PowerBook Updates in September?

[eric67]

completely nuts

30 F***KIN MHZ IS NOT AN UPDATE!!!!!!! BOOOOOOOOOOOOOOOOOOOOOOO!!!!!!!!

I thought that Macusers did not believe into MHz myth??

Just look at the specs!!

the 7448 will allow better battery lifespan, boost performance around 15% thanks to a larger cache and higher bus. So what do you want?? you will get a cooler and more performing processor?? the best way to wait for the Yonah dualcore 2.13GHz-based powerbook in June 2006.

 

[MikeAtari]

http://www.theinquirer.net/?article=25388

If the INQUIRER's story is true, we can only hope, the FSB will run at 533mhz, making the DDR2 usable.

Update: Actually the bus speed could be 566 * 3 = 1698 or 1.7ghz.

 

[MikeAtari]

And Another Thing...

IF Apple can get their hands on the MPC8641, then does that imply that the MPC8641D is available? The 8641 runs at 15 watts, which is nice, but 8641D would be a great LAST PowerPC laptop, if possible.

Come On Steve, leave in a hail of GLORY!

 

[Platform]

Yeah it has been a while. We should see something in this space.

Nah......iLife and iWork will proberly not get updated until MWSF 06.......iLife 06 and iWork 06

 

[eric67]

hard to believe

If the Register's story is true, we can only hope, the FSB will run at 533mhz, making the DDR2 usable.

I do not believe it, the inquirer is not also reliable.
Freescale has announced the 7448 in mass production with availability in October. So having the MPC8641 in the next PB I do not believe it because it would imply also that Apple would have run into the design and the production of a new motherboard. indeed the MPC8641 is not compatible with the current motherboard of PB; whereas the 7448 is pin-to-pin compatible.

in addition, the MPC8641 requres a RapidI/O interface, something not developed by Apple yet. So knowing that the PB will probably be the first Apple model to be equipped with Intel Yonah dualcore processor as early as June 2006. so Do you really think that Apple will invest in hardware design and development for only a couple of months?? I do not think so.
So the update will be based on 7448, toping officially at 1.7GHz (Freescale website) but Apple could probably overclock it to 1.8 or 2.0GHz only if it does not impact too much heat release and battery lifespan.

 

[shawnce]

I got $50 that says P4's or P5's and Xeons will still be alive and kicking in 2007. It won't be Netburst...

Which is exactly what I am saying. NetBurst is dead come the end of 2006 and in many ways is dead from a development point-of-view already. I am also saying that Apple won't use NetBurst CPUs in shipping MacIntels.

Intel has a new desktop core design (family) that they are switching over to use (know more about it next week from the horses mouth). It will run at clock rates lower (I believe in the 2-3GHz range) then current P4s yet out perform them and burn less power. It appears to differ more from NetBurst then similar to NetBurst given what information is out in the public currently.

No dual processor Pentium M only dual core in the future (which puts them far behind todays Opteron and Xeon platforms)

The ability to build multiprocessor systems is primarily a fact of the interconnect on and off die not so much the core itself. Intel chooses what the interconnect will support for a given chip family, they could choose to support multi processor (talking separate chips not multi-core) with really any of their cores if they wanted.

No cheap dual core like Intel offers with the P4. Pentium M's are simply more expensive to make.

Don't fool yourself with pricing of current chips, for one a lot of that is a construct of the marketing and volumes not actual production costs. For example the die size of the Pentium M is 84 mm^2, and the die size for the Pentium D is 206 mm^2 (both in 90nm process). It is cheaper to crank out the Pentium M then the Pentium D because you can fit more per silicon wafer and the packaging is similar.

reference 1
reference 2
reference 3

 

[MikeAtari]

I do not believe it, the inquirer is not also reliable.
Freescale has announced the 7448 in mass production with availability in October. So having the MPC8641 in the next PB I do not believe it because it would imply also that Apple would have run into the design and the production of a new motherboard. indeed the MPC8641 is not compatible with the current motherboard of PB; whereas the 7448 is pin-to-pin compatible.

in addition, the MPC8641 requres a RapidI/O interface, something not developed by Apple yet. So knowing that the PB will probably be the first Apple model to be equipped with Intel Yonah dualcore processor as early as June 2006. so Do you really think that Apple will invest in hardware design and development for only a couple of months?? I do not think so.
So the update will be based on 7448, toping officially at 1.7GHz (Freescale website) but Apple could probably overclock it to 1.8 or 2.0GHz only if it does not impact too much heat release and battery lifespan.

