[OreoCookie]

So what's next from Apple?

In the new enclosure, there is a power supply with at least 650 W and a cooling system that could be safely put into a mid-sized workstation.

Apple wouldn't do that to have a power supply with 300 W to spare (350 W are sufficient for PC with power-guzzling CPUs).

So what's next and when will it be disclosed?

My speculation (top configuration):

2 Power4 derivatives (aka Power4Desktop) at 2 GHz, enough of DDR memory (dual channel) connected to a (modified) nForce2 chipset (which sports the dual DDR interface, at least for PCs), 2 (maybe SerialATA) HDs, and a new kind of grafx card (ATI Radeon 9700 or derivative). USB 2.0 + Gigawire.

This seems like a day dream at first, but all of this technology exists or will soon exist. The Power4Desktop is already announced (IBM has never announced a CPU that they didn't produce); it hopes to reach 2 GHz in fall and is slated for volume production in April 2003. There are enough discussions about that little baby.
The nForce2 chipset will eventually emerge for the PC market. It sports a dual-DDR RAM interface.
The first mobos with Serial ATA for PCs should ship now. Of course, the PowerMac G5 (or whatever they will call it) will still have classical ATA for �old' peripherials.
The Radeon 9700 is already out for sale and currently the fastest Grafx card that you can get your hands on. A higher clocked version is believed to be dubbed Radeon 10000.
USB 2.0 is out, and the Gigawire (aka IEEE1394b) specs have been nodded off in spring this year.

So all technology will probably be available (in one form or another) in June next year, i. e. MacWorld Boston, or was that New York?

My guesstimate is that we will see one last refresh of the current G4 line (speed bump, maybe the 7470, who cares) before the real big whummer.
I don't think they will go for x86, as argued before. The new enclosure clearly indicates that the Big Thing will happen within the next 8 months, which is too little time for an x86 switch (no apps for the pros, ...).

What's your scoop?

[tkmd]

I would like to believe that the mac will switch over to the IBM chip- and although the pieces of info seem to fit - I dont think its going to happen. If this was going to be apples next big thing wouldn't of SJ forced silence?
My feeling is that apple and moto will be together for a long time.

[OreoCookie]

IBM has always announced its CPUs, the G3 used in iBooks and iMacs (aka Sahara) has been announced. IBM is a lot bigger than Apple, and Steve Jobs knows that.
He hasn't shut up IBM in the past, and for the time being, he probably won't.

[vasu]

I don't know where the rumor got started that these new PowerMac (mirrored doors) have these hugs power supplies, I think it was a SpyMac article.

The New PowerMac G4 (mirrored doors) has a 400W Power Supply, only 50watts more than before, probably just to handle that 2nd optical drive and extra hard drives.

400W, not 650, not 700, or any bigger.

-vasu

[OreoCookie]

Nope, haven't read that from spymac, I have read it in a German Mac magazine that has actually tested the smallest one. The specs of the power supply are

3.5 Amps @ 220 V. They have assumed a power factor of 0.85 which is perfectly normal.

3.5 A * 220 V * 0.85 \approx 650 VA

[biscuit]

I read the SpyMac article and the resulting forum disscussion. There was some pretty convincing info there that these new PMacs DO NOT have an overly large power supply. It was all getting pretty technical, but it does seem like this rumor is a bag of hot air. The assumed 'power factor' makes a big difference and theres no way to know if the assumptions are viable.

On the IBM front, if these new chips are destined for Apple then Steve would have no chance getting IBM to keep quiet. None at all. IBM makes the chip, they can annouce it when they like.

biscuit

[OreoCookie]

That's not the point. Fact is that the power supply as well as the ventilation system is way oversized, be it one or two G4s.

My little brother's PC power supply with its 350 W is fine for his power guzzling Athlon @1500 MHz (real MHz), GeForce 4Ti, DVD-ROM, burner, 5+1 sound card ...
What will be put into this enclosure has a great need of power and a lot of heat dissipation.

There is method to this madness.

Shall we focus on the wish list?

[raferx]

DP 2.4 GHz Power 4 chips
new mobo (RAPID I/O? nvidia board?)
ATA 133 160 gig HD
DDR RAM
FW2
USB 2.0
same case...
MWNY 2003

[Cipher13]

Okay, people are saying that "this technology should be out soon, so let's take this... and this... and this..."

Have you stopped to consider this: how long has DDR been mainstream? How long has ATA/133 been mainstream? How long have 5.1 soundcards been mainsream? How long have higher-than-100 MHz bus speeds been mainstream?

