Yes, I always thought such a mother-ship unit would be great for portable devices, especially for video cards where even MacBook Pros have limitations due to space and heat issues.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Acer and Asus to Provide Momentum to Thunderbolt Adoption
- Thread starter MacRumors
- Start date
- Sort by reaction score
Yes, I always thought such a mother-ship unit would be great for portable devices, especially for video cards where even MacBook Pros have limitations due to space and heat issues.
Exactly, tech companies don't test the functionality of new products during research and development.
(Disclaimer: this post contains sarcasm.)
But, since there were no TBolt devices on the market, how could any company test TBolt during research and development?
The answer is obvious that no testing could have been done with production devices. Therefore, we're seeing delay after delay, firmware updates, and *still* few if any items actually shipping. The only "in stock" TBolt item at the Apple store is a $50 cable.
(This post contains no sarcasm.)
Agreed. FireWire 800 is great. I am looking forward to Thunderbolt!
Ooops.
Please explain the "oops", if you can.
The other poster wasn't able to point out my errors either.
We simply need to agree to disagree on this one.
Someday, when it actually ships, imagine that I plug a USB mouse into an Apple TBolt Display.
If I look at the device characteristics or the PCIe map - there's no "TBolt mouse" on my system.
It shows up as a "USB mouse".
If I connect a TBolt disk enclosure, I don't see any "TBolt disks" - I see SATA disks connected to a PCIe SATA controller (or virtual disk LUNs connected to a PCIe RAID controller).
When I say that there are no TBolt devices, I mean that a remote root user on the system might not be able to tell if the system has any devices connected via TBolt. To the system, they are all devices connected through PCIe controllers - like all or almost all of the other devices on the system.
"Native" would mean that the device directly connects to the TBolt controller, not through a TBolt -> PCIe bridge -> PCIe controller -> device.
From the system viewpoint, TBolt doesn't exist.
It passes DisplayPort signals through untouched.
It allows PCIe controllers (either slot-based or embedded) that are outside the box to be used, and devices attached to those PCIe controllers to be seen. They are seen as PCIe devices, though, not as TBolt devices.
I can't wait for someone to actually receive a TBolt device and post the PCI mapping for the system.
Last edited:
Year 1 for USB3 was 2008. First usb3 device shipped in 2010. :scratch: Just over a year apparently. Like they say, the more things change the more they stay the same. Facts, people, facts; an important part of reality.
Addendum: Really you were right; I only needed to go back to 2008 to demonstrate TBolt is reasonable and could still succeed.
Last edited:
Wow, it took me a few reads to understand this, but once I envisioned Yoda saying this, it made sense. Thanks for the explanation.
Wirelessly posted (Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_5 like Mac OS X; en-us) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8L1 Safari/6533.18.5)
Good. So long USB, you were great. Tb will go thru 2 or 3 revisions before adding optical and will only get better... If something else doesn't replace it first.
Good. So long USB, you were great. Tb will go thru 2 or 3 revisions before adding optical and will only get better... If something else doesn't replace it first.
Easy, you make some devices to test TBolt and the devices themselves prior to being released.
You have already replied to that post. I have already replied to your initial reply to that post. See below:
Intel doesn't seem to agree with you in relation to your belief that TBolt is not a native external interface for PCIe. See below:
Nope.
But, I said in the next sentence "The answer is obvious that no testing could have been done with production devices."
Any person with any QA sense would know that testing with a couple of hand-built prototypes is not a substitute for testing with multiple off-the-shelf devices. The recent spate of firmware upgrades, and the mysterious delays in shipping TBolt monitors and devices are proof of that.
When I read your slide, I see "native PCIe", not "native TBolt". Intel agrees with me.
Then please describe the TBolt daisy-chain topology if I'm wrong about saying 6 external TBolt controllers in a daisy-chain.
Last edited:
This sentence doesn't seem very ambiguous to me: "Thunderbolt I/O technology allows you to daisy-chain up to six Thunderbolt peripherals"…
http://www.apple.com/thunderbolt/
That said, AidenShaw: I can't believe most of this drivel you're saying here. You seriously don't think that the manufacturers of the upcoming production devices haven't or aren't going to test it themselves? Give me a break.
That's why you introduce computers with ports that no devices exist for: so people can make devices that work with them.
If a hard drive is connected via Firewire 800, it's a Firewire hard drive. It might also have a USB 2.0 port on it, or an eSATA port on it, as mine do, so it's all of the above. If a drive can be connected to a Thunderbolt-equipped computer via a Thunderbolt cable, it's a Thunderbolt hard drive. It doesn't matter what protocol it uses. The interface is what matters.
http://www.lacie.com/products/product.htm?id=10549
Interface: 2 x Thunderbolt 10Gbits
What is the point of insisting otherwise? At best, you're just splitting hairs. At worst… I can't even type it, because I'd get a warning from a mod.
It's Apple that shipped computers without testing TBolt with production devices.
