Liveblog Of Apple's October Event

HardOCP News

[H] News
Joined
Dec 31, 1969
Messages
0
If you want the up-to-the-minute info on Apple's event this afternoon, we definitely recommend Engadget's liveblog. The event kicks off at 10am PT, the throngs of media are gathering outside the event as we speak.
 
ouch!
So Microsoft is pretty much the only one still trying to make money off their operating system?
 
Nice one Apple. Awesome job "re-innovating" the sham RAM compression technology. Straight off their slides, "6GB of data in 4GB of RAM".


All aboard the train to 1995.

s8XxEku.jpg
 
I like the physical design, if it were about $2K less.

The problem is, it is a computer aimed at businesses that doesn't have anything businesses want like standardized parts, repairability, or manageability. Home users don't tend to buy Xeon-equipped workstations nor do they need things like ECC ram.
 
mavericks is using the WKdm compression algorithm
http://terpconnect.umd.edu/~barua/matt-compress-tr.pdf

It's not a matter of algorithm speed.

Each program running has its own memory space. To be fair, I don't have the details of Apple's scheme but it's probably something like: compresses least recently used RAM or background process RAM. When the system/user wants to access a certain running program it switches currently running programs to compressed RAM and restores so the application you are switching to into uncompressed space.

The speed loss is in the compressing/decompressing and switching memory space around. Bottom line, there is no substitute for real RAM.
 
"The competition doesn't seem to have a clear idea about where the PC industry should go" Cook stated, "They're confused; they chased after netbooks. They tried to make tablets into PCs and PCs into tablets." Cook made these comments while the screen behind him showed a graphic of a street sign that was filled with a squiggly street icon."

Boy if that ain't the truth. This household will continue to remain Apple-free but man tells the truth about MS.

Michael-Scott-Failing-to-Hold-In-Laughter.gif
 
ouch!
So Microsoft is pretty much the only one still trying to make money off their operating system?

No you pay for OSX. You just pay for it when you buy the hardware.

With the exception of Surface, Microsoft doesn't sell the hardware, so they have to charge for the OS.
 
What is the right direction, then? Computers that look like toilet brush holders with proprietary connectors that no one uses?
 
I'll underline and bold the important factor since you missed it.

switching memory space around.

Can't edit posts,damnit i'll just add on here:

It's also important to think about costs of RAM relative to the time. Today changing from 4GB RAM to 16 GB RAM when building a new PC is about a $60 difference (I'm assuming you are building your own). In 1995 jumping from the average 4 MB RAM to 16 MB RAM was a $370 dollar price increase... or in 2012 dollars, $550. Now that's a huge difference. It's easy to see why promising that you could install a $39.99 piece of software and instantly "double your RAM" was enticing. Today, not so much, just throw a couple more sticks in and call it a day. Plus as mentioned, you get the benefits of having real RAM.

Sources:
http://www.jcmit.com/memoryprice.htm
http://www.westegg.com/inflation/
 
Can't edit posts,damnit i'll just add on here:

It's also important to think about costs of RAM relative to the time. Today changing from 4GB RAM to 16 GB RAM when building a new PC is about a $60 difference (I'm assuming you are building your own). In 1995 jumping from the average 4 MB RAM to 16 MB RAM was a $370 dollar price increase... or in 2012 dollars, $550. Now that's a huge difference. It's easy to see why promising that you could install a $39.99 piece of software and instantly "double your RAM" was enticing. Today, not so much, just throw a couple more sticks in and call it a day. Plus as mentioned, you get the benefits of having real RAM.

Sources:
http://www.jcmit.com/memoryprice.htm
http://www.westegg.com/inflation/

It is only a matter of time before Macs require new iRAM™ with new and improved Retina™ bits.
 
Can't edit posts,damnit i'll just add on here:

It's also important to think about costs of RAM relative to the time. Today changing from 4GB RAM to 16 GB RAM when building a new PC is about a $60 difference (I'm assuming you are building your own). In 1995 jumping from the average 4 MB RAM to 16 MB RAM was a $370 dollar price increase... or in 2012 dollars, $550. Now that's a huge difference. It's easy to see why promising that you could install a $39.99 piece of software and instantly "double your RAM" was enticing. Today, not so much, just throw a couple more sticks in and call it a day. Plus as mentioned, you get the benefits of having real RAM.

Sources:
http://www.jcmit.com/memoryprice.htm
http://www.westegg.com/inflation/
no one can just "throw a couple more sticks" in to their MacBooks and call it a day
ram doubler wasn't a sham/scam. some products did make promises they couldn't deliver but Connectix' Ram Doubler was not one of them.

extra, unused RAM is a waste of money and power draw
modern computing is moving toward greater efficiency rather than brute force, raw power

I'd like to see where you are getting the information that there is a significant/noticeable performance hit from "switching memory space around." I did provide a peer reviewed article substantiated my point that compression/decompression speed is a function of the algorithm used.
 
