I like the touch/trackpad idea - reinvent the mouse but it's not really a transition although HP have rushed out the SmartTouch which smacks of "we got there first"
I'm interested to see what a touch interface would look like on a computer (as opposed to a phone or a tablet). I have a hard time imagining that it would be practical and would work well. Still, if anyone can make a good job of it I suppose that Apple can.
I ought to be clear about my thoughts on Apple made chipsets... the timing is wrong. With the CPU interface about to change radically in the next generation of Intel chips, and with OpenCL/GrandCentral not set to arrive until mid-09 the timing of this just doesn't sit well. If they were to do it I don't think marketing would be a stumbling block, and that certainly wouldn't stop them from doing it. So I don't really know what to expect in September.
I am very much liking the vector chip idea, as this would fit into all new computers, and possibly all new devices, while allowing scaling. iMac only has 1 or 2 of these chips while the Power Mac has 4-8. This would also fit with Snow Leopard. This would be something competition could not match, and would put Apple years ahead of everyone, in every market. Apple seems to do things in broad stokes, even if it only offers technology, such as touch, on limited devices at first.
Thanks for the read Programmer
When you look at a specialized vector engine as being for the only for the Macs I think you are being way to limiting. I'm taking Apples comments to heart here and enjoying the thought of a highly developed SOC for the pocket devices. OK so maybe they haven't gone into detail as to what those PA chips will contain.
The thing is we can imagine! Look at the video decoder chip as a starting point for an optimized SOC. Now imagine you are Apple putting all this functionality on one chip. Do you go the route of a specialized function block for video decode or do you take all your previous AltVec experience and design a new somewhat more generalized co processor? I know if it was me and I had access to some of the best low power tech going I'd go for a processor I could leverage in a number of ways.
Think about it this way a loosely coupled DSP/Vector Processor/Media Engine offers up multiple opportunities for Apple. It can provide for signal processing for communications, acceleration of video play back, enhancement to graphics and a number of other features.
We are not talking a lot of power here in the watts sense either. Each vector unit on Cell isn't extremely power hungery and that is on a IBM built device. Given that the feature set and speed would be markedly different and the nature of SOC I could easily see such a co processor going into these new devices. The big question would be the number of cores. All Apple really needs to do is to come up with a processor that doesn't burden the system any more than the current external video decoder. Correctly implemented this could give Apple that edge others couldn't touch. At least not very quickly.
So what is this idea I'm throwing in the ring that might be that secrete sauce that Apple has hinted at? Well I'm thinking a vector engine some what similar in concept to those that Cell employs but grafted onto an ARM core. The cores would be just a bit more capable and might be able to do their own I/O on a local bus. Don't think of these as the massive floating point processors that are the vector units in Cell but more as an array of specialized DSPs.
I like this idea too... if future macbooks INCLUDED an iPod touch that just snaps into the bottom of the laptop where the touchpad normally sits and becomes the trackpad when the laptop is turned on, and can be charged along with the laptop battery when the computer is plugged in. I like where this idea is headed...
Apple, are you listening?
OR included in the macbook was an iphone 3G that does all the above, plus uses the data and GPS capabilities of the iphone 3G from any program running on the mac! Holy crap that would be awesome!
That has to be one of the dumbest things I ever read.
Who would want a laptop with a big hole where the trackpad is supposed to be when their iPod is in use somewhere else?
Not to mention the computer would have to be quite thick to accommodate such a thing, and it would be damn near impossible to make it "flush" with the system or keep it steadily in place so that it does not move at all, which would create a very uncomfortable experience.
It would also increase wear on the computer and the ipod from all that taking in and out, especially if the iPod were used often.
Oh, and lastly, it would just look tacky. Definitely a step down from Apple's polished/professional look and feel.
Boy do I hope that was a sarcastic post :-)
Anyways, I think a new chipset to lower power consumption and maybe increase the capabilities of the integrated graphics on the regular Macbook so that it can at least play SOME games (even some older ones) would be nice.
