Just happened to see this kickstarter project. Fiber Optic HDMI?

[yakitatefreak]

Seems like a bold claim coming from the people doing the Kickstarter project. Here's the link for the Kickstarter project (By Vivify). I'm wondering what to think of this, especially since monitors have been pretty crazy as of late.

[GDRRiley]

we have it with USB, it is likely to be done yep; now is there some marketing BS in their yeah.

I'd trust corning tho to get it first over this kickstarter. 

[Dujith]

https://www.kvm-shop.nl/nl/aten-ve872-hdmi-active-optical-cable-15m-4kx2k-plug-play/

https://www.amazon.com/FIBBR-UltraPro-18Gbps-Speed-Active/dp/B074SFKK21/ref=pd_lpo_sbs_23_img_2?_encoding=UTF8&psc=1&refRID=Z5E4S3J5C5R2S4AXWQ14

 

I think they might have a problem. This already exists  

Plus 80% of their tech is HDMI spec  

[corrado33]

Bunch of bull shit. 

 

HDMI is digital, no need for fancy cables. HDMI 2.0 will already support 4k at 60 fps. 

 

Uses bullshit marketing terms. Electrons in wires move slow, but that's no matter because it's like water in a pipe. Push one electron in one side and an electron comes out the other side very quickly. 

 

The cables are lighter... oh... that's great.

 

Fiber optics are more durable than copper cables.... HAHAHAHAH no. 

 

They even contradict themselves. They say that copper's limitation is 10 Gb/s, yet in the bar graph right below that they show that copper HDMI 2.1 at 48 Gb/s.

 

DO NOT BUY EXPENSIVE HDMI CABLES, THEY ALL WORK THE SAME. (Provided you bought the correct version.)

[WereCatf]

A typical money-grab.

Quote

Fiber optic transmission is faster.

Irrelevant, the HDMI-spec has already been certified to work with copper-based cables and thus fiber's higher theoretical bandwidth doesn't do jack shit.
 

Quote

 

Vivid Full Spectrum Color : YCbCr 4.4.4 compliance  

Four times to sixteen times the color pixels than current cable

 

Blatant lie. As long as the current cable is up to spec, it supports the full color-spectrum.

Quote

Super Fast FPS for Games - 240 FPS compliance in Full HD  

Any HDMI 2.1 - certified cable does, and any HDMI 2.0a or HDM I2.0b does in 4:2:0, too.

 

And so on and so forth; it's just a bunch of actual lies, some intentionally misleading stuff and lots of marketing-bullshit.

[Dujith]

1 minute ago, corrado33 said:

Bunch of bull shit. 

This Kickstarter is bull yes. The idea behind the fiber cables is length. Beyond 15 meters you are going to have issues. Fiber allows much longer distances.

[Mira Yurizaki]

The problem I'm seeing is if the magic sauce is actually capable of handling the data rates needed to support higher end features.

 

But even then, I would only find this useful for running long cables where signal degradation does start to become a problem in the digital domain.

 

EDIT: This claim is more or less false

Quote

2. Fiber optic transmission is faster. Fiber optic (speed of photons) vs copper wire (speed of electrons).    

According to https://en.wikipedia.org/wiki/Velocity_factor for at least networking cables, the velocity factor for optic fiber and twisted pair wiring (say Cat6) are about the same.

[corrado33]

3 minutes ago, Dujith said:

This Kickstarter is bull yes. The idea behind the fiber cables is length. Beyond 15 meters you are going to have issues. Fiber allows much longer distances.

Active HDMI cables (75 m)  are still cheaper than this bullcrap. 

Besides, 99% of people are not going to need longer than a 15 m HDMI cable. Even people with centralized media systems. 

 

Also, HDMI over Ethernet exists. 

[corrado33]

Makes me wonder of the latency of transferring a digital signal to optical, then back again. Must be slower than just keeping it as electrons. 

 

Linus should test this. 

[Sauron]

Here's why it's bullshit and you shouldn't fund it:

 

1) HDMI is a fixed standard, any cable that meets the standard will perform exactly the same as every other cable.

2) Since HDMI is a digital connection, cable quality is even less of a factor than it would be otherwise.

3) Fibre optic does nothing differently from copper, it's just potentially faster. "Purity of light" means nothing and is "pure" snake oil.

 

4) This image:

has nothing  to do with the point it's supposed to illustrate (that fibre optic cables are faster, which as I already explained is bullshit for its own reasons).

 

5) A fiber optic cable is NOWHERE NEAR worth 1500$ unless it's 100 metres long.

 

6)

Quote

5. Light cannot catch on fire. Benefit of fiber optic cables is that they are not a fire hazard. There is no electric current traveling through the core! 

excuse me, WHAT? First of all, copper HDMI cables are NOT a fire hazard - ever. There isn't nearly enough power going through them to be dangerous. Secondly, just because the current isn't going through the whole wire doesn't mean it's magically not using any power, and the part that is most likely to get damaged is the connector, which DEFINITELY has electrical circuitry in it.

 

Which means nothing, because this cable HAS LEDS IN IT! ARE THEY HIGH? HOW DO THEY PLAN ON POWERING THOSE LEDS WITHOUT A COPPER WIRE????

 

7)

Quote

Four times to sixteen times the color pixels than current cable

Once again, utter bullshit. HDMI is a standard, either everyone has 4:4:4 chroma support or nobody does - and if the TV doesn't support it, it's the latter.

 

8 )

Quote

4:4:4 close up ( left ) | 4:2:0 close up ( right )

bullshit, you're watching this on your monitor and if your monitor doesn't support 4:4:4 you shouldn't be able to see the difference - and yet you can. Why? Because the image on the right is modified to look worse regardless of your monitor's colour support.

 

9)

Quote

Super Fast FPS for Games - 240 FPS compliance in Full HD  

Gamers can play their gaming titles in PURE 4K 60hz form and push their frames to 240 fps in full HD

Nothing that other cables compliant with the same standard can't do.

 

So yeah, complete and utter bullshit, and borderline a scam.

 

[Sauron]

3 minutes ago, corrado33 said:

Makes me wonder of the latency of transferring a digital signal to optical, then back again. Must be slower than just keeping it as electrons. 

 

Linus should test this. 

It depends on the distance, but it really doesn't matter - your screen has a higher latency than any HDMI cable.

[yakitatefreak]

2 minutes ago, corrado33 said:

Makes me wonder of the latency of transferring a digital signal to optical, then back again. Must be slower than just keeping it as electrons. 

Depends on cable length? But all I see is a bunch of marketing schemes right now. Though with that said, this is the type of content that LTT does for review, where on one hand, there's this marketing jargon, and on the other hand, there's the actual delivery. 

[corrado33]

Just now, Sauron said:

It depends on the distance, but it really doesn't matter - your screen has a higher latency than any HDMI cable.

No I mean the chip that changes the electrons into photons. That's definitely not "lag free". I'm not talking about transmission lag for longer cables. 

[WereCatf]

6 minutes ago, corrado33 said:

No I mean the chip that changes the electrons into photons. That's definitely not "lag free". I'm not talking about transmission lag for longer cables. 

Um, no. You literally turn electricity into photons with e.g. LEDs, and that doesn't cause any latency. The fact that this happens inside a chip, instead of using discrete components, doesn't change that. This is what optocouplers do: inside the chip-package, there's an LED-like device that emits light and a phototransistor that detects the light and acts on it, opening or closing depending on whether there is light shining on it or not.

[corrado33]

2 minutes ago, WereCatf said:

Um, no. You literally turn electricity into photons with e.g. LEDs, and that doesn't cause any latency. The fact that this happens inside a chip, instead of using discrete components, doesn't change that.

But there has to be some sort of processing to send it down the fiber optic.... right? You can't just connect an LED to the signal and expect it to arrive at the other side.

[WereCatf]

1 minute ago, corrado33 said:

But there has to be some sort of processing to send it down the fiber optic.... right? You can't just connect an LED to the signal and expect it to arrive at the other side.

Yes, you literally can.

