Thanks mate, for taking your time.
I'm a typical end user, artist creator that demands fast performance for graphics and video creation in high resolutions. I'm stuck with semi professional tools from Adobe, Toon Boom and Apple etc.
I'm of the opinion that it will be a very long time before computers are fast enough. Like everybody currently reading this forum has retired six feet under.
They say it's for professionals, and it is, but when looking at the performance of most software it's somewhat sad. I just know how well they should run if they was coded right.
How do you look at such performance? If you look at benchmark figures online you are not doing yourself any favors unless they are openly documented and you understand what is being relayed.
I was asking for years about proper multi threaded rendering in the animation software I'm using, and finally they did it. Great now. Perhaps everything can't be multi threaded, but I'd expect the developers to be a bit more creative when threading their apps.
Multi threading an app can either be extremely simple or next to impossible. Some apps will never multithread well and that is where you need to focus on single core performance if those apps are critical to your work load.
And Apple should just nail their software. An app like Compressor should be faster on the Mac Pro. Right now it's faster on a MacBook Pro. It's not right.
There are no right nor wrongs here, the Mac Book Pro may be faster in carefully crafted benchmarks but it is a total dog for people that know how to leverage the Mac Pro in a work environment. The point here is that a Mac Pro can be doing many things at once and not suffer the slow down seen in Many of todays low end systems.
It's confusing for an end user like me.
It is pretty simple really if you are doing professional work, seriously consider the Mac Pro. This especially if your work habits have you doing multiple things at once. Multi threading is nice but that isn't the only way to leverage all the core possible in a Mac Pro. Learn to keep many things in flight on a Mac Pro and you will begin to understand how powerful the machine is. If your work habits are such that you can't keep multiple task running productively on the mac Pro then the machine isn't for you.
Now given all of that I'd not buy any Apple hardware right now if I could help it. The upgrades to various platforms with SkyLake or other new Intel processors, will mean far better performance for many of Apples machines. The same thing goes for the Mac Pro even though corralling dates for a new hardware release is a bit more difficult.
Those OpenCL Intel tasks are 1.1 compliant and some 1.2. The AMD line up is 100% 2.x compliant OpenCL and a lot of features in the 2.x range matter to Apple improving system-wide performance.
The only point I was trying to make is that Intels GPU's are often better coprocessors for OpenCL acceleration than they are GPU's for video acceleration.
Nothing Intel produces with OpenCL can touch AMD's position. The designs are long since done for the Zen FX and Zen APU. The latest news from China is that HBM2 is exclusive again in 2016 for AMD and that means not 4GB (4 x 1GB) HBM limits, but 32GB HBM2 (4 x 8GB) configuration limits.
Well like things intel we have to wait until something is actually released before we pass judgement on it. Both AMD and Intel have failed to deliver on leaked performance metrics so I take a wait and see attitude.
The interesting thing here is that 32 GB of HBM RAM in an APU design would mean it would be possible to forego conventional system RAM. This right there would be an impressive performance boost for both the GPU and the CPU. The funny thing here is that DDR4 may end up having a very short shelf life compared to older technologies. The way to maximum speed with the RAM subsystem is to incorporate RAM right into the CPU package.
The key will be manufacturing moving forward with Samsung/GloFo 14nm FinFET, already announced by AMD and also leveraging TSMC 16nm FinFET.
Lets hope AMD can stay around long enough to place this stuff on the market. I know I've been advocating a AMD Mini for years now and have yet to see one, so maybe there is hope.
If you look at the clock speed of the Xeons, you'll notice as the core count goes up the clock speed comes down. In small quick tasks, the higher clocked iMac can actually be faster than an 8-12 core Mac Pro, but in tasks such as rendering full length videos or shading 3D animated scenes, the Mac Pro will eat through that much faster while the iMac starts to overheat.
True. However the trick with any many core machine is to have work constantly being done in background. From the standpoint of productivity it is hard to beat a Mac Pro if the user has the capability to leverage the machine in this way.
Interestingly professionals in the same general field may or may not be able to leverage the Mac Pro. For example software developers come in many varieties, somebody build Python Apps may never see an advantage with the Mac Pro but someone doing builds of large C++ apps might be able to keep all cores going in a Mac Pro. What I'm after here is the idea that ones ability to leverage the Mac Pro depends not only on what that person does but how they do it.
As an aside, The new Mac pro could really use a new chipset based on the smaller process technologies. This would go a long ways to making clock rates less of an issue.
Because when they were using it NetApp was suing Sun then Oracle over file system patents (lawsuit dropped in 2010). Sun/Oracle would not indemnify Apple from infringement so Apple dropped it. When Oracle bought Sun it whacked the last nail in the coffin. As a result the original ZFS work was removed from OS X and never released.
Yes, but after all that drama didn't Apple bring a known file-system guru in-house and the speculation was that they were going to build ZFS-like tech into Mac OS X all by themselves?
I note that all the Skylake-H processors are rated at 45W -- no more 35W parts. This may increase Apple's desire to finally drop the last vestige of discrete GPUs from the MacBook Pro line.
If they do that they better drop the "pro" designation.
Shameful for a platform founded on graphical prowess that so few machines are offered with GPUs and that there is only one machine in the entire line not made with laptop parts
I'm just salivating at the 8-core CPU. I hope Apple put's that as a high-end option for the iMac. I'd put them all to good use! Hopefully this means I finally get to retire my 2009 iMac!
If there is no 8 core laptop part I wouldn't hold your breath :rolleyes:
There are many good reasons to prefer AMD in things like the Mini.
For one it gives manufactures leverage with Intel.
Two; intel needs the competition, the last couple of years should make that obvious.
Three; GPU performance is very important in machines like the Mini. For many users and uses for the Mini the GPU is far mor important than the CPU.
Four; Apple could easily knock a $100 or so of the price of the Mini keeping the machine competitive in the marketplace.
Competition is always good. But AMD is a failure at keeping their promises on performance and schedules. Apple went through that with Motorola in their later days, and then, after Motorola sold their semiconductor unit, with Freescale. Even IBM failed to product a mobile G5, which was the major reason why Apple moved to Intel.
No matter how you look at it, AMD is behind Intel in most every area. Where they aren't, the difference isn't much. But Apple is never again going to use a company that they don't have confidence in. If Intel is running late, the entire industry is running late. If AMD is running late, their customers end up behind everyone else.
I also don't see how Apple could knock $100 off the price off the Mini. The pricing between the equivalent chips of AMD and Intel isn't that different.
r OS X that uses sky lake processors? Second there is the Apple TV update that is not a hobby. Gaming on one of these processors could actually be enjoyable. An A9 processor, plus Metal plus a huge library of software from iOS. I am not sure what is coming, but I suspect there are going to be some disappointed Apple Bears in 7 days.
Competition is always good. But AMD is a failure at keeping their promises on performance and schedules. Apple went through that with Motorola in their later days, and then, after Motorola sold their semiconductor unit, with Freescale. Even IBM failed to product a mobile G5, which was the major reason why Apple moved to Intel.
The problem is Intel isn't anymore reliable. From what I can see they still haven't mastered 14nm. The issues surrounding Broadwell and SkyLake have become a joke in the industry. In fact Intel has slipped more here than AMD as far as promised release dates.
No matter how you look at it, AMD is behind Intel in most every area. Where they aren't, the difference isn't much. But Apple is never again going to use a company that they don't have confidence in. If Intel is running late, the entire industry is running late. If AMD is running late, their customers end up behind everyone else.
Apple uses AMD parts extensively. They are using AMD GPUs in both the laptops and the desktops.
I also don't see how Apple could knock $100 off the price off the Mini. The pricing between the equivalent chips of AMD and Intel isn't that different.
I haven't looked recently but a $100 delta wasn't unheard of in the past. In a cost sensitive machine like the Mini AMD makes good sense, I'm not advocating using AMD chips everywhere. What I'm advocating is squeezing every bit of cost out of the Mini that they can. The idea being to keep the machine competitive with alternative solutions.
In the sense of the Mac going AMD is nothing like the switch to PowerPC and then to Intel so I'm not sure why this is brought up so often.
The problem is Intel isn't anymore reliable. From what I can see they still haven't mastered 14nm. The issues surrounding Broadwell and SkyLake have become a joke in the industry. In fact Intel has slipped more here than AMD as far as promised release dates.
Apple uses AMD parts extensively. They are using AMD GPUs in both the laptops and the desktops.
I haven't looked recently but a $100 delta wasn't unheard of in the past. In a cost sensitive machine like the Mini AMD makes good sense, I'm not advocating using AMD chips everywhere. What I'm advocating is squeezing every bit of cost out of the Mini that they can. The idea being to keep the machine competitive with alternative solutions.
In the sense of the Mac going AMD is nothing like the switch to PowerPC and then to Intel so I'm not sure why this is brought up so often.
How well is AMD doing with 14nm? Not too well. But despite what you say, Intel has gotten 14nm pretty much nailed. The problems with smaller process cycles is that we're reaching the limits of what can be done. If Intel is behind and they are, as I've been staring here for years, then everyone else is even further behind. And AMD can't even control their own process, because they don't have one. That's even worse.
AMD graphics is a completely different division, and is doing much better technically, than the rest of the company is.
The Mini doesn't use high end chips, so the delta, as you put it, is much less. And the reason for that delta isn't because AMD is a good guy, it's because their chips are less desirable, and so can't be sold for the same markup.
How well is AMD doing with 14nm? Not too well. But despite what you say, Intel has gotten 14nm pretty much nailed. The problems with smaller process cycles is that we're reaching the limits of what can be done. If Intel is behind and they are, as I've been staring here for years, then everyone else is even further behind. And AMD can't even control their own process, because they don't have one. That's even worse.
The reports are that both Samsung and TSMC are both up and running around the 14nm node. We might even hear about a 14 nm processor from Apple tomorrow. That would mean that Intel has less than a years lead on the competition which would be the shortest in recent memory.
The good thing for AMD is that they don't have a process of their own, they couldn't possible sustain the R&D right now. Being free of manufacturing allows them to select the best foundry possible.
AMD graphics is a completely different division, and is doing much better technically, than the rest of the company is.
They are still getting beat up in the market and frankly the last GPU release wasn't all that impressive.
The Mini doesn't use high end chips, so the delta, as you put it, is much less. And the reason for that delta isn't because AMD is a good guy, it's because their chips are less desirable, and so can't be sold for the same markup.
The Mini doesn't use high end chips so there is less of a performance delta going with AMD. That was my point. The other point is that AMD needs the support, actually the whole industry needs AMD, I'd hate to see an Intel only world when it comes to X86 processors. The Mini is the only x86 platform Apple has where it wouldn't make a huge difference if an AMD chip was put in the box.
The reports are that both Samsung and TSMC are both up and running around the 14nm node. We might even hear about a 14 nm processor from Apple tomorrow. That would mean that Intel has less than a years lead on the competition which would be the shortest in recent memory.
