The iMac will supplant the Pro. It will come with optional dual processors. They may be limited to quad cores, but possibly hex or octo in a few years.
For those "Pro" users who need the "breakout" capability of the pro frame there may be solace in the form of keeping existing pro boxes but controlled from an iMac/laptop via thunderbolt or USB 3.
Are you a troll or just ignorant? Consumer devices are not going the direction of multi socket configurations. They're headed in the opposing direction. You need dual socket compliant cpus to make this work with Xeons. Outside of Xeons they're going the direction of integrating more and more into the cpu package to cut costs. Using two discrete cpu packages is just not going to happen unless they completely reverse their direction. Core count is a matter of programming, yields, and other things. You may see it increase, but I think they'll allocate the space differently before increasing core count again given the typically less than linear scaling and the desire to preserve margins on their higher end chips. You can debate how much computer is needed for one user or another. Higher end configurations will still exist for many years as server farms still exist today.
Just give me a MacBook Pro update and I will be happy. I may have said this before here or perhaps I mentioned it on another site. I don't want Apple to go the way that E3 did for me.
E3 used to introduce the latest games that people actually wanted and in the past few years has been terrible aside from a few things. I get that some people are interested in iOS 6 and ML and I want to see stuff on that too even though I don't own an iPhone (yet) and probably won't upgrade to ML until next year. Though I always get excited for product updates now and new designs. It's been that way since 2008 which is when I really started to take notice of Apple with the unibody MacBooks.
The comparison though I draw to E3 is if WWDC is just iOS 6 and ML, it'll be like expecting say Nintendo to have the next big Mario or Zelda game only to get Guitar Hero and Wii Fit. Or when the PS3 was introduced, what everyone remembers there (R-r-r-r-r-ridge racer and $599 $599 $599 US Dollars).
Are you a troll or just ignorant? Consumer devices are not going the direction of multi socket configurations. They're headed in the opposing direction. You need dual socket compliant cpus to make this work with Xeons. Outside of Xeons they're going the direction of integrating more and more into the cpu package to cut costs. Using two discrete cpu packages is just not going to happen unless they completely reverse their direction. Core count is a matter of programming, yields, and other things. You may see it increase, but I think they'll allocate the space differently before increasing core count again given the typically less than linear scaling and the desire to preserve margins on their higher end chips. You can debate how much computer is needed for one user or another. Higher end configurations will still exist for many years as server farms still exist today.
Why do you leap to ask if someone is a troll or blast them for ignorance just because they present a different path for discussion? Just answer it without the ad hominem.
Apple abandoned professional servers as their first volley in balancing their offering.
Most appliances we attach to our macs are serial interfaced.
The iMac's are now at their physical largest (27") and sport dual disks (one is SSD to be sure). Why not dual CPU's? While (IMO) there may not be very much need for the MP's monster case, there is always need by video and other compute intense types for high performance machines. The casing of the MP is ridiculously large and the 27" iMac (even if needing a slightly deeper case) can accommodate 2 CPU sockets.
Another reason for the monster MP case is disk drives. With Thunderbolt there is not need to have all those disks inside the cabinet - push them out.
I'm not claiming I'm right just that Apple have a lot more options on paths coupled to a long standing (since the return of Jobs) of streamlining the offerings. That's why the Xserve went away. Are there enough MP sales to justify a new version?
I see there's another thread where someone predicts that the Pro will keep on going. Great if it does.
Because it's physically impossible to put the dual chip platform in the iMac case and no new chips suited for this purpose are slated to be created… ever?
Quote:
With Thunderbolt there is not need to have all those disks inside the cabinet - push them out.
Because it's physically impossible to put the dual chip platform in the iMac case and no new chips suited for this purpose are slated to be created… ever?
Actually it gets easier every year to make te IMac a multiprocessor machine. As integration goes up and feature size goes down space will be freed up for a multi processor iMac.
Now I don't believe Apple will go this route, just that technically it becomes easier to do every year. Especially now that customers are demanding lower thermal expenses from server solutions. Heat in a data center is nothing but waste so it wouldn't be impossible to have a suitable set of work station (Xeons) available for the iMac.
Cabling mess, I believe is the issue there.
Actually cabling is a huge issue here. It isn't just the mess either, the TB connector just doesn't inspire lots of confidence. You have a significantly greater risk of accidental disconnect.
That being said I suspect Apple will try to split off the bulk storage into a separate box. Doing so would allow them to lower the price of the "CPU" box significantly. I can see Apple searching for an answer to the sluggish sales problem the Pro has and one factor there is base system cost.
Beyond that TB isn't magic. If you try to push to much data throught that hose you will slow down your computer. This is a more serious problem than many are willing to acknowledge.
Actually cabling is a huge issue here. It isn't just the mess either, the TB connector just doesn't inspire lots of confidence. You have a significantly greater risk of accidental disconnect.
That being said I suspect Apple will try to split off the bulk storage into a separate box. Doing so would allow them to lower the price of the "CPU" box significantly. I can see Apple searching for an answer to the sluggish sales problem the Pro has and one factor there is base system cost.
Beyond that TB isn't magic. If you try to push to much data throught that hose you will slow down your computer. This is a more serious problem than many are willing to acknowledge.
The fact remains you'd have to rethink the hardware used completely, especially in terms of logic board design. This would be more likely on a tock cycle given that they could use the board for two generations that way.
I've been looking at TB, and to me it's still a first generation product. Coupling with a display is one issue that hinders its use in place of a desktop with a lot of ports and bandwidth. I don't know what the total PCI lane allocation is on mobile boards, so I don't know if they could output both thunderbolt and display signals on separate ports with independent bandwidth pools. You mentioned slowing down the computer. I thought thunderbolt carried a very low cpu overhead, or is the concern that it can saturate the QPI on a laptop? Or are you talking about trying to push too many devices through one port?
The mac pro base system cost is out of line for what really goes in it. They used the internals comparable to budget configurations from other retailers. As you go toward higher end configurations and dual cpu package models, it evens out. Overall though I don't see the imac adopting that socket type anytime soon. I doubt it would fit, and Intel's went with 130W cpus this round.
Anyway the current macbook pro actually pushes some impressive numbers, and I was looking at one because I look at everything. I'd own a laptop either way. It just gets updated less frequently. Bandwidth and ports were an unfortunate problem. When at my desk, I like to use 1-2 large displays. I have some basic hot swappable JBOD storage + backups. I have two dongle key licenses, a wide format printer, a large intuos tablet, random thumb drives, and a couple idevices to synch. In certain cases there are things that have to be plugged into certain ports. With my displays, I'm not sure if they can be calibrated running off a hub. The colorimeter plugs into the display on one, which is a nice feature. The tablet has to plug into a port with full power. It won't run off a hub. Overall my bandwidth needs aren't too terrible outside of saving files. Not wanting to deal with raid means that I accept slower speeds there, but I may be forced to go to a RAID configuration in the near future for bandwidth reasons. SSDs are not appropriate for this yet. They have too many issues, especially in RAID 5. I've banged my head against the wall many times trying to explain to others not to use cheap drives in their RAID 5 setups or that RAID 1 does not cover them for backups.
The point of all of this was that even from a cpu perspective they're often pretty close, the capability to run cool for long periods of time and deal with higher bandwidth needs when necessary is still kind of limited to desktops in a lot of ways, especially when it comes to shared bandwidth.
Even if Apple goes to usb3 this year, will it be like firewire 800 +400 on older imacs where they were running off a single bridge? If you don't remember, it used to be that if both were populated, you ran at firewire 400 speeds due to a shared bridge.
I really don't mind a trend toward smaller hardware. With a lot of software that I've used, it doesn't always address all cores efficiently outside of really long tasks. The point was that Apple would need to change many things before the ceiling available to one becomes identical to the other. If instead they just go for thinner is better and continue to strip bandwidth in favor of better battery life due to fewer parts receiving power, then it will take much longer for such a thing to be feasible. It's important to look at the direction of each product line. If the focus of the lower one stayed on maximizing throughput, it would be different. As of right now the macbook pros still only have 2 usb ports, and i'm not sure if they're on a shared bandwidth pool.
I don't like the idea of losing Ethernet, but 802.11ac would make that a moot point. However, aren't you guys forgetting something??? With new WiFi, there will have to be new AirPort Extreme routers and new Time Capsules. I'm going to guess 4TB in the $499 version and still 2TB (as opposed to 3TB) in the $299 version, so as to make the higher-priced TC a better value.
As for the Retina display, I have one "issue": We've heard a lot about 2880x1800 for the 15" display and while that would be awesome, the amount of SPACE on the desktop will still only be 1440x900, which I consider way too little. The pricier 15" model as it currently stands gives us the option of upgrading to 1680x1050, which, in Retina terms, would have to be a 3360x2100 screen. This is assuming that Apple retains the 16:10 aspect ratio, which I'm almost willing to bet that they won't and instead go with the industry standard, 16:9. If we do see the 16:9 aspect ratio, we will already lose 150 horizontal lines, and in the process valuable desktop space. This would result in a 2880x1536 Retina display (1440x768 doubled). Will we see only one Retina display and will the option for the higher pixel count at the standard "non-Retina" level or will it also be a crazy beautiful high-res high pixel density high ppi proposition? I, for one, sure hope so.
I also predict that the "One more thing..." will still be around even w/o Jobs.
The fact remains you'd have to rethink the hardware used completely, especially in terms of logic board design. This would be more likely on a tock cycle given that they could use the board for two generations that way.
I've been looking at TB, and to me it's still a first generation product.
Exactly! It has barely left the cradle, this is something many on this forum don't grasp.
Coupling with a display is one issue that hinders its use in place of a desktop with a lot of ports and bandwidth. I don't know what the total PCI lane allocation is on mobile boards, so I don't know if they could output both thunderbolt and display signals on separate ports with independent bandwidth pools.
this is a very interesting question as Apple is know to have put the PCI Express bridge to TB on different ports. This could conceivable impact performance but I've yet to see sound benchmarking. If TB sits on the other side of the chipset interface then it has to contend with all traffic across QPI.
You mentioned slowing down the computer. I thought thunderbolt carried a very low cpu overhead, or is the concern that it can saturate the QPI on a laptop? Or are you talking about trying to push too many devices through one port?
Actually to much data. This would especially be an issue if data has to go both ways over QPI.
The mac pro base system cost is out of line for what really goes in it. They used the internals comparable to budget configurations from other retailers. As you go toward higher end configurations and dual cpu package models, it evens out. Overall though I don't see the imac adopting that socket type anytime soon. I doubt it would fit, and Intel's went with 130W cpus this round.
I don't disagree with this but the problem with the low end model is that it still has to cover the same infrastructure as the higher end machines.
Anyway the current macbook pro actually pushes some impressive numbers, and I was looking at one because I look at everything. I'd own a laptop either way. It just gets updated less frequently. Bandwidth and ports were an unfortunate problem. When at my desk, I like to use 1-2 large displays. I have some basic hot swappable JBOD storage + backups. I have two dongle key licenses, a wide format printer, a large intuos tablet, random thumb drives, and a couple idevices to synch. In certain cases there are things that have to be plugged into certain ports.
This is another issue that supports mymposition that USB won't go away anytime soon. Those dongle keys can be a big pain in the rear.
Interestingly you seem to describe needs that will mesh well with the coming new Mac Book Pros. In any event in little over a week you will get to see just how good the new machines mesh.
With my displays, I'm not sure if they can be calibrated running off a hub. The colorimeter plugs into the display on one, which is a nice feature. The tablet has to plug into a port with full power. It won't run off a hub. Overall my bandwidth needs aren't too terrible outside of saving files. Not wanting to deal with raid means that I accept slower speeds there, but I may be forced to go to a RAID configuration in the near future for bandwidth reasons. SSDs are not appropriate for this yet. They have too many issues, especially in RAID 5. I've banged my head against the wall many times trying to explain to others not to use cheap drives in their RAID 5 setups or that RAID 1 does not cover them for backups.
A single RAID in and of itself should never be considered as ones only backup solution. At is something else people don't understand which leads to more head banging.
The point of all of this was that even from a cpu perspective they're often pretty close, the capability to run cool for long periods of time and deal with higher bandwidth needs when necessary is still kind of limited to desktops in a lot of ways, especially when it comes to shared bandwidth.
