Say Hello to the PPC NextGen (G5?)

[drastik]

All Right, this could be very cool. I'd like to See it by Next year, just in time for me to buy, probably cost a bundle though. Maybe IBM can produce enough chips to keep em cheap though, moto can't seem too, but the Saharas are cheap, and peopl keep saying on par with G4 sans Altivec.

 

[Chryx]

Originally posted by topicolo
Not to my knowledge (or any of the hardware sites' either)

How about straight from the Horses mouth?

Nforce page on Nvidia.com

The SPP/IGP features the TwinBank™ Memory Architecture for efficient memory processing, a Dynamic Adaptive Speculative Pre-processor™ (DASP) to help boost CPU performance, and the AMD HyperTransport, a high-performance I/O bus interface.

For the record, the interconnect between chipset components on Nforce is Hypertransport.

 

[Faeylyn]

Originally posted by jadam
not really sneed, its PPC compatible, no reason why it cant boot into OS0.

While getting OS9 to work on newer hardware like this is technically possible, it also would be a big resource drain on Apple. And it's not just booting, but all the drivers for all the different goodies out there. The more radical the change to the machine, the more time it will take. Why bother?

 

[bobindashadows]

Originally posted by sneed


Wouldn't there be an issue with 64 bit?

There could be a 64-bit/backwards compat-32 bit proc to run classic in os x, but when Apple drops classic bootup their just going to take out the boot handlers/(drivers whatever you want to call them) for os 9. Apple isn't going to want to hold on to OS 9 that long, and I'm pretty sure that if there was no backwards compat- chip and classic didn't work on 64 bit (dunno if it wouldn't but i don't think so) then Apple would take this chip over classic.

Just my non-refundable $0.02 USD Restrictions Apply Results may vary.

 

[sneed]

Originally posted by bobindashadows

There could be a 64-bit/backwards compat-32 bit proc to run classic in os x, but when Apple drops classic bootup their just going to take out the boot handlers/(drivers whatever you want to call them) for os 9. Apple isn't going to want to hold on to OS 9 that long, and I'm pretty sure that if there was no backwards compat- chip and classic didn't work on 64 bit (dunno if it wouldn't but i don't think so) then Apple would take this chip over classic.

Just my non-refundable $0.02 USD Restrictions Apply Results may vary.

Can you imagine a better carrot for developers to go to X over 9?

 

[rice_web]

How many pipeline stages in this puppy? If it's 4, that will be three less than the current batch of G4s, and result in more work per clock--just like the 750FX is now.

And that vector processing... yowser, this could be great.

 

[digital1]

Originally posted by bobindashadows

There could be a 64-bit/backwards compat-32 bit proc to run classic in os x, but when Apple drops classic bootup their just going to take out the boot handlers/(drivers whatever you want to call them) for os 9. Apple isn't going to want to hold on to OS 9 that long, and I'm pretty sure that if there was no backwards compat- chip and classic didn't work on 64 bit (dunno if it wouldn't but i don't think so) then Apple would take this chip over classic.

Just my non-refundable $0.02 USD Restrictions Apply Results may vary.

------------------------------------------------------------------------------------

Also, there could be special handlers directly on the processor that could handle backwards compatibility. Basically you throw seperate pipes within the processor core to handle this. And since this could be based on the Power4 processor, and the Power4 processor was designed to handle multiple cores, hardware wise this could be possible. But also, I think it would be apple shooting itself in the foot. They are talking so much about putting OS 9.2.x to rest and letting OS X take the torch. But hardware wise this could be possible.

 

[ddtlm]

rice_web:

Well over 4 stages, since it's Power4 related. I don't know how many Power4 has, but at 1.3ghz and 130nm, it's not any 4-stage chip.

Chryx:

Bingo about hypertransport.

mischief:

A nice dream, but I don't know how you'd cool a "G5" in a case without the right fans in the right places.

macmunch:

Give up on the nForce2. It's not even very good, all it offers is onboard graphics almost as good as GF4MX. AFAIK, it fails to outperform more down-to-earth alternatives such as VIA's KT400.

everyone:

There is nothing to suggest this will be released anytime real soon, but I bet it will show up in 2003 and give Apple a fighting chance again.

 

[Hawthorne]

Not to dampen the parade, but,

What is the heat output associated with this chip? Does it mean the installation of so many fans that a Mac would sound like the flight line of the John C. Stennis?

 

[StephenBeckwith]

Future MAC's and HT

In case you all missed it, when AMD showed off their "Hammer" parts back during the IDF (Intel Developer Forum) in Feb 2002, the test systems were running with API (the old Alpha Semi guys) HT silicon, primarily for CPU to PCI bridge functionality. The Hammer (or Opteron, or whatever they call them these days) are built strictly with HT interfaces. So, this technology has been around for quite some time. It will be interesting to WHEN (not IF) HT makes it's debut in a "mainstream" PC System (Mac or otherwise).

 

[sneed]

Re: Not to dampen the parade, but,

Originally posted by Hawthorne
What is the heat output associated with this chip? Does it mean the installation of so many fans that a Mac would sound like the flight line of the John C. Stennis?

wouldn't the larger heat sink take care of that?

 

[Rocketman]

Re: sweet

Originally posted by dongmin


We know IBM's desktop hardware presence is small (if any at all) so this new desktop-targeted PPC chip can't be for themselves. And if it's PPC-based, it's probably not for the x86 crowd. Who does that leave? Linux? Nah, it's gotta be Macs.
[/B]

TV commercials in the LA market for IBM tout their hardware with linux os.

Rocketman

 

[bousozoku]

Originally posted by Faeylyn
This processor was announced last October. I doubt IBM has been sitting around playing solitare since then. Even if it doesn't make it into this next round of Macs (and I think it will), my guess is it will be out in less than 12 months.

if they are playing solitaire, they're probably playing it really fast.

 

[digital1]

Think about it this way. If any of you have seen the huge layer of Heat sink for the PS2,than you can speak too. But I think if you have a decent enough heat sink, and Decent fans, and just plain good ol' manufacturing processes. By the time these processors come out we will .13 and .1 manufacturing processes, and that will dramatically cut down heat dissipation.

 

[Chryx]

Originally posted by ddtlm

Give up on the nForce2. It's not even very good, all it offers is onboard graphics almost as good as GF4MX. AFAIK, it fails to outperform more down-to-earth alternatives such as VIA's KT400.

Point of contradiction #1 : VIA chipsets tend to be fast, but twitchy

Point of contradiction #2 : Nforce also offers a pretty powerful DSP for Audio purposes, I can see that being useful in a mac

Point of contradiction #3 : remove the onboard video (which I can still see as a useful thing for consumer lines) and you have a BIG FAT MEMORY SUBSYSTEM to feed whatever processor you hang off it, dual PC3200 channels and a processor with an FSB big enough to actually utilise that bandwidth would be somewhat astonishing

the current Athlon's only have a 133Mhz DDR fsb, which is obviously the bottleneck,

Processor < 2.1GB/s > Memory controller in 'northbridge' < 6.4GB/s > memory

crank the FSB of the processor up to match and you have a serious powerhouse

 

[dongmin]

Originally posted by Faeylyn
This processor was announced last October. I doubt IBM has been sitting around playing solitare since then. Even if it doesn't make it into this next round of Macs (and I think it will), my guess is it will be out in less than 12 months.

Can you provide a link of this announcement?

Assuming the rumored new motherboard and case makes an appearance next week, I would say 12 months MAX before the G5 replaces the G4.

My bet is for MWNY, July 2003.

 

[Chryx]

Originally posted by rice_web
How many pipeline stages in this puppy? If it's 4, that will be three less than the current batch of G4s, and result in more work per clock--just like the 750FX is now.

For the record, the number of stages isn't directly tied to the IPC.

notice how the 12-stage AthlonXP manages to land fairly evenly with the 7-Stage 7455 on a clock-for-clock basis. (unless Altivec and/or L3 cache come into the equation)

 

[mischief]

Originally posted by ddtlm
rice_web:


mischief:

A nice dream, but I don't know how you'd cool a "G5" in a case without the right fans in the right places.

Perhaps you read too quickly: I was saying that the FRAME was effectively unchanged and that it would be possible to swap out BASICALLY EVERYTHING. If you saw the P58 images and really looked at them you'd see that it's an El Capitan with it's guts completely reworked, new faceplate/backplate assemblies and a fan hacked into the stationary side. Really it would be quite possible for all but the dual Optical drives.