The good thing for AMD is that they don't have a process of their own, they couldn't possible sustain the R&D right now. Being free of manufacturing allows them to select the best foundry possible.
They are still getting beat up in the market and frankly the last GPU release wasn't all that impressive.
The Mini doesn't use high end chips so there is less of a performance delta going with AMD. That was my point. The other point is that AMD needs the support, actually the whole industry needs AMD, I'd hate to see an Intel only world when it comes to X86 processors. The Mini is the only x86 platform Apple has where it wouldn't make a huge difference if an AMD chip was put in the box.
Except that Intel's 14nm is more 14nm than those others are. They are also ahead in sophistication in the process as compared to those others. Both Samsung and TSMC's 14nm is being called, by some as 14.5nm, as the density is lower than Intel's, due to the larger line width they are using for the on chip circuit wiring, which is as large as 20-28nm in width.
All processes aren't the same, even if they are at the same node. And let's not forget that Intel has been at a full 14nm process for over a year, while they are just now getting started.
You just seem to have contradicted yourself about AMD. Among other things, you said that the delta for a Mini would be $100, while I said it would be much lower, and now you are saying that it would be much lower, and that was your point? I don't get that.
And if AMD is being beat up in the market, why would Apple have any interest in them, and that was my point. Apple is simply not going to go with a secondary company that's doing poorly. There's no advantage for them in that. A few bucks difference in pricing isn't enough. Nobody can rely on AMD's x86 chip designs or production any more. They are always late, with poorer performance than promised, often by a large amount.
I don't care in the slightest about AMD, and its chances. Intel doesn't either. Intel competes against themselves, not AMD. Let Dell buy all of their chips. The small difference might make a difference to their customers, though historically, people have avoided computers with AMD chips inside, unless the computers were REALLY cheap.
And if AMD is being beat up in the market, why would Apple have any interest in them, and that was my point. Apple is simply not going to go with a secondary company that's doing poorly. There's no advantage for them in that. A few bucks difference in pricing isn't enough. Nobody can rely on AMD's x86 chip designs or production any more. They are always late, with poorer performance than promised, often by a large amount.
I don't care in the slightest about AMD, and its chances. Intel doesn't either. Intel competes against themselves, not AMD. Let Dell buy all of their chips. The small difference might make a difference to their customers, though historically, people have avoided computers with AMD chips inside, unless the computers were REALLY cheap.
AMD has promised the world for the last 10+years and it never really delivers a really good reason to chose "the clone". I have written of AMD for CPUs a long time ago.
I am more interested in advances they are making with ARM processors. If Apple's bragging is to be believed the A9X 4 core geekbench 3 score will be around 8000 (big if)..... that would be a quite amazing jump and would mean the gap is closing rather rapidly. (personally would still need Intel since I really want to be able to run Oracle Database in a VM).
The user experience is pretty good with the latest tablet, like the I pad air 2. The increase of performance of the ARM chips are way faster, than the one of Intel. If nothing changes, at this speed, the two curves will cross
Now, one part of the good experience of the I pad is related to IOS . IOS is more light and less a waste of RAW Power than let's say mac OS X or Windows. I fear that an A9X on an I book air, will be not that good for running mac os X, with a worse user experience than the one the I pad Pro.
Apple's A series of ARM processors are the state of the art period. Intel has nothing that is even comparable.
It is inconceivable that Apple isn't putting a great deal of resources into OS X. They are making incredible profits and the Mac lines are doing well in and of themselves. OS X isn't the resource hog Windows is and should be able to perform well on the A9/A9X. I suspect that all of the available A9/A9X production is going to go into the iPhone/iPad leaving little available for the other lines.
The Apple TV doesn't sell in large numbers, but will be based on the A8 CPU. If the A9 were readily available, Apple very likely would have used the chip instead.
As Samsung and TSMC overtake Intel, the supply constraints for the A series will improve. It isn't hard to foresee Apple putting an ARM chip into a laptop or the mini running OS X.
Intel's lead really isn't much of one in reality anyhow. When the A8X has much higher transistor density at a larger process node and performs comparably to the Core M while consuming less energy, Intel's so-called superior 14 nm process doesn't amount to much.
Samsung and TSMC will be producing far more 14 and 16 nm FF CPUs than Intel will over the next 2 years. I myself am purchasing an iPad Pro. Don't need any of the Intel based machines for what I do. I doubt that the Core M based MacBook can edit 3 simultaneous 4k video streams either.
TSMC and Samsung are now competing with each other. Intel isn't the company driving them. And Apple's CPU/GPU design team is superior to Intel's also.
I will wait for the Mac mini with an ARM CPU priced significantly less than an Intel powered one. It will be coming. There is no way that Intel will be able to keep up going forward charging far higher prices for an x86 CPU than an ARM one that offers better performance.
Apple has shifted CPU platforms before and can easily do it again, especially with their own knowledge of their in house designed CPUs. Going forward, there really isn't anything limiting them also from designing a high performance multi core variant of an ARM CPU that could perform comparably to Intel's Xeons running OS X. Though somewhat unlikely, it is well within the realm of possibility. Especially with TSMC and Samsung attempting to get to 10 nm and 7 nm nodes as quickly as they can in order to win Apple's business. They will likely get there before Intel does. With Apple's design team and the very substantial cost advantages of ARM over X86, why would Apple not want to put out a high end ARM powered machine? They could still sell X86 powered machines at the same time for those who still require Windows or legacy software.
X86 is stale. The market is moving to ARM in a very substantial way. If Intel's offerings are no longer compelling, AMD's won't be either.