...how many of these technologies do we have? Two - 133-167 MHz buses, and watered-down DDR.

The rest of them? We're still at ATA/100, using stereo sound. Of the techs we *do* have, they're mutilated versions of it.

What the hell makes you all think Apple's gonna jump to Serial ATA (example) all beacause they've got a big power supply in there? The fact that we *are* so behind in itself?

[The Ancient One]

Hmmm... Look at this:

http://zdnet.com.com/2100-1103-956036.html

Looks like it could be in the next Power Mac.

[olePigeon]

Originally posted by OreoCookie:
USB 2.0 is out, and the Gigawire (aka IEEE1394b) specs have been nodded off in spring this year.

It wouldn't surprise me if Apple never supports USB 2. USB 2 is a deadend technology. USB is a great for replacing serial, ADB, and to some extent parallel. Mice, keyboards, printers, scanners, that's fine. But USB has no business being developed for highend media technology such as hard drives, video/audio capture equipment, etc.

Leave that to FireWire.

[fulmer]

Originally posted by olePigeon:


Leave that to FireWire.

absolutely. all the reviews I've seen say USB 2 is a nice boost to the spec, but that it has bad sustained and real-world rates compared to firewire.

my research also tells me that ata/133 matters little when an (e)ide 7200 rpm drive can't even take advantage of an ata/100.

[businezguy]

The only company that has used ATA 133 remains Western Digital and in this case, they are the company that developed the technology. The rest of the hard drive manufacturers will remain with ATA 100 until Serial ATA comes out (which should be relatively soon). With that said, ATA 100 is a fine technology which is plenty fast for today's hard drives.

Measuring Apple's technology based on the ATA standard is fairly silly. It just isn't a big deal. It is extremely probable that Apple, just as with virtually every other PC maker, will skip and go directly with Serial ATA.

On the processor end, it is very unlikely that Apple will go more than 1 year without introducing a new processor from some company. Much of this speculation definately has a basis. The problem is, whe Apple comes out with the "G5", I'm not quite sure I'll care too much. Computers are pretty fast as things stand today.

Just my opinion.

[fulmer]

Originally posted by businezguy:
The problem is, whe Apple comes out with the "G5", I'm not quite sure I'll care too much. Computers are pretty fast as things stand today.

Just my opinion.

well, you're correct, in a sense. for email and web browsing, they were fast enough a long time ago. however, we've seen home computers make huge strides into photo editing and--especially with Macs--video editing. when you're a pro in photoshop, or anyone with video, those encodings can't be fast enough. I do a lot with iMovie and iDVD, and it'd be great to have speedier machines that could shave scores of minutes off encoding times which, over just a couple days, would save me hours of down-time.

[OreoCookie]

Originally posted by fulmer:


well, you're correct, in a sense. for email and web browsing, they were fast enough a long time ago. however, we've seen home computers make huge strides into photo editing and--especially with Macs--video editing. when you're a pro in photoshop, or anyone with video, those encodings can't be fast enough. I do a lot with iMovie and iDVD, and it'd be great to have speedier machines that could shave scores of minutes off encoding times which, over just a couple days, would save me hours of down-time.

I'd be more than thrilled to do my work (numerical simulations of solids) with my Mac, but this is a no-go. If one calculation takes 6 hours on an AlphaStation (a fairly new one) or 4 hours on a spankin new Athlon XP@1533 MHz, I don't even wanna try it on my iBook.

On the other hand, if I had a hammer, a huge hammer, I am sure I'd hit the nail pretty hard at home.

So -- in my case -- there is a need for more CPU power (besides, I'd already be happy, with a GUI that flows like aqua).

But I don't think I am that far off the mark. Apple knows about the problem and I am sure they are working on a fix.
Good to know that they have an emergency exit strategy (aka Marklar).

[Graymalkin]

OreoCookie's original specs are for hard core LSD users to speculate over. Someone picked up a bunch of buzzwords out of a ZD owned rag and suggested they be put in Macintoshes.

First, the processors in current Macs are by no means slow or inherently under powered. I think a part of the problem is in writing portable code and code meant to be compliant with all PowerPC chips developers needs to stay away from heavy AltiVec usage. The AltiVec unit on the G4 has enough capability to act as a second FP unit and do a ton more work than it is currently doing. It isn't good for JUST redundant matrix math. If you've got a bunch of code written before the G4 was released and you continue to use it, why break something that works trying to make it work on a vector processing unit? This is quite unfortunate for the G4's reputation and the Mac in general.

