Since when did a resistor, diode or capacitor
become "a chip"?
I count no more than 3 chips at each end, the rest of the devices in the photograph are dirt cheap passive components.
Want to know why Apple is charging £39 - $50 in the US - for its Thunderbolt cable? For once it's not entirely the company's price-high-for-fanboys policy - the cable is an active transmitter. Apple's Thunderbolt cable Apple's Thunderbolt cable: on the outside... The kit disassemblers at iFixit.com have found that the …
But I'll bet that those chips cost pennies to make, probably on a par with the copper in the cable. Assuming it is copper cored of course.
Although to be fair, It's probably better value for money than some of the 'solder some connectors on the ends of a length of copper' hi-fi interconnect manufacturers...
...the port can also handle forthcoming optical cables - and the "gubbins" for sending and receiving the optical signal is also in the cable. If the gubbins were in the port the port would not be compatible with optical cables and you'd be whining about that instead.
Read up on the subject.
And even further in the future Intel are planning to turn the port into optical only, so read deeper into the subject before mocking - the port is already doomed to obsolescence.
The port can't natively support optical cables - that's why the gubbins will exist (and had been explicitly stated in the case of optical) but there is no reason the first thunderbolt interface couldn't be natively copper with the expected gubbins in the future optical cable.
It's also easily possible for a port to be both - check out mini-TOSLINK 3.5mm jacks that already exist in Apple kit. There's no reason Thunderbolt couldn't have taken a similar approach.
They take one serial protocol (PCI express) and another (Display Port - though unidirectional and four(?) lane) and add in a bunch of expensive chips just to save on a few extra pairs of wires and no increase in cable length. And a requirement for power. And the connector is already multi pole.
That makes no sense, even for Apple volumes.
If a new protocol isn't reliable enough to run over 2 metres of cable without expensive circuitry then the thing is broken. That's not surprising seeing as Light Peak was supposed to use fibreoptic cable originally and ended up using copper and was gimped in other ways. Presumably if the system had been optical as promised the cable could have been 10 metres in length and still worked.
Apple shouldn't have included the tech at all in its current state.
You're talking out of your arse. Thunderbolt is implemented over copper cable, not optical cable. The tech was gimped before it was released and what was released is copper only. It may well be that when Thunderbolt 2 turns up it will finally offer optical but that will be zero use to people stuck with the copper only version 1. Which is the case for every existing Mac implementation.
No, actually you are.
Thunderbolt was as "gimped" as USB3 (remember how everyone said it would come as optical too) and even DisplayPort (VESA's been talking about optical DisplayPort since 2007), trouble is when it comes to getting ports out it's always too expensive so it gets dropped time and time again.
Thunderbolt is also not the first electrical protocol that needs active cables, DisplayPort actually already supported it. It makes sense for thin wires and thin connectors.
When optical Thunderbolt comes around it's quite likely there will be a way of connecting that to then legacy devices and ports.
I've followed Light Peak long before it was released and was extremely optimistic for the tech. It was precisely because USB3.0 was such a fudge of extra pins and wires that I was hoping Light Peak would become the natural successor. It was just such a clean concept where it could even act as a unified bus inside PCs with legacy protocols like USB / SATA etc. all travelling over fibre instead of a mess of cables like now.
And the clue of how it was supposed to work is in its codename - Light Peak. Light. It was always meant to be optical with the potential for insane data transfer rates over long and short distances. Then suddenly it turns up in copper and is rebranded Thunderbolt. Not only that but as we can see it's been gimped, and requires expensive cables.
"When optical Thunderbolt comes around it's quite likely there will be a way of connecting that to then legacy devices and ports."
Yeah, a horrifically expensive hub or dongle.
One thing is certain, in it's present incarnation this thing is not going to replace USB 3.0. Maybe some niche users will find a purpose for Thunderbolt as it is but niches don't drive mainstream acceptance as we can see from Firewire.
The reason they went away from lasers? They couldn't get the reliability they wanted and still be cheap. While high-bitrate optical setups exist, they're all industrial-grade, meant for backhaul (translation: expensive as hell). Intel's been working on laser multiplexing, lasers in ICs, and other tech meant for Light Peak, but they've been slow going. I would imagine Intel's trying to steer people away from USB before it develops an all-optical USB4 that could take the shine away from any future true-optical solution.
That would be LEDs, not Lasers. Optical links usually run on LEDs, unless you're going distance.
Multimode = LED lit (usually)
Monomode = Laser lit (usually)
You may have a point, cost wise, FC ports for instance do cost an arm and a leg but LEDs are much cheaper and you really don't need a Laser unless you're going over a long distance.
Maybe in the future it would be possible to replace the 2m of copper with a longer fibre, just change the chippery for suitable alternatives. The connection to the pooter would remain electrical so there wouldn't be any need to change the port.
Also aren't electrical connections more resistant to dirt etc than optical ones.
Clearly the author doesn't have experience buying Fibre Channel and external SCSI/SAS cables, both of which are more expensive and usually come in speeds slower than this.
The problem is the bidirectional bit, it's much easier to just stuff the bits all to one direction, but when you have them going back and forth at the same time it makes everything much harder, at these speeds it's an nightmare.
As for optical fiber, yeah would have been good, but would anyone then buy the ports at 50x times the cost? (even then the optical cables wouldn't be that much cheaper either)
Question is will El Reg be reviewing this cable as they did for the iPad HDMI?
Considering PC world have a 3M ethernet cable (HP branded) @ £44.99 (real copper though) I don't think the apple cable is that poor a price.
Lets see people we have the following taxes to apply to this cable
The 'early adaptor tax'
The 'no other supplier tax'
The 'apple tax'
It's a reasonably fast bus and it's in the early stages why wouldn't it be expensive at the minute.
Have any of you tried to purchase a decent quality USB OTG (Mini A in my case) cable at a reasonable price that on the whole will be a thoroughly unreasonable price.
Biting the hand that feeds IT © 1998–2019