[Sauron]

6 minutes ago, corrado33 said:

No I mean the chip that changes the electrons into photons. That's definitely not "lag free". I'm not talking about transmission lag for longer cables. 

 

2 minutes ago, WereCatf said:

Um, no. You literally turn electricity into photons with e.g. LEDs, and that doesn't cause any latency. The fact that this happens inside a chip, instead of using discrete components, doesn't change that.

To be clear: ANY circuit has latency. LED or not, it takes some time to turn that light on and off. And it takes some time for light to go through the cable. Electricity through copper is already damn fast; fibre optic is not "faster" because of light - it's faster because you can fire off more data beams without fear of interference in a FO cable than you can in a copper cable.

[corrado33]

3 minutes ago, WereCatf said:

Yes, you literally can.

But HOW. Are there the same number of optical conductors inside the optical cable as in the copper cable? Or do they send signals at different frequencies and use Fourier transform at the other end to split it all back out?

 

And the HDMI signals would have to be run through an amplifier anyway to power any sort of LED. 

[WereCatf]

1 minute ago, Sauron said:

To be clear: ANY circuit has latency. LED or not, it takes some time to turn that light on and off. And it takes some time for light to go through the cable.

Well, technically, literally everything that exists has an amount of inherent latency. Whether the latency is actually measurable or means anything in practice is another thing and, when it comes to LEDs and such, the turn-on-time has no bearing in even terabit-speed communications.

[WereCatf]

1 minute ago, corrado33 said:

But HOW. Are there the same number of optical conductors inside the optical cable as in the copper cable? Or do they send signals at different frequencies and use Fourier transform at the other end to split it all back out?

That all depends on the specific cable; they can manufacture a cable with the same number of individual optical fibers, or, if they want to go fancy, they can include a chip that combines all the signals from the copper-cable and encodes them for transmission over a single fiber, in which case there would, indeed, be some latency added. For something like this it wouldn't really make much sense, though, unless it was cheaper to manufacture it like that.

[Sauron]

1 minute ago, WereCatf said:

Well, technically, literally everything that exists has an amount of inherent latency. Whether the latency is actually measurable or means anything in practice is another thing and, when it comes to LEDs and such, the turn-on-time has no bearing in even terabit-speed communications.

Copper is no worse in this sense - not measurably. But hey, THEY are the ones who claimed "no latency", and therefore it's fair to slam them for it.

3 minutes ago, corrado33 said:

But HOW. Are there the same number of optical conductors inside the optical cable as in the copper cable? Or do they send signals at different frequencies and use Fourier transform at the other end to split it all back out?

Fibre optics can carry multiple signals in the same glass tube.

4 minutes ago, corrado33 said:

And the HDMI signals would have to be run through an amplifier anyway to power any sort of LED. 

It depends on how powerful the LEDs are. They don't need a lot of power.

[corrado33]

2 minutes ago, WereCatf said:

That all depends on the specific cable; they can manufacture a cable with the same number of individual optical fibers, or, if they want to go fancy, they can include a chip that combines all the signals from the copper-cable and encodes them for transmission over a single fiber, in which case there would, indeed, be some latency added. For something like this it wouldn't really make much sense, though, unless it was cheaper to manufacture it like that.

That's my point. We're not talking about on-board electrical vs optical transmission. We're talking about on-board electrical converted to optical via a cable. There MUST be some additional processing inside the cable itself, even if it IS only amplification. 

 

Powering an LED with signal level... signals seems like a great way to mess up the signals. Looking at some optical transceivers on digikey, they do indeed include an amplifier. 

[WereCatf]

1 minute ago, Sauron said:

Copper is no worse in this sense - not measurably.

Hey, I never implied anything to the contrary, so I don't really know why you're going at me about it. Like I said already in my first post in this thread, HDMI-spec is already designed for copper, so none of this fiber-bullshit is relevant for speed and/or latency.

[WereCatf]

3 minutes ago, corrado33 said:

There MUST be some additional processing inside the cable itself, even if it IS only amplification. 

Why would you think that? All the fibers are inside the cable, there's no light-bleed, so a couple of milliwatts is plenty.

[Glenwing]