Even if Apple goes to usb3 this year, will it be like firewire 800 +400 on older imacs where they were running off a single bridge? If you don't remember, it used to be that if both were populated, you ran at firewire 400 speeds due to a shared bridge.
Whatever the configuration it I'll be defined by the Intel hardware if they go Intel. Sadly it seems like Intels USB 3 efforts are half hearted at best.
I really don't mind a trend toward smaller hardware. With a lot of software that I've used, it doesn't always address all cores efficiently outside of really long tasks. The point was that Apple would need to change many things before the ceiling available to one becomes identical to the other. If instead they just go for thinner is better and continue to strip bandwidth in favor of better battery life due to fewer parts receiving power, then it will take much longer for such a thing to be feasible. It's important to look at the direction of each product line. If the focus of the lower one stayed on maximizing throughput, it would be different. As of right now the macbook pros still only have 2 usb ports, and i'm not sure if they're on a shared bandwidth pool.
My point is that Apple needs a different approach to Pro hardware if they ever expect to get the base model within a reasonable price range. Many users can benefit from a higher performance machine but don't need an expensive solution like the Mac Pro. Apple really needs to hit $1200 or so for a base performance machine. That may mean a desktop processor with integrated graphics, with the jump to XEONs occurring further up the ladder.
I don't like the idea of losing Ethernet, but 802.11ac would make that a moot point. However, aren't you guys forgetting something??? With new WiFi, there will have to be new AirPort Extreme routers and new Time Capsules. I'm going to guess 4TB in the $499 version and still 2TB (as opposed to 3TB) in the $299 version, so as to make the higher-priced TC a better value.
no Ethernet I was in a hotel that had in room Ethernet with out wifi.
I don't like the idea of losing Ethernet, but 802.11ac would make that a moot point.
I don't like loosing Ethernet either. More importantly WiFi of any sort is not an acceptable replacement.
However, aren't you guys forgetting something??? With new WiFi, there will have to be new AirPort Extreme routers and new Time Capsules. I'm going to guess 4TB in the $499 version and still 2TB (as opposed to 3TB) in the $299 version, so as to make the higher-priced TC a better value.
Possibly. I've not been able to justify Time Capsules thought. LAN based storage should be expandable otherwise what is the point.
As for the Retina display, I have one "issue": We've heard a lot about 2880x1800 for the 15" display and while that would be awesome, the amount of SPACE on the desktop will still only be 1440x900, which I consider way too little.
That is complete and utter bull crap. The space on a 15" screen is the same if it is composed of one pixel or a hundred billion. As to GUI elements Apple has little choice here as the interface would become unusable at 2X resolution increase. Not that is the GUI, apps can and will make use of the pixel density as they see fit. Just grab an iPad 3 to see the advantage.
The pricier 15" model as it currently stands gives us the option of upgrading to 1680x1050, which, in Retina terms, would have to be a 3360x2100 screen. This is assuming that Apple retains the 16:10 aspect ratio, which I'm almost willing to bet that they won't and instead go with the industry standard, 16:9.
Apple is one of the few companies with enough influence to do as they see fit and a pricing structure that allows them to be successful. It is funny you brought this up as Imwas at WallMart last night and happened through the electronics area. Most of the really cheap laptops are 16:9, that seldom works out well for power users though. I wouldn't be surprised to see Apple stay with 16:10 even if that means $100 extra for each screen.
If we do see the 16:9 aspect ratio, we will already lose 150 horizontal lines, and in the process valuable desktop space. This would result in a 2880x1536 Retina display (1440x768 doubled). Will we see only one Retina display and will the option for the higher pixel count at the standard "non-Retina" level or will it also be a crazy beautiful high-res high pixel density high ppi proposition? I, for one, sure hope so.
I don't think you understand what you are talking about here.
I also predict that the "One more thing..." will still be around even w/o Jobs.
Maybe. I wouldn't be surprised to see a new twist put on WWDC keynotes.
My point is that Apple needs a different approach to Pro hardware if they ever expect to get the base model within a reasonable price range. Many users can benefit from a higher performance machine but don't need an expensive solution like the Mac Pro. Apple really needs to hit $1200 or so for a base performance machine. That may mean a desktop processor with integrated graphics, with the jump to XEONs occurring further up the ladder.
