Saturday, June 7th 2008

NVIDIA and AMD Vie with Intel Over USB 3.0

First-tier makers NVIDIA and AMD are dissatisfied with Intel's ownership of the USB 3.0 specification, CNET reported yesterday. The next-generation high-speed USB 3.0 specification is aiming to become standard in 2009. It will be 10 times faster than USB 2.0 and at the same time as widespread as the current USB 2.0. The USB 3.0 will also retain full compatibility with USB 2.0 and presumably USB 1.0. The main investitors behind the USB 3.0 specs include Intel as well as Microsoft, Hewlett-Packard, Texas Instruments, NEC, and NXP Semiconductors. "The challenge is that Intel is not...giving the specification to anybody that competes with CPUs and chipsets," said a source close to AMD who is familiar with the dispute. As a result AMD, NVIDIA, VIA Technologies and probably others could be driven to create their own USB 3.0 specification that may lead to a lot of frustration and money lost on development. "We are going to be forced to create a secondary specification ... new open host controller standard for USB 3.0." the AMD source added. "They could spend the time, engineers and money developing their own host controller spec," an Intel source said. "In the past they have chosen to let us do the work and then benefit from the fruit of our labor." Continue reading the full story here.
Source: CNET News
Add your own comment

43 Comments on NVIDIA and AMD Vie with Intel Over USB 3.0

#1
OzzmanFloyd120
Lame, just Microsoft and Intel being greedy again.
Posted on Reply
#2
Wile E
Power User
Meh, just push 1394b (Firewire 800) instead. It will probably perform better anyway. lol.
Posted on Reply
#4
Megasty
Just how incompetent can MS & Intel get. If the others do make their own 3.0 spec then it'll just serve those 2 BOECs right.
Posted on Reply
#5
1c3d0g
Damn it. I don't want this to happen. Fragmentation of something as important as USB 3.0 must not be allowed. It's exactly this type of splintering that has divided the Linux community and caused them to get stuck in a never-ending battle, for instance (.rpm, .deb, GNOME/KDE, etc. etc.), or the stupid HD format wars (Blu-ray/HD-DVD). :mad: If everyone worked together and played nice with each other this crap could be avoided.
Posted on Reply
#6
beyond_amusia
Well, this will kill Firewire off for good unless Apple feels a needs to out do the 5 gigabits USB 3 is going to bring.

What's interesting is that AMD allows Crossfire to run on Intel platforms, but Intel is going to be a stinker about this... I do hope that AMD does not rush into making another USB spec though, because they will be able to get Intel's eventually, and it will take a while for USB 3 to replace USB 2, so I imagine everyone will have it before it's even replaced USB 2.

I am personally looking forward to getting a PCI=E USB 3 card for my tower and a USB 3 thumb drive for ReadyBoost. =)
Posted on Reply
#7
OzzmanFloyd120
1c3d0gDamn it. I don't want this to happen. Fragmentation of something as important as USB 3.0 must not be allowed. It's exactly this type of splintering that has divided the Linux community and caused them to get stuck in a never-ending battle, for instance (.rpm, .deb, GNOME/KDE, etc. etc.), or the stupid HD format wars (Blu-ray/HD-DVD). :mad: If everyone worked together and played nice with each other this crap could be avoided.
That will never happen because there's too much money involved.
Posted on Reply
#8
beyond_amusia
1c3d0gDamn it. I don't want this to happen. Fragmentation of something as important as USB 3.0 must not be allowed. It's exactly this type of splintering that has divided the Linux community and caused them to get stuck in a never-ending battle, for instance (.rpm, .deb, GNOME/KDE, etc. etc.), or the stupid HD format wars (Blu-ray/HD-DVD). :mad: If everyone worked together and played nice with each other this crap could be avoided.
On the flip side, there would be no compition, but then again, to argue your point, look at DVD. The DVD standard is the result of merging the best specs from 2 initial DVD formats into one.

Now DVD -R and +R are another story... =/
Posted on Reply
#9
Wile E
Power User
beyond_amusiaWell, this will kill Firewire off for good unless Apple feels a needs to out do the 5 gigabits USB 3 is going to bring.

What's interesting is that AMD allows Crossfire to run on Intel platforms, but Intel is going to be a stinker about this... I do hope that AMD does not rush into making another USB spec though, because they will be able to get Intel's eventually, and it will take a while for USB 3 to replace USB 2, so I imagine everyone will have it before it's even replaced USB 2.

I am personally looking forward to getting a PCI=E USB 3 card for my tower and a USB 3 thumb drive for ReadyBoost. =)
It's a proposed 5Gb. USB 2.0 is is rated at 480Mb, yet Firewire 400 shames it. I doubt USB3 is gonna perform anywhere near their claims. That would make it theoretically faster than SATA II. Not gonna happen.
Posted on Reply
#10
OzzmanFloyd120
Also, for it to perform at those speeds, wouldn't the USB have to be linked right to the NB instead of being routed though the SB?
Posted on Reply
#11
beyond_amusia
Wile EIt's a proposed 5Gb. USB 2.0 is is rated at 480Mb, yet Firewire 400 shames it. I doubt USB3 is gonna perform anywhere near their claims. That would make it theoretically faster than SATA II. Not gonna happen.
But most people that use Firewire also use SCSI drives, so naturally it will seem faster than USB 2.0

Oddly enough, USB was first made to kill off PS/2, and then they realizedthat Apple was really greedy by charging per Firewire controller, so the USB people decided they neededto break some chains. If you noticed, most of Apple's products now lack Firewire support.
Posted on Reply
#12
beyond_amusia
OzzmanFloyd120Also, for it to perform at those speeds, wouldn't the USB have to be linked right to the NB instead of being routed though the SB?
It'll use the PCI interface, and I imagine Intel's QuickPath will play a role as well to achive those speeds, but until hard disks catch up we will not ever use all that speed in a normal situation.
Posted on Reply
#13
Wile E
Power User
beyond_amusiaBut most people that use Firewire also use SCSI drives, so naturally it will seem faster than USB 2.0

Oddly enough, USB was first made to kill off PS/2, and then they realizedthat Apple was really greedy by charging per Firewire controller, so the USB people decided they neededto break some chains. If you noticed, most of Apple's products now lack Firewire support.
No, it's been proven time and time again, Firewire 400 flat out outperforms USB 2.0. SCSI has nothing to do with it. Neither interface even has enough bandwidth to max out modern IDE drives, let alone SCSI.
Posted on Reply
#14
beyond_amusia
Wile ENo, it's been proven time and time again, Firewire 400 flat out outperforms USB 2.0. SCSI has nothing to do with it. Neither interface even has enough bandwidth to max out modern IDE drives, let alone SCSI.
Not even the upcoming USB 3? =P I forgot how to convert mb/s into MB/s, otherwise I'd do the math.

And yes, I read up on it and you are right.

Maybe USB 2 has more latency than Firewire 400?
Posted on Reply
#15
Wile E
Power User
beyond_amusiaNot even the upcoming USB 3? =P I forgot how to convert mb/s into MB/s, otherwise I'd do the math.

And yes, I read up on it and you are right.

Maybe USB 2 has more latency than Firewire 400?
The difference is Firewire is a true streaming protocol, whereas USB is a "burst" (can't think of the proper term) protocol.
Posted on Reply
#16
beyond_amusia
USB: speak only when spoken to, vs. Firewire: all devices freely communicate with each other.
Posted on Reply
#17
Wile E
Power User
beyond_amusiaUSB: speak only when spoken to, vs. Firewire: all devices freely communicate with each other.
Sort of, but not really. lol. USB communicates in bursts. It doesn't stream the data constantly. It sends a little, then stops, then sends a little, then stops, etc., etc.

Makes an especially noticeable difference when dealing with large chunks of data. Whereas, when you move something over the firewire interface, it constantly streams the data, without the pauses.
Posted on Reply
#18
WarEagleAU
Bird of Prey
Wow, I didnt know that. I thought USB 2.0 was faster than firewire. Learn something new everyday.
Posted on Reply
#19
Darknova
WarEagleAUWow, I didnt know that. I thought USB 2.0 was faster than firewire. Learn something new everyday.
You don't say...

Still, as great as firewire seems to be, USB is number one with consumers (guess it's kinda like VHS winning over Betmax) and this kind of segregation will do nothing for the industry as a whole.

Imagine have USB 3a and USB 3b, and having to have all things compatible with each, or having to look out for devices specifically for each spec.

It will cause chaos with Joe Public.
Posted on Reply
#20
spacejunky
OzzmanFloyd120Lame, just Microsoft and Intel being greedy again.
Why are you dragging Microsoft into this? Just because they have to work with the new spec to make sure drivers will work on their OS? Why not blame TI as well? This is all Intel.

Don't forget Apple uses Intel boards and CPUs so you can't count on them for Firewire support anymore. They'll use what ever Intel gives them the best deal on.

You can nderstand Intel's frustration in making these specs and developing a new standard and not realizing a proprietized gain from it but this is the business they're in and what made them great are the open standards that separated them from the likes of Apple hardware back in the day.


It is looking like another mess to add to the non-sli Intel boards.
Posted on Reply
#21
candle_86
actully AMD doesn't have to do anything here, Intel has tried this before and AMD took them to court and won. All AMD has to do at most is a direct reverse engineer
Posted on Reply
#22
OzzmanFloyd120
True, but if there are two different technology standards then companies making USB peripherals will be forced to do more work to make their products, hence raising prices.
At least that's how I see it.
Posted on Reply
#23
KBD
god forbid AMD and Intel work together on this, but no they have tranfer their little squabble into this too and the consumer will suffer in the end.
Posted on Reply
#24
driver66
So let me understand this. Intel develops the standard and is just supposed to give it away?:shadedshu I really am not understanding your guys logic on this one :confused:
It is Intel's IP ........ they ae supposed to just share it with everyone??............
explain plz :toast:
Posted on Reply
Add your own comment
Dec 27th, 2024 18:02 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts