This is where I believe Samsungs long term goals lay: http://www.edn.com/article/512298-AR...p_to_14_nm.php. That is 14nm hardware. Of course there will be nodes in between. The question for us is will Apple push hard to skip 32 nm to move to 22nm. The performance and power usage considerations are huge and for the near future would allow for some incredible hand held devices.
As a complete distraction one should look at all the job opening for engineers and technicians in up state NY. Either for Global Foundries or the ISDA.
As to Samsung here is Ana talking away: http://www10.edacafe.com/video/Samsu...185/media.html. She has obviously mastered the art of answering questions without saying much. To highlight just now much this is breaking news here is another tidbit: http://www.voicesatsamsungsemiconduc...gy-forum-2011/ In any event consider dinning at http://www.edacafe.com/ if your hunger for process tech is strong. As far as Samsung it appears to have 32/28 nm ready to roll, thus my strong suspicion that Apple will deliver a new SoC that will likely more than double performance in a very similar power profile. some of the numbers I've seen for shrinks to these nodes are very impressive.
Obviously considering Apples volume demands, it would not be uncommon for the companies to look at this transition as a partnership. Thus I would not be surprised at all to find out that Apple is throwing cash Samsungs way to secure capacity at this node and future nodes. I have a hard time believing that the entire 3.9 billion will go Samsungs way though. Previous release indicate that a few hundred million can take care of things like Flash and displays. In the end it is probably a safe way for Apple to secure production without building a fab itself.
The 32/28 nm processes are ready to go!!! Or exactly "Samsung?s 32/28nm Processes for SoCs are Ready to Go". Note the obvious reference to SoCs.
20% Higher Performance
30% Lower Dynamic Power
50% Less Leakage Power
What is very notable is that no matter where you look there is never ever a mention of Apple anywhere on these sites never. Apples secrecy mentality has yet to soften. In any event these numbers lead me to believe that iPad2 could be running dual core with both the GPU and the CPU in the next rev with only a minor increase in power which at the same time can be offset in other ways. Things are looking good.
If this is all true it would certainly lend credibility to the recent stories of dual core ipads with dual core gpus. 2048x1536 displays would still be a possibility if it's also true that the power expense comes from the backlight and not the pixel density, which is independent of the pixels.
It would seem that 14nm is a couple-few years off from being a mass market reality, at least there would seem to be a clear path for future hardware and room for it to grow. The innovation is clearly in the mobile space and not in the desktop space these days. It would appear that we will very shortly be running mobile OS's and hardware that rivals today's desktops and notebooks in just about every facet. Probably also within 3 years.
At the moment I'm of the opinion that the earlier rumors of the upcoming iPad 2 are true, and it really will be an across the board quadrupling of power (cpu, gpu, pixel density) with additional features that are expected and a few unexpected. The Fireball 'tamp down' counter-rumors are just trying to make sure that the iPad2 announcement, when it happens, will still be able to wow us.
I wonder if it could be solid state drives, ala the Macbook air?
People generally seem to be raving about how fast the macbook air is too boot and run the Air normally shows us where Apple are taking the mac.
E.g. using the aluminium body first debuted on the Air as did the lack of CD drive. Perhaps the next macs will also use SSD?
Just a guess but it could make sense considering how expensive SSD's currently are. By investing in a dedicated production line, they could bring down the cost and secure supply. Also, they way they use SSD in the Air is not the same as normal SSD's, which normally are encased in one unit.
We'll see...
I'll go you one further... if Apple can get the cost on SSD down far enough, and also manage to convince people that disc drives are silly and antiquated (part way there, mac app store helps) then they can effectively eliminate the MBP line as we know it. I think if Apple has their way future MBP notebooks will look more like the MBA than how the MBP looks today.
They obviously can't do this while the MBA still has to make some compromises. Someday however they'll be able to make a high end MBA with 15 or 17" screen, 1TB SSD, no disc drive (because the demand will die off) and sufficient processing power, while having this new machine maintain a more MBA style weight and enclosure.
If they can do that sooner than later, then yes there's no reason they couldn't move the entire line over to SSD (except the minority of pro stuff that wants truly large storage). In fact we would all be using SSD, and wax poetic about the days of spinning disc hard drives and how silly and antiquated that technology was just like how we think of the floppy disk and how at one point those were so critical to our daily lives. Anything not solid state will be looked down on by Zoll as old, retarded stuff his dad used to use (you mean this has to 'spin up'?).
If this is all true it would certainly lend credibility to the recent stories of dual core ipads with dual core gpus. 2048x1536 displays would still be a possibility if it's also true that the power expense comes from the backlight and not the pixel density, which is independent of the pixels.
It would seem that 14nm is a couple-few years off from being a mass market reality, at least there would seem to be a clear path for future hardware and room for it to grow. The innovation is clearly in the mobile space and not in the desktop space these days. It would appear that we will very shortly be running mobile OS's and hardware that rivals today's desktops and notebooks in just about every facet. Probably also within 3 years.
At the moment I'm of the opinion that the earlier rumors of the upcoming iPad 2 are true, and it really will be an across the board quadrupling of power (cpu, gpu, pixel density) with additional features that are expected and a few unexpected. The Fireball 'tamp down' counter-rumors are just trying to make sure that the iPad2 announcement, when it happens, will still be able to wow us.
It looks as though 14-16nm will be here, assuming there are no problems at that process, which is very possible, in three years. Intel is starting production in 22 nm in early 2012. They've got samples out now. Samsung will likely be around that time as well. Of course, Intel has been at 32nm for some time now.
TSMc, which is at the 40nm level, and makes chips for Nvidia and ATI (a brand name about to die), has just had a fire at a wafer plant.
I'll go you one further... if Apple can get the cost on SSD down far enough, and also manage to convince people that disc drives are silly and antiquated (part way there, mac app store helps) then they can effectively eliminate the MBP line as we know it. I think if Apple has their way future MBP notebooks will look more like the MBA than how the MBP looks today.
Let's say my guess that 2 billion goes to Samsung for processors and development work is correct or at least plausible for this discussion. That leave almost 2 billion to go after other components so maybe the throw 0.5 billion each at displays and flash, like they have in the past. That still leaves close to a billion to spread around. These numbers are huge and represents incredible leverage for Apple.
Quote:
They obviously can't do this while the MBA still has to make some compromises. Someday however they'll be able to make a high end MBA with 15 or 17" screen, 1TB SSD, no disc drive (because the demand will die off) and sufficient processing power, while having this new machine maintain a more MBA style weight and enclosure.
They can do that today if they really wanted to. However a 15" inch AIR would never be a power users machine. Since the current MBPs are marketed towards power users Apple will need to offer up performance machines.
Having two 15" machines is no more of a problem than having two 13" machines. The MBP 15" machine would simply use the extra space to implement higher performance chips (bigger heat sinks and fans), larger batteries and a better selection of ports.
Quote:
If they can do that sooner than later, then yes there's no reason they couldn't move the entire line over to SSD (except the minority of pro stuff that wants truly large storage).
This I consider twisted because my pro storage needs are rather simple. However I'd love massive amounts of storage for media. So yeah skinny SSDs do have their negatives. Pro class machines though would be able to easily support hybrid systems. AIRs can't really do that well due to the thin design and limited space. For me though a 15" AIR that didn't have to support my Media collection would work well, given that they can up performance past my early 2008.
AIR as a concept is great because it allows them to serve more markets. That doesn't mean that the markets for the MBPs or the Mac Books go away.
Quote:
In fact we would all be using SSD, and wax poetic about the days of spinning disc hard drives and how silly and antiquated that technology was just like how we think of the floppy disk and how at one point those were so critical to our daily lives. Anything not solid state will be looked down on by Zoll as old, retarded stuff his dad used to use (you mean this has to 'spin up'?).
Well if not this year it isn't far off at all. The really big question though is how long will flash be the SolidState process of choice. I have a feeling flash memory will not be in the spotlight as long as magnetic drives have been. The thing is other SolidState tech would seem to be a better fit in the long run, especially for boot drives.
In any event I have to agree with the idea that rapid changes are coming.
This I consider twisted because my pro storage needs are rather simple. However I'd love massive amounts of storage for media. So yeah skinny SSDs do have their negatives. Pro class machines though would be able to easily support hybrid systems.
I need at least a 240GB startup drive for my Mac Pro. But as I'm retired, I no longer want to spend really big bucks for large SSD's. I bought a 60GB from OWC to play with as that was only $134 on sale. So I installed 10.6.6, CS5 and some other apps and large files.
It really is fast. Apps such as Safari and Mail open almost as fast as I raise my finger from the key. Photoshop takes longer, but about a quarter the time it usually does. Startup times are very fast, but it's not a fair test, as I have a lot of startup files and apps normally that aren't present on this.
Opening a 100MB photo takes vey little time, as does saving it.
I'm waiting for the $525 to come down to about $300 later this year before I pop for a 240. The problem is that due to high density Flash prices, the 480 costs $1,500 right now. There was a time when I wouldn't have cared what it costs, but not now. We're too used to low prices for things.
In addition, the new Sandforce 2000 controller will be out soon, and tests have shown that speeds are close to double the present record of 285MB/s which the OWc models do with the Sandforce 1220. We'll need SATA 3 with 6GB/s to really see those speeds though. I've got a 6GB/s card, so I hope to get one of those.
I need at least a 240GB startup drive for my Mac Pro. But as I'm retired, I no longer want to spend really big bucks for large SSD's. I bought a 60GB from OWC to play with as that was only $134 on sale. So I installed 10.6.6, CS5 and some other apps and large files.
I was actually thinking of up dating me early 2008 MBP with a SSD. As you note the price ballon pretty quick and one does have a minimal size need. They other thing is that the old MBPs are a bear to disassemble.
Quote:
It really is fast. Apps such as Safari and Mail open almost as fast as I raise my finger from the key. Photoshop takes longer, but about a quarter the time it usually does. Startup times are very fast, but it's not a fair test, as I have a lot of startup files and apps normally that aren't present on this.
I get far more beach balling on my MBP than I'd like. I blame a lot of it on slow disk drives, both internal and external. I suspect a SSD would help a lot here. My concern is early wear out due to running software builds which read and write files in ways the average user doesn't.
Quote:
Opening a 100MB photo takes vey little time, as does saving it.
I'm waiting for the $525 to come down to about $300 later this year before I pop for a 240. The problem is that due to high density Flash prices, the 480 costs $1,500 right now. There was a time when I wouldn't have cared what it costs, but not now. We're too used to low prices for things.
I'm actually hoping Apple can speed up the price drops. I still work for a living, but even so find that buying personal use hardware to be in conflict with many other things that interest me. Flash prices will be under pressure from both old and new technology so maybe their is hope. I'm also concerned about the reliability of the latest generation of flash chips.
Quote:
In addition, the new Sandforce 2000 controller will be out soon, and tests have shown that speeds are close to double the present record of 285MB/s which the OWc models do with the Sandforce 1220. We'll need SATA 3 with 6GB/s to really see those speeds though. I've got a 6GB/s card, so I hope to get one of those.
That would be nice, faster is generally good. What I don't understand is why the industry is taking so long to abandon legacy SATA. It would seem to me that PCI-Express should be a no brainer.
I was actually thinking of up dating me early 2008 MBP with a SSD. As you note the price ballon pretty quick and one does have a minimal size need. They other thing is that the old MBPs are a bear to disassemble.
I get far more beach balling on my MBP than I'd like. I blame a lot of it on slow disk drives, both internal and external. I suspect a SSD would help a lot here. My concern is early wear out due to running software builds which read and write files in ways the average user doesn't.
Got one of those half-terrabyte momentus hybrid SSD drives, very good price and power. Installed in a 2007 15" MBP, did wonders for it.
Pretty sure it does NOT involve batteries... the tech there is actually moving a bit after stagnating for many years. My "guess" would be in displays and SSDs. AND I would not be surprised if part of that also involves memory.
AND I also would NOT be surprised if this is also partially a very m$ kind of move, lock up component pricing as a subtle way to put the screws on others... if everyone else HAS to price their stuff high, Apple's very high pricing actually looks reasonable (take a guess how they have amassed almost 60 BILLION in cash in such a short time... wasn't too long ago their overall GM was approaching 45%...).
Nah, they bought the liquid metal company outright several months ago. They own it lock, stock, and barrel.
Apple didn't buy liquid metal, they only bought the rights to the intellectual property for consumer electronics worldwide. I personally hope they do decide to buy the company in the future as I own some shares..
Comments
Not a problem as no one wants a "gigantic" 10" screen. Everyone only requires 7" screen.
What are you, some kind of comedian?
This is where I believe Samsungs long term goals lay: http://www.edn.com/article/512298-AR...p_to_14_nm.php. That is 14nm hardware. Of course there will be nodes in between. The question for us is will Apple push hard to skip 32 nm to move to 22nm. The performance and power usage considerations are huge and for the near future would allow for some incredible hand held devices.
As a complete distraction one should look at all the job opening for engineers and technicians in up state NY. Either for Global Foundries or the ISDA.
As to Samsung here is Ana talking away: http://www10.edacafe.com/video/Samsu...185/media.html. She has obviously mastered the art of answering questions without saying much. To highlight just now much this is breaking news here is another tidbit: http://www.voicesatsamsungsemiconduc...gy-forum-2011/ In any event consider dinning at http://www.edacafe.com/ if your hunger for process tech is strong. As far as Samsung it appears to have 32/28 nm ready to roll, thus my strong suspicion that Apple will deliver a new SoC that will likely more than double performance in a very similar power profile. some of the numbers I've seen for shrinks to these nodes are very impressive.
Obviously considering Apples volume demands, it would not be uncommon for the companies to look at this transition as a partnership. Thus I would not be surprised at all to find out that Apple is throwing cash Samsungs way to secure capacity at this node and future nodes. I have a hard time believing that the entire 3.9 billion will go Samsungs way though. Previous release indicate that a few hundred million can take care of things like Flash and displays. In the end it is probably a safe way for Apple to secure production without building a fab itself.
As to Samsung specific processes here is info form the horses mouth: http://www.samsung.com/us/business/c...r/foundry.html. Note the following:
What is very notable is that no matter where you look there is never ever a mention of Apple anywhere on these sites never. Apples secrecy mentality has yet to soften. In any event these numbers lead me to believe that iPad2 could be running dual core with both the GPU and the CPU in the next rev with only a minor increase in power which at the same time can be offset in other ways. Things are looking good.
If this is all true it would certainly lend credibility to the recent stories of dual core ipads with dual core gpus. 2048x1536 displays would still be a possibility if it's also true that the power expense comes from the backlight and not the pixel density, which is independent of the pixels.
It would seem that 14nm is a couple-few years off from being a mass market reality, at least there would seem to be a clear path for future hardware and room for it to grow. The innovation is clearly in the mobile space and not in the desktop space these days. It would appear that we will very shortly be running mobile OS's and hardware that rivals today's desktops and notebooks in just about every facet. Probably also within 3 years.
At the moment I'm of the opinion that the earlier rumors of the upcoming iPad 2 are true, and it really will be an across the board quadrupling of power (cpu, gpu, pixel density) with additional features that are expected and a few unexpected. The Fireball 'tamp down' counter-rumors are just trying to make sure that the iPad2 announcement, when it happens, will still be able to wow us.
I wonder if it could be solid state drives, ala the Macbook air?
People generally seem to be raving about how fast the macbook air is too boot and run the Air normally shows us where Apple are taking the mac.
E.g. using the aluminium body first debuted on the Air as did the lack of CD drive. Perhaps the next macs will also use SSD?
Just a guess but it could make sense considering how expensive SSD's currently are. By investing in a dedicated production line, they could bring down the cost and secure supply. Also, they way they use SSD in the Air is not the same as normal SSD's, which normally are encased in one unit.
We'll see...
I'll go you one further... if Apple can get the cost on SSD down far enough, and also manage to convince people that disc drives are silly and antiquated (part way there, mac app store helps) then they can effectively eliminate the MBP line as we know it. I think if Apple has their way future MBP notebooks will look more like the MBA than how the MBP looks today.
They obviously can't do this while the MBA still has to make some compromises. Someday however they'll be able to make a high end MBA with 15 or 17" screen, 1TB SSD, no disc drive (because the demand will die off) and sufficient processing power, while having this new machine maintain a more MBA style weight and enclosure.
If they can do that sooner than later, then yes there's no reason they couldn't move the entire line over to SSD (except the minority of pro stuff that wants truly large storage). In fact we would all be using SSD, and wax poetic about the days of spinning disc hard drives and how silly and antiquated that technology was just like how we think of the floppy disk and how at one point those were so critical to our daily lives. Anything not solid state will be looked down on by Zoll as old, retarded stuff his dad used to use (you mean this has to 'spin up'?).
If this is all true it would certainly lend credibility to the recent stories of dual core ipads with dual core gpus. 2048x1536 displays would still be a possibility if it's also true that the power expense comes from the backlight and not the pixel density, which is independent of the pixels.
It would seem that 14nm is a couple-few years off from being a mass market reality, at least there would seem to be a clear path for future hardware and room for it to grow. The innovation is clearly in the mobile space and not in the desktop space these days. It would appear that we will very shortly be running mobile OS's and hardware that rivals today's desktops and notebooks in just about every facet. Probably also within 3 years.
At the moment I'm of the opinion that the earlier rumors of the upcoming iPad 2 are true, and it really will be an across the board quadrupling of power (cpu, gpu, pixel density) with additional features that are expected and a few unexpected. The Fireball 'tamp down' counter-rumors are just trying to make sure that the iPad2 announcement, when it happens, will still be able to wow us.
It looks as though 14-16nm will be here, assuming there are no problems at that process, which is very possible, in three years. Intel is starting production in 22 nm in early 2012. They've got samples out now. Samsung will likely be around that time as well. Of course, Intel has been at 32nm for some time now.
TSMc, which is at the 40nm level, and makes chips for Nvidia and ATI (a brand name about to die), has just had a fire at a wafer plant.
I'll go you one further... if Apple can get the cost on SSD down far enough, and also manage to convince people that disc drives are silly and antiquated (part way there, mac app store helps) then they can effectively eliminate the MBP line as we know it. I think if Apple has their way future MBP notebooks will look more like the MBA than how the MBP looks today.
Let's say my guess that 2 billion goes to Samsung for processors and development work is correct or at least plausible for this discussion. That leave almost 2 billion to go after other components so maybe the throw 0.5 billion each at displays and flash, like they have in the past. That still leaves close to a billion to spread around. These numbers are huge and represents incredible leverage for Apple.
They obviously can't do this while the MBA still has to make some compromises. Someday however they'll be able to make a high end MBA with 15 or 17" screen, 1TB SSD, no disc drive (because the demand will die off) and sufficient processing power, while having this new machine maintain a more MBA style weight and enclosure.
They can do that today if they really wanted to. However a 15" inch AIR would never be a power users machine. Since the current MBPs are marketed towards power users Apple will need to offer up performance machines.
Having two 15" machines is no more of a problem than having two 13" machines. The MBP 15" machine would simply use the extra space to implement higher performance chips (bigger heat sinks and fans), larger batteries and a better selection of ports.
If they can do that sooner than later, then yes there's no reason they couldn't move the entire line over to SSD (except the minority of pro stuff that wants truly large storage).
This I consider twisted because my pro storage needs are rather simple. However I'd love massive amounts of storage for media. So yeah skinny SSDs do have their negatives. Pro class machines though would be able to easily support hybrid systems. AIRs can't really do that well due to the thin design and limited space. For me though a 15" AIR that didn't have to support my Media collection would work well, given that they can up performance past my early 2008.
AIR as a concept is great because it allows them to serve more markets. That doesn't mean that the markets for the MBPs or the Mac Books go away.
In fact we would all be using SSD, and wax poetic about the days of spinning disc hard drives and how silly and antiquated that technology was just like how we think of the floppy disk and how at one point those were so critical to our daily lives. Anything not solid state will be looked down on by Zoll as old, retarded stuff his dad used to use (you mean this has to 'spin up'?).
Well if not this year it isn't far off at all. The really big question though is how long will flash be the SolidState process of choice. I have a feeling flash memory will not be in the spotlight as long as magnetic drives have been. The thing is other SolidState tech would seem to be a better fit in the long run, especially for boot drives.
In any event I have to agree with the idea that rapid changes are coming.
This I consider twisted because my pro storage needs are rather simple. However I'd love massive amounts of storage for media. So yeah skinny SSDs do have their negatives. Pro class machines though would be able to easily support hybrid systems.