Quoting system is still broken for me for some reason. The problem I frequently find with Apple is configuring a machine that can adequately hit each of my needs. I don't see an imac as likely for me at all. It presents many compromises, yet it doesn't really cover me much better than a laptop (which I'd own in addition anyway). I'm not going to replace my Eizo with the imac display, and 27" is too big for a secondary display. It makes for a very awkward work area. In terms of considering options, okay the quad macbook pro did quite well on some of the tests conducted by barefeats, but I tend to wonder if it can do certain things off the charger without leaning on the battery for extra power. Can it go for longer periods of time without throttling? Is it reliable under heavy use? Can I figure out a port solution for everything? I look at everything before determining a solution. It's certainly easier to configure with a mac pro or any given workstation. In terms of Macs specifically, the mac pro doesn't always get enough attention for its price. Gpu drivers aren't always tuned as well as it's a niche product. They have a raid ROC card that gets nothing but complaints, yet it's still sold as a configuration option. If I can make one of the new macbook pros work, I'll probably buy one of those. That covers me for laptops. If my needs jump, I can add on down the line. The point of contention is more in configuration and functionality than raw power. It's entirely possible to point out individual solutions, but I view it as a checklist. Beyond the basic requirements it must not be prone to failure when subjected to a high duty cycle. I'm aware any electronics can fail, but multiple repair issues would dissuade me in the future.
Because it's physically impossible to put the dual chip platform in the iMac case and no new chips suited for this purpose are slated to be created… ever?
Cabling mess, I believe is the issue there.
Are you suggesting XEON is dead?
Cabling messes can't be avoided. IAC, look at the offerings out there: you can have 4 and more disks in a chassis with a single TB port. Daisy chain MP to disks to displays and it's done. (Not sure how much BW is going to suffer with two large displays and doing disk2disk transfers...).
When you suggested "Because it's physically impossible to put the dual chip platform in the iMac case and no new chips suited for this purpose are slated to be created… ever?"
When you suggested "Because it's physically impossible to put the dual chip platform in the iMac case and no new chips suited for this purpose are slated to be created… ever?"
I'll finish that sentence: "…I was not talking about the Xeon, but rather a mythical chip that could somehow provide the power of a Xeon in the case of an iMac". Intel, as far as I know, isn't planning to create workstation chips that fit in all-in-one desktops.
Comments
Quote:
Originally Posted by alanbrowne
The iMac will supplant the Pro. It will come with optional dual processors. They may be limited to quad cores, but possibly hex or octo in a few years.
For those "Pro" users who need the "breakout" capability of the pro frame there may be solace in the form of keeping existing pro boxes but controlled from an iMac/laptop via thunderbolt or USB 3.
Are you a troll or just ignorant? Consumer devices are not going the direction of multi socket configurations. They're headed in the opposing direction. You need dual socket compliant cpus to make this work with Xeons. Outside of Xeons they're going the direction of integrating more and more into the cpu package to cut costs. Using two discrete cpu packages is just not going to happen unless they completely reverse their direction. Core count is a matter of programming, yields, and other things. You may see it increase, but I think they'll allocate the space differently before increasing core count again given the typically less than linear scaling and the desire to preserve margins on their higher end chips. You can debate how much computer is needed for one user or another. Higher end configurations will still exist for many years as server farms still exist today.
Just give me a MacBook Pro update and I will be happy. I may have said this before here or perhaps I mentioned it on another site. I don't want Apple to go the way that E3 did for me.
E3 used to introduce the latest games that people actually wanted and in the past few years has been terrible aside from a few things. I get that some people are interested in iOS 6 and ML and I want to see stuff on that too even though I don't own an iPhone (yet) and probably won't upgrade to ML until next year. Though I always get excited for product updates now and new designs. It's been that way since 2008 which is when I really started to take notice of Apple with the unibody MacBooks.
The comparison though I draw to E3 is if WWDC is just iOS 6 and ML, it'll be like expecting say Nintendo to have the next big Mario or Zelda game only to get Guitar Hero and Wii Fit. Or when the PS3 was introduced, what everyone remembers there (R-r-r-r-r-ridge racer and $599 $599 $599 US Dollars).
Quote:
Originally Posted by hmm
Are you a troll or just ignorant? Consumer devices are not going the direction of multi socket configurations. They're headed in the opposing direction. You need dual socket compliant cpus to make this work with Xeons. Outside of Xeons they're going the direction of integrating more and more into the cpu package to cut costs. Using two discrete cpu packages is just not going to happen unless they completely reverse their direction. Core count is a matter of programming, yields, and other things. You may see it increase, but I think they'll allocate the space differently before increasing core count again given the typically less than linear scaling and the desire to preserve margins on their higher end chips. You can debate how much computer is needed for one user or another. Higher end configurations will still exist for many years as server farms still exist today.
Why do you leap to ask if someone is a troll or blast them for ignorance just because they present a different path for discussion? Just answer it without the ad hominem.
Apple abandoned professional servers as their first volley in balancing their offering.
Most appliances we attach to our macs are serial interfaced.
The iMac's are now at their physical largest (27") and sport dual disks (one is SSD to be sure). Why not dual CPU's? While (IMO) there may not be very much need for the MP's monster case, there is always need by video and other compute intense types for high performance machines. The casing of the MP is ridiculously large and the 27" iMac (even if needing a slightly deeper case) can accommodate 2 CPU sockets.
Another reason for the monster MP case is disk drives. With Thunderbolt there is not need to have all those disks inside the cabinet - push them out.
I'm not claiming I'm right just that Apple have a lot more options on paths coupled to a long standing (since the return of Jobs) of streamlining the offerings. That's why the Xserve went away. Are there enough MP sales to justify a new version?
I see there's another thread where someone predicts that the Pro will keep on going. Great if it does.
Quote:
Originally Posted by alanbrowne
Why not dual CPU's?
Because it's physically impossible to put the dual chip platform in the iMac case and no new chips suited for this purpose are slated to be created… ever?
Quote:
With Thunderbolt there is not need to have all those disks inside the cabinet - push them out.
Cabling mess, I believe is the issue there.
Now I don't believe Apple will go this route, just that technically it becomes easier to do every year. Especially now that customers are demanding lower thermal expenses from server solutions. Heat in a data center is nothing but waste so it wouldn't be impossible to have a suitable set of work station (Xeons) available for the iMac.
Actually cabling is a huge issue here. It isn't just the mess either, the TB connector just doesn't inspire lots of confidence. You have a significantly greater risk of accidental disconnect.
That being said I suspect Apple will try to split off the bulk storage into a separate box. Doing so would allow them to lower the price of the "CPU" box significantly. I can see Apple searching for an answer to the sluggish sales problem the Pro has and one factor there is base system cost.
Beyond that TB isn't magic. If you try to push to much data throught that hose you will slow down your computer. This is a more serious problem than many are willing to acknowledge.
Quote:
Originally Posted by wizard69
Actually cabling is a huge issue here. It isn't just the mess either, the TB connector just doesn't inspire lots of confidence. You have a significantly greater risk of accidental disconnect.
That being said I suspect Apple will try to split off the bulk storage into a separate box. Doing so would allow them to lower the price of the "CPU" box significantly. I can see Apple searching for an answer to the sluggish sales problem the Pro has and one factor there is base system cost.
Beyond that TB isn't magic. If you try to push to much data throught that hose you will slow down your computer. This is a more serious problem than many are willing to acknowledge.
The fact remains you'd have to rethink the hardware used completely, especially in terms of logic board design. This would be more likely on a tock cycle given that they could use the board for two generations that way.
I've been looking at TB, and to me it's still a first generation product. Coupling with a display is one issue that hinders its use in place of a desktop with a lot of ports and bandwidth. I don't know what the total PCI lane allocation is on mobile boards, so I don't know if they could output both thunderbolt and display signals on separate ports with independent bandwidth pools. You mentioned slowing down the computer. I thought thunderbolt carried a very low cpu overhead, or is the concern that it can saturate the QPI on a laptop? Or are you talking about trying to push too many devices through one port?
The mac pro base system cost is out of line for what really goes in it. They used the internals comparable to budget configurations from other retailers. As you go toward higher end configurations and dual cpu package models, it evens out. Overall though I don't see the imac adopting that socket type anytime soon. I doubt it would fit, and Intel's went with 130W cpus this round.