- Yes, the 7448 is pin-for-pin compatible, but, the fsb has increased to 200mhz, what work needs to be done on a motherboard to accomadate that change?
- The 8641 requires RapidI/O but, this isn't yesterday's news to Apple, surely they knew about this last year?
- If it's just the 7448, then there's no reason to limit the speed to 1.7 as the 7447A can hit 2ghz, but, if it's the 8641 with a 566 bus then 566 * 3 makes some sense.

But, yes I know all this is pure BS speculation, and the only real indiaction of what Apple WILL Do, is when they actaully ship a product.

 

[Doctor Q]

I thought that Macusers did not believe into MHz myth??

Perhaps you were joking, but for the record, the myth is that it is useful to compare megahertz of one processor to megahertz of another processor. When you compare two speeds of the same processor, such as a rumored 1.67GHz to 1.7GHz update for the PowerBook G4, it's science, not myth. And science says that 1.67 is not much different than 1.7. It would take other improvements, such as higher resolution, to make this update anything to get excited about.

 

[eric67]

7448 rather than MPC8641

- Yes, the 7448 is pin-for-pin compatible, but, the fsb has increased to 200mhz, what work needs to be done on a motherboard to accomadate that change?
- The 8641 requires RapidI/O but, this isn't yesterday's news to Apple, surely they knew about this last year?
- If it's just the 7448, then there's no reason to limit the speed to 1.7 as the 7447A can hit 2ghz, but, if it's the 8641 with a 566 bus then 566 * 3 makes some sense.
But, yes I know all this is pure BS speculation, and the only real indiaction of what Apple WILL Do, is when they actaully ship a product.

answer:
- Apple can overclock the chipset controling the clockspeed of the motherboard from 166 to 200MHz.
- there is a reason to limit to 1.7GHz. Freescale has announced that mas production of the 7448 is only available for 7448 clocked from 1GHz to 1.7GHz. so the only way for apple to propose a higher clockspeed-based PB is only to overclock the fastest 7448. We do not know what is the yield of this processor in Freescale production line.
- yes RapidI/O is not new for Apple hardware specialists but it is also not new that RapidI/O has been difficult for them to be implemented because it is different from the hypertransport normally used by Apple as by AMD.

 

[eric67]

yes and no

Perhaps you were joking, but for the record, the myth is that it is useful to compare megahertz of one processor to megahertz of another processor. When you compare two speeds of the same processor, such as a rumored 1.67GHz to 1.7GHz update for the PowerBook G4, it's science, not myth. And science says that 1.67 is not much different than 1.7. It would take other improvements, such as higher resolution, to make this update anything to get excited about.

for your info , I am a scientist.
2nd point. We are not comparing the same processor my friend. we are comapring a 1.67GHz 7447 with a 1.7GHz 7448, something really different as you might consider when looking at Freescale technical data.
I think an update to a 1.7GHz 7448 PB is much more exciting than the last PB revision with a 1.5GHz to a 1.67GHz speed bump while remaining with a 7447.

 

[shawnce]

I do not believe it, the inquirer is not also reliable.
Freescale has announced the 7448 in mass production with availability in October. So having the MPC8641 in the next PB I do not believe it because it would imply also that Apple would have run into the design and the production of a new motherboard. indeed the MPC8641 is not compatible with the current motherboard of PB; whereas the 7448 is pin-to-pin compatible.

in addition, the MPC8641 requres a RapidI/O interface, something not developed by Apple yet.

Well the MPC8641/D has an on die memory controller supporting dual 64 bit DDR2 channels (533MHz), two PCIe interconnects (1 to 8 lane I believe it says), an on die ethernet controller, and it says an MPX bus (weird). Those exist in addition to the RapidIO interface. In other words Apple doesn't need to use RapidIO at all (it exist primarily for chip to chip communication in a multi-chip system).

I will say Freescale has a nice little thing in the e600 core and the MPC8641 package (supports dual and single core options) better then I had originally realized.

Apple likely knew about what IBM and Freescale had in the works years in advance and likely had mother boards being developed based on the options they saw. So in all likely hood Apple has mother boards ready to go for either the IBM low power G5 or the Freescale stuff (maybe even both). As I said before Apple made the decision to jump to Intel relatively recently and many of the PPC teams had little to know idea until it was publicly announced (based on what I have heard from Apple folks, granted a lot indirectly).

It is down to what options out of their developed mother boards, etc. do they want to use as the main filler for the next 9+ months.

If it is MPC8641 based I will be buying one.

