backtop


Print 24 comment(s) - last by superstition.. on Feb 25 at 1:48 AM


A supposedly leaked icon for "ThunderBolt" the rumored name for Apple's upcoming Intel Light Peak implementation.  (Source: Engadget)

A leaked image of uncertain credibility shows ThunderBolt co-inhabiting with the DisplayPort adapter.  (Source: Engadget)

The 13-inch MBP appears to be ditching the discrete GPU, if this leak is to be believed.  (Source: fscklog.com via Engadget)
Is Apple preparing to hurl a Thunderbolt at its users?

Apple typically updates its MacBook Pro lineup at least once a year.  After an update in July, Apple is looking to kick things off a bit early, with a refresh coming later this week according to numerous reports.

Engadget claims to have a leaked spec sheet of the 13-inch model, along with photos of the shiny new Apple notebook.

What is presumably the entry-level model at 13 inches is powered by an Intel 2.3GHz Sandy Bridge i5 processor with 3 MB of L3 cache.  That's down ever so slightly speed-wise from the current generation, but given Sandy Bridge's improvement it should be a bit faster while offering more battery life.

Apple looks to be bumping its DDR3 memory interface from 1066 MHz in the current model to 1333 MHz in the new model.  The hard drive gets a slight capacity bump, as we anticipated, jumping from 250 GB to 320 GB.

The screen resolution (1280x800) remains unchanged, as does the FaceTime camera, wireless adapters, and optical drive.

A minor addition is the inclusion of an SDXC reader, which supports both normal SD cards and the extended capacity (XC) models.

But two things really stand out about the spec sheet.  The first is the fact that Apple has ditched a discrete graphics chip, opting to go with only Sandy Bridge's built in Intel 3000 HD GPU. Hopefully its larger and higher-end models (15", 17") will have new discrete GPUs.

Graphics aside, the other intriguing note is a new port dubbed "Thunderbolt" that apparent co-inhabits the DisplayPort adaptor.  This is supposedly the implementation of Light Peak.

Light Peak is Intel's answer to USB 3.0.  Despite claims of spectacular performance, based on the fiber optics implementation, the communications format in its current form should offer little if any speed gain from USB 3.0 given that it's being reportedly implemented on copper wires.  And where USB 3.0 is an open standard that anyone can use or contribute to, LightPeak is a proprietary standard, which OEMs will have to license from Intel.

If Apple supports Light Peak, it would be the first major OEM to embrace the format.

If the specs sheet holds true, it creates an interesting dilemma for Apple buyers.  The new-ish 13-inch MacBook Air offers a superior discrete GPU, a higher resolutions screen, and is thinner/lighter.  But it comes with a much slower Intel Core 2 Duo processor and lacks built-in optical media capabilities.



Comments     Threshold


This article is over a month old, voting and posting comments is disabled

GeForce 320M is not a discrete solution
By Flunk on 2/23/2011 9:38:27 AM , Rating: 3
The NVIDIA GeForce 320M the comes with the current MacBook Air is a integrated solution that has the gpu on the northbridge. Not a discrete GPU as mentioned above.

Notebookcheck.com has the Geforce 320M rated lower than the Intel HD Graphics 3000 that the supposed spec sheet listed describes.




RE: GeForce 320M is not a discrete solution
By Leper Messiah on 2/23/2011 9:59:48 AM , Rating: 1
Notbookcheck.com's benchmarking process is pretty worthless though. I don't think I've seen any real world benchmarks that say that the HD3000 is better than the 320M.


RE: GeForce 320M is not a discrete solution
By chaos386 on 2/23/2011 10:21:59 AM , Rating: 2
Anandtech has a comparison of the 320M and HD 3000 in Bench: http://www.anandtech.com/bench/Product/241?vs=327&...

The HD 3000 wins most of the benchmarks, although it's paired with a much faster CPU (2.3 GHz quad core Sandy Bridge vs a 2.4 GHz Core 2 Duo). It's not the best comparison, but it's something.


RE: GeForce 320M is not a discrete solution
By Ushio01 on 2/23/2011 2:20:42 PM , Rating: 2
Using your link it shows 6 games being tested each of which has a lowest and medium/highest depending on the game and shows the 320M winning with 4 games at the higher setting looks like a downgrade to me.


By wielander on 2/23/2011 6:38:13 PM , Rating: 2
In each of those cases neither is capable of decent framerates so the benchmarks at the lower settings are far more important. Sandy Bridge wins solidly in synthetics and is better in 2/3rds of the benchmarks above 30fps. The built-in video encoding/decoding tech is far more significant in my opinion, though (cuda video coding is still worthless for consumers and has practically been abandoned).


RE: GeForce 320M is not a discrete solution
By omnicronx on 2/23/2011 11:27:38 AM , Rating: 3
Don't belive the hype, this iteration is a small downgrade from the 320M.. I'm going to guess Apple only now finds its performance acceptable for its use, and having the GPU on die allows for smaller form factors.

In depth tests have also shown that its not exactly faultless either, i.e as usual Intel is going to have to work out the kinks in the drivers.

Also does not truly support OpenCL which I find interesting considering how entrenched it is in OSX.

If anything the fact they can now use newer Intel CPU's is the real story here.. I don't think the switch back to Intel graphics should excite anyone considering their history.


By omnicronx on 2/23/2011 11:39:53 AM , Rating: 2
I would also like to point out that those benchmarks are most likely for Windows based notebooks, none of which I can see are using any sort of LV/ULV variants which could easily have a lower clocked GPU.


RE: GeForce 320M is not a discrete solution
By Flunk on 2/23/2011 1:24:36 PM , Rating: 2
Not the point of my post. My point is that it is not, as described above, a discrete solution. The performance is immaterial to that point.


By omnicronx on 2/23/2011 1:36:51 PM , Rating: 2
You made two points, I replied to one of them.. So clearly performance is of relevance to one of your points. If you don't want someone to comment on one of your points, then stick to one thing at a time ;)

Though you are correct, certainly not a discrete solution with it being on the north bridge and utilizing shared memory and all..


By Solandri on 2/23/2011 3:53:48 PM , Rating: 2
Yeah, I still can't get my notebook to display anything higher than 1366x768 on my 1080p HDTV when it's using the Intel GMA HD. I've tried upgrading drivers, manually tweaking config files you're not supposed to manually config, etc. It displays fine on my 1920x1200 monitor, but it refuses to believe the TV can display 1920x1080. I have to switch it to the nVidia GT 330M to get 1920x1080 on the TV.

That said, it was a really sleazy move by nVidia to release the GT 320M as a discrete graphics chipset, then name their integrated chipset the 320M.


By KoolAidMan1 on 2/23/2011 5:19:17 PM , Rating: 2
Incorrect, the 320M is a little slower than the HD 3000: http://www.anandtech.com/bench/Product/241?vs=327&...

As for OpenCL, apparently Intel has hacked in some sort of OpenCL compute on the CPU itself. It obviously would have additional overhead instead of having it on the GPU (GPU compute is the whole point), but given the choice to go with a Core 2 Duo/NVIDIA solution again or Sandy Bridge, I think Apple made the right call. This will be a stopgap solution until Ivy Bridge (which does have OpenCL) drops next year.


dumb move
By Gungel on 2/23/2011 10:48:34 AM , Rating: 3
Instead of wasting time and money on a new proprietary standard they could have just improved upon USB 3.0 and triple the speed to 12Gbit/s and call it a day. Or wait until an optical solution is actually ready. So know we have LightPeak over copper and in a couple of years throw it all out and get the new LightPeak over optical and 10 times the speed.




RE: dumb move
By mcnabney on 2/23/2011 10:52:50 AM , Rating: 2
I don't see how Light Peak is going to avoid copper entirely for external devices. One of the key benefits to USB is the ability to send power as well as signal. So even if the signal moves to fiber, the connector and cable will still have to deal with having copper in it to provide power (and the limitations of that copper connection).

Now replacing SATA inside the computer with fiber, now that will be nice.


RE: dumb move
By omnicronx on 2/23/2011 1:30:43 PM , Rating: 2
All speculation at this point.. I've heard some many different stories about lightpeak (all which contradict each other) that there is little point even guessing until it is released.


