectively for USB. Until DP 1.4 appeared, this meant you were quite limited…

What is the benefit of adding compression here? Does that mean that the bandwidth might or might not be sufficient depending on which pictures are being shown on the screen? If so, that doesn't seem very reliable and if not, what's the point of compression?

> DSC is most often used in a Constant Bit Rate (CBR) mode, where the number of coded bits generated for a slice is always a constant that is generally equal to the bit rate multiplied by the number of pixels within the slice

https://vesa.org/wp-content/uploads/2020/08/VESA-DSC-Source-...

Re: Displayport: A Better Video Interface

#213

Post by mschuster91 »

ShadowBanThis01 wrote: post #203
mschuster91 wrote: post #54

Earlier quoted context omitted.

SDI can't do any communication between the source and sink. You can do a limited form if you use two SDI cables (one upstream, one downstream) which is what stuff like BlackMagic's bi-directional micro converter does, but IIRC that's only used for talkback to the camera operator and HDMI CEC messages which is how they implement things like their ATEM Mini consoles triggering recording - not for DDC-style format and f…

BMD has done some innovative things through their SDI port, specifically enabling camera control with an Arduino-add-on board. Someone finally did what I've wanted to see for years: built a follow-focus unit that controls the focusing motors in the lenses, so you don't have to bolt a ridiculous contraption (and janky focusing-ring adapters) onto a lens to turn its (non-mechanical) focusing ring manually.

What I want to see is someone dump the FPGA in these Micro Converters and hack it to add an USB interface. The Micro Converters are all just the same in the interior: SerDes units and clock/redrivers on the I/O ports and an FPGA that additionally has an USB connection which is used to power the converter but could also be used to do other things.

That FPGA should be powerful enough to do a lot of interesting things - anything from fooling around with commands (like with the Arduino board) to image manipulation (e.g. a watermark embed).

Re: Displayport: A Better Video Interface

#215

Post by msie »

Ajedi32 wrote: post #210

> A carefully shaped money-backed lever over the market is absolutely part of reason you never see DisplayPort inputs on consumer TVs, where HDMI group reigns supreme, even if the TV might use DisplayPort internally for the display panel connection. Curious about this. Is the HDMI standards group engaging in anti-competitive behavior to prevent DisplayPort from taking over on TVs? I've always assumed it was just mome…

I was just thinking about USB-C vs Lightning... What if the EU were to mandate HDMI over DisplayPort?

Re: Displayport: A Better Video Interface

#216

Post by chx »

emilyst wrote: post #177
chx wrote: post #168

> However, if you don’t need the full bandwidth that four lanes can give you, a DP link can consist of only one or two main lanes resulting in two This is how many USB C docks operate. The USB C connector also has four high speed lanes and there's a mode where two are assigned to carry DisplayPort data and two are assigned to be TX and RX respectively for USB. Until DP 1.4 appeared, this meant you were quite limited…

> Ps.: When I say Thunderbolt, I am well aware of how Thunderbolt 4 is just USB4 with optional features made mandatory. It's not relevant to the discuss at hand. Dear God, I hope this situation settles down in the near future. As it is I have years of USB-C-looking cables that all do different things but are visually indistinguishable.

I wish the manufacturers would just adopt the USB IF marketing names and logos. https://i.imgur.com/H3unbD5.png would be a lot simpler.

I also wish the USB IF defined colors for high speed lanes absent vs 5/10/20 gbps capable high speed lanes and then 60W/100W/240W power. All it needed were two color bands on the plastic part of the plug. If colors are too gaudy then go Braille-style, have a 2x2 grid on top and bottom of the plug where bit 0 is a little hole and bit 1 is a little bump. That's 16 possible data speeds and 16 possible power levels and so far we have only needed 4 for data and 3 for power.

Intel could've added a separate row for Thunderbolt.

Re: Displayport: A Better Video Interface

#217

Post by someplaceguy »

chx wrote: post #212
someplaceguy wrote: post #189

Earlier quoted context omitted.

What is the benefit of adding compression here? Does that mean that the bandwidth might or might not be sufficient depending on which pictures are being shown on the screen? If so, that doesn't seem very reliable and if not, what's the point of compression?

> DSC is most often used in a Constant Bit Rate (CBR) mode, where the number of coded bits generated for a slice is always a constant that is generally equal to the bit rate multiplied by the number of pixels within the slice https://vesa.org/wp-content/uploads/2020/08/VESA-DSC-Source-...

Ah, so basically this is typically used as lossy compression with a constant bit rate, which makes the bandwidth usage predictable/constant but depending on the pictures being shown (and the noise level?), it can lead to some visual degradation (which they say is usually imperceptible).

That's interesting and surprising to me.

Re: Displayport: A Better Video Interface

#218

Post by baby_souffle »

pxc wrote: post #137
jchw wrote: post #121

Earlier quoted context omitted.

None of them are perfect, but I've heard good things about the DP switch from Level1techs. The thing is, all of them are a little tricky, but they mostly differ in how quirky they are, and I suspect the reason why people like the Level1techs DP switch is that they seem to at least try to alleviate some of the issues DP switches tend to get into. https://store.level1techs.com/products/14-kvm-switch-dual-mo... The star…

The Level1Techs KVM switches are rebranded Startech switches with 'dumber' firmwares whose dumbness affords better compatibility with niche DisplayPort features. I have a bunch of them and I like them pretty well, but getting a bunch of computers all plugged in turns out to be a bit of a nightmare, especially when you need some long-ish cable runs or you are daisy-chaining devices (e.g., multiple KVM switches, adding…

> The Level1Techs KVM switches are rebranded Startech switches with 'dumber' firmwares whose dumbness affords better compatibility with niche DisplayPort features.

Rextron [1] is the actual ODM. They don't do any direct to consumer sales, though. That's why L1 / Startech / other "brands" sell them on amazon and the like.

Last I spoke with the L1 guy, they were still having some issues with the high speed USB-C switching chips on the 1x4 "all USB-C" model that he's got a wait-list for.

[1] https://www.rextron.com/KVM-Switches.html

Re: Displayport: A Better Video Interface

#219

Post by Dr4kn »

msie wrote: post #215
Ajedi32 wrote: post #210

> A carefully shaped money-backed lever over the market is absolutely part of reason you never see DisplayPort inputs on consumer TVs, where HDMI group reigns supreme, even if the TV might use DisplayPort internally for the display panel connection. Curious about this. Is the HDMI standards group engaging in anti-competitive behavior to prevent DisplayPort from taking over on TVs? I've always assumed it was just mome…

I was just thinking about USB-C vs Lightning... What if the EU were to mandate HDMI over DisplayPort?

The EU Generally chooses the more open and cheaper alternative. They did this with phone charging, car charging and would probably do the same with display connectors.

USB C is the cheaper and more capable connector

Re: Displayport: A Better Video Interface

#220

Post by 6D794163636F756 »

Ajedi32 wrote: post #210

> A carefully shaped money-backed lever over the market is absolutely part of reason you never see DisplayPort inputs on consumer TVs, where HDMI group reigns supreme, even if the TV might use DisplayPort internally for the display panel connection. Curious about this. Is the HDMI standards group engaging in anti-competitive behavior to prevent DisplayPort from taking over on TVs? I've always assumed it was just mome…

Another key is that HDMI carries more power than display port. There are dongles for your tv that can get all the power they need from an HDMI cable. Not so far display port.

When it comes to use case, display port is better when connecting a fully independent device to a display and HDMI is better for connecting a peripheral to a system

Post reply on HN