I need at least a 240GB startup drive for my Mac Pro. But as I'm retired, I no longer want to spend really big bucks for large SSD's. I bought a 60GB from OWC to play with as that was only $134 on sale. So I installed 10.6.6, CS5 and some other apps and large files.
It really is fast. Apps such as Safari and Mail open almost as fast as I raise my finger from the key. Photoshop takes longer, but about a quarter the time it usually does. Startup times are very fast, but it's not a fair test, as I have a lot of startup files and apps normally that aren't present on this.
Opening a 100MB photo takes vey little time, as does saving it.
I'm waiting for the $525 to come down to about $300 later this year before I pop for a 240. The problem is that due to high density Flash prices, the 480 costs $1,500 right now. There was a time when I wouldn't have cared what it costs, but not now. We're too used to low prices for things.
In addition, the new Sandforce 2000 controller will be out soon, and tests have shown that speeds are close to double the present record of 285MB/s which the OWc models do with the Sandforce 1220. We'll need SATA 3 with 6GB/s to really see those speeds though. I've got a 6GB/s card, so I hope to get one of those.
I need at least a 240GB startup drive for my Mac Pro. But as I'm retired, I no longer want to spend really big bucks for large SSD's. I bought a 60GB from OWC to play with as that was only $134 on sale. So I installed 10.6.6, CS5 and some other apps and large files.
I was actually thinking of up dating me early 2008 MBP with a SSD. As you note the price ballon pretty quick and one does have a minimal size need. They other thing is that the old MBPs are a bear to disassemble.
It really is fast. Apps such as Safari and Mail open almost as fast as I raise my finger from the key. Photoshop takes longer, but about a quarter the time it usually does. Startup times are very fast, but it's not a fair test, as I have a lot of startup files and apps normally that aren't present on this.
I get far more beach balling on my MBP than I'd like. I blame a lot of it on slow disk drives, both internal and external. I suspect a SSD would help a lot here. My concern is early wear out due to running software builds which read and write files in ways the average user doesn't.
Opening a 100MB photo takes vey little time, as does saving it.
I'm waiting for the $525 to come down to about $300 later this year before I pop for a 240. The problem is that due to high density Flash prices, the 480 costs $1,500 right now. There was a time when I wouldn't have cared what it costs, but not now. We're too used to low prices for things.
I'm actually hoping Apple can speed up the price drops. I still work for a living, but even so find that buying personal use hardware to be in conflict with many other things that interest me. Flash prices will be under pressure from both old and new technology so maybe their is hope. I'm also concerned about the reliability of the latest generation of flash chips.
In addition, the new Sandforce 2000 controller will be out soon, and tests have shown that speeds are close to double the present record of 285MB/s which the OWc models do with the Sandforce 1220. We'll need SATA 3 with 6GB/s to really see those speeds though. I've got a 6GB/s card, so I hope to get one of those.
That would be nice, faster is generally good. What I don't understand is why the industry is taking so long to abandon legacy SATA. It would seem to me that PCI-Express should be a no brainer.
I was actually thinking of up dating me early 2008 MBP with a SSD. As you note the price ballon pretty quick and one does have a minimal size need. They other thing is that the old MBPs are a bear to disassemble.
I get far more beach balling on my MBP than I'd like. I blame a lot of it on slow disk drives, both internal and external. I suspect a SSD would help a lot here. My concern is early wear out due to running software builds which read and write files in ways the average user doesn't.
Got one of those half-terrabyte momentus hybrid SSD drives, very good price and power. Installed in a 2007 15" MBP, did wonders for it.
http://www.google.com/products/catal...d=0CCoQ8wIwAQ#
AND I also would NOT be surprised if this is also partially a very m$ kind of move, lock up component pricing as a subtle way to put the screws on others... if everyone else HAS to price their stuff high, Apple's very high pricing actually looks reasonable (take a guess how they have amassed almost 60 BILLION in cash in such a short time... wasn't too long ago their overall GM was approaching 45%...).
Nah, they bought the liquid metal company outright several months ago. They own it lock, stock, and barrel.
Apple didn't buy liquid metal, they only bought the rights to the intellectual property for consumer electronics worldwide. I personally hope they do decide to buy the company in the future as I own some shares..