My evidence is quite simple to understand. You have to think like a criminal to be one, right? Look at the bottom of the first page of this thread. There are two links which were posted by DanPCamera. These links reveal many articles written by The Register. This is where Mastergateing turned for the rest of his information. He copied and pasted his bullshit right from the register. Take a LONG look.
PowerPC G5 redux
By Tony Smith
Posted: 03/12/2001 at 16:46 GMT
So, how likely is Apple to release the Power Mac G5 next January, as our regular correspondent, who claims to be close to the company's hardware division, has said? With the anticipated launch date now barely more than a month away, it's time to take a closer look at what's being claimed and what picture of Apple's product plans can be drawn from the leaked information.
According to our source, Apple CEO Steve Jobs will take to the stage at Macworld Expo San Francisco and unveil desktop Power Mac G5s running at 1.2, 1.4 and 1.6GHz. They will be based on a processor codenamed G5 - we've also heard the word 'Goldfish' - but officially believed to be monikered the Power PC 8500 on the basis of Motorola's most-recent roadmap, which gives that name to the first fifth-generation PowerPC chip. Our Apple source, it should be noted, has used that name only once, referring to the part almost exclusively as the G5.
One caveat that our source offers is that while Motorola is now ramping up volume production of the G5 and the silicon is running across a wide range of clock frequencies, yield problems remain - always an issue in the early days of a chip's life - and Apple may be forced to opt for a 1GHz machine instead of the 1.6GHz part.
Other sources suggest that may not be necessary. Another correspondent, who claims to be from Motorola, has told us that the company is producing G5s "in quantity" at all four clock speeds. Other correspondents have said in the past that enough 1.6GHz G5s have come off the production line for Apple to give real consideration to shipping a machine at that clock frequency.
Our Motorola source - if such he is - gives the following specifications, among others, for the G5: 256-bit on-chip bus, 64-bit integer and floating-point units (four to five times faster than their G4 equivalents), 400MHz "effective bus speed", 128KB L1 cache, 512KB on-die L2 cache, support for 2-8MB of external L3 cache, Book E "core instruction set" and a ten-stage pipeline.
To that we can add 64-bit addressing with full 32-bit backward compatibility.
The chip is being produced using Motorola's 0.13 micron HIP 7.0 process and contains 58 million transistors, claims our Motorola source. The die-size comes in at 192mm©. It ship in a 575-pin package. By contrast, the second-generation G4-class Power PC 7450 is fabbed at 0.18 micron, is 106mm© and uses a 483-pin package.
Many of these figures match those provided by our Apple source - and others, including other Register informants and the sources of information used by other Web sites beyond those we know our sources also communicate with.
But here's a caveat of our own: we can't, hand on heart, say that we're one hundred per cent certain that the information we have is genuine. Superficial similarities between the emails we've received from a couple of apparently unrelated sources could indicate that they all come from the same person. That would clearly question their authenticity. However, other sources' information, which also offers some of what our Apple mole has told us, doesn't follow the pattern, so we're willing - with a due sense of caution - to take what we've received at face value.
And don't forget, that Motorola European Marketing Communications Manager, Paul Clark, in an interview with Macworld UK has gone on record and effectively confirmed the existence of the G5, its use of 64-bit addressing, 32-bit compatibility, Rapid IO support and its Book E basis - though many of the details have already appeared on the company's roadmap.
Power Mac predictions
Comparing all the information we've received, it's possible to come up with a core specification for the upcoming Power Macs Apple is preparing. The machines will ship with 266MHz DDR SDRAM and contain 2MB of external L3 cache. A new version of FireWire, dubbed GigaWire, is expected to be included to that, as its name suggests, offers file transfers at 1.6Gbps, a speed made possible by the IEEE 1394b standard.
You'll notice that we haven't given the new machines a name. Why? Because that's one of the two central questions regarding the boxes that have yet to be answered. The other being, of course, what processor will they use?
We have to ask this question because the G5 isn't the only next-generation PowerPC that Motorola is working on. In October 2000, at the Microprocessor Forum, held in San Jose, Motorola unveiled Apollo, the chip it said would take the PowerPC family to 1GHz and beyond. To do so, Motorola said it would employ a new chip design and a shift toward silicon-on-insulator chip construction technology. The company's representative, David Bearden, said Apollo would be fabricated on a 0.18 micron process. It would feature 256KB of on-die L2 cache, with support for up to 2MB of external L3 cache and a 256-bit bus hooking them all together. Bearden cited power consumption of under 23W at 1GHz-plus frequencies.
Apollo is believed to be officially known as the PowerPC 7460. In short, it's the third-generation of the G4 and the successor to the 7450 Apple's is using in its current top-end desktop Macs.
So, on one hand we have a third-generation G4 in the pipeline. On the other we have a completely new PowerPC, the G5. Together we have Motorola and Apple developing two next-generation PowerPC processors at the same time. We know for a fact that at least one of these - Apollo - is under development, because Motorola has gone on record to say so. And while development schedules that overlap are not without precedent, they are not common. So are Apollo and the G5 one and the same?
Apollo = G5
There's certainly circumstantial evidence to suggest that. For Apple, the most important feature that any next-generation PowerPC can deliver is a clock speed in excess of 1GHz. The company can talk until it's blue in the face about how unsuited clock speed is to the process of comparing processors and measuring their relative performance, but when Intel is at 2GHz (2.2GHz in January 2002, 2.4GHz in April) and it's only at 867MHz, it needs to break that (largely psychological) gigahertz barrier as soon as possible.
If Apollo is designed to get PowerPC to 1GHz and beyond, that's arguably enough for Apple. Of course, the company will want to shout about such an landmark achievement as the first 1GHz-plus Mac, and how better to do that than give it a new name: G5. Not only would such a move highlight the significance of the desktop's clock speed, it would also allow Apple to bring the G4 to is consumer machines. Quite apart from that, the G4 brand, first used by Apple nearly two-and-a-half years ago, is starting to show its age.
If Apple is going to call its next desktop machines Power Mac G5s, that might well explain how the rumour mill has ended up with two processors: Apollo has been dubbed the G5, not because its a fifth--generation CPU, but because that's the machine it's going into.
Apollo isn't G5
Equating G5 and Apollo is tempting, but it's worth bearing in mind that there are strong counter arguments. Apple may, for example, be unwilling to risk any of the consumer backlash that might arise from claiming a G4-class processor (Apollo) is a G5, which most experienced Mac users - ie. the Power Mac's target audience - would expect to be a genuine fifth-generation PowerPC.
Equally, with high speed G3-class CPUs in the pipeline from IBM, Apple has less need to upgrade the iMac to G4 status than it might at first seem. And Apple does want more from its next CPU than an increased clock frequency.
Then there's the basic point that our sources say that Apollo and G5 are not the same processor. Apollo is said to operate at 1.0, 1.13 and 1.26GHz. Other specs, such as frontside bus speed and L2 cache, are different.
That said, Apollo also appears to have been finished before the G5. In which case, why hasn't Motorola announced its availability yet? Its silence on the G5 is said to be at Apple's behest, the Mac maker having contributed significantly to its design. But that surely doesn't apply to Apollo, which Motorola did launch in public?
Motorola's newly redesigned Web site may herald just such an announcement. In the meantime, the company has declined to respond to our enquiries for further details of Apollo's development progress and release schedule.
As for Apple, its refusal to comment on rumour and unannounced product means we won't know the truth until Steve Jobs gets on stage early next month as Macworld Expo. New, faster Power Macs have to be on the cards, though there remains the possibility that that perennial favourite, the next-generation iMac, will appear alongside it or instead. Ditto Apple's Mac OS X Server rack unit, as far from Apple's target markets as that is. The PowerBook G4 is unlikely to be updated before Apple's traditional PowerBook announcement slot, Macworld Expo Tokyo, in February.
However, if our various correspondents are correct - or even if we take only the points on which their accounts coincide - Apple will unveil new Power Macs. Whether they're G4s or G5s will, we think, depend on whether they're based on Apollo or the G5 - assuming they're not one and the same - and that's too close to call. ®