Anyway the current macbook pro actually pushes some impressive numbers, and I was looking at one because I look at everything. I'd own a laptop either way. It just gets updated less frequently. Bandwidth and ports were an unfortunate problem. When at my desk, I like to use 1-2 large displays. I have some basic hot swappable JBOD storage + backups. I have two dongle key licenses, a wide format printer, a large intuos tablet, random thumb drives, and a couple idevices to synch. In certain cases there are things that have to be plugged into certain ports. With my displays, I'm not sure if they can be calibrated running off a hub. The colorimeter plugs into the display on one, which is a nice feature. The tablet has to plug into a port with full power. It won't run off a hub. Overall my bandwidth needs aren't too terrible outside of saving files. Not wanting to deal with raid means that I accept slower speeds there, but I may be forced to go to a RAID configuration in the near future for bandwidth reasons. SSDs are not appropriate for this yet. They have too many issues, especially in RAID 5. I've banged my head against the wall many times trying to explain to others not to use cheap drives in their RAID 5 setups or that RAID 1 does not cover them for backups.
The point of all of this was that even from a cpu perspective they're often pretty close, the capability to run cool for long periods of time and deal with higher bandwidth needs when necessary is still kind of limited to desktops in a lot of ways, especially when it comes to shared bandwidth.
Even if Apple goes to usb3 this year, will it be like firewire 800 +400 on older imacs where they were running off a single bridge? If you don't remember, it used to be that if both were populated, you ran at firewire 400 speeds due to a shared bridge.
I really don't mind a trend toward smaller hardware. With a lot of software that I've used, it doesn't always address all cores efficiently outside of really long tasks. The point was that Apple would need to change many things before the ceiling available to one becomes identical to the other. If instead they just go for thinner is better and continue to strip bandwidth in favor of better battery life due to fewer parts receiving power, then it will take much longer for such a thing to be feasible. It's important to look at the direction of each product line. If the focus of the lower one stayed on maximizing throughput, it would be different. As of right now the macbook pros still only have 2 usb ports, and i'm not sure if they're on a shared bandwidth pool.
New Pro announced. Uses many, many ARM processors.
I don't like the idea of losing Ethernet, but 802.11ac would make that a moot point. However, aren't you guys forgetting something??? With new WiFi, there will have to be new AirPort Extreme routers and new Time Capsules. I'm going to guess 4TB in the $499 version and still 2TB (as opposed to 3TB) in the $299 version, so as to make the higher-priced TC a better value.
As for the Retina display, I have one "issue": We've heard a lot about 2880x1800 for the 15" display and while that would be awesome, the amount of SPACE on the desktop will still only be 1440x900, which I consider way too little. The pricier 15" model as it currently stands gives us the option of upgrading to 1680x1050, which, in Retina terms, would have to be a 3360x2100 screen. This is assuming that Apple retains the 16:10 aspect ratio, which I'm almost willing to bet that they won't and instead go with the industry standard, 16:9. If we do see the 16:9 aspect ratio, we will already lose 150 horizontal lines, and in the process valuable desktop space. This would result in a 2880x1536 Retina display (1440x768 doubled). Will we see only one Retina display and will the option for the higher pixel count at the standard "non-Retina" level or will it also be a crazy beautiful high-res high pixel density high ppi proposition? I, for one, sure hope so.
I also predict that the "One more thing..." will still be around even w/o Jobs.
Interestingly you seem to describe needs that will mesh well with the coming new Mac Book Pros. In any event in little over a week you will get to see just how good the new machines mesh. A single RAID in and of itself should never be considered as ones only backup solution. At is something else people don't understand which leads to more head banging. Whatever the configuration it I'll be defined by the Intel hardware if they go Intel. Sadly it seems like Intels USB 3 efforts are half hearted at best.
My point is that Apple needs a different approach to Pro hardware if they ever expect to get the base model within a reasonable price range. Many users can benefit from a higher performance machine but don't need an expensive solution like the Mac Pro. Apple really needs to hit $1200 or so for a base performance machine. That may mean a desktop processor with integrated graphics, with the jump to XEONs occurring further up the ladder.