 

[Chryx]

Originally posted by dongmin
Can you provide a link of this announcement?

I think he means the POWER4 (which was announced/released late last year) ?

 

[bobindashadows]

Re: Re: Not to dampen the parade, but,

Originally posted by sneed


wouldn't the larger heat sink take care of that?

The heat sink wouldn't dampen the sound of the fans, well maybe a little if the fans are close enough to the heat sink, though i doubt Apple would release a machine that loud ;-)

 

[sneed]

I was under the impression that a heat sink reduces the need for fans.

 

[bobindashadows]

.1 micron? don't think so

Originally posted by digital1
Think about it this way. If any of you have seen the huge layer of Heat sink for the PS2,than you can speak too. But I think if you have a decent enough heat sink, and Decent fans, and just plain good ol' manufacturing processes. By the time these processors come out we will .13 and .1 manufacturing processes, and that will dramatically cut down heat dissipation.

I recently had a little lesson in transistor construction with a man by the name of Dr. Murarka - the person titled as being the father of the copper manufacturing process, and i believe (believe) he said that the smallest theoretical space we are going to be able to use w/ a practical amount of leakage is .11 microns... however i might be mistaken.

 

[ddtlm]

Chryx:

IBM will have their own chipsets as they always do... I would be amazed if the nForce2 was even on their radar. People *love* to believe unsubstantiated rumors round these forums, but just cause the nForce2 exists, just cause it has two channels of DDR memory, and just cause it sounds like it could be cool... we have not been provided any reason to believe that it will ever show up in a Mac.

The Athlon's FSB "which is obviously the bottleneck" is not really standing in the way of progress, see http://www.tech-report.com/reviews/2002q3/athlon-333bus/index.x?pg=1
for some hard numbers. At anandtech there is an article http://www.anandtech.com/showdoc.html?i=1344 which shows that lower speed Athlons (up to 1.2ghz) do not benefit much from DDR vs SDR. Since the G4 core outside of AltiVec is arguably weaker than the Athlon core, and since the G4 additionally has an L3 cache which reduces main memory traffic, there is no reason to believe that DDR is a big deal for the G4 or that dual-channel DDR would be a big deal for this "G5". I think that overall, bandwidth is overated, and has become nothing more than a scapegoat for the G4's performance.

As I've said many times, my 1.53ghz Athlon on KT133A (PC133) kicks the crud out of my G4-800-DP on PC133.

 

[Chryx]

Re: .1 micron? don't think so

Originally posted by bobindashadows


I recently had a little lesson in transistor construction with a man by the name of Dr. Murarka - the person titled as being the father of the copper manufacturing process, and i believe (believe) he said that the smallest theoretical space we are going to be able to use w/ a practical amount of leakage is .11 microns... however i might be mistaken.

I think Intel have already got .09 stuff working in the labs?

 

[dongmin]

Remember this quote by Steve Jobs:
"Now, as you point out, once our transition to Mac OS 10 is complete, which I expect will be around the end of this year or sometime early next year and we get the top 20% of our installed base running 10, and I think the next 20 will come very rapidly after that. Then we'll have options, and we like to have options. But right now, between Motorola and IBM, the roadmap looks pretty decent."

He seems to be hinting at two things here: first, dropping OS 9 support and second, introducing new architecture(s), made possible by going fully OS X. And even though it's ambiguous, he seems fairly comfortable with staying with the PPC for a while. (that means no to OS X on Intel)

He makes the point of mentioning both Moto and IBM so IBM will obviously play a big role in the future, and we know that's not going to be with G3s.

"The roadmap on the PowerPC actually looks pretty good and there are some advantages to it. As an example, the PowerPC has something in it called AltiVec, we call the Velocity Engine -- it's a vector engine -- it dramatically accelerates media, much better than, as an example, the Intel processors or the AMD processors... so we actually eke out a fair amount of performance from these things when all is said and done."

Yeah, Altivec is here to stay. IBM must be licensing it from Moto. Or maybe Apple bought out the PPC assets from Moto. It's one aspect of the G4 that really shines. So it's good to see Apple keeping it around and committing to it more. Hopefully developers are hearing this and following suit.