PPC 970 Powerbooks and Powermacs in Production?

Thadddius_Brinks writes "MacWispers.com is reporting here that apple is currently in production of a redesigned single processor PowerPC 970 Powermac system and a 15.4 inch Powerbook. They (MacWhispers.com) are also standing by their earlier claims about the speed of the new processor." This article consolidates many of the major rumors surrounding WWDC including the rumor of a new case for the Powermacs, but it raises the ultimate question: 17" Powerbook, or PPC 970 Powerbook?

Comments

Performance claims need clarification

[John Harrison]

It isn't clear from the last paragraph of the article if they are comparing a dual G4 to a single 970. The other option is a single G4 to a single 970. A dual 970 doesn't seem to be an option since they don't seem to be in production currently.

Does anyone have any clarification on this? With the rumours that the 970 chip is actually less expensive for Apple than the G4 I was hoping for dual 970 boxes at price points similar to the current crop of PowerMacs.

Re:Performance claims need clarification

[Dylan Zimmerman]

They are comparing to a single G4. Due to the G4's bad architecture, putting two of them into a computer doesn't improve performance nearly as much as it should. IIRC, you only see roughly a 25%-50% improvement. Therefore, a single 970 is about on par with a dual G4 in integer based operations and it's much faster running Altivec code.

Re:Performance claims need clarification

[Lank]

I don't think the ArsTechnica report said that the implementation was supposed to be worse - just that it was more of a hack. Just because something is a hack doesn't necessarily mean it can't perform well. In fact, most hacks are done for none other than performance reasons.

Re:Performance claims need clarification

[John Harrison]

I think that the increase in the bus could really help here. If you can keep feeding data through a slow altivec unit you will might get better performance than with a fast one that is starved for data.

Re:Performance claims need clarification

[Anonymous Coward]

Speaking as the owner of a dual G4... er, no. Code that's written to scale linearly scales pretty much linearly. The Maya renderer is a good example.

Re:Performance claims need clarification

[azav]

Actually, from what I have heard, the duals "balance the drain". By that I mean that though you don't feel it being a load faster in day to day activities, you feel less of a slowdown when one process would bog the processor. - so overall, the entire system feels smoother - nere I say snappier.

My own bets

[Dark Paladin]

1. That the 970's are being produced, but so far, there's still small numbers, so either only folks at the Developer's Conference will first crack, or you'll have to wait for a bit of time before you can actually get your hands on one (kind of like when the 12" and 17" first started shipping).

2. The major OS Upgrade to 64 bits will happen in a few more months - either way, I expect that OS 10.3 will cost another $50 - $100 (depending on how they do it).

3. The G3 iBook line will be slowly phased out, and replaced with G4 based systems.

4. Dual processor systems by Christmas or so.

5. iTunes for Windows sooner than we thought.

6. Somewhere in this timeframe, new Xserves will start to appear with the 970 chip and the 64-bit server operating system (which should be interesting for folks running "big ass" database/graphical rendering farms.

So either way, I'd say we'll "see" the devices, a few "first adopters" will play with them, pass judgement, Ars Techana [SIC] will write a big ass article on them, and "everybody else" will pick them up later.

Hopefully somebody can convince Valve that Half Life 2 would really run rather nicely on these boxes so I don't have to spend money upgrading my old Wintel Game Box.

Re:My own bets

[mrpuffypants]

Actually, the 12" pbook was out pretty damn fast.....my parents picked one up about 2 weeks after macworld.

the 17" just took about 2 months to fully ramp up production.

Re:My own bets

[Dylan Zimmerman]

The 970 is already being produced en masse. Because of Apple's contract with IBM, they get a bunch of them pretty early.

According to ThinkSecret, Panther is going to launch sometime in September. They also say that a special 64-bit version of Jaguar called 'Sméagol' is being built for the new PowerMacs (known internally as Q37). It could be finalized as early as WWDC. Why would they have a special version of Jaguar if they didn't have the 970's rolling off the production lines already? Seriously. If Apple thought that they couldn't get them to people quickly, they wouldn't bother recompiling the entire OS for them. They would just wait a month to release them. They don't want a repeat of the bad publicity they got as a result of the 17" slowness.

Apple is moving away from Motorola chips entirely. IBM still makes the G3. In all likelihood, Apple is going to keep the G3 in the iBook and just start using IBM chips instead of Motorola ones.

I sure hope so.

Again, I really hope so. I've been itching to spend all of my money at the iTunes Music Store, but I don't have a Mac.

I would imagine that the Xserves would use the real thing. The Power4 uses more power and produces more heat than the 970, but it also has an insane MTBF. Besides, you wouldn't need a dual Power4 since it's already two cores on a single die. Pretty cool design, really.

Re:My own bets

[Anonymous Coward]

"They also say that a special 64-bit version of Jaguar called 'Sméagol' is being built for the new PowerMacs (known internally as Q37)."

oooh ... We wants it! ... my precious ...

G5 Powerbook?

[spookysuicide]

If they release a 15" g5 powerbook, what would happen to sales of their 17" g4 powerbook? I don't believe apple would have a powerbook line with their midrange model having such a radically better architecture/processor then their high end model.

40h bit, not 64 bit!

[Thinkit3]

Isn't much more elegant to use hexadecimal?

Re:40h bit, not 64 bit!

[TeknoHog]

Agreed. It does look more 0x539 that way.

Re:40h bit, not 64 bit!

[Drakonian]

I will take this oppourtunity to gain easy karma and inform you that 0x539 is hexadecimal for "1337" which is hacker-speak for "elite".

Re:I'm wondering...

[gwernol]

Do you think the new "G5"s will sport new enclosures too..?

The current design is long in the tooth to say the least and is highly associated with the G4 processor... yet I've heard nothing about new enclosures at all..

Dude, RTFA:

"- The new Power Mac has a new case design with "metallic look plastics," and a front panel "mostly made with the same anodized aluminum surface" as the newest Powerbooks."

Its only one short page. Its not slashdotted (yet). How hard was that?

Shullbit

[Graymalkin]

I would have believed 970 PowerMacs in production but definitely not 970 based Powerbooks. That goes double because there's even more rumors of 15" Powerbooks based on the G4 but all aluninumized waiting to be shipped to Apple's stores and other retailers. If you've noticed there's a dearth of 15" Powerbooks in stock anywhere that sells them.

Besides June production doesn't mean a June release or even announcement date. Apple likes to build up stocks of computers before selling them. Building and shipping computers in the same month would be a ridiculous strain on their resources. As for a June announcement, see the Osborne computer company.

WWDC isn't exactly a place Steve Jobs likes to announce hardware products, it is really the wrong venue for such announcements. MacWorld Expo is a much better place to do things like that and is only two months away. It's not really a secret Panther developer previews are going to be released at WWDC which will likely be SJ's keynote subject. MacOS and related software ought to be and typically is the subject of SJ's WWDC keynotes. Not hardware announcements.

Re:Shullbit

[haunebu]

Apple likes to build up stocks of computers before selling them. Building and shipping computers in the same month would be a ridiculous strain on their resources.

Er, no. Apple prides itself on having a low inventory (a couple of weeks, at the most) - all personal computer manufacturers do. If they had any more, they'd wind up in the December 2001 "5+ weeks of unsellable inventory" glut problem.

WWDC isn't exactly a place Steve Jobs likes to announce hardware products, it is really the wrong venue for such announcements. MacWorld Expo is a much better place to do things like that and is only two months away.

Except that Apple isn't going to be at MWNY this year.

Re:Shullbit

[gamgee5273]

Don't forget: Jobs intro'd the original bondi blue iMac at WWDC five years ago. He will do hardware at WWDC if he thinks it is cool enough to show off.

this makes sense

[b17bmbr]

apple is offering sweet $300 dollar rebates to students, and they have just dropped the price on many models like the powerbook. might be a good time to jump on one.

MacWhispers...

