Andrew Richards
Forum Replies Created
-
[Chris Harlan] “And why–after all this time–would Apple finally embrace USB3? Not that I don’t think they should. They should have done that a couple of years ago. But why would they now?”
I’m not saying they would, just that they could. Apple doesn’t need to wait for Intel to integrate it into their system controllers. Apple was not an early adopter of USB2 either. Plus, USB3 is hardly ubiquitous even now.
Heck, given how much Apple is outpacing the rest of the industry with Mac sales growth since Thunderbolt came out, and given how USB3 is still rather uncommon on PCs, I wouldn’t be surprised to see them neck and neck in adoption rates despite USB3’s age.
Best,
Andy -
[Tapio Haaja] “I’d really love to see new Macpro or EOL information soon so we can stop this speculating where Apple is heading :)”
Amen to that!
Best,
Andy -
Andrew Richards
March 29, 2012 at 1:44 pm in reply to: What’s your favorite file recovery app for a Mac? -
[Tapio Haaja] “What I read Ivy Bridge server processors are coming out next fall…”
Given how much trouble Intel had shipping the Sandy Bridge E5 Xeons, and the fact that Ivy Bridge is using their new 3D tri-gate transistors, I wouldn’t bet on an on-time launch for Ivy Bridge E5 Xeons. Unless by “next fall” you mean Fall 2013, then I’d be pretty confident they’ll ship by then.
[Tapio Haaja] “…so I guess there is possibility that Apple don’t want to release new MacPro now with much delayed Sandy bridge server processor but instead will wait to Ivy Bridge server processors because with Ivy Bridge they can have USB3 ports and possible next-gen thunderbolt ports.”
I don’t see the logic. They waited this long for the E5 Xeon and would just skip a generation? Why? Either they’re in the market or they aren’t. Not shipping a Sandy Bridge Xeon Mac Pro is a tacit EOL, IMHO. Intel is’t going to continue producing the Gainestown and Westmere Xeons found in the current (2010) Mac Pro, so if the Mac Pro has a future, it lies with the chips that Intel is manufacturing.
Thunderbolt is of little consequence on a tower equipped with PCIe 3.0 slots that offer more than an order of magnitude greater bandwidth than Thunderbolt.
[Tapio Haaja] “And I guess Macbook Pro refresh what we’ll propably see in a month will include USB3 ports. So do they wan’t to release new MacPro now with features that are somewhat worse than in their Macbook Pro?”
USB3 can be added with a third party controller if Apple is worried about USB feature parity.
[Tapio Haaja] “So I guess there’s possibility that we won’t know until next fall if MacPro is EOL.”
As far as I’m concerned, we’ll know this month (April). If HP and other vendors are shipping E5-based workstations and Apple isn’t, then I’d say Apple is clearing the channel before quietly EOL-ing the Mac Pro as we know it.
Best,
Andy -
[Liam Hall] “You won’t get new Mac Pros until Intel’s new Ivy Bridge processors are released – pos last week of April.”
The Ivy Bridge chips coming this year are not Xeons, thus not dual processor. If Apple were going to abandon the Xeon in the Mac Pro, they could have done so last year with Sandy Bridge-E. Either there will be a new Mac Pro based on the E5-2600 Xeon that Intel just started shipping (the same chips that will go into the HP Z820), or there won’t be a new one at all.
Marco Arment lays it out well.
Best,
Andy -
I don’t mean this to seem glib, because it really, really, really sucks to lose data (I have), but this is an excellent example of why RAID is not a backup.
RAID can protect you from hardware failure of HDD(s). It does not protect you from (among other risks) file system corruption, accidental deletion, or in this case, accidental reformatting of the filesystem. I know no one in this thread claimed otherwise, but for any future readers, I think the scenario witnessed in this thread can be instructive.
Best,
Andy -
Try DiskWarrior and cross your fingers. If he really did just reformat the volume and immediately power down, then the data should still be there. Be prepared for the recovery attempt to take a very loooooooong time.
Best,
Andy -
Andrew Richards
March 26, 2012 at 1:14 pm in reply to: What do you REALLY want to see at NAB? (Non-Apple edition)[Bernhard Grininger] “The current workaround sending a live signal out of the computer
to scale and re-capture it is simply a pain! It doesn’t fit into the 21st century!”THIS!
Long overdue!
Best,
Andy -
[Walter Soyka] “It has six times the number of CUDA cores, but I’d also note that it has the same amount of RAM (2 GB) — so while performance will almost certainly be higher across the board, there may be some cases where memory will cause a bottleneck.”
They are also PCIe 3.0, so the data can get in and out twice as fast and might require less RAM for buffering. I guess.
Best,
Andy -
[Lance Bachelder] “Well FCPX and Motion need OpenCL which the nVidia does support.”
…barely. They use it the same way PPro uses CUDA- to improve performance and enable more real-time processing of effects and codecs. I agree it is a poke in the eye, but it is in keeping with Apple’s MO to build for their own offerings. It could happen; maybe offering a BTO card that suits third party apps while hamstringing Apple’s own is just the kind of deviation from Apple’s norm we’ll see more of under Tim Cook. It would be refreshing, for sure. It’s one thing to not put them in iMacs, but to pretend that Mac Pros are only used for Apple’s software? Silly.
Plus it could be that OpenCL isn’t a weakness, just that particular OpenCL benchmark is a weakness.
Best,
Andy