I am so incredibly happy I put off purchasing the recently released MacBooks. I already have the money saved up for the iPad Pro. And to be totally honest, I don't ever plan on purchasing another Intel powered machine again. I plan on reclaiming the desktop space my current iMac is taking up. Once Apple sells a mini with an A series CPU or the Apple TV becomes powerful enough, I plan on hooking it up to LG's 4k OLED monitor. I honestly don't see a need for anything from Intel as a consumer.
For all the talk of Apple's products being expensive, the X86 CPU is obscenely over priced.
Apple's A series of ARM processors are the state of the art period. Intel has nothing that is even comparable.
It is inconceivable that Apple isn't putting a great deal of resources into OS X. They are making incredible profits and the Mac lines are doing well in and of themselves. OS X isn't the resource hog Windows is and should be able to perform well on the A9/A9X. I suspect that all of the available A9/A9X production is going to go into the iPhone/iPad leaving little available for the other lines.
The Apple TV doesn't sell in large numbers, but will be based on the A8 CPU. If the A9 were readily available, Apple very likely would have used the chip instead.
As Samsung and TSMC overtake Intel, the supply constraints for the A series will improve. It isn't hard to foresee Apple putting an ARM chip into a laptop or the mini running OS X.
Intel's lead really isn't much of one in reality anyhow. When the A8X has much higher transistor density at a larger process node and performs comparably to the Core M while consuming less energy, Intel's so-called superior 14 nm process doesn't amount to much.
Samsung and TSMC will be producing far more 14 and 16 nm FF CPUs than Intel will over the next 2 years. I myself am purchasing an iPad Pro. Don't need any of the Intel based machines for what I do. I doubt that the Core M based MacBook can edit 3 simultaneous 4k video streams either.
TSMC and Samsung are now competing with each other. Intel isn't the company driving them. And Apple's CPU/GPU design team is superior to Intel's also.
I will wait for the Mac mini with an ARM CPU priced significantly less than an Intel powered one. It will be coming. There is no way that Intel will be able to keep up going forward charging far higher prices for an x86 CPU than an ARM one that offers better performance.
Apple has shifted CPU platforms before and can easily do it again, especially with their own knowledge of their in house designed CPUs. Going forward, there really isn't anything limiting them also from designing a high performance multi core variant of an ARM CPU that could perform comparably to Intel's Xeons running OS X. Though somewhat unlikely, it is well within the realm of possibility. Especially with TSMC and Samsung attempting to get to 10 nm and 7 nm nodes as quickly as they can in order to win Apple's business. They will likely get there before Intel does. With Apple's design team and the very substantial cost advantages of ARM over X86, why would Apple not want to put out a high end ARM powered machine? They could still sell X86 powered machines at the same time for those who still require Windows or legacy software.
X86 is stale. The market is moving to ARM in a very substantial way. If Intel's offerings are no longer compelling, AMD's won't be either.
I am so incredibly happy I put off purchasing the recently released MacBooks. I already have the money saved up for the iPad Pro. And to be totally honest, I don't ever plan on purchasing another Intel powered machine again. I plan on reclaiming the desktop space my current iMac is taking up. Once Apple sells a mini with an A series CPU or the Apple TV becomes powerful enough, I plan on hooking it up to LG's 4k OLED monitor. I honestly don't see a need for anything from Intel as a consumer.
For all the talk of Apple's products being expensive, the X86 CPU is obscenely over priced.
The new A9X definitely an interesting beast... based only on comments by Apple, the A9 processor benchmarks and the last processor that they compared to I expect the A9X processor in the iPad Pro to come in at around 8,000 on the Geekbench 3 benchmarks which would put it above the lowest end of the Macbook pro lineup (but far below the top end). If I were not tied to wanting to run VMWare with Linux/86 for Oracle RDBMS I would not care if they dumped the Intel chips for the low end - but unfortunately there is no ARM processor that compares to the top end of the lineup for Apple. It would mean having a mixed lineup which would only end up confusing the customer - will my computer run this application or the other version.... this was the reason why the Microsoft foray into the Windows RT platform had no chance.
But, with applications being submitted to the app store in bitcode (transportable assembly language) once Apple has the processes in place it could easily make the processor architecture meaningless and hidden from the customer.... which means Apple could have the top end of the lineup on Xeon processors and the low end on ARM and there would be nothing for the customer to have to know about - and would not be confusing (with the exception of running VMWare or Parallels).
I believe it is coming, but it is probably a year or two away from being possible.
Apple's design team could design a high performance ARM processor to compete with the Xeon. It is a relatively straight forward task and likely less complicated than designing a low power 64 bit design using the ARM ISA.
Such a chip would allow for further customization of the Mac apart from the general PC market and Apple could do things that Intel is simply not willing to do for sake of preserving their margins. And Apple could release new machines on their own schedule, freeing them from the Intel development cycle. Lowering the cost of the machines while increasing performance is a win win situation. Especially now with the GPUs in Apple's mobile devices outclassing Intel integrated graphics.
Apple's design team could design a high performance ARM processor to compete with the Xeon. It is a relatively straight forward task and likely less complicated than designing a low power 64 bit design using the ARM ISA.
Such a chip would allow for further customization of the Mac apart from the general PC market and Apple could do things that Intel is simply not willing to do for sake of preserving their margins. And Apple could release new machines on their own schedule, freeing them from the Intel development cycle. Lowering the cost of the machines while increasing performance is a win win situation. Especially now with the GPUs in Apple's mobile devices outclassing Intel integrated graphics.
Lol. No. If it were that easy we'd see more ARM servers. It isn't easy and ARM inroads into the server market are far below expectations as micro servers didn't explode like folks were claiming it would.