Comments
I like the touch/trackpad idea - reinvent the mouse but it's not really a transition although HP have rushed out the SmartTouch which smacks of "we got there first"
I'm interested to see what a touch interface would look like on a computer (as opposed to a phone or a tablet). I have a hard time imagining that it would be practical and would work well. Still, if anyone can make a good job of it I suppose that Apple can.
I ought to be clear about my thoughts on Apple made chipsets... the timing is wrong. With the CPU interface about to change radically in the next generation of Intel chips, and with OpenCL/GrandCentral not set to arrive until mid-09 the timing of this just doesn't sit well. If they were to do it I don't think marketing would be a stumbling block, and that certainly wouldn't stop them from doing it. So I don't really know what to expect in September.
I am very much liking the vector chip idea, as this would fit into all new computers, and possibly all new devices, while allowing scaling. iMac only has 1 or 2 of these chips while the Power Mac has 4-8. This would also fit with Snow Leopard. This would be something competition could not match, and would put Apple years ahead of everyone, in every market. Apple seems to do things in broad stokes, even if it only offers technology, such as touch, on limited devices at first.
Thanks for the read Programmer
When you look at a specialized vector engine as being for the only for the Macs I think you are being way to limiting. I'm taking Apples comments to heart here and enjoying the thought of a highly developed SOC for the pocket devices. OK so maybe they haven't gone into detail as to what those PA chips will contain.
The thing is we can imagine! Look at the video decoder chip as a starting point for an optimized SOC. Now imagine you are Apple putting all this functionality on one chip. Do you go the route of a specialized function block for video decode or do you take all your previous AltVec experience and design a new somewhat more generalized co processor? I know if it was me and I had access to some of the best low power tech going I'd go for a processor I could leverage in a number of ways.
Think about it this way a loosely coupled DSP/Vector Processor/Media Engine offers up multiple opportunities for Apple. It can provide for signal processing for communications, acceleration of video play back, enhancement to graphics and a number of other features.
We are not talking a lot of power here in the watts sense either. Each vector unit on Cell isn't extremely power hungery and that is on a IBM built device. Given that the feature set and speed would be markedly different and the nature of SOC I could easily see such a co processor going into these new devices. The big question would be the number of cores. All Apple really needs to do is to come up with a processor that doesn't burden the system any more than the current external video decoder. Correctly implemented this could give Apple that edge others couldn't touch. At least not very quickly.
So what is this idea I'm throwing in the ring that might be that secrete sauce that Apple has hinted at? Well I'm thinking a vector engine some what similar in concept to those that Cell employs but grafted onto an ARM core. The cores would be just a bit more capable and might be able to do their own I/O on a local bus. Don't think of these as the massive floating point processors that are the vector units in Cell but more as an array of specialized DSPs.
Dave
That's exactly what we have been talking about.
I like this idea too... if future macbooks INCLUDED an iPod touch that just snaps into the bottom of the laptop where the touchpad normally sits and becomes the trackpad when the laptop is turned on, and can be charged along with the laptop battery when the computer is plugged in. I like where this idea is headed...
Apple, are you listening?
OR included in the macbook was an iphone 3G that does all the above, plus uses the data and GPS capabilities of the iphone 3G from any program running on the mac! Holy crap that would be awesome!
That has to be one of the dumbest things I ever read.
Who would want a laptop with a big hole where the trackpad is supposed to be when their iPod is in use somewhere else?
Not to mention the computer would have to be quite thick to accommodate such a thing, and it would be damn near impossible to make it "flush" with the system or keep it steadily in place so that it does not move at all, which would create a very uncomfortable experience.
It would also increase wear on the computer and the ipod from all that taking in and out, especially if the iPod were used often.
Oh, and lastly, it would just look tacky. Definitely a step down from Apple's polished/professional look and feel.
Boy do I hope that was a sarcastic post :-)
Anyways, I think a new chipset to lower power consumption and maybe increase the capabilities of the integrated graphics on the regular Macbook so that it can at least play SOME games (even some older ones) would be nice.