When the system/user wants to access a certain running program it switches currently running programs to compressed RAM and restores so the application you are switching to into uncompressed space.
When your alternative is to page to a disk or to simply add more memory (which you, in some cases, simply cannot do), this starts looking like a reasonably attractive option, assuming it's faster than paging.

It's also important to think about costs of RAM relative to the time. Today changing from 4GB RAM to 16 GB RAM when building a new PC is about a $60 difference (I'm assuming you are building your own).
In the past year, DDR3 pricing has increased by a factor of around 1.5. It's still relatively cheap, but it still has a cost.
 
"The competition doesn't seem to have a clear idea about where the PC industry should go" Cook stated, "They're confused; they chased after netbooks. They tried to make tablets into PCs and PCs into tablets." Cook made these comments while the screen behind him showed a graphic of a street sign that was filled with a squiggly street icon."

Boy if that ain't the truth. This household will continue to remain Apple-free but man tells the truth about MS.

This is how I feel, Microsoft simply lacks a coherent vision of the future of PCs. Honestly I think they are in the best position, but have no clue where they want to go...At least Apples products show some innovation. I am not always sure I like the thought, but at least they move the bar forward.

Lenovo seems to be the only PC firm really trying to innovate. I thought Asus was going to do well but they quickly became lost. Sometimes you have to wonder if the PC's markets focus on cost is the reason we lack innovation. That anything innovative we are rarely willing to pay for and only firms with lower cost structures can afford to spend anything on R&D.
 
I think there are different measures of innovation. What Microsoft did with Windows 8, for better or for worse, is kind of innovative. What Apple is doing with the Mac Pro, for better or for worse, is also innovative (and not just because it comes in a cylinder).

I'd argue that Microsoft simultaneously knows where they are and where they're going and lost. There are players in the PC industry who are totally clueless and there are those who have heads on their shoulders: it's just very company-specific.
 
I can't believe they fit all that into that small package.

I can; Take an mITX motherboard, chop off the expansion slots, split it into thirds and you have a system smaller than the TrashCan Mac.
 
This is how I feel, Microsoft simply lacks a coherent vision of the future of PCs. Honestly I think they are in the best position, but have no clue where they want to go...At least Apples products show some innovation. I am not always sure I like the thought, but at least they move the bar forward.

One may not like what Microsoft did with Windows 8 but the vision it pretty clear, a single platform that can run on all of today's computing form factors and input methods. There is the issue about the future of the conventional desktop, and while many are proclaiming that Microsoft is trying to get rid of it, that simply doesn't make a lot of sense nor is even feasible anytime soon. Now it may be that certain devices in the future may not come with the traditional desktop, like future Windows RT devices or phones running full RT, but x86 devices aren't going to loose the desktop for many years to come. But modern apps will be able to run natively on phones, desktops, tablets and probably consoles. Right now the only thing missing is phones and consoles though the Xbox

Lenovo seems to be the only PC firm really trying to innovate. I thought Asus was going to do well but they quickly became lost. Sometimes you have to wonder if the PC's markets focus on cost is the reason we lack innovation. That anything innovative we are rarely willing to pay for and only firms with lower cost structures can afford to spend anything on R&D.

There are some new things happening with PCs, they are getting very thin, light and battery efficient. The new 8" Bay Trail devices aren't about the same size and weight as the iPad Minis with equivalent battery life and cheaper in some cases. Hook one of the Bay Trail devices up to an external monitor, mouse and keyboard and there's enough there to serve as a workstation for many folks.
 
I still like the Mac Pro G5 case. In the process of building a "Haswell" mac using that case.

PC Cases look like shit compared to Macs. They always have since the 1st mac showed up.
 
no one can just "throw a couple more sticks" in to their MacBooks and call it a day

Sounds like a design problem to me. Even my shitty Gateway laptop allows memory expansion and provides simple access to the HD. For the price Apple commands it should be a premium product, with maximum expansion options.



ram doubler wasn't a sham/scam. some products did make promises they couldn't deliver but Connectix' Ram Doubler was not one of them.

It wasn't the same as literally doubling your RAM which is what's printed on the box. In cases where you had lots of small applications open it did provide an improvement for sure, but it did NOT "Double your RAM".

RAMDoubler2-276x300.jpg



extra, unused RAM is a waste of money and power draw
modern computing is moving toward greater efficiency rather than brute force, raw power

I suppose it's a "waste" if you don't need it, but if you do need more RAM instead of trying software tricks just buy more RAM, RAM is cheap now.