[evil carrot]

I'm no fan of Jack Campbell and honestly do not believe anything his site spouts. The only time I ever hear anything about it is when (semi-)legitimate news sources pick up "scoops" from his site.

To read more about how cool a guy he is, check out the MacTable report at Macintouch:

http://www.macintouch.com/mactable.html

From the past...

[Nexum]

Well, from past knowledge of how Apple has done things recently, I'd say...

PPC 970 Single 1.4 Ghz shipping July.

PPC 970 Duallies shipping within 4 weeks of the single.

OSX 10.3 Late August... and I would bet my kidneys you WILL have to pay for it (~$129), but don't moan... apparently there is a LOT of new/improved stuff, and this is only the beginning as Apple have found that they can build on the code very easily *indeed* due to the quality and clenaliness of it... exactly the problem MS seems to have with Windows ATM.

Re:MacWhispers and Macrumors etc

[Anonymous Coward]

Oh sure. Would that be the Thinksecret that, in 1999, predicted Apple was about to release an Intel version of the iMac that ran Windows?

Or the one in June 2000 that predicted that Apple was about to drop the entire PowerMac line and just sell laptops and all-in-ones?

Or the one in August 2001 that predicted Apple and AMD had collaborated on a "secret" CPU design code-named "Twostone", a 48 bit CPU with 16384 registers, that was going to replace the G3 and run it in emulation just as the PPCs had emulated the 68000?

Or maybe the one that, in September 2002, predicted Apple was about to release a cordless phone, FOR NO REASON WHATSOEVER.

I hardly would describe them as a dependable source of unfounded rumours.

Re:No need to click, article text.. NOT a karma wh

[Rude Turnip]

"...an OS rampant with Rob Malda's homosexuality..."

Hmmm...and this was posted as an Anon Coward. Trying to tell us something Taco?

Mac rumor sites vs. Slashdot

[Gizzmonic]

Slashdot is not a place for rumors.

To top it off, the person who runs Macwhispers is completely morally bankrupt, and is most likely fabricating the entire story.

I'd like to see some real competition for high performance CPUs as much as the next guy, but let's not lose our heads to con artists like this guy...we will know for sure in about 2 weeks ;)

Everyone seems to misunderstand MacWhispers

[Elwood P Dowd]

Everyone's immediate reaction to MacWhispers is always negative.

"Oh, I'll believe it when I see it."

Obviously. MacWhispers has given up on making release date predictions. You'll notice that they have *not* given a specific timeframe for the release of these machines. They have said that they are being built.

So, now, when WWDC makes no mention of the 970, everyone will say "See! MacWhispers are a bunch of damn fools." and no one will remember, two months from now, when these machines surface, that it fits perfectly with MacWhispers' information.

If you take them completely literally, they are a valuable source of information. They cannot divine the future, and they don't seem to be trying to do so, either.

Re:Everyone seems to misunderstand MacWhispers

[gorsh]

But using this logic, they're always going to be right - everyone knows that Apple will most likely announce a PPC970 system at some point in the future.

Predicting something that everyone knows will happen eventually is not so significant as correctly predicting *when* it will happen.

I doubt the speed predictions at least.

[laertes]

Going from the Ars Technica article a while back, I don't buy the claims about the speed of the new chip.

Specifically:

On Altivec optimized tasks, these machines have as much as 2 to 2.5 times the through performance as a similarly clocked G4.

Unlike the G4, where the AltiVec unit is integrated rather nicely into the issue unit, and can issue several types of vector instructions in parallel, the 970 can only do a permute in parallel with another instruction. Hence, for some tasks, I would expect the G4 to be almost twice as fast as a similarly clocked 970.

They make a similar claim about the non-AltiVec speed, which I tend to believe. The compiler has to be a little smarter (but Apple did add a bunch of G4 optimizations to gcc anyway), but the 970 can do more per clock the the G4 can, under many circumstances. Not only can it have more instructions in-flight, but it has a much more advanced reordering unit than the G4.

Oh yeah, and when did /. just copy over Apple rumor stories? </obligatory>

Re:I doubt the speed predictions at least.

[gsfprez]

cache performance, bus performance increases.

the real reason G4's don't perform is that they are usually waiting on data... the G4's may be fast and have great AltiVec, the whole issue of still running (essentially) PC133 memory is the bottleneck.. no matter how fast your CPU is, if you can't get it lots of data - not just the data in the L2 cache - its just gonna sit there.

the 970 systems should, by any means, at least keep the CPU(s) busy. that alone will greatly enhance the performance of the new machines when doing things like 3D rendering, video transcoding, etc.

Its like why my Powerbook rips mp3's from CD's at only 10x, while my slower desktop rips them at 14x... the desktop has a 52x CD-ROM drive and my Powerbook has a slow-as-ass Superdrive. I can't keep the machine busy because I can't get it the data. The bottleneck in that case is the CD read.

In the G4's vs. the 970's discussion, the bottleneck is the pathetic (compared to Intel mobos) G4 motherboard because the mobo's running the 970's are all around faster.

Re:I doubt the speed predictions at least.

[illumin8]

Yes, but you aren't taking into account the FSB speed of the processor which is really choking the G4 right now. 167 mhz. is just not fast enough to feed the processor as fast as the Altivec unit can process instructions.

The 1.8 ghz. PPC 970 should have a 900 mhz. FSB, fed by dual bank DDR400 (PC3200) memory it will really cook!

Re:I doubt the speed predictions at least.

[stripes]

Unlike the G4, where the AltiVec unit is integrated rather nicely into the issue unit, and can issue several types of vector instructions in parallel, the 970 can only do a permute in parallel with another instruction. Hence, for some tasks, I would expect the G4 to be almost twice as fast as a similarly clocked 970.

The Ars review used the CPU documentation to decide the 970's Altivec was slow. It might have missed something, like maybe the G4's Altivec is always memory starved and thus runs a 25% of it's maximum speed while the 970 has lots of memory bandwidth and thus while in thery is only 50% as fast as the G4's, that manages to be 2x what the G4 really does.

Of corse it may well be slower (or merely "about the same"), because I'm not able to tell you how fast a CPU is from the pre-release docs either (I thought the TI SuperSPARC would be fast, and the P-II would be slow). Basically I'm saying "who can really tell here".

Re:Yikes

[JudgeFurious]

I haven't looked but do you get an equal number of PC rumour sites if you go Googling for that?

If not then you've got to ask yourself why is it that so many people care about what may or may not be coming along next from Apple and so few give the proverbial rats ass about the next offering from HP, Dell, or eMachines.

Apple seems to have perfected getting attention to an art. You can love them or hate them but almost nobody ignores them. That kind of PR is priceless.

Re:SCOOP: PPC 990, "G6" on the Way now!

[TheCrazyFinn]

It's actually called the PPC980, and is due next year.

Yes, the PPC980 is in the IBM roadmap, it's to the Power5 as the 970 is to the Power4.

(Re)Stating the Obvious

[buckhead_buddy]

I think that at this point it's quite obvious _something_ is in production, but that doesn't mean the time frame you can by an updated Mac is close at hand though.

Major changes that effect developers will be announced at WWDC. It's likely that whatever new directions in CPUs, APIs, or Market segments will be announced there because developers will have to react to the news. That doesn't mean anything will be available for sale or even that we'll get the whole picture of what cases, prices, user interface changes, or iApps will be released. Not only don't developers need to know this stuff, but traditionally they've been a very conservative, non-spontaneous, purchasing crowd where such surprises would be wasted.

People keep talking about having the whole Mac product line refreshed at WWDC and nonsense like that. My guess is far more conservative. We may get a timetable to expect new PowerMacs, but we probably aren't even going to see the new machines in final plastics.

During the transition from 68000 to PowerPC, Apple bent over backwards to give developers access to emulators, test labs, and even loaned machines to big developers. But they didn't start commercially selling anything until eight or nine months after the WWDC announcements.

I don't think Apple will wait quite that long to introduce new chips if such plans are really on the horizon, but I think there will be some non-trivial lag from WWDC to new consumer-marketed debut of new hardware.

Re:PPC 970 Powerbook

[__past__]

Why wait for 64-bit laptops? Get yours today!

OK, that might addmittedly not be for the same audience as a 64-bit powerbook, but still... drool indeed!

Memory bound

[benwaggoner]

Assuming you aren't memory bound. You've only got 166 Mbps TOTAL between the two processors, which well tuned AltiVec codec can saturate with a single processor, let alone two.

PC Rumours ...

[Anonymous Coward]

"I haven't looked but do you get an equal number of PC rumour sites if you go Googling for that?"

To determine what features will be on the leading edge PCs of the coming years, all you have to do is go to www.apple.com.

Egads, I've turned into a troll!

Re:MacWhispers and Macrumors etc

[foniksonik]

"I hardly would describe them as a dependable source of unfounded rumours."

What are you trying to say, they are a great source of unfounded rumours! Not to mention dependable!

No 64-bit OS for some time - my prediction

[gsfprez]

I don't get why its believed that Mac OS X has to be 64-bit to run on the 970. We've been told in numerous places (Ars, IBM's frickin website) that the 970 runs in 32-bit just fine.

SO... that being said... my WWDC announcment predictions along with what Apple has already stated

- New Macs with PPC 970
- New PowerBook 15 with PPC 970 (the 15" PowerBook is the workhorse of the line, always will be, sorry)
- Preview of Panther
- Macintosh Roadmap Roadmap showing the future... and this is the big-ass news that everyone's been talking about.

Roadmap steps will look something like this...

July 2003:
10.2.7 running on new 970-based Macs practically unmodified because the 970 handles 32-bit operation just totally fine. x.x.+1 updates between WWDC and October 2003. Developer-only release of 64-bit SDK which will not be ready for prime time, but will allow developers to make the swtich, if necessary (similar to Mac OS 9 -> Carbon transition)

October 2003:
10.3 release - all those cool updates in iApps, updates in performance and operation of Mac OS X UI that were shown at WWDC. Panther Will NOT BE a 64-Bit OS!! - why not? Does not need to be because there are no 64-bit apps! Where are the apps? They are still being worked on with the 64-bit SDK, see you at Mac World San Fran with first 64-bit apps.

Jan 2004 (MWSF):
10.3.5 release. Mac OS X will run 64-bit applications. Only apps that NEED to be recomplied 64-bit clean will be recompiled 64-bit clean (iChat, for example, does NOT need to be 64-bit). Finder will be first app to be 64-bit clean because it needs to be.

beyond that, its non-speculateable.

But i think that the real news at this WWDC is going to be the first major Macintosh Roadmap since we saw the Rhapsody one in 1997(8?). You will see where the Mac is going hardware-wise and software wise.

Apple is going to push into the small-medium server market in a hard way.... 64-bit XServes which can run horkin Oracle databases, huge fileservers, and be the backbone of big-ass renderfarms... all with Mac OS X moron-simple UI and none of the pain of cost with Windows servers or admin headache of Linus servers? Puhlease... Apple is going to kick ass and move in where Intel and AMD are just simply lagging behind.

(yes, half the guys in my wedding party have apple.com email addresses.. no, none of this information was gleaned from them.)

My predictions

[Phroggy]

The PowerMac G5 will be announced at WWDC, based on the IBM PowerPC 970. The name of the machine will be "PowerMac G5" or "Power Macintosh G5", but Apple will also advertise the processor as the "IBM PowerPC 970 processor with Velocity Engine". It's 64-bit, and they'll hype that up as much as they can.

The low-end (1.4GHz?) model will be available immediately, or within two weeks and Apple will be taking pre-orders. The mid-range and high-end (dual 1.6 and dual 1.8?) models will be shipping within a month after that. Photoshop comparisons with the latest Compaq PC will be mind-blowing, for the types of people who get excited about Photoshop performance. USB2, Bluetooth, FireWire 800 and 400 and Gigabit Ethernet will be standard features, with a slot for an Airport Extreme card.

The Aluminum 15" PowerBook will be released. We will not see a PowerBook G5 before January '04 and maybe not until March '04.

The PowerMac G5 will ship with a hacked version of Mac OS X 10.2, which will not be fully optimized to take advantage of the new processor. However, the PPC970 is designed to run 32-bit code just as well as 64-bit code, so it will still be amazingly fast. Anyone who buys a G5 will be entitled to a "free" ($29 S&H) upgrade to Mac OS X 10.3, which will ship in September for $129.

The new OS will be 64-bit native, optimized for the PPC970, and compiled with gcc 3.3. Large chunks of the Finder will be rewritten for performance and better UI, and there will be a ton of little system-wide UI improvements (adding up to a significantly better experience). One convenient new feature will be support for multiple users being logged in locally at the same time, like Windows XP (go to a login screen without quitting all your apps, second person logs in, first person's apps stay running hidden in the background, can switch back and forth between users).

Mac OS X 10.3 will include WebCore, Apple's Aquafied version of KHTML, available for any application to use. Safari will be the default browser. I suspect Internet Explorer will not be included, although of course you can download it from Microsoft. Help Viewer will be replaced (thank god) with a version that uses WebCore. Now that WebCore is available, it'll be possible for Apple to support PAC and WPAD for automatic proxy server discovery, although I don't know whether these features will make it into 10.3.

Did I miss anything? We'll see how accurate my predictions are next month...

From past experience...

[jriskin]

Ways of knowing Apple is going to release something.

1. Supply chain starts drying up.

2. Apple allows upgrade makers to catch up.

On the supply chain i've heard conflicting rumors, but on the upgrade side both OWC and Powerlogix have now released 1.42Ghz upgrades, this matches the top of the line G4 desktops.

So there is definitely SOMETHING coming, whether it is a new 970 based machine or simply speed bumped G4's only time will tell.

Great setup, but you forgot the punch line:

[inertia187]

"Banging your head against the shiny new Dell Inspiron 600m keyboard...priceless. There are some things money can't buy. For everything else, there's Microsoft."

What we can expect

[customjake]

There are doubts in my mind that Apple will use the WWDC as a launching pad for the new 970s. I do belive that Apple is currently in production of 970 machines, and plans to release them alongside Panther, whenever it releases...probably August. After all, why release an OS that's 64 bit if you don't have hardware to run 64 bit. The rest of the computer world is waiting for a viable 64 bit desktop system....there are other 64 bit processors out there.....

I do belive that Apple will have some sort of 970 based machine for the developers to look at, as well as some benchmarks for us all to drool over.

So what can we expect? Previewing Panther on a 970 based system, probably an update to the developer software (64 bit Project Builder), benchmarks of the 970, overview of 970 based hardware. Probably annouce a PPC 970 based Powerbook in the near future...END OF YEAR? Possibly a Naming change to hardware lineup, and Adjustable Displays

What is less likely? Windows verison of iTunes, PPC 970s shipping, PPC 970 Powerbooks shipping, iCam,

What is NOT going to happen? Apple PDA, Apple Cell phone,

As for the whole product line being revamped, i don't think we'll see a whole new product line at WWDC, but i think that Apple is going to be moving all it's lines away from the G4/G3 based systems. I don't even see the iBooks being converted to G4 systems, as IBM's chip lineup is known to be 20-30% cheaper than Motorola's line...

Instead of seeing G4s as the low end chips, i'd probably wager that IBMs will develop a PPC off of the 750 chip and this will probably replace the G3 systems. But i wouldn't expect to see a PPC 750 until you see a 980 in a powermac.... I think if Apple is going to IBM processors, it would be prudent that they went to an all IBM chip lineup.

Re:So is this The End of 32-bit Apple OS

[psyconaut]

They wouldn't be supporting two operating system, it's the same operating system!

64-bit PPC architecture can run 32-bit PPC code just fine as-is. Which is part of the reason that there are claims of stock MacOS X 10.2.6 being run on these PPC970 machines.

-psy

Re:You're right

[Anonymous Coward]

Apple can waffle on about the Mhz myth all they want, but I don't see them REDUCING clock rates generation after generation.

Pentium 4: 3 GHz

Xeon: 2.6 GHz

Pentium M: 1.6 GHz

Itanium 2: 1 GHz

Re:Next from Apple...

[phillymjs]

Chill out, rabid fanboy!

I was not slamming/mocking Apple for only shipping a single-button mouse. I'm typing this post on my G4. All I was saying is, based on their nomenclature changes they might be about to ship one with their new towers.

~Philly

Re:You may have to buy an Intel machine anyway

[kalidasa]

Cost of Mac laptop: $1000. Half-Life license: $25. Cost of Intel laptop to play Half-Life on because Half-Life doesn't work on Macs: $1000

Cost of a real life so you don't waste all of your time on a Pentium laptop playing Half-Life: Priceless.

Re:PPC 970 Powerbook

[statusbar]

"32 bits ought to be enough for anybody"

right?

BTW, a 64 bit processor means more than just the amount of ram you can stuff in it... Think virtual.

--jeff++

Re:Yikes

[pi42]

I think PCs do get equal attention, but since all the parts in Dell's, HP's, etc. machines are all industry standard parts manufactured by other companies, the speculation isn't about their offerings but the offerings of their suppliers.

There's a lot of speculation about AMD's chips, or new motherboard chipsets, or nVidia or ATI's new graphics cards -- probably just as much or more in total as that which surrounds Apple's products. Apple just has a lot of relative speculation surrounding it since it's the only provider for a particular platform.

Re:You're right

[Toraz Chryx]

" well, I've read some moderately convincing arguments that - because of the low latency, 4GB/sec L3 cache that the current top end G4s sport - memory bandwidth is not the bottleneck that we'd all like to believe. It seems that the REAL bottleneck is simply the low clock speed and lack of integer and FP execution resources that the 7455 has available."

The only problem with that argument is that the bus tops out whenever you're churning on a large dataset.. like video editing...

There's no doubt that the L3 helps, but actual memory bandwidth would help a whole lot more.

Re:PPC 970 Powerbook

[Sandor at the Zoo]

Anyway, as I was going to bed last night, I got to wonder about the CRLF (DOS) vs CR (Unix) vs LF (Macs) incompatibility.

You've got that backwards. Macs use the carriage return, decimal 13 (0x0D) as their line break character. Unix uses the line feed, decimal 10 (0x0A). DOS (and therefore Windows) uses the asinine CR-LF combination, which is also the standard marker for line and command endings in most text-based internet protocols, such as SMTP.

For those youngsters not in the know, these "control" characters really did control things. On a paper teletype machine, a carriage return would move the print head back to the left side of the carriage. A line feed would advance the roller one line. The combination of the two would prepare the teletype for printing another line, so in a way Windows does it right.

In reality, using two characters to mark an end of line is a major pain in the ass. I wish everyone would standardize on using just an LF.

Re:PPC 970 Powerbook

[Phishpin]

How bout in an automobile?

"Those arn't pillows!

Re:PPC 970 Powerbook

[tbone1]

I don't know that the 970 in the Powerbook will happen this soon. I recall a discussion on The Mac Observer about the towers being Apple's best money-makers. They have the highest margins, in terms of dollars and in percentage, of Apple's computer hardware. Tower sales have been lagging for a few years, so don't be surprised if there is a big push to sell the towers, particularly in the sense of "The 970 is available, but not in the consumer products and only in our desktop".

Also, Quark has just announced that it is shipping for OS X any day now. Lack of a native version of Quark, along with a sluggish market in publishing, has kept a lot of the tower buyers (ie print shops, etc) from upgrading. Quark is ready to ship, and publishing seems to be on its way to recovery. Those are two big pushes on the demand side.

Also, it would seem silly to have the 970 in the towers and not in the XServe, so expect them there as well; that would further reduce the supply side even further. In fact, I suspect that the 970 in the XServe might drive XServe sales higher, thus reducing the pool of 970s available for laptops.

A 970 Powerbook will be here, just don't expect it within the next couple-three months.