RE: dumb move
By kattanna on 2/23/2011 12:37:55 PM , Rating: 2
quote:
And where USB 3.0 is an open standard that anyone can use or contribute to, LightPeak is a proprietary standard, which OEMs will have to license from Intel


and there is the death knell of it right there.

there is no compelling need above and beyond USB 3.0 for the vast majority of users. so why would the OEMS pay money for something not needed?


RE: dumb move
By XZerg on 2/23/2011 1:43:45 PM , Rating: 2
Ah the good ol' supposedly Billy quotes:

640K ought to be enough for anybody.

I can see many reasons to have faster speed available - imagine daisy chaining multiple monitors with one cable or even being done away with port replicators (docking stations) that require a special port and design for each laptops? You would understand the pain of having a laptop that has no docking station port or being limited to ONE model that cost quarter the price of the laptop.

A faster port could allow for a docking station/port replicator that can have dual monitor output, usb3, esata, network, sound, ... all running simultaneously without any speed issues.

So I definitely disagree with you on USB3.0 provides all the necessary bandwidth which it does not for purpose I just mentioned.


RE: dumb move
By Johnmcl7 on 2/23/2011 2:42:06 PM , Rating: 2
Except there's no evidence Bill Gates or anyone else ever actually said that.


RE: dumb move
By XZerg on 2/23/2011 3:19:55 PM , Rating: 2
I know but it is a very known myth that it was said or is attributed to Billy.


RE: dumb move
By MGSsancho on 2/23/2011 4:33:57 PM , Rating: 2
http://en.wikipedia.org/wiki/DisplayPort

Displayport already support daisy chaining of displays and allows for bidirectional communication, granted at what appears to be USB2.0.

UDI is dead so I will ignore that. http://en.wikipedia.org/wiki/Unified_Display_Inter...

Possibly HDBaseT? http://en.wikipedia.org/wiki/HDBaseT however the position of the thunderbolt in the leaked picture is not next to the ethernet interface so it is not that.

There are many companies backing DisplayPort and I am too lazy to look through all their press releases and product lineups. Apple was one the first major manufacturers to ship displayport and expecially their own open mini-displayport connector. Apple usealy jumps on newer technologies that are arguably better and cheaper to their bottom line. Maybe their implementing more of the spec than competitors or pushing it forward. Look at their SDXC support. Granted there have been linux, windows 7 and official solaris support for a while so this makes sense.

Perhapse someone else can add to or breakdown my theory.


RE: dumb move
By XZerg on 2/23/2011 10:22:37 PM , Rating: 2
i know about the daisy chaining with DP. However my other point was that need a standard port to which docking stations regardless of make or model can connect to and yet have enough bandwidth to cater to esata, usb3, hdmix2 (for 2 screen), network... Power charge would be nice to have but can be optional.


displayport and lightpeak on same port = fail
By zyren on 2/23/2011 12:27:15 PM , Rating: 2
The idea to combine them on the same port is a HUGE mistake. People who use external monitors while at a desk are going to be screwed because they cannot use lightpeak, unless, of course, Apple decides to sell a splitter for $100. The only people i can see taking advantage of lightpeak (or are willing to invest in something that has it) would be professionals, and they would most likely have an external monitor setup.

Way to fail Apple.




By XZerg on 2/23/2011 2:17:53 PM , Rating: 2
Apple doesn't "fail" without a purpose - the obvious need to gouge was more important than what is better for the user. And the users actually like it being screwed in the *ss and are so much more willing to abide by what their magical master wants them to do.


pic
By jonc1028 on 2/23/2011 1:48:26 PM , Rating: 2
Nice use of Zeus from Age of Mythology :)




Firewire 800 is good enough
By superstition on 2/25/2011 1:48:57 AM , Rating: 2
Firewire 800 is good enough for most users. It's unfortunate that the standard didn't catch on very well. Part of the blame rests with Apple, as the company withheld it from some of its product line to try to induce people to spend more.




"Death Is Very Likely The Single Best Invention Of Life" -- Steve Jobs














botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki