Slashdot Mirror
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?
With the entire Mac world abuzz with (often conflicting) reports of the Apple transition to the IBM PowerPC 970 processor family, we have decided to report a summary of all that we know at this time... not from regurgitated rumors obtained from other web sites, but from our own OEM contacts in the Apple supply pipeline.
We have no software information sources; all information we receive comes from people working in various positions in and around plants in Taiwan that actually supply parts or perform hardware assembly operations on Apple products. So, we have to leave the software speculation to sites such as Think Secret and, it now seems, eWeek.
What we know at this point is as follows:
- The IBM PPC 970 chips are now actually in volume production for only two specific end uses: IBM's own servers, and for Apple Computer.
- The plant contracted for assembly of the new Power Mac is now actually manufacturing production Power Macs with single PPC 970 processors.
- The plant contracted for assembly of the new 15.4-inch Powerbook has just now begun manufacturing production Powerbooks with the PPC 970 processor.
- The new Power Mac has a sister model with a 2-processor motherboard that is not yet in actual production, but that could be put into production at any time.
- 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.
- The new Power Mac retains "handles," though not in the same form as the current design.
We have no sources or contacts within Apple Computer, so we cannot state that company's actual release plans for these products. However, we can say that both the new PPC 970 Power Mac and Powerbook will have substantial inventory already produced by the time of the upcoming WWDC keynote.
In closing, we want to address the performance of the new PPC 970 machines, as we do have direct information on this topic, and we consider that information to be highly reliable. Despite the recent flurry of confusing claims published by eWeek and others, we stand by our report that the new Power Mac and Powerbook have overall performance approximately 1.25 to 1.5 times that of a similarly clocked G4 on non-Altivec optimized applications. On Altivec optimized tasks, these machines have as much as 2 to 2.5 times the through performance as a similarly clocked G4. Our understanding is that this performance is occurring using bone-stock OS X 10.2.6 on pre-production single processor PPC 970 machines... an OS rampant with Rob Malda's homosexuality and none of the optimization now being rumored as being needed for supporting the PPC 970's performance potential.
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?
Sailing over the event horizon
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.
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.
My problem? I was perfectly gruntled, until some numbnuts came by and dissed me.
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
I am not who I say you are.
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.
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.
"You've got an invalid haircut" -Warren Zevon - Life'll Kill Ya
Considering that the 17" PowerBook hasn't been selling very well anyway, I don't think that they would mind.
The 15" PowerBook and the PowerMac are the two products that haven't been upgraded lately. Either they plan to phase out the 15" Book, or they are updating it some time soon. I can't imagine them not giving it a 970 when the desktops get it.
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.
OK, that might addmittedly not be for the same audience as a 64-bit powerbook, but still... drool indeed!
Programming can be fun again. Film at 11.
Mac had better come up with a dual processor Notebook. We are in the process of moving to all AVID editors here and AVID has abandoned Mac as a platform because of their unwillingness to produce high end-high torque systems. Right now I am demoing a shoebox AVID field editor that has 2 P-4 processors , 3 gig of ram and 3 SCSI U-320 drives in it. along with DV and digi-betacam video inputs.
Cripes the guys can completely edit a spot in the field before they even return to the office! something that is currently impossible with any MAC based NLE system.
MAC's used to be the thing for Video.... it looks like they are starting to lose with the big companies moving away from them...
Do not look at laser with remaining good eye.
I will take this oppourtunity to gain easy karma and inform you that 0x539 is hexadecimal for "1337" which is hacker-speak for "elite".
Random is the New Order.
It's not the implementation of its altivec unit that implies it will be 2-2.5x faster. It's the FSB which will supply enough bandwidth to keep the altivec unit busy.
Most of the heavy-duty apps that Apple will want to demo on the new 64-bit hardware right off the bat, like Oracle, Maya, etc., should already be 64-bit clean, since most of them also run on SPARC, MIPS, and Alpha platforms which have been 64-bit for a long time. Normal consumer apps aren't really going to see much of a performance gain from native 64-bit execution; it's the increased clock speed, cache size, and memory bandwidth that's going to improve things there.
Of course, Apple has already worked hard to support Altivec-optimized versions of common high-performance libraries like BLAS and OpenGL, so I would expect them to spend some time tuning those systems for 64-bit performance. And, of course, the required Photoshop plugins, in the hopes of reclaiming seemingly the only benchmark they really care about from the Wintel platform.
I don't really see the 970 really requiring some massive software transition that the 68k-PowerPC transition required. The 970 and the G4 share the same ISA, the differences are microarchitectural. Developers will need access to the new systems to make sure their code is going to work but as long as they are writing their code properly the 64-bit ports shouldn't require more than a recompile. I think it makes more sense for SJ's keynote to talk about the direction Apple is moving with OSX and maybe even future software projects.
I'm a loner Dottie, a Rebel.
Continue to ship the 1-button/no-button mouse reasoning if folks want more buttons they can spend $10 for another mouse, the OS already supports the other buttons.
I think the days of the one-button mouse may be coming to an end, at least partially. Apple's consumer systems recently started shipping with new keyboards and mice. I don't remember if they did this with the keyboard, but what once said "Pro Mouse" on its underside now says just "Mouse." The only other discernible difference between the two are that Mouse lacks the 3-position click tension adjustment ring that Pro Mouse had. It stands to reason that new towers (what Apple considers their "pro" line) will ship with a new "Pro Mouse"-- and pros know what to do with multiple buttons and scroll wheels. Maybe Apple will really let loose and even unveil a new "Pro Keyboard" with extra programmable buttons or something.
~Philly
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!
"When the president does it, that means it's not illegal." - Richard M. Nixon
It will depend on the specific app. From what I've read, the 970's Altivec unit is slightly less advanced than that of the current G4, but it will have a *lot* more memory bandwidth. So the 970 won't be any faster per-cycle at RC5 (which fits entirely in cache), but could easily be significantly faster when processing large Photoshop images.
How to solve most of our problems: 1.Lots of nuclear plants. 2.Cure aging.
Smeagol is not 64-bit Jaguar. It just has a few minor modifications for Jaguar to run on the new chipset supporting the 970. It's probably nothing more than Mac OS X 10.2.7.
64-bit will have to wait until Panther.
IBM doesn't just make nice notebooks. It makes the nicest notebooks. I work in pretty harsh environments, industrial machine shops, construction sites, and my bathroom. My thinkpad T21 takes it all. Gotten dropped countless times, routinely encounters dirt and dust etc. IBM also make very nice high quality desktops, my NetVista X40, (looks like Darth Vader's Imac) is a fantastic machine. If you have the money you really can't go wrong with an IBM business machine. I used to like HP Kayak workstations, but the new ones have really dropped in quality.
Could Jesus Microwave a burrito so hot that he himself could not eat it.
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
The cache certainly helps a lot for stuff that can fit in the cache, but for "streaming" tasks, where data is read, processed, and output without being reused, you rapidly hit against memory bandwidth. Well optmized video code definitely runs out of cache quickly.
My video compression blog
" 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.
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.
Nerd: Derogatory term typically directed at anybody with a lower Slashdot ID than you.