Usually there is a coordination so that there are at least a few "launch partners" with devices ready. Instead, it's mid-September and for all intensive porpoises there are still no TBolt devices available for the computers that Apple starting shipping in February. We are getting a lot of firmware updates, though.
If you want to connect a 3 TB drive to your Apple TBolt display, do you buy a USB drive, a 1394 drive, or a TBolt drive?
Yes, the interface matters.
As I said, I'm looking forward to seeing the PCI maps for these devices - I doubt that the system will see any "Thunderbolt" devices.
Attachments
Last edited:
I feel a disturbance in the Force.
It's like a million engineers screamed out and started beating their heads against the wall at the same time.
I honestly can't tell if you're agreeing with me, or making fun of me!
The engineers say "We can't test it", and the turtlenecked pointy-haired managers say "Ship it!".
Last edited:
I honestly can't tell if you're agreeing with me, or making fun of me!
The engineers say "We can't test it", and the turtlenecked pointy-haired managers say "Ship it!".
By the same logic there can't be devices to test with without some one releasing a device with a host controller. It quickly becomes a chicken and egg thing. This equally applies to USB3.0 or anything new that requires a group of things. Sure USB3.0 devices could test they are backwards compatible with a USB2.0 host or the reverse but in either case one of them has to ship (or at least be shippable) for there to be production equipment to test USB3.0 on USB3.0.
Someone has taken a punt on at least one side of cable or nothing happens.
Apples price point model allows them to use savings as tech gets older to take those sorts of punts. Yet they still have to laid off the risk, and to be really selective about which to back to give them advantage in their markets.
It's not like couldn't test it at all. They could test many parts of the system, and model others to come up with a risk cost. If that is in the budget then it ships. It would make sense then it always has to be the host that ships first as it is a much smaller part of their value equation.
It's not like Apple just took a blind punt, they did have the PA Semi guys after all.
This is great. The more momentum the better. Thunderbolt has a lot of potential... so hopefully we'll be able to use that potential! Now Apple just has to add USB 3 as well... I think the 2 technologies can coexist very well together...
As I said:
It's unusual to ship systems for 7 months with untested silicon.
Earlier I called these "teething pains" - worrisome, but most likely will be fixed over time.
TBolt won't be a "train wreck" unless Apple has to replace the early motherboards with revised silicon.
Wast just looking at the dock for the sony z series, the one which has a external blu-ray burner and a dedicated graphics card in it.
Are there any plans for something similar for mac products? I personally think it's a simple yet amazing idea.
Are there any plans for something similar for mac products? I personally think it's a simple yet amazing idea.
Your statement shown above doesn't define any limit to only TBolt devices.
A daisy chain allows 6 TBolt devices and 1 DP device for a total of 7 devices.
Interesting, you include the limit of only TBolt devices only after being shown that slide which reveals your error.
This doesn't make a difference given that production devices are built to specification of the standard.
Any issues of production devices not functioning properly are due to an error on the part of the device manufacturer not building to the specifications of the standard. This can occur at any time after a standard is released to the public and is just as likely to occur in relation to any standard.
TBolt is the native external interface for PCIe.
I think it is great we are beginning see a wider adaptation of Thunderbolt. I really hope for this port to survive and thrive.
All the USB 3.0 vs Thunderbolt discussion is a bit pointless. These ports are vastly different and should not be compared.
The beauty of Thunderbolt is its ability to work as a docking port. In the era of mobile devices like iPad or MacBook Air, it is a very important aspect. There are a lot of industries out there that require huge processing power (be it GPU, CPU or even sound cards) that is just not present in these mobile devices, forcing people to use 'pro' machines at work, mobile machines for commuting and gaming machines for entertainment. Thunderbolt is the first development I know of that promises to change it. I would love to be able to use a light MacBook Air when I commute, then plug it via Thunderbolt into an OpenCL compatible graphics card at work, daisy chained to a RAID matrix, allowing me to do heavy duty processing work required for my job, and then when I get back home plug it into a good dedicated graphics card daisy chained to a large LCD for gaming purposes.
All the USB 3.0 vs Thunderbolt discussion is a bit pointless. These ports are vastly different and should not be compared.
The beauty of Thunderbolt is its ability to work as a docking port. In the era of mobile devices like iPad or MacBook Air, it is a very important aspect. There are a lot of industries out there that require huge processing power (be it GPU, CPU or even sound cards) that is just not present in these mobile devices, forcing people to use 'pro' machines at work, mobile machines for commuting and gaming machines for entertainment. Thunderbolt is the first development I know of that promises to change it. I would love to be able to use a light MacBook Air when I commute, then plug it via Thunderbolt into an OpenCL compatible graphics card at work, daisy chained to a RAID matrix, allowing me to do heavy duty processing work required for my job, and then when I get back home plug it into a good dedicated graphics card daisy chained to a large LCD for gaming purposes.