Will the next Powerbook G4 have true DDR implementation?

That depends on good'ole Motorola. If they finally are shipping their famous 745x-RM-chip, this might be a possibility, else, you'll probably end up with the same kind of DDR implementation as it's currently.

Well if the use the same bus protocol and just up rate it to 200MHz then DDR won't mean anything. In the end the performance increase would be marginal as we are only talking 33MHz.

Now if a new improved G4 came out with an independant memory interface to DDR ram of any speed it would be a very nice improvement. If the complement of I/O was DDR memory, a Hypertransport interface, and maybe the system I/O we would have a potentially awsome upgrade. There are how ever no indication that such a chip is coming from Motorola.

The issue becomes is there enough of a performance increase with the current chip to justify a revision to a 200 MHz bus. I would have to say no especially relative to the performance increase we are seeing every where else. So I think an improved G4 is in order or a complete replacement for that chip.

The bigger possibility is a complete replacement for the G4 from IBM, desinged especially for portables. But I think that is a year or two off.

Dave

Just buy one now.
Dont expect any new cpu's from Moto targeted at Apple. The new chip company will be struggling with their embedded customers to turn a profit.

Errr, I must be misunderstanding your post somehow.

How exactly is it that the current G4 Powerbooks are not using "true" DDR? Just because they're using a slower standard of DDR-RAM doesn't mean that it is not DDR. Also worth mentioning that DDR-400 is relatively new in the industry.

Chagi - because the processor cannot actually use the DDR

Ummm, yes you are misunderstanding, ... maybe you should check out Apple's G4 motherboard implementation. What he's referring to is Apple's current G4 motherboard implentation using DDR. While the bus between the memory and system controller is a double pumped bus, the bus (FSB) between the system controller and CPU is a normal bus (It isn't double-pumped - it's a single 167Mhz bus). Therefore the FSB cripples or bottlenecks the CPU. Apple is using true DDR RAM, but the FSB cripples the full benefits of using DDR RAM.

IMHO, we won't see any new CPUs from Motorola, Faster G4s, maybe, but nothing with on-board DDR support.

What on earth do you mean with "the processor cant use the DDR" the L3 cache uses DDR, your statement does not make any sense. What would you call it if the fsb was suddenly bumped to 400Mhz without DDR?

If you mean crippeled IO then yes I'll agree that the G4 machines has got a crippeled IO system in the sense that the CPU can't use the full bandwidht provided by the memory sub-system.
And I'll agree on that the current G4's has got, by modern standards, sub par IO.

It should be mentioned thought, that the people that I know who work with Moto equipped system have been hinting for quite some time about G4's with better IO (due to info from Moto sales drones). But then again with all the spin-off rumors and the monolithical "screw up culture" that seems to reside within Motorola I wouldn't even trust them to take care of my pet cat.
So I'm taking those rumors whith a suitable truckload with salt.

1) The PowerBook G4 has a true DDR implementation. I $$$$ing hate this bullshit about what's "true" DDR and what's not. Yes, the memory bus and the CPU bus are asynconous, but that's the case on almost every Athlon computer in the world, and you don't see people calling them "hacks" or "false DDR". Hell, the G5 has asyncronous busses.

2) Insofar as the 745x G4s ever getting a faster front side bus, no way, not going to happen, never. Maybe a new chip will be designed with a G4 core and Rapid-I/O and an onboard memory controller (as Amorph suspects), but you won't see a faster FSB on the 745x.

Barto

That's correct, and also COMPLETELY irrelevent to Apple's various G4 architectures.


Believe me, that's totally obvious reading your post

No CPU other than those with onboard memory controllers (eg POWERs, Athlon64s) ever sends and recieves directly to and from memory. In between the CPU and the memory is a northbridge (or system controller, or system intergrated circuit). The CPU makes requests to the northbridge for data from the memory, and the northbridge fetches the data from memory and sends it to the CPU.

It IS true that less data can be sent to the G4 alone than can be fetched from memory, but it is NOT half. System busses aren't that simple, you can't just look at the MHz and make a judgment. The G4's bus is very efficient for it's speed, and DDR is highly inefficient.