Quote:
Originally Posted by Crunch
I don't like the idea of losing Ethernet, but 802.11ac would make that a moot point. However, aren't you guys forgetting something??? With new WiFi, there will have to be new AirPort Extreme routers and new Time Capsules. I'm going to guess 4TB in the $499 version and still 2TB (as opposed to 3TB) in the $299 version, so as to make the higher-priced TC a better value.
no Ethernet I was in a hotel that had in room Ethernet with out wifi.
Maybe. I wouldn't be surprised to see a new twist put on WWDC keynotes.
Even when they have WiFi it often has serious issues. There are many reasons beyond this one though to justify the desire for Ethernet.
Quote:
Originally Posted by wizard69
My point is that Apple needs a different approach to Pro hardware if they ever expect to get the base model within a reasonable price range. Many users can benefit from a higher performance machine but don't need an expensive solution like the Mac Pro. Apple really needs to hit $1200 or so for a base performance machine. That may mean a desktop processor with integrated graphics, with the jump to XEONs occurring further up the ladder.
Quoting system is still broken for me for some reason. The problem I frequently find with Apple is configuring a machine that can adequately hit each of my needs. I don't see an imac as likely for me at all. It presents many compromises, yet it doesn't really cover me much better than a laptop (which I'd own in addition anyway). I'm not going to replace my Eizo with the imac display, and 27" is too big for a secondary display. It makes for a very awkward work area. In terms of considering options, okay the quad macbook pro did quite well on some of the tests conducted by barefeats, but I tend to wonder if it can do certain things off the charger without leaning on the battery for extra power. Can it go for longer periods of time without throttling? Is it reliable under heavy use? Can I figure out a port solution for everything? I look at everything before determining a solution. It's certainly easier to configure with a mac pro or any given workstation. In terms of Macs specifically, the mac pro doesn't always get enough attention for its price. Gpu drivers aren't always tuned as well as it's a niche product. They have a raid ROC card that gets nothing but complaints, yet it's still sold as a configuration option. If I can make one of the new macbook pros work, I'll probably buy one of those. That covers me for laptops. If my needs jump, I can add on down the line. The point of contention is more in configuration and functionality than raw power. It's entirely possible to point out individual solutions, but I view it as a checklist. Beyond the basic requirements it must not be prone to failure when subjected to a high duty cycle. I'm aware any electronics can fail, but multiple repair issues would dissuade me in the future.
My prediction (aka wild assed guess based on my own hopes and desires):
All those TBA dev sessions are about Apps on the aTV and new BT gaming controller API in the iOS 6 SDK...
Quote:
Originally Posted by shawkin
New Pro announced. Uses many, many ARM processors.
I wouldn't laugh. I think very soon we'll see a "thin client" OS X from Apple on top of a multicore ARM.
Quote:
Originally Posted by Tallest Skil
Because it's physically impossible to put the dual chip platform in the iMac case and no new chips suited for this purpose are slated to be created… ever?
Cabling mess, I believe is the issue there.
Are you suggesting XEON is dead?
Cabling messes can't be avoided. IAC, look at the offerings out there: you can have 4 and more disks in a chassis with a single TB port. Daisy chain MP to disks to displays and it's done. (Not sure how much BW is going to suffer with two large displays and doing disk2disk transfers...).
Quote:
Originally Posted by alanbrowne
Are you suggesting XEON is dead?
Nowhere did I say that in any capacity.
Quote:
Originally Posted by Tallest Skil
Nowhere did I say that in any capacity.
When you suggested "Because it's physically impossible to put the dual chip platform in the iMac case and no new chips suited for this purpose are slated to be created… ever?"
While I've riled a few spirits with the notion that it's the end of the MPro, today I saw the following:
http://www.latimes.com/business/technology/la-fi-tn-apple-parts-leaked-20120606,0,4312732.story
(their source is 9to5Mac)
indicating a new MPro is a coming as well as MBP, MBA and iMac (21.5/27) and a new Airport Express.
Quote:
Originally Posted by alanbrowne
When you suggested "Because it's physically impossible to put the dual chip platform in the iMac case and no new chips suited for this purpose are slated to be created… ever?"
I'll finish that sentence: "…I was not talking about the Xeon, but rather a mythical chip that could somehow provide the power of a Xeon in the case of an iMac". Intel, as far as I know, isn't planning to create workstation chips that fit in all-in-one desktops.