Intel sold 100 million CPUs in Q4 2014. 2015 has been better than 2014 for them from what I remember. They are doing fine and still ahead in process.
There is no way Apple moves from Intel for the MacBooks in the next few years.
Apple's A series of ARM processors are the state of the art period. Intel has nothing that is even comparable.
It is inconceivable that Apple isn't putting a great deal of resources into OS X. They are making incredible profits and the Mac lines are doing well in and of themselves. OS X isn't the resource hog Windows is and should be able to perform well on the A9/A9X. I suspect that all of the available A9/A9X production is going to go into the iPhone/iPad leaving little available for the other lines.
The Apple TV doesn't sell in large numbers, but will be based on the A8 CPU. If the A9 were readily available, Apple very likely would have used the chip instead.
As Samsung and TSMC overtake Intel, the supply constraints for the A series will improve. It isn't hard to foresee Apple putting an ARM chip into a laptop or the mini running OS X.
Intel's lead really isn't much of one in reality anyhow. When the A8X has much higher transistor density at a larger process node and performs comparably to the Core M while consuming less energy, Intel's so-called superior 14 nm process doesn't amount to much.
Samsung and TSMC will be producing far more 14 and 16 nm FF CPUs than Intel will over the next 2 years. I myself am purchasing an iPad Pro. Don't need any of the Intel based machines for what I do. I doubt that the Core M based MacBook can edit 3 simultaneous 4k video streams either.
TSMC and Samsung are now competing with each other. Intel isn't the company driving them. And Apple's CPU/GPU design team is superior to Intel's also.
I will wait for the Mac mini with an ARM CPU priced significantly less than an Intel powered one. It will be coming. There is no way that Intel will be able to keep up going forward charging far higher prices for an x86 CPU than an ARM one that offers better performance.
Apple has shifted CPU platforms before and can easily do it again, especially with their own knowledge of their in house designed CPUs. Going forward, there really isn't anything limiting them also from designing a high performance multi core variant of an ARM CPU that could perform comparably to Intel's Xeons running OS X. Though somewhat unlikely, it is well within the realm of possibility. Especially with TSMC and Samsung attempting to get to 10 nm and 7 nm nodes as quickly as they can in order to win Apple's business. They will likely get there before Intel does. With Apple's design team and the very substantial cost advantages of ARM over X86, why would Apple not want to put out a high end ARM powered machine? They could still sell X86 powered machines at the same time for those who still require Windows or legacy software.
X86 is stale. The market is moving to ARM in a very substantial way. If Intel's offerings are no longer compelling, AMD's won't be either.
I am so incredibly happy I put off purchasing the recently released MacBooks. I already have the money saved up for the iPad Pro. And to be totally honest, I don't ever plan on purchasing another Intel powered machine again. I plan on reclaiming the desktop space my current iMac is taking up. Once Apple sells a mini with an A series CPU or the Apple TV becomes powerful enough, I plan on hooking it up to LG's 4k OLED monitor. I honestly don't see a need for anything from Intel as a consumer.
For all the talk of Apple's products being expensive, the X86 CPU is obscenely over priced.
I'll be getting rid of my iMac and getting a 13 inch iPad Pro (The latter of which can handle 3x4k video? Can the Macbook do that? Can an iMac from just a few years ago do that?) for conceptualisation work with Procreate 3 (which looks like being a belter of a release..)
I can wait for a retina iMac that can really 'throw' that 5k display around. ...and then use 'astro' to enable the iPad to act as a tablet for the Mac.
Comments
It is pretty simple really if you are doing professional work, seriously consider the Mac Pro. This especially if your work habits have you doing multiple things at once. Multi threading is nice but that isn't the only way to leverage all the core possible in a Mac Pro. Learn to keep many things in flight on a Mac Pro and you will begin to understand how powerful the machine is. If your work habits are such that you can't keep multiple task running productively on the mac Pro then the machine isn't for you.
Now given all of that I'd not buy any Apple hardware right now if I could help it. The upgrades to various platforms with SkyLake or other new Intel processors, will mean far better performance for many of Apples machines. The same thing goes for the Mac Pro even though corralling dates for a new hardware release is a bit more difficult.
The interesting thing here is that 32 GB of HBM RAM in an APU design would mean it would be possible to forego conventional system RAM. This right there would be an impressive performance boost for both the GPU and the CPU. The funny thing here is that DDR4 may end up having a very short shelf life compared to older technologies. The way to maximum speed with the RAM subsystem is to incorporate RAM right into the CPU package.
Lets hope AMD can stay around long enough to place this stuff on the market. I know I've been advocating a AMD Mini for years now and have yet to see one, so maybe there is hope.
True. However the trick with any many core machine is to have work constantly being done in background. From the standpoint of productivity it is hard to beat a Mac Pro if the user has the capability to leverage the machine in this way.
Interestingly professionals in the same general field may or may not be able to leverage the Mac Pro. For example software developers come in many varieties, somebody build Python Apps may never see an advantage with the Mac Pro but someone doing builds of large C++ apps might be able to keep all cores going in a Mac Pro. What I'm after here is the idea that ones ability to leverage the Mac Pro depends not only on what that person does but how they do it.
As an aside, The new Mac pro could really use a new chipset based on the smaller process technologies. This would go a long ways to making clock rates less of an issue.
Because when they were using it NetApp was suing Sun then Oracle over file system patents (lawsuit dropped in 2010). Sun/Oracle would not indemnify Apple from infringement so Apple dropped it. When Oracle bought Sun it whacked the last nail in the coffin. As a result the original ZFS work was removed from OS X and never released.
Yes, but after all that drama didn't Apple bring a known file-system guru in-house and the speculation was that they were going to build ZFS-like tech into Mac OS X all by themselves?
If they do that they better drop the "pro" designation.
Shameful for a platform founded on graphical prowess that so few machines are offered with GPUs and that there is only one machine in the entire line not made with laptop parts
If there is no 8 core laptop part I wouldn't hold your breath :rolleyes:
Competition is always good. But AMD is a failure at keeping their promises on performance and schedules. Apple went through that with Motorola in their later days, and then, after Motorola sold their semiconductor unit, with Freescale. Even IBM failed to product a mobile G5, which was the major reason why Apple moved to Intel.
No matter how you look at it, AMD is behind Intel in most every area. Where they aren't, the difference isn't much. But Apple is never again going to use a company that they don't have confidence in. If Intel is running late, the entire industry is running late. If AMD is running late, their customers end up behind everyone else.
I also don't see how Apple could knock $100 off the price off the Mini. The pricing between the equivalent chips of AMD and Intel isn't that different.
r OS X that uses sky lake processors? Second there is the Apple TV update that is not a hobby. Gaming on one of these processors could actually be enjoyable. An A9 processor, plus Metal plus a huge library of software from iOS. I am not sure what is coming, but I suspect there are going to be some disappointed Apple Bears in 7 days.
For more info visit at =========>>> http://healthproductsource.com/crazy-mass-reviews/ and http://www.bulkcrazymass.com/
Apple uses AMD parts extensively. They are using AMD GPUs in both the laptops and the desktops.
I haven't looked recently but a $100 delta wasn't unheard of in the past. In a cost sensitive machine like the Mini AMD makes good sense, I'm not advocating using AMD chips everywhere. What I'm advocating is squeezing every bit of cost out of the Mini that they can. The idea being to keep the machine competitive with alternative solutions.
In the sense of the Mac going AMD is nothing like the switch to PowerPC and then to Intel so I'm not sure why this is brought up so often.
How well is AMD doing with 14nm? Not too well. But despite what you say, Intel has gotten 14nm pretty much nailed. The problems with smaller process cycles is that we're reaching the limits of what can be done. If Intel is behind and they are, as I've been staring here for years, then everyone else is even further behind. And AMD can't even control their own process, because they don't have one. That's even worse.
AMD graphics is a completely different division, and is doing much better technically, than the rest of the company is.
The Mini doesn't use high end chips, so the delta, as you put it, is much less. And the reason for that delta isn't because AMD is a good guy, it's because their chips are less desirable, and so can't be sold for the same markup.
The good thing for AMD is that they don't have a process of their own, they couldn't possible sustain the R&D right now. Being free of manufacturing allows them to select the best foundry possible. They are still getting beat up in the market and frankly the last GPU release wasn't all that impressive.
The Mini doesn't use high end chips so there is less of a performance delta going with AMD. That was my point. The other point is that AMD needs the support, actually the whole industry needs AMD, I'd hate to see an Intel only world when it comes to X86 processors. The Mini is the only x86 platform Apple has where it wouldn't make a huge difference if an AMD chip was put in the box.
Except that Intel's 14nm is more 14nm than those others are. They are also ahead in sophistication in the process as compared to those others. Both Samsung and TSMC's 14nm is being called, by some as 14.5nm, as the density is lower than Intel's, due to the larger line width they are using for the on chip circuit wiring, which is as large as 20-28nm in width.
All processes aren't the same, even if they are at the same node. And let's not forget that Intel has been at a full 14nm process for over a year, while they are just now getting started.
You just seem to have contradicted yourself about AMD. Among other things, you said that the delta for a Mini would be $100, while I said it would be much lower, and now you are saying that it would be much lower, and that was your point? I don't get that.
And if AMD is being beat up in the market, why would Apple have any interest in them, and that was my point. Apple is simply not going to go with a secondary company that's doing poorly. There's no advantage for them in that. A few bucks difference in pricing isn't enough. Nobody can rely on AMD's x86 chip designs or production any more. They are always late, with poorer performance than promised, often by a large amount.
I don't care in the slightest about AMD, and its chances. Intel doesn't either. Intel competes against themselves, not AMD. Let Dell buy all of their chips. The small difference might make a difference to their customers, though historically, people have avoided computers with AMD chips inside, unless the computers were REALLY cheap.
And if AMD is being beat up in the market, why would Apple have any interest in them, and that was my point. Apple is simply not going to go with a secondary company that's doing poorly. There's no advantage for them in that. A few bucks difference in pricing isn't enough. Nobody can rely on AMD's x86 chip designs or production any more. They are always late, with poorer performance than promised, often by a large amount.
I don't care in the slightest about AMD, and its chances. Intel doesn't either. Intel competes against themselves, not AMD. Let Dell buy all of their chips. The small difference might make a difference to their customers, though historically, people have avoided computers with AMD chips inside, unless the computers were REALLY cheap.
AMD has promised the world for the last 10+years and it never really delivers a really good reason to chose "the clone". I have written of AMD for CPUs a long time ago.
I am more interested in advances they are making with ARM processors. If Apple's bragging is to be believed the A9X 4 core geekbench 3 score will be around 8000 (big if)..... that would be a quite amazing jump and would mean the gap is closing rather rapidly. (personally would still need Intel since I really want to be able to run Oracle Database in a VM).
The user experience is pretty good with the latest tablet, like the I pad air 2. The increase of performance of the ARM chips are way faster, than the one of Intel. If nothing changes, at this speed, the two curves will cross
Now, one part of the good experience of the I pad is related to IOS . IOS is more light and less a waste of RAW Power than let's say mac OS X or Windows. I fear that an A9X on an I book air, will be not that good for running mac os X, with a worse user experience than the one the I pad Pro.
It is inconceivable that Apple isn't putting a great deal of resources into OS X. They are making incredible profits and the Mac lines are doing well in and of themselves. OS X isn't the resource hog Windows is and should be able to perform well on the A9/A9X. I suspect that all of the available A9/A9X production is going to go into the iPhone/iPad leaving little available for the other lines.
The Apple TV doesn't sell in large numbers, but will be based on the A8 CPU. If the A9 were readily available, Apple very likely would have used the chip instead.
As Samsung and TSMC overtake Intel, the supply constraints for the A series will improve. It isn't hard to foresee Apple putting an ARM chip into a laptop or the mini running OS X.
Intel's lead really isn't much of one in reality anyhow. When the A8X has much higher transistor density at a larger process node and performs comparably to the Core M while consuming less energy, Intel's so-called superior 14 nm process doesn't amount to much.
Samsung and TSMC will be producing far more 14 and 16 nm FF CPUs than Intel will over the next 2 years. I myself am purchasing an iPad Pro. Don't need any of the Intel based machines for what I do. I doubt that the Core M based MacBook can edit 3 simultaneous 4k video streams either.
TSMC and Samsung are now competing with each other. Intel isn't the company driving them. And Apple's CPU/GPU design team is superior to Intel's also.
I will wait for the Mac mini with an ARM CPU priced significantly less than an Intel powered one. It will be coming. There is no way that Intel will be able to keep up going forward charging far higher prices for an x86 CPU than an ARM one that offers better performance.
Apple has shifted CPU platforms before and can easily do it again, especially with their own knowledge of their in house designed CPUs. Going forward, there really isn't anything limiting them also from designing a high performance multi core variant of an ARM CPU that could perform comparably to Intel's Xeons running OS X. Though somewhat unlikely, it is well within the realm of possibility. Especially with TSMC and Samsung attempting to get to 10 nm and 7 nm nodes as quickly as they can in order to win Apple's business. They will likely get there before Intel does. With Apple's design team and the very substantial cost advantages of ARM over X86, why would Apple not want to put out a high end ARM powered machine? They could still sell X86 powered machines at the same time for those who still require Windows or legacy software.
X86 is stale. The market is moving to ARM in a very substantial way. If Intel's offerings are no longer compelling, AMD's won't be either.
I am so incredibly happy I put off purchasing the recently released MacBooks. I already have the money saved up for the iPad Pro. And to be totally honest, I don't ever plan on purchasing another Intel powered machine again. I plan on reclaiming the desktop space my current iMac is taking up. Once Apple sells a mini with an A series CPU or the Apple TV becomes powerful enough, I plan on hooking it up to LG's 4k OLED monitor. I honestly don't see a need for anything from Intel as a consumer.
For all the talk of Apple's products being expensive, the X86 CPU is obscenely over priced.
Apple's A series of ARM processors are the state of the art period. Intel has nothing that is even comparable.
It is inconceivable that Apple isn't putting a great deal of resources into OS X. They are making incredible profits and the Mac lines are doing well in and of themselves. OS X isn't the resource hog Windows is and should be able to perform well on the A9/A9X. I suspect that all of the available A9/A9X production is going to go into the iPhone/iPad leaving little available for the other lines.
The Apple TV doesn't sell in large numbers, but will be based on the A8 CPU. If the A9 were readily available, Apple very likely would have used the chip instead.
As Samsung and TSMC overtake Intel, the supply constraints for the A series will improve. It isn't hard to foresee Apple putting an ARM chip into a laptop or the mini running OS X.
Intel's lead really isn't much of one in reality anyhow. When the A8X has much higher transistor density at a larger process node and performs comparably to the Core M while consuming less energy, Intel's so-called superior 14 nm process doesn't amount to much.
Samsung and TSMC will be producing far more 14 and 16 nm FF CPUs than Intel will over the next 2 years. I myself am purchasing an iPad Pro. Don't need any of the Intel based machines for what I do. I doubt that the Core M based MacBook can edit 3 simultaneous 4k video streams either.
TSMC and Samsung are now competing with each other. Intel isn't the company driving them. And Apple's CPU/GPU design team is superior to Intel's also.
I will wait for the Mac mini with an ARM CPU priced significantly less than an Intel powered one. It will be coming. There is no way that Intel will be able to keep up going forward charging far higher prices for an x86 CPU than an ARM one that offers better performance.
Apple has shifted CPU platforms before and can easily do it again, especially with their own knowledge of their in house designed CPUs. Going forward, there really isn't anything limiting them also from designing a high performance multi core variant of an ARM CPU that could perform comparably to Intel's Xeons running OS X. Though somewhat unlikely, it is well within the realm of possibility. Especially with TSMC and Samsung attempting to get to 10 nm and 7 nm nodes as quickly as they can in order to win Apple's business. They will likely get there before Intel does. With Apple's design team and the very substantial cost advantages of ARM over X86, why would Apple not want to put out a high end ARM powered machine? They could still sell X86 powered machines at the same time for those who still require Windows or legacy software.
X86 is stale. The market is moving to ARM in a very substantial way. If Intel's offerings are no longer compelling, AMD's won't be either.
I am so incredibly happy I put off purchasing the recently released MacBooks. I already have the money saved up for the iPad Pro. And to be totally honest, I don't ever plan on purchasing another Intel powered machine again. I plan on reclaiming the desktop space my current iMac is taking up. Once Apple sells a mini with an A series CPU or the Apple TV becomes powerful enough, I plan on hooking it up to LG's 4k OLED monitor. I honestly don't see a need for anything from Intel as a consumer.
For all the talk of Apple's products being expensive, the X86 CPU is obscenely over priced.
The new A9X definitely an interesting beast... based only on comments by Apple, the A9 processor benchmarks and the last processor that they compared to I expect the A9X processor in the iPad Pro to come in at around 8,000 on the Geekbench 3 benchmarks which would put it above the lowest end of the Macbook pro lineup (but far below the top end). If I were not tied to wanting to run VMWare with Linux/86 for Oracle RDBMS I would not care if they dumped the Intel chips for the low end - but unfortunately there is no ARM processor that compares to the top end of the lineup for Apple. It would mean having a mixed lineup which would only end up confusing the customer - will my computer run this application or the other version.... this was the reason why the Microsoft foray into the Windows RT platform had no chance.
But, with applications being submitted to the app store in bitcode (transportable assembly language) once Apple has the processes in place it could easily make the processor architecture meaningless and hidden from the customer.... which means Apple could have the top end of the lineup on Xeon processors and the low end on ARM and there would be nothing for the customer to have to know about - and would not be confusing (with the exception of running VMWare or Parallels).
I believe it is coming, but it is probably a year or two away from being possible.
Such a chip would allow for further customization of the Mac apart from the general PC market and Apple could do things that Intel is simply not willing to do for sake of preserving their margins. And Apple could release new machines on their own schedule, freeing them from the Intel development cycle. Lowering the cost of the machines while increasing performance is a win win situation. Especially now with the GPUs in Apple's mobile devices outclassing Intel integrated graphics.
Lol. No. If it were that easy we'd see more ARM servers. It isn't easy and ARM inroads into the server market are far below expectations as micro servers didn't explode like folks were claiming it would.
Intel sold 100 million CPUs in Q4 2014. 2015 has been better than 2014 for them from what I remember. They are doing fine and still ahead in process.
There is no way Apple moves from Intel for the MacBooks in the next few years.
Apple's A series of ARM processors are the state of the art period. Intel has nothing that is even comparable.
It is inconceivable that Apple isn't putting a great deal of resources into OS X. They are making incredible profits and the Mac lines are doing well in and of themselves. OS X isn't the resource hog Windows is and should be able to perform well on the A9/A9X. I suspect that all of the available A9/A9X production is going to go into the iPhone/iPad leaving little available for the other lines.
The Apple TV doesn't sell in large numbers, but will be based on the A8 CPU. If the A9 were readily available, Apple very likely would have used the chip instead.
As Samsung and TSMC overtake Intel, the supply constraints for the A series will improve. It isn't hard to foresee Apple putting an ARM chip into a laptop or the mini running OS X.
Intel's lead really isn't much of one in reality anyhow. When the A8X has much higher transistor density at a larger process node and performs comparably to the Core M while consuming less energy, Intel's so-called superior 14 nm process doesn't amount to much.
Samsung and TSMC will be producing far more 14 and 16 nm FF CPUs than Intel will over the next 2 years. I myself am purchasing an iPad Pro. Don't need any of the Intel based machines for what I do. I doubt that the Core M based MacBook can edit 3 simultaneous 4k video streams either.
TSMC and Samsung are now competing with each other. Intel isn't the company driving them. And Apple's CPU/GPU design team is superior to Intel's also.
I will wait for the Mac mini with an ARM CPU priced significantly less than an Intel powered one. It will be coming. There is no way that Intel will be able to keep up going forward charging far higher prices for an x86 CPU than an ARM one that offers better performance.
Apple has shifted CPU platforms before and can easily do it again, especially with their own knowledge of their in house designed CPUs. Going forward, there really isn't anything limiting them also from designing a high performance multi core variant of an ARM CPU that could perform comparably to Intel's Xeons running OS X. Though somewhat unlikely, it is well within the realm of possibility. Especially with TSMC and Samsung attempting to get to 10 nm and 7 nm nodes as quickly as they can in order to win Apple's business. They will likely get there before Intel does. With Apple's design team and the very substantial cost advantages of ARM over X86, why would Apple not want to put out a high end ARM powered machine? They could still sell X86 powered machines at the same time for those who still require Windows or legacy software.
X86 is stale. The market is moving to ARM in a very substantial way. If Intel's offerings are no longer compelling, AMD's won't be either.
I am so incredibly happy I put off purchasing the recently released MacBooks. I already have the money saved up for the iPad Pro. And to be totally honest, I don't ever plan on purchasing another Intel powered machine again. I plan on reclaiming the desktop space my current iMac is taking up. Once Apple sells a mini with an A series CPU or the Apple TV becomes powerful enough, I plan on hooking it up to LG's 4k OLED monitor. I honestly don't see a need for anything from Intel as a consumer.
For all the talk of Apple's products being expensive, the X86 CPU is obscenely over priced.
Great post.
Lemon Bon Bon.
I'll be getting rid of my iMac and getting a 13 inch iPad Pro (The latter of which can handle 3x4k video? Can the Macbook do that? Can an iMac from just a few years ago do that?) for conceptualisation work with Procreate 3 (which looks like being a belter of a release..)
I can wait for a retina iMac that can really 'throw' that 5k display around. ...and then use 'astro' to enable the iPad to act as a tablet for the Mac.
Hmm. Can't wait.
Lemon Bon Bon.