 

[MikeAtari]

answer:
- there is a reason to limit to 1.7GHz. Freescale has announced that mas production of the 7448 is only available for 7448 clocked from 1GHz to 1.7GHz. so the only way for apple to propose a higher clockspeed-based PB is only to overclock the fastest 7448. We do not know what is the yield of this processor in Freescale production line.

PowerLogix has announced the immediate availability of the 2GHz single-processor PowerForce G4 7447A processor upgrade. Utilizing the fastest available FreeScale PowerPC 7447A processor with 512K L2 cache, this tasty gem is ready to ship now at Other World Computing for $399.

I don't think PowerLogix is selling "Overclocked" 7447A's.
So, Freescale is continuing it's "Tradition" of not releasing info to the public about what it's fastest available clock speed is.

Since the CORE of the 7448 is Exactly the same as the 7447A's except for an Altivec improvement, there should be no reason why the 7448 couldn't go to 2ghz. Except for heat in a laptop.

 

[MikeAtari]

If it is MPC8641 based I will be buying one.

Me Too.
This is one GOOD way Apple could preserve it's sales targets while building a MacTel laptop.

 

[kcwookie]

Who Cares

I doubt that Apple is going to put much more under the hood now. What they put now won't be something the can crow about when the new Mactels come out.

I just bought a new PB and am going to be happy for quite a while.

Kcwookie

 

[ryannel2003]

+ a possible FSB speed boost
+ a possible increase in cache
+ a possible increase in memory bandwidth (DDR2)
etc.

If those happen, the gains would easily exceed what a 30MHz clock speed bump implies. Heck we don't even know what the CPU core is... X Ghz in one core could be radically different then Y GHz with a different core.

All true. But it's still sad that Moto. is still pumping out chips that only give you a 30MHz increase. Now if they were going from a 1.67 to maybe a 1.8 or 1.9GHz, that might be better. I would also hope they upgrade the 12" ers nVidia GeForce FX5200 Ultra to a Radeon 9700 or something. Even tho it is a ok card, even the eMac has a better graphics card and it starts at $799.

 

[ryannel2003]

I thought that Macusers did not believe into MHz myth??

Just look at the specs!!

the 7448 will allow better battery lifespan, boost performance around 15% thanks to a larger cache and higher bus. So what do you want?? you will get a cooler and more performing processor?? the best way to wait for the Yonah dualcore 2.13GHz-based powerbook in June 2006.

I've never really bought the whole "MHz Myth" thing. I have always known that my Dell Dimension 4700 desktop (w/ a Intel Pentium 4 2.8GHz with HT) blows my eMac (PowerPC G4 1.25GHz) out of the water performance wise. However, it's not really the hardware that attracted me to the Mac platform. It was mostly OS X and how much of an improvement it was over OS 9. Before OS X, I really would've never considered a Mac for myself or anyone I knew but Panther all changed that for me. Now if a person asks me what kind of computer to buy, I tell them a Apple.

 

[eric67]

heat is the reason

Since the CORE of the 7448 is Exactly the same as the 7447A's except for an Altivec improvement, there should be no reason why the 7448 couldn't go to 2ghz. Except for heat in a laptop.

the 7448 is also featuring better power managment tools compare to the 7447, running on a faster bus, twice the cache of the 7477.
The reason is the capacity of Freescale to deliver such processor.
officially it tops at 1.7GHz. As I already mention it Appel could propsos it at 1.8-2.0GHz depending on the yield and overclocking capacity.

 

[QCassidy352]

ugh, why can't they release these now... I'm going to end up with an ibook for school because the powerbook updates will be just a few weeks too late. Does apple realize that this is prime computer buying season?

 

[eric67]

can not wait??

ugh, why can't they release these now... I'm going to end up with an ibook for school because the powerbook updates will be just a few weeks too late. Does apple realize that this is prime computer buying season?

you can not wait until the end of september??

 

[kcwookie]

The current PB is wonderful...

ugh, why can't they release these now... I'm going to end up with an ibook for school because the powerbook updates will be just a few weeks too late. Does apple realize that this is prime computer buying season?

I don't see any upgrade being all that much. They are going to wait until the new Mactels come out for a big upgrade.

Mark

 

[RHMMMM]

This thread is going to keep going and going until they get released, LOL. I don't know if I can take another month of waiting. I'll just have to suck it up.

Whatever they release, it's definitely going to be better than what is out right now, so don't worry about that - whether it be marginal or a bigger leap. I'm personally hoping for a 8651, but obviously time will tell. A 7448 will be just fine, too.

Notice the misspellings and grammar errors in the Inquirer article. Do we really trust them? My bet goes to the upgrades that the AppleInsider article talks about.

 

[shyataroo]

http://www.powerlogix.com/press/releases/2005/050809.html
http://www.insanelygreatmac.com/news.php?id=5140

Test of the 2ghz http://www.barefeats.com/g4up.html

That is a 2ghz CPU. Apple just dont use them. Why not how knows? I dont understand, the cost of a 1,42 and a 1,8ghz g4 must be the same. So why apple dont speed up there mac mini and emac? They coulde speed away at 2ghz if apple wanted them to. But on the laptop side I understand that the exstra heat would not be that good.


PS: NO intel CPU is comming even CLOSE to the exstream low watt output that the g4 has. AND pci-e uses more power then agp. So apple should avoid pci-e if they can when they make a p-m powerbook. And the thing with ddr2 is this at 533 mhz they have crap latenzy, and the higher the fsb the more heat. The g4 is a amacing laptop cpu, it is rahter fast and it take almost no power.

Freescale has a reason to suck, on acout that they should not spend money on a cpu that they will never sell. On acount that the g4 is EOL. Cant blame a man for not being stupid.

I know about the G4 that has reached 2GHz but, that was not made by Motorola which just proves my point that Motorola sucks at chip making or... The Conspiracy Theorist part of me says that Intel/AMD secretly struck a deal with IBM and Motorola to make apple's chips suck so they would be forced to switch to either Intel or AMD and from there MSFT having a partnership with Intel and AMD will learn the secrets of OS X and copy them.

 

[Lord Kythe]

I think Apple's move of waiting until right after the back-to-school rush to introduce a new powerbook is pretty irritating, especially since laptops are such a necessity in universities. People have every right to complain that Apple appears to by introducing a relatively minor update at such a big event, which is usually reserved for items of bigger caliber (the iMac, anyone?) So there's no reason to be so cold, Lord Kythe.

Ok, so you're saying Apple should ignore the millions of Mac users who are not in school and put students first (which do NOT constitute a majority of users by the way) by releasing new/upgraded lines based on school rather than on major business events so that they make less impact and less money but satisfy students...

I've never thought of it this way... Nah, let's face it, this is crap: updates will happen on major business events, and it's better this way.

And you have no idea of the caliber of this update. For all you/we know, it could be a new CPU family (G4/G5, but let's face it, probably not G5), the video card could be better, HD screen, anyone? etc. You're basing your assumptions on the alleged 30 MHz upgrade, which is, by the way, a rumor, and we're here to discuss probability, not take rumors as facts. Nor have we any idea of when they'll be able to produce the updated line, maybe Paris was the soonest possible time. Bitchin' seems pretty stupid to me without having the info. Give them the benefit of the doubt!

I'm not cold, I'm just defending Apple. Someone ought to. They can never be good enough, can they? A lot of people think they could do a better job than Apple; most of those people never even ran a business! I can bet you anything that Steve and his staff are doing a better job than 99.9% of all of us could with Apple.

Come on, let's stop bitchin' and enjoy the ride! Steve knows what he's doing and 'til now, he hasn't let us down.

 

[Durendal]

1.7ghz, my ass. Moto is ramping up 7448 production in October. The 7448 eats less power at 1.7ghz than the 7447A does at 1.42ghz. Since they have the 7447A running at 1.67ghz, Then I find it very reasonable to expect 2ghz in this next update. Why would Apple saddle it at 1.7ghz? That's not an update! If they're using 7448s, then there's no reason to clock it that low. If they're using 7447As, then they can wait another month for faster chips. This article is bogus.

 

[Rza]

There seems to be some confusion about the MPC8641 and how it will be implemented.
Think about the Athlon64. It has a integrated memory controller.
Now also integrate the PCI Express. Then you wouldn't need a northbridge anymore, only a southbridge will suffice.
Now do the same thing with the MPC7447A in a PowerBook, add a memory and pci-e controller, ... yes you won't need a northbridge anymore.

So what Eric is suggesting about the bus or chipset or ... is thought wrong.
Apple only needs to connect a southbridge to the cpu. There isn't a bus like before. So no RapidIO or MPX needed at all.

There are two pci-e contollers. Apple can use one to connect to the videocard and one to connect to the southbridge.
Apple can take any pci-e videocard and use any pci-e southbridge.
Couple of examples of pci-e southbridges: ULI M1573, ULI M1575, ATI SB400, Intel ICH, ... just to name a couple.

So Apple can design a completly new motherboard without to much trouble.
All chip are available. No new chips have to be designed just a new motherboard and that's not too hard.

I think you can forget about the MPC7448 because that will be even harder to implement with DDR2. That way they would have to design a new chipset and they won't do that.