Now guess what. The Athlon is the same. The majority of Athlon computers in the world have a CPU bus slower than the memory bus. Guess what, PC user's don't call them "hacks" or "false". Do you know why? Because there's nothing wrong with asyncronous busses.

But wait, there's more.

In modern architectures, more and more devices have DMA. That's Direct Memory Access, the ability to ask for data from the memory, without going through the CPU. On the PowerBook, DMA devices include gigabit ethernet, the hard drive and most importantly the graphics chip. Using Quartz Extreme, the Radeon 9600 in the PowerBook G4 transfers buttloads of data from meory. So it's not JUST the CPU that consumes the memory bus, it's the networking, drives and graphics chip too.

Barto

Hey, you aren't wrong that the G4 suffers because of the limited bandwidth available to it.

A faster front side bus or an onboard DDR memory controller would vastly improve the G4's performance in many applications.

It isn't a hack BUT it is bad and the slowest of all the desktop CPUs out there.

Barto

From memory we all got whipped into a frenzy by the 7457 RM, and then Motorola cancelled it (after probably never doing any actual work on it). Amorph and others keep saying Motorola will probably introduce a G5 (the BookE embedded G5s, not the POWER derived 970 G5s) with a G4 core and onboard DDR controller. If that happened, I'm almost sure Apple would use it in low cost Macs.

Of course, there's nothing stopping IBM from doing the same thing, only using the G5's VMX unit and a G3 core to create basically a 7400 (the original G4). Combine that with an onboard DDR memory controller and you have a Motorola killer. That isn't as simple as it sounds but IBM thought there was a market for the G5, and they built that didn't they

Barto

I have too read somewhere that this project has indeed been canceled.


Once upon a time there was something like 85xx series of Motorola processors. If I remember well, this was supposed to be the G5 in the PowerMacs (is this the BookE G5 you mention?). This G5-PowerMac project was also killed since, allegedly, the boxes... exploded! But what happened to 85xx since then? Is there any use of them today? Or improvements to make them usable in a notebook?

that wasn't the 85xx. the 75xx had to be the rumored G5 (back in october/november 2001) canned by motorola somewhere early 2002.

The Motorola 85xx is real. It is a BookE chip, but IBM's competing BookE based PowerPC 440 owns Motorola (ie, IBM delievers). BookE is basically the next generation architecture developed by IBM and Motorola, designed to replace the G3 series chips (although it could replace the G4).

Now, there was a G4 replacement BookE chip suitable for inclusion in Macs. It was called the 7500, it was real, it was on the Motorola roadmap. Sometime late 2001 or early 2002 it got cancelled.

Then a bunch of crappy rumor web sites like MOSR and TheReg (and, ahem, Dorsal) started talking about the 85xx being the G5. Which is totally and utterly bogus, and ignores all common sense. Then when the 85xx G5 didn't materialize, they all made up bogus stories to cover their tracks, like "boxes were exploding".

Barto

Funny, I mean really sad... It is not however clear to me if it is reasonable to expect some replacement or, at least, substantial improvement for the G4 next year (powerbooks and iMacs; I expect the iBooks will be using current G4 for a while). And what would be the candidates?

I would be surprised if the 9xx series chips are the only options for Apple next year.

That said, the G4 is probably dead. The standard G4 has stalled at 1.33GHz for a year or so. I don't see Motorola spending time on a product they consider has no future.

Looking to the future, there is nothing concrete at all. Certainly anything new will most likely be a BookE chip with a fast, Velocity Engine enabled core and onboard memory controller. An IBM BookE chip is the most likely candidate with Apple's relationship with Motorola seemingly over.

Barto

It would be smart for motorola to keep as many customers as possible given their situation at the moment. It seems they have lost to IBM for the desktops; however, if they could pump out the 74x7 -RM in a timely manner they may be able to keep Apple's portable market. I am no hardware engineer but it seems as if it can't be too bard to add DDR support into the G4. I mean really? Would it not be worth it to moto to do this considering how many people are using DDR now? I am sure it would help them sell their chips to other companies besides apple...

I MEAN COME ON!!!!!!!!!!

Thanks for the post. A lot of consumers don't realize the RAM subsystem services more than the processor.

As far as FSB goes, while it would be nice to have a faster FSB, you have to ask yourself why? What will it really do for me? On the 7455 Processors, for almost all Office Productivity tasks, the FSB speed is invisible. The instructions and size of the data being worked on is within the performance envelop of the FSB. The question is where does a higher FSB make sense. Well Video editing, complex graphics work, SIMD operations on huge data sets. You get the point. For most of us what is there is more than enough, though it would be nice to start at 400MHz FSB to future proof for Digital Lifestyle apps.

But, one of the biggest problems with most computers is that they don't use what they got now. Software isn't developed very efficiently either. With simple techniques where Load operations are executed in parallel with Integer, FPU and SIMD instructions, a significant performance increase can be appreciated. Too, Apple continuing to optimize the GCC compiler, we are seeing significant performance increases with memory transfers as borne out by xBench with Panther compared to Jaguar.

Using Quartz Extreme many of the Aqua operations no longer use the FSB for display operations, but do DMA from the GPU. So a lot of the OS work that can clog up the FSB has been offloaded to the GPU subsystem, like shading, moving, resizing, window drawing, transparencies, etc. Panther uses even more of the GPU for many of its operations than Jaguar.

So what you have with Apples design is a better balanced system, were all the components are contributing to consistently good performance. The FSB will not be as bottleneck if less cars are driving down its road.

There are other advantages to the Apple implementation of the FSB. It is a fully duplexed FSB, which means that a Load and a Store operation can be performed simultaneously. So 167 MHz in either direction which would consume the entire DDR bandwidth.

Basically, my point is that it is not as grim as some make it out to be. The current FSB is more than enough and well balanced with the system it is on, especially when you include such other enhancements such as QE. Can it be better? Yes, it can always be better. It would be better in the Windows World as well if they had QE and a faster per clock CPU like the G4 as well as an orthogonal SIMD unit like Mac users have.

If the G4 is not enough for you, Apple has the G5 which addresses all the FSB issues you can shake a stick at.

Sorry to be nitpicky, but it's called "DDR SDRAM". as opposed to just plain "SDRAM", with the "SD" standing for "Synchronous Dynamic", not "Single Data".

I thought benchmarks of the DDR PowerMac G4s showed it wasn't a real win over SDR. The primary advantage DDR has for Apple is that SDR is basically obsolete and will soon cost more (if it doesn't already).

Also, Apple obviously thinks a fast system bus is important because the G5s have a incredibly fast bus.

Of course a fast fsb is important in a workstation, that and it's "a good thing TM" for marketing.

Our front page just chimed in with the news that IBM is making the 750VX (the long-rumored "G3+Altivec") especially for Apple, and that it's in the final stages of development now.

This is a likely candidate, because the 7457 is the end of the line for the Motorola G4. The 7457-RM got canned, and Motorola has only made noises about a "high-performance PowerPC" next year, which could be anything. Dual core offerings are an outside shot; RapidIO busses to memory controllers are a near lock, as is 90nm. The question, as always, is will Motorola be able to produce these things in any quantity, and in a timely fashion? If their experience with 90nm is anything like their experience with 130nm (and keep in mind, this time it's not their fab they're using, and they have help from Philips and STM) we won't see it for a good long time.

So the IBM CPU is a good candidate for a replacement. Since the general description (G3 with longer pipelines and AltiVec) matches the current G4 closely, Apple's claims that the G4 has a roadmap with Apple can survive the jump across vendors and models.

Well maybe it sounds grim because it is. Not so much from the perspective of todays machines, but form the perspective of stagnation in the laptop lines. Much as the MAC community has gone through this issue with the desktops in the past, the fear is that no growth is due for the laptop lines.

On ething you did miss that can be an issue with bus bandwidth is all the fast serial ports now attached to a MAC. USB, Firewire, and gigabit ethernet can soak up a bit of that bandwidth. So while you are correct that Apple has attempted to balance system performance as best they can, and has done very well I might add, there can be problems with the current designs. Like all things it is question of how you use it, for many the systems are fine, but better performance does open up new opportunities.

If there is a peception that the line has stagnated and those performance increase aren't to be seen then Apple will have a bit of a problem.

Dave

There's a Media Access Controller community? Huh?