Then everyone brings up USB 2. Why do you need it again? For the ten devices that will ever use it? USB 2 is a dead end technology and in the long run will make little impact the world. Why do I spout such blasphemy? USB 2 is not needed. The original USB spec was desperately needed. USB 2 does not solve any glaring problem that currently exists. Originally USB solved the periphrial problem. USB 2 just runs faster so Intel can make noise about being faster at something than somebody else. USB 2 is braille on the keypad of a drive up ATM machine. Anything it can do Firewire can do better or USB is already quite proficient at.

Gigawire, what are you plugging into your Mac that needs FW2 class bandwidth? Methinks it is more about a numerical pissing contest much like the one involving USB2. Gigawire will come along eventually, it need not be in the next generation Powermac to make said Powermac worth buying. Some day something will really necessitate the sort of bandwidth FW2 can provide, that day will probably be another couple Powermac generations away however.

Serial ATA is still a good year away from widespread acceptance and will probably never hold a dominant position. There's hundreds of millions of PCs around, the number with serial ATA support is so small it doesn't even garner a percentage of a percentage point. Even if 10 million PCs were sold next year using it the number of serial ATA supporting computers would lack the numbers to be more than a single percentage point of all PCs. It is a great idea but will be very very slowly accepted. If Apple went whole hog with serial ATA users would be stuck with a handful of serial ATA devices. Do you want to see more "macs are incompatible" PC trolls?

The only logical things on your list are the Radeon 9700 et all graphics cards and the nForce chipset. From what I've read nForce is very agnostic in terms of what sort of processor you attach to the northbridge. The XBox uses a Pentium chip on an "nForce" board while there are one or two Athlon sporting boards available. A PowerPC chip stuck to one of those puppies is actually not that far fetched. I think it'd be a fine idea as it would really boost the performance of the Mac's periphrial subsystems above even what the unified XServe architecture offers. Same with the Radeon. Both new DirectX9 chips from nVidia and ATi are going to be damn impressive. The R300 and NV30 cores are screamers and will make quite an impact on the sort of fanciness that will be available with QuartzExtreme not to mention games.

These predictions are pipe dreams, peace pipe dreams.

[fulmer]

Originally posted by OreoCookie:

there is a need for more CPU power

that's what I was trying to say.

[OreoCookie]

Originally posted by Graymalkin:
OreoCookie's original specs are for hard core LSD users to speculate over. Someone picked up a bunch of buzzwords out of a ZD owned rag and suggested they be put in Macintoshes.

First, the processors in current Macs are by no means slow or inherently under powered. I think a part of the problem is in writing portable code and code meant to be compliant with all PowerPC chips developers needs to stay away from heavy AltiVec usage. The AltiVec unit on the G4 has enough capability to act as a second FP unit and do a ton more work than it is currently doing. It isn't good for JUST redundant matrix math. If you've got a bunch of code written before the G4 was released and you continue to use it, why break something that works trying to make it work on a vector processing unit? This is quite unfortunate for the G4's reputation and the Mac in general.

Then everyone brings up USB 2. Why do you need it again? For the ten devices that will ever use it? USB 2 is a dead end technology and in the long run will make little impact the world. Why do I spout such blasphemy? USB 2 is not needed. The original USB spec was desperately needed. USB 2 does not solve any glaring problem that currently exists. Originally USB solved the periphrial problem. USB 2 just runs faster so Intel can make noise about being faster at something than somebody else. USB 2 is braille on the keypad of a drive up ATM machine. Anything it can do Firewire can do better or USB is already quite proficient at.

Gigawire, what are you plugging into your Mac that needs FW2 class bandwidth? Methinks it is more about a numerical pissing contest much like the one involving USB2. Gigawire will come along eventually, it need not be in the next generation Powermac to make said Powermac worth buying. Some day something will really necessitate the sort of bandwidth FW2 can provide, that day will probably be another couple Powermac generations away however.

Serial ATA is still a good year away from widespread acceptance and will probably never hold a dominant position. There's hundreds of millions of PCs around, the number with serial ATA support is so small it doesn't even garner a percentage of a percentage point. Even if 10 million PCs were sold next year using it the number of serial ATA supporting computers would lack the numbers to be more than a single percentage point of all PCs. It is a great idea but will be very very slowly accepted. If Apple went whole hog with serial ATA users would be stuck with a handful of serial ATA devices. Do you want to see more "macs are incompatible" PC trolls?

The only logical things on your list are the Radeon 9700 et all graphics cards and the nForce chipset. From what I've read nForce is very agnostic in terms of what sort of processor you attach to the northbridge. The XBox uses a Pentium chip on an "nForce" board while there are one or two Athlon sporting boards available. A PowerPC chip stuck to one of those puppies is actually not that far fetched. I think it'd be a fine idea as it would really boost the performance of the Mac's periphrial subsystems above even what the unified XServe architecture offers. Same with the Radeon. Both new DirectX9 chips from nVidia and ATi are going to be damn impressive. The R300 and NV30 cores are screamers and will make quite an impact on the sort of fanciness that will be available with QuartzExtreme not to mention games.

These predictions are pipe dreams, peace pipe dreams.

You are right that this stuff is merely speculation. But I have just compiled stuff that were run stories on various sites. But a lot of information is openly available. Go to a store and buy an ATI 9700 today (for a PC, of course).
IBM has announced a CPU that will be produced -- and it will be powerful. My speculation is that Apple will use this chip.

That's what I am trying -- gather available information and try to see a little in the future.

Example: The new PowerMac. Why did everyone know that Apple will use DDR in this baby? Because they have presented the XServe and you drew the conclusion that this is technology the PowerMac will also eventually get.

Same here. Take an enclosure that is able to cope with a lot more heat and suffice the power need of a workstation.

Hey, it's about dreaming

[Metzen]

Originally posted by OreoCookie:
Same here. Take an enclosure that is able to cope with a lot more heat and suffice the power need of a workstation.

Hey, it's about dreaming

Or a P4

[adheyes]

I doubt Apple will go for the nForce in the near future. They have already gone to a 167MHz FSB before Intel and AMD who use double and quad pumped 133 FSB's respecitvely to get 533 and 266 "effective" The only thing holding back a full DDR implemenation is the G4's ability to utilise DDR.

What would be interesting is AIM licencing AMD/Digital's EV6 bus as used on the Alpha and now Athlon which allows MP systems to have concurrent access to RAM, rather than having one CPU suck all the bandwidth. Maybe IBM has something in place for their POWER 4 Derived desktop chip?

USB2 is not needed as we have Firewire and most USB2 devices are also available in Firewire. Also a USB1 device on a USB2 bus drags the entire bus down to USB1 speeds. "Mom, why can't I edit my movie? Oh son, you plugged in the keyboard as well as the camera!"

I really hope they don't use the name Gigawire as it just sounds awful. Firewire 800 would be better, if they have to give it a name at all. They may just stick with plain Firewire to emphasise that all your devices can coexist and operate at whatever speed they need to on the bus (ie 100, 200, 400, 800).

Finally, Apple won't ditch Moto altogether. IBM for the workstations, Moto for laptops & consumer.

[OreoCookie]

Originally posted by adheyes:
I doubt Apple will go for the nForce in the near future. They have already gone to a 167MHz FSB before Intel and AMD who use double and quad pumped 133 FSB's respecitvely to get 533 and 266 "effective" The only thing holding back a full DDR implemenation is the G4's ability to utilise DDR.

What would be interesting is AIM licencing AMD/Digital's EV6 bus as used on the Alpha and now Athlon which allows MP systems to have concurrent access to RAM, rather than having one CPU suck all the bandwidth. Maybe IBM has something in place for their POWER 4 Derived desktop chip?

USB2 is not needed as we have Firewire and most USB2 devices are also available in Firewire. Also a USB1 device on a USB2 bus drags the entire bus down to USB1 speeds. "Mom, why can't I edit my movie? Oh son, you plugged in the keyboard as well as the camera!"

I really hope they don't use the name Gigawire as it just sounds awful. Firewire 800 would be better, if they have to give it a name at all. They may just stick with plain Firewire to emphasise that all your devices can coexist and operate at whatever speed they need to on the bus (ie 100, 200, 400, 800).

Finally, Apple won't ditch Moto altogether. IBM for the workstations, Moto for laptops & consumer.

AMD uses HyperTransport (that is the basic technology behind nForce) for future chips too.
But the CPU mix sounds good to me.

[Metzen]

Originally posted by adheyes:
USB2 is not needed as we have Firewire and most USB2 devices are also available in Firewire. Also a USB1 device on a USB2 bus drags the entire bus down to USB1 speeds. "Mom, why can't I edit my movie? Oh son, you plugged in the keyboard as well as the camera!"

I thought some USB2 controllers could operate both speeds concurrently, no?

[Metzen]

Originally posted by adheyes:
What would be interesting is AIM licencing AMD/Digital's EV6 bus as used on the Alpha and now Athlon which allows MP systems to have concurrent access to RAM, rather than having one CPU suck all the bandwidth. Maybe IBM has something in place for their POWER 4 Derived desktop chip?

Sorry for the double posts, I'm lazy...

Have you heard about "multicores"?

They don't have the "CPU death struggle for bandwidth" either...

[OreoCookie]

Originally posted by Metzen:


Sorry for the double posts, I'm lazy...

Have you heard about "multicores"?

They don't have the "CPU death struggle for bandwidth" either...

Multicores are not going to happen for the cut-down version (far too expensive in manufacturing).

[OreoCookie]

New evidence pops up (just read it). It's not much by itself, by the evidence adds up that the new Power4Desktop CPUs will support AltiVec. At least that is what a post on gcc.org suugests (ppc64 AltiVec issue for the gcc compiler suite).

After all, if it quacks like a duck, walks like a duck, ... then it's probably not a parrot.

That would be like christmas. Nope, christmas would be when I could afford one of those MacOS X workstations when they come out

[Metzen]

Originally posted by OreoCookie:
Multicores are not going to happen for the cut-down version (far too expensive in manufacturing).

Would you have ever figured that a "Multi-Core" machine would be under that "Ultimate" category Apple makes, hmmm?

I certaintly don't expect it in the "Fast, Faster, Fastest" configuration, but that's not to say Apple wouldn't charge a premium and put it in the "Ultimate" config...

At least, until the manufacturing costs came down...

[OreoCookie]

Originally posted by Metzen:


Would you have ever figured that a "Multi-Core" machine would be under that "Ultimate" category Apple makes, hmmm?

I certaintly don't expect it in the "Fast, Faster, Fastest" configuration, but that's not to say Apple wouldn't charge a premium and put it in the "Ultimate" config...

At least, until the manufacturing costs came down...

Too expensive to manufacture. Plus there have to be some features left to the big brother.

Even IBM is now selling �cut-down' versions of the Power4 where one core is not functional. Raised the yields quite a bit and made them able to offer Power4s at a much lower price.

[gumby5647]

ok, so Motorola is looking like it's going to skip .13nm completely for the G4 and possibly move it to .9nm

Question....,if said die shrink does happen, how much on-chip L2 could they put on a G3/G4?

[adheyes]

Originally posted by OreoCookie:


AMD uses HyperTransport (that is the basic technology behind nForce) for future chips too.
But the CPU mix sounds good to me.

I think we're both right :-) nForce uses HyperTransport for its on chip interconnects. Its CPU bus is a regular 133/166MHz bus using dual or quad pumped DDR (for P4 and Athlon).

In MP P4 systems (Xeon), I believe the CPU memory bus bandwidth is shared between all CPU's. In MP Athlon systems, each CPU effectively has it's own bus to the RAM, using the EV6 protocol designed by Digital for the Alpha. Following the Digital -> Compaq -> HP mess, some Alpha engineers headed to AMD.

Apple along with AMD are members of the HyperTransport group IIRC. I believe Motorola is in the RapidIO camp.

On the subject of "defective" POWER4's being sold at a "cut down" price. Whilst these are being labelled as cheap POWER4's, they are not the same as the processor that will be announced in the autumn, nor are they really that cheap, they are just relatively cheap compared to a $10,000 POWER4 workstation.

[OreoCookie]

Originally posted by adheyes:


I think we're both right :-) nForce uses HyperTransport for its on chip interconnects. Its CPU bus is a regular 133/166MHz bus using dual or quad pumped DDR (for P4 and Athlon).

In MP P4 systems (Xeon), I believe the CPU memory bus bandwidth is shared between all CPU's. In MP Athlon systems, each CPU effectively has it's own bus to the RAM, using the EV6 protocol designed by Digital for the Alpha. Following the Digital -> Compaq -> HP mess, some Alpha engineers headed to AMD.

Apple along with AMD are members of the HyperTransport group IIRC. I believe Motorola is in the RapidIO camp.

On the subject of "defective" POWER4's being sold at a "cut down" price. Whilst these are being labelled as cheap POWER4's, they are not the same as the processor that will be announced in the autumn, nor are they really that cheap, they are just relatively cheap compared to a $10,000 POWER4 workstation.

I know, cheap is relative
If the other systems don't sell below �50.000, then it's cheap. Well, not for me anyways.

[JamesDP]

Just a little info for all the people hoping for Serial ATA. I work for a cable manufacturer developing SATA, and I can tell you that it's going to be a while. Just recently there was a gathering of companies on the SATA committee, both cable manufacturers and OEMs, to test *prototype* cables. We have a *prototype* non-functional hard drive from one of the manufacturers in our possession. There's definitely no large scale production going on. Hell, the spec isn't even complete. You'll be lucky if you see product at COMDEX in November, and even then it'll still be months before it starts showing up in PCs, let alone as standard equipment. So if you're holding off for that, you may as well buy now and sell in a year.

[Superchicken]

What's all thi explaining that to me?

Oh and by the way, I seriously hope that maybe a speed bump or two after the Power Macs go IBM the iMacs do too, so they can offer slower speed, but if this new IBM chip is gona be that much faster, it'd kinda be dumb attaching an expencive LCD to a machine that costs almost the same as a low end power mac that tottaly blows it out of the water.
not to mention then those moronic profile 4 commercials will be forced to stop saying they're faster :-P

[Metzen]

Originally posted by OreoCookie:
Even IBM is now selling �cut-down' versions of the Power4 where one core is not functional. Raised the yields quite a bit and made them able to offer Power4s at a much lower price.

Again, your making my point for me

Anyways, I suspect Apple/IBM will be able to pull it off. Right now IBM is sitting @ 1.3Ghz for the Power4. Apple is 1.25Ghz. Apple is ripe to introduce a slightly faster PowerMac @ say, 1.4Ghz/1.5Ghz (Power4-style since we know the that Apple made that chip to scale for clock speeds). The timing just seems to be right.

Again, I'm not saying a "true" Power4, just that desktop-style version that everyone is speculating about.

[OreoCookie]

Originally posted by Metzen:


Again, your making my point for me

Anyways, I suspect Apple/IBM will be able to pull it off. Right now IBM is sitting @ 1.3Ghz for the Power4. Apple is 1.25Ghz. Apple is ripe to introduce a slightly faster PowerMac @ say, 1.4Ghz/1.5Ghz (Power4-style since we know the that Apple made that chip to scale for clock speeds). The timing just seems to be right.

Again, I'm not saying a "true" Power4, just that desktop-style version that everyone is speculating about.

IBM plans to have this chip running at up to 2 GHz when it is released. Personally, 1.4 would be fine for the start (comparable or better SPECmarks Power4Desktop vs. P4)

[Pojo]

Serial ATA spec has been finished and there are controller chips already shipping, there are even motherboards that already have Serial ATA on board. There is a wave of new motherboards that have many or all of these features:

AGP8X
5.1 Sound
FireWire
USB2.0
10/100/1000 Lan
ATA133
Serial ATA
533Mhz Bus (133x4)
DDR400 Support

Depending on what you need, you can get motherboards that have all the features built in. So you still have 5-6 PCI slots open. Apple is so far behind that its just plain sad. No need to make excuses for Apple as to why they don�t implement something. The only reason they don�t bother staying current with the industry is because there a large user base that sees apple do no wrong. They will take it up the ass and like it too...

As long as you keep excusing their incompetence they will never change.

Hard drive manufacturers don�t need to change the entire drive just upgrade them with the new serial interface. Of course they rather do it with a new generation of drives of 60gig/plater or higher. It�s more about marketing at this point.

Looking for real info on Serial ATA then visit: http://www.serialata.org/index.html

[OreoCookie]

As I said, most of that technology is out there or is already announced and will come within the next 5-8 months seriously.

That's why I said �It's not all daydreaming.'

[Superchicken]

for small PC companies it might be worth it to ship oumother boards with all kindsa stuff that the majority of people won't even care about. But why should apple deveote the R&D to making a product that less than 1% of Mac users will care about having for another 6 years, and then they'll have to slow of a proccessor to care about having all that crap.

Right now Apple is focusing on things that the majority care about.

[OreoCookie]

Well, it's no distraction at all. E. g. Apple is a (founding?) member of the Hypertransport consortium, it seems natural to me that they will eventually incorporate this into future products. It's not a distraction, it is a consequence of years of activity in certain fields. Now that Apple also uses nVidia cards, it is at least not totally out of question that they will use nVidia Hypertransport technology.

[Pojo]

The HyperTransport� Technology Consortium was formed on July 24, 2001 by a coalition of high-tech industry leaders including AMD, Apple, Broadcom, Cisco Systems, NVIDIA Corporation, PMC-Sierra, Sun Microsystems, SGI, and Transmeta.

For the latest list of members check http://www.hypertransport.org/org_members.html

Here is the list of notable products that use HyperTransport.. http://www.hypertransport.org/featur.../products.html

Nice specs...
http://www.hypertransport.org/technology.html

[cdrgonzo20]

This was over at Xlr8yourmac.com general mac forum. Hate to burst your bubbles.

----------------------------------------------------
From: Craig Hunter <[email protected]>
Date: Thu Sep 5, 2002 5:30:18 PM US/Eastern
To: <[email protected]>, <[email protected]>
Subject: IBM Power4

Folks,

I attended a briefing today on IBM's high performance computing technology,
which is hinged on their Power4 CPU (this CPU has awesome performance in the
various real-world benchmarks I have seen). After the briefing, I asked the
presenter (a chief engineering manager from IBM) about the Power4 derivative
for desktops and low end servers to be announced in October. You may recall
that there has been speculation that this CPU would find its way into
PowerMacs in the future. Well, it sounds like this CPU is not in Apple's
future -- the "over 160" vector instructions are not AltiVec (even though
AltiVec has 162 instructions), and there are technical issues that would
prevent AltiVec from ever marrying with Power4 or its successors.
Furthermore, the guy came right out and said that they have pitched the
desktop Power4 to Apple, but Apple was not interested.

So, although Power4-based PowerMacs seemed like a promising (and likely)
possibility, it looks like it won't happen. I guess we will have to wait
and see what Apple has in store for the future . . . .

Craig

--
Dr. Craig Hunter
NASA Langley Research Center
AAAC/Configuration Aerodynamics Branch
[email protected]
(757) 864-3020
(G4 Cube - OS X)
------------------------------------------------
I figured this would be the case, and its really not that serious of a problem. Power has been on the back burner for Apple for quite some time, and it will obviously stay that way. It makes sense, and is a reasonable business decision if there weren't two serious issues I had for it....

1) They are consuming professional grade film/audio companies to broaden their professional appeal, yet are stuck with processors that are vastly inferior to PC processors... Let's not kid ourselves, in terms of raw, system wide performance it is not even close.

2) If indeed we are stuck with slower chips, then shouldn't we be paying at least a little bit less for them? I plopped a total of around 4000 to get my 733 w/ superdrive, and while I have no regrets, and have thoroughly enjoyed the experience, that is still quite a bit for a computer that is laughably slow in comparison to the dark side.

OS X is fantastic, and I refuse to spend money on Microsoft ever again, but we are a dying breed. Computers have made it to the point where speed is mostly academic, but the fact is that consumers are still sold on speed, and Apple does not have sufficient presence to change that paradigm... It will be interesting to see what happens. Whether or not they can really make a run at it almost entirely based on thier brilliant line of software.

[Apple Pro Underwear]

from reading the above post...

I have said repeatedly the power4 thing is a rumor and not to get your hopes up for it until you see some more concrete evidence...

Having said that...IF APPLE TURNED DOWN THE POWER4...they must have something better in the works because it seems like now that OSX is in the house, they must address other issues. Like speed for the Powermacs. I say this becuuse if they want to be in professional creative fields...they must have superior equipment.

But meh...new processor is years away. We are in the OSX years for awhile, where nothing happens except OSX becomes soaked in by the public.

[Nimisys]

Originally posted by OreoCookie:
nVidia Hypertransport technology.

try amd, it was their lightning data transport bus, they finished the spec and then open up a consortium under hypertransport name

motorola did the r&d for rapid IO as well

and Intel did 3GIO

3GIO+Hypertransport should = arapahoe

[OreoCookie]

Originally posted by cdrgonzo20:
This was over at Xlr8yourmac.com general mac forum. Hate to burst your bubbles.

----------------------------------------------------
From: Craig Hunter <[email protected]>
Date: Thu Sep 5, 2002 5:30:18 PM US/Eastern
To: <[email protected]>, <[email protected]>
Subject: IBM Power4

Folks,

I attended a briefing today on IBM's high performance computing technology,
which is hinged on their Power4 CPU (this CPU has awesome performance in the
various real-world benchmarks I have seen). After the briefing, I asked the
presenter (a chief engineering manager from IBM) about the Power4 derivative
for desktops and low end servers to be announced in October. You may recall
that there has been speculation that this CPU would find its way into
PowerMacs in the future. Well, it sounds like this CPU is not in Apple's
future -- the "over 160" vector instructions are not AltiVec (even though
AltiVec has 162 instructions), and there are technical issues that would
prevent AltiVec from ever marrying with Power4 or its successors.
Furthermore, the guy came right out and said that they have pitched the
desktop Power4 to Apple, but Apple was not interested.

So, although Power4-based PowerMacs seemed like a promising (and likely)
possibility, it looks like it won't happen. I guess we will have to wait
and see what Apple has in store for the future . . . .

Craig

--
Dr. Craig Hunter
NASA Langley Research Center
AAAC/Configuration Aerodynamics Branch
[email protected]
(757) 864-3020
(G4 Cube - OS X)
------------------------------------------------
I figured this would be the case, and its really not that serious of a problem. Power has been on the back burner for Apple for quite some time, and it will obviously stay that way. It makes sense, and is a reasonable business decision if there weren't two serious issues I had for it....

1) They are consuming professional grade film/audio companies to broaden their professional appeal, yet are stuck with processors that are vastly inferior to PC processors... Let's not kid ourselves, in terms of raw, system wide performance it is not even close.

2) If indeed we are stuck with slower chips, then shouldn't we be paying at least a little bit less for them? I plopped a total of around 4000 to get my 733 w/ superdrive, and while I have no regrets, and have thoroughly enjoyed the experience, that is still quite a bit for a computer that is laughably slow in comparison to the dark side.

OS X is fantastic, and I refuse to spend money on Microsoft ever again, but we are a dying breed. Computers have made it to the point where speed is mostly academic, but the fact is that consumers are still sold on speed, and Apple does not have sufficient presence to change that paradigm... It will be interesting to see what happens. Whether or not they can really make a run at it almost entirely based on thier brilliant line of software.

Well, haven't heard of that, but we'll see what happens. Till now, all pieces of evidence seemed to fit ...

This casts some doubts, but I nevertheless would bet some money on it, because it is the most probable solution.

BTW, could you post the link? Thanx.

[Pojo]

I sure like being annoying and spamming threads.

[jrramsey]

why did I quote the whole spam post, just to make a one line comment?

[Pojo]

Scotttheking come on man that presses release had lots of relevant info to this thread. I didn�t have the time to just clip out important information so I posted the entire press release.

So in the future please refrain from mindlessly editing post, can I quote you on that?

I sure like being annoying and spamming threads.
Last edited by Scotttheking on 09-06-2002 at 08:49 PM

[KidRed]

mm, because of the editing, I can't make out what the last 2 posts were about. What article?

[piracy]

Originally posted by KidRed:
mm, because of the editing, I can't make out what the last 2 posts were about. What article?

This "article" (actually, a post to Apple's Scitech mailing list):

http://archives:[email protected].../msg01219.html

[Stephane]

the webmaster from a well known Belgian OS X focused website
has written an article in Macgeneration (in french) : Stephane Rieppi writes that despites craig hunter sayings about the Power4 desktop being non Altivec able, there is a new patch recently added in the GCC compiler . This patch added on the 25 August 2002 by a IBM programmer is titled : [BOLD]powerpc64-linux altivec support[/BOLD] !

I think you read like me : PowerPC 64 and ALTIVEC SUPPORT which means only one thing for me. IBM has recently readied a POWERPC 64 bits proc Altivec ready : be it the Power4 Desktop or its little sister, I don't mind. And "Linux" : who are you kidding ?
Since when Linux makes a bigger use of Altivec PPC processors than Apple's Unix ?

So it seems than the Power4 is still in the race to be the G4 successor...

[OreoCookie]

Originally posted by Stephane:
the webmaster from a well known Belgian OS X focused website
has written an article in Macgeneration (in french) : Stephane Rieppi writes that despites craig hunter sayings about the Power4 desktop being non Altivec able, there is a new patch recently added in the GCC compiler . This patch added on the 25 August 2002 by a IBM programmer is titled : [BOLD]powerpc64-linux altivec support[/BOLD] !

I think you read like me : PowerPC 64 and ALTIVEC SUPPORT which means only one thing for me. IBM has recently readied a POWERPC 64 bits proc Altivec ready : be it the Power4 Desktop or its little sister, I don't mind. And "Linux" : who are you kidding ?
Since when Linux makes a bigger use of Altivec PPC processors than Apple's Unix ?

So it seems than the Power4 is still in the race to be the G4 successor...

I agree. The market would be too small for IBM itself. They know that they got a chance now to get the people to switch to �their' 64 bit architecture.