PowerPC 975 or Not?

[NNO-Stephen]

there are three possible scenarios the way I see it.

1. IBM was targeting the 3Ghz part to be the 975. The manufacturing problems were with the design of the 970FX, and not the 90nm process as some have suggested. Thus, they can still meet the 3Ghz projection by summer end.

2. IBM was targeting the 3Ghz part to be the 975. The manufacturing problems were with the 90nm process, and not the design of the 970FX as some have suggested. Thus, they can not meet the 3Ghz projection by summer end, and will instead release aprox. 2.6Ghz processors at best.

3. IBM was targeting the 3Ghz part to be a ramped 970FX. the manufacturing problems were with either the 90nm process or the processor itself (it is a moot point in this scenario.) They have solved these issues and will release aprox. 2.6Ghz processors at best.

I believe the 2nd one may be true. they could have possibly worked out the issues with the process, and then made some massive runs for Apple, but I just don't know. we'll see. Either way, IBM must be busting their collective asses right now. the POWER 5 introduction, nVidia chip manufacturing, and Apple is undoubtedly putting great pressure on IBM to deliver the 3Ghz promise.

but either way, those are the three scenarios I see. take your pick.

 

[wrldwzrd89]

Why has everyone forgotten about rumors past:

<Quote><I>originally posted by <B>Macrumors</B></I>
The PowerPC 980 Processor is a rumored chip to be coming from IBM as a followup to the PowerPC 970. Just as the PowerPC 970 is based on the Power4 Processor, the PowerPC 980 is rumored to be based on the upcoming Power5 Processor.

First hints of a new derivative chip came from this extensive roadmap of the PowerPC, Apple, Motorola and IBM's relationships from November 2002. At that time, the PowerPC 970 had been announced by IBM but not yet publicly endorsed by Apple. Apple later announced PowerPC 970 based PowerMacs in June of the following year (2003).

Today, Appleinsider reports that the PowerPC 970 will max out at 2.6GHz with the Power 980 coming due in Q3 of 2004 to bring the PowerMac line up to and beyond 3.0GHz. The 980 is reportedly a 90nm chip and may be branded as the G6. Appleinsider also comments on updates to the subsystems of the upcoming PowerMacs offering significant performance gains.

This "980 @ 3GHz" timeframe was echoed by a detailed report from July 2003. In that report, Apple and IBM were said to be working on parallel development of the Power5 and PowerPC 980. That report also claimed that the 3GHz PowerMac would not arrive by way of the 970 -- which was said to top out at 2.6-2.8GHz, while the 980 was expected to start at 2.6-3GHz (from the July report).

A subsequent report noted that Apple was already given PowerPC 980 samples in October, and also made predictions about 90nm 970 based PowerMac updates in February at 2.5-2.8GHz. The also offered significant details regarding cache, SPEC scores, and bus multipliers on the as-yet-unannounced PowerPC 980.

Without any tangible releases from these rumors, it's hard to gauge their accuracy, though the information appears to be consistent.</Quote>

This was posted on November 20th of last year by Arn himself. As you can note IBM themselves stated that the 980 was the chip being developed alongside the POWER5. Nowhere has there been any real info about a 975 chip and now everyone is abuzz with new rumors.

A look at the release dates above does seem to fit: the expected release of a 90nm 970 in February was correct, but because of fabrication problems was only released at 2.0Ghz and only in the XServe. The 980 that was rumored to be released "quarter 3 of 2004" is also supposed to be 90nm and my guess is that IBM is probably having the same fabbing problems with this chip as it did with the 970FX. <B>IF</B> there is a "975" it would seem likely that this was a step back from the 980 and not likely to scale as high (perhaps a 130nm derivitive of the POWER5). This might be an interim step for the company until they can work out the problems they are having at 90nm.

I'm not a hugely technical computer person, so please tell me if my thinking is not logical. It just seems everyone is so caught up in the rumors that they forget about reality.

Now i'm going to Thailand to practice my cantonese withe the Thaiwanese people.

Did you not read the earlier posts in the thread? If you didn't, I made a post about this very issue - that the planned replacement for the PowerPC 970 was the PowerPC 980, not the 975.

 

[Soire]

God forbid...

What would happen if no G5s were released at WWDC? Or just as bad nothing was going to be shipped for months?

Flying Pigs?

The apocalypse?

Steve announces he's selling out to Bill Gates?

 

[mklos]

It is quite offputting.

Whereas Intel and AMD are very adroit in releasing detailed roadmaps indicating exactly what they're researching, what strategies they are adopting, and what performances they expect to attain (how, and when!) IBM, Motorola and Apple enjoy keeping their own consumers in the dark and indulging in all this cloak-and-dagger stuff.

Do they consider us consumers (and especially, "prosumers") so stupid that we can't be trusted with a clear view of the future? As if the fact that next year computers will be faster and cheaper than now is some sort of secret!

I'm sure Steve is kicking himself morally for "announcing" 3GHz machines by Summer 2004 - and consequentially looking at sales of only 60,000 PowerMacs. It's not the issue of imminent new machines - it's the inherent uncertainty about what is going on.

The PowerPC really needs to issue clear roadmaps because at the moment their collective behaviour is really pathetic. In order to make informed purchasing decisions I demand clear information.

Not to offend anybody, but the whole "Apple Rumours" phenomenon is merely an indication of how confused, unclear, and downright paranoid Apple is about letting anybody understand what they're up to. That is not becoming of a mature company. It reminds me of school-kids playing their CIA games with code-wheels and secret salutes.

Apple never discloses what its going to introduce. They've never done that, and never will do that. If they do then people will know that new products are coming out and nobody will buy the currently models. This makes it very hard for Apple and Apple Resellers to get rid of their current inventory of the old machines to make room for the new ones. I think its hard to predict where things will be down the road. Steve Jobs made a promise of 3 GHz and could be wishing he didn't make that promise. Not that I really care if he does fulfill that promise. He doesn't have any power over whether or not IBm can make a 3 GHz G5.

As far as the chip makers go...well both Motorola and IBM have announced where they want to be down the road. All you gotta do is look! IBM announced their potential roadmap I believe sometime last year, and Motorola just announced their new chip lineups a few weeks ago.

As far as rumors go...well its actually bad for Apple. Some ignorant people believe everything they read and get their hope up so high then when Apple only announces some of the rumors that are only partially correct, or doesn't announce anything that had to do with the rumors they get all pissed off and of course they blame Apple. Thats exactly what the rumor sites do for Apple. You just wait and watch after the WWDC Keynote how many people start bitching and complaining about Apple was supposed to announce this, or Apple didn't do this, or this didn't have this feature in it. So the rumor sites don't do anything but hurt Apple.

 

[wizard]

HI Arn;

Flights of fantasy it may be but something has to be up. The reason I say that is that the 970/FX simply isn't up to the task. That is from the standpoint of being a chip one would want in their flagship model its time has already come and passed.

The rumored chip addresses on of the 970's biggest problems, especially as clock rates increase, and that is its very limited cache size. Hopefully other issues are addressed also, but as clock rate starts to our strip memory again cache becomes very important.

That doesn't mean though that the 975 is exactly as rumored, it could be an upgraded 970 instead of a down graded Power5 device. Either way the chip would have to address performance scaling at higher clock rates. Otherwise what is the sense.

Dave

if anyone has/digs up any official IBM docuemntation that the 975 exists... let me know.

This thing keeps getting mentioned as if it's fact. I think it may just be entirely made up.

arn

 

[danbirchall]

I'm sure Steve is kicking himself morally for "announcