Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
There is *no* optical version of TBolt

The optical versions of TBolt will probably multiplex DP and the PCI-e data and then demultiplex it back out before it hits the controller. Obviously there are not 4 separate sets of wires inside of that cable. :)

Note that there is *no* optical version of TBolt specified.

In the current spec there are hints (but no existing hardware) about copper cables that have Cu<->fibre transceivers in each connector, so that the copper signals are transcoded to optical fibres at one end, then transcoded back to copper signals on the other end.

As far as any of the devices are concerned, it's a copper cable (but it's able to go longer distances than copper allows).

And this won't be "optical TBolt", it will be copper TBolt with very expensive cables that can be longer than standard cables. It will have no advantages over copper cables other than the ability to have longer cables.
 
Wirelessly posted (Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_4 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8K2 Safari/6533.18.5)

crw said:
I hope someone who is a bit more savvy with i/o technology can explain clearly how TB would be able to run two 27" ACD monitors without completely hogging all of the bandwidth available to a TB bus, especially with a new or next generation MBP, which will have only one GPU and no ability to add another.

Doesn't Apple's Thunderbolt implementation multiplex in the PCIe 10Gb bus with a DisplayPort signal? That is why a standard DisplayPort monitor can be used if it is the last in a TB chain -- the last TB device pulls out the DisplayPort signal and passes it along. If this is indeed the case then the first display would get its video via a signal on the Thunderbolt bus and the second display could get the DisplayPort signal. This would also imply that the displays can continue to accept a standard DisplayPort signal from non-TB systems as the second display would be getting straight DP.

So it sounds like a thunderbolt cinema *will* work with a display port equipped mac? If the display also has USB ports on it, could display port carry that? Or as with the current cinema, do you need a USB cable coming off the cinema display to the mac? I have a 2010 MacBook pro with display port.
 
where is the 3200x2000 monitor?? that should be out together with os x lion. i will be very mad if there is no such monitor.

God, I'd hate to imagine how expensive that'd be, especially since the current one is already a thousand dollars.
 
God, I'd hate to imagine how expensive that'd be, especially since the current one is already a thousand dollars.

Barco's Coronis Fusion 6MP DL is a 30" 3280x2048 display and you can get them for just over $15,000. The 42"-64" 3840x2160 displays that are around weren't really useful for editing or working on due to being more like a TV and are in excess of $50,000. Eizo have a 36" 4096x2160 monitor now, but that is over $35,000. That is the closet thing to a real step up from the traditional 2560x1600 30" IMO.

There just isn't the demand to bring these resolutions to the desktop for a few thousand dollars.
 
Last edited:
Sold my two year old cinema display a week or two ago and got back 80% of what I paid for it. Bought a 23" Dell IPS and the best part about it is that I'm no longer confined to a single connector that isn't used by the majority of hardware manufacturers.

Put the other £300 I got from the sale towards an iPad.

Oh, and I LOVE having a matte display again :D

I have a 23" cinema from 2006. I love it. I'll be sad when I have to replace it with NEC, Dell or some other non apple brand. That 27 glossy is just not for me.
 
Note that there is *no* optical version of TBolt specified.

" Developed by Intel (under the code name Light Peak),
...
... Electrical or optical cables ... "
http://www.intel.com/technology/io/thunderbolt/index.htm

There is no optical version deployed in production. The notion that that there is no optical component in a technology previously nicknamed "Lightpeak" is farcical.


so that the copper signals are transcoded to optical fibres at one end, then transcoded back to copper signals on the other end.

Since the rest of the chips inside the computer are connected by copper at some point, this is going to happen. So the light peak demos had this inside the computer on a card. The TB solutions are pushing it out to the electronics in the cable. The issue is that the copper path is going to be kept to single digit number of inches or less.

A pragmatic design issue with putting the transceiver into the cable connector housing is that there isn't going to be room for 2 or 3 transceivers. (at least for a while longer). A 30-50Gbps single optical thread can easily multiplex two 10Gbps signals into a single stream and back out to multiple copper streams on the other end. Similarly, a future 50Gbps TB controller will take multiple copper PCI-e lanes in the computer and turn then into a a fewer number of copper lanes to the connector , where they may or may not be multiplexed onto a single optical stand.


It will have no advantages over copper cables other than the ability to have longer cables.

"... A new technology was announced at Intel Developer Forum (IDF) which provides initial data rates of 10 Gigabits and potential scalability to 100 Gigabits and beyond; something copper IO won’t be able to achieve. ..."
http://blogs.intel.com/technology/2009/09/lighting_fast_-_high_speed_opt.php

You may be able to run copper 5-12 inches at 100Gbps still pass FCC Class B, but is that a practical external connectivity I/O solution when others run 3-6+ft or better? There are EM spectrum leakage/interference advantages too. A simple way to blow away a Mac Pro's FCC Class B pass is to stick a very high end copper I/O card into it ( 100Gb copper ethernet).

Going optical gives you a lower cost way of getting rid problems by not adding to the problem (and then having to work around it).

It is more than just length.

http://www.zurich.ibm.com/st/photonics/interconnects.html


Plus the economic factor of whether the devices can be shipped off to cheap Chinese chop shops for production. That also played a big factor in the backslide of Light Peak into Thunderbird. Instead of innovative fiber, a copper solution that can be farmed out to cheap labor. The copper cables cost $50. Optical ones would have been higher still.
 

So it sounds like a thunderbolt cinema *will* work with a display port equipped mac?


This is dependent on how complete the backward compatibilities are of the TB controller chip. However, in pure Display Port backward compatible mode you loose the additional features of TB.


If the display also has USB ports on it, could display port carry that?

No they are not part of the Display Port spec. There is video and optional audio traffic only.



Or as with the current cinema, do you need a USB cable coming off the cinema display to the mac?

Again apple could put in a kluge to make it work but likely will increase costs. With TB running PCI-e out to the monitor the USB device in the monitor is a host controller (e.g., it is like you place an addition USB PCI-e card into the computer if there was a PCI-e slot). The host controller is at the "center" of the USB star-tiered chain.

What is inside of most monitors is a USB hub. A hub is not at the center of a star-tiered chain. What they would need is some "config on power up" logic in the USB controller/hub or some way of toggling it externally to put the USB handler into a different mode when it can't find a PCI-e connection to a host.

The simpler thing to do would be to just have a host controller and set the webcam, usb socket, etc. go dormant if you don't plug into a TB connector. Same issue as if you had only plugged in 2 of the three leads coming off the older display cable ( leaving USB unplugged). Display Post is not forward compatible with TB. Jury rigging things so that it looks like it is will likely leads to causing as many problems in the long term (miss set expectations ) as it solves in the short term ( appears to do more than it does ). Want the ports to work? ... use TB cable to TB ports. If don't use TB plugs/connectors that is the same thing as removing USB PCI-e card from your computer. You would not expect it to work anymore.
 
Last edited:
Originally Posted by AidenShaw
Note that there is *no* optical version of TBolt specified.

There is no optical version deployed in production.

We agree completely.

There is no "specification" for an optical connection for the current TBolt implementation. The only connector "specified" is the copper mini DisplayPort connector.

That's all I was saying - sometimes things are unclear when the vaporware "Thunderbolt optical" cables are discussed. People want them to be something that they won't be.

The rest of your post described things that Intel has demoed, and speculation/predictions of future enhancements. That will wait for Thunderbolt 2.0 and Thunderbolt 3.0, and new Apples with the new ports.
 
Last edited:
Makes sense but the picture shows a MacBook Pro connected to the Cinema Display with two connections to the MacBook Pro, just like the current model.

I think the problem with the website is they posted the images showing Lion on the display a week in advance. I doubt there is a new Cinema Display at all and there will be a new MacBook as originally predicted.

The image shows only the power and thunderbolt connections to the macbook pro. Thus, the display, isight, speakers, microphone, and usb hub will all be connected via the single Thunderbolt connection. i love you apple.
 
I am also curious about this. Any thoughts on whether the impending 27" Thunderbolt Cinema display will work with a 2010 Displayport equipped MacBook Pro?

If there are ports on the back of the display, such as USB, would those work as well?

I assume it will only function as minidisplayport
 
I guess the question is if the new displays draw power via the Thunderbolt interface. If so then ordinary mDP ports wouldn't work with them.

The images show no cables other than those between the computers and the monitors, so one might suspect this is the case. Unless Apple puts an ordinary power cord connector on the back of the monitor in addition to the Thunderbolt connector, of course.
 
I guess the question is if the new displays draw power via the Thunderbolt interface. If so then ordinary mDP ports wouldn't work with them.

The images show no cables other than those between the computers and the monitors, so one might suspect this is the case. Unless Apple puts an ordinary power cord connector on the back of the monitor in addition to the Thunderbolt connector, of course.

The monitor has its own power cord, just like today.

TBolt only supplies 10 watts max and shared, far too small for a monitor, and not even enough for a 4 port USB hub. The current LED monitor draws up to 250w - and it supplies power to the laptop, not vice-versa.
 
Last edited:
A sudden lull in rumors... is it the calm before the storm?
Or just the calm before the reeeeeally long calm before a little drizzle?
 
or perhaps weirder

A sudden lull in rumors... is it the calm before the storm?
Or just the calm before the reeeeeally long calm before a little drizzle?

Maybe Apple is thinking differently ;)

cloudy-meatballs.jpg
 
The monitor has its own power cord, just like today.

TBolt only supplies 10 watts max and shared, far too small for a monitor, and not even enough for a 4 port USB hub. The current LED monitor draws up to 250w - and it supplies power to the laptop, not vice-versa.

Heh, silly me.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.