On the power draw, technically it's true, the draw is so trivial that extra RAM power consumption isn't worth worrying about, IIRC when I added an extra 8 GB into my Sandy Bridge build my power draw went up 3 watts. If we are thinking about mobile applications and battery life, I'd guess that running the CPU cycles for compression/decompression probably offsets the power savings of less RAM, but without seeing data I couldn't definitively say either way.

I'd like to see where you are getting the information that there is a significant/noticeable performance hit from "switching memory space around." I did provide a peer reviewed article substantiated my point that compression/decompression speed is a function of the algorithm used.

I remember reading a very in depth article about it eons ago. I'd find it and link it to you but after ~18 years good luck tracking it down, especially since it was physical print, lol. The testing was quite in depth, they took a machine, ran tests without the "RAM doubling" applications, with the "RAM doubling" applications, and then without the RAM doubling applications but actually with double the RAM. The "RAM doublers" did next to nothing performance wise except when there were lots of small programs open, in which chase there was a performance gain but still not as much as just doubling the RAM.

From memory, essentially the issue was partly due to the actual compression and decompression taking time, but an even larger factor was the fact that a PC can't run programs in compressed memory. Keep in mind a program is set to a certain address space, that's where the OS expects it to be when using it. So the "RAM Doubler" would have to swap/move any stored/compressed data that was in that programs address space, then compress the currently running application and store it somewhere, and uncompress the application that you wanted to use to the memory address ranges that the PC though the program was running in. Keep in mind, when you are stuffing more things into the same space, there is going to be overlap in the address space compared to what the PC thinks is there. Handling these issues through a piece of software so it's transparent to the host OS is going to create a ton of memory writes/moves, which takes up time.
 
Sounds like a design problem to me. Even my shitty Gateway laptop allows memory expansion and provides simple access to the HD.
How much larger is your Gateway laptop? Expandable memory is a space constraint more than it is a cost constraint.
 
How much larger is your Gateway laptop? Expandable memory is a space constraint more than it is a cost constraint.

Then make it bigger.

I have a 17 inch laptop that I tote on the road between multiple offices and I have no problem carrying it. I have four memory slots, two SATA bays, an MSATA slot, two mini PCIe slots, an optical drive bay, Thunderbolt/DP, four USB 3.0, HDMI, VGA and gigabit ethernet. No Macbook even comes close to offering that level of connectivity or expandability. Why anyone would want to work on some dinky 12 inch screen is beyond me unless they are out of shape because I don't find the 17 inch very heavy.
 
I don't make Apple laptops. You'll want to send your feedback to Tim Cook. I'm certain he'd take great interest in your suggestions to make their popular notebook products larger and to include VGA ports.
 
I can; Take an mITX motherboard, chop off the expansion slots, split it into thirds and you have a system smaller than the TrashCan Mac.

And it would have Xeons, ECC RAM, dual GPU, Thunderbolt 2, and drive three 4K monitors.

Oh, wait....
 
And it would have Xeons, ECC RAM, dual GPU, Thunderbolt 2, and drive three 4K monitors.

Oh, wait....

All of which can be supported via the chipset. When it really comes down to it, it doesn't take any more traces than it already does to support memory sticks, 8+4 USB2/3 ports, DVI port, VGA port, HDMI port, Serial, parallel, gigabit, optical out, Dual PS2...

All of which are built into my mITX system.

Cut out the traces needed for half that shit, you have enough for the rest... and even some left over.

It really isn't the engineering marvel that people are making it out to be. Same shit, different package.
 
Techrat says it doesn't matter even though nobody has done it before

Don't be a dumbass.

Silent computing HAS been done before.

Reducing footprints HAS been done before.

Making proprietary components HAS been done before.

Triangular shaped heatsinks HAS been done before.

Hell. Even cylindrical computers have been done before.

Apple lists the entire Mac Pro at 9.9 inches tall. Judging from the gut pics, about 25% of it is the case frame, fan and vents. This places the height of the board at about 7.43 inches, rounding up to 7.5 just for simplicity.

Wtkxbar.jpg


The largest of the three boards is almost square. I would say it's slimmer by approx 25% again... so 25% off 7.5" is 5.625... Rounding down to 5.6"

Main board is 7.5 x 5.625"

Mini-ITX 6.7 x 6.7"

And this is just ONE of the three pieces inside that Mac Pro.

Bonus: A lot of those pieces are double sided with the chips facing the heatsink and other components on the other side of the board. That's additional utilized surface area compared with traditional motherboards.

Let me recap what I originally responded to and said.

I can't believe they fit all that into that small package.

I can; Take an mITX motherboard, chop off the expansion slots, split it into thirds and you have a system smaller than the TrashCan Mac.

Science, bitch.
 
Right, which is why so many OEMs have made such powerful workstations so quiet and compact.

They're literally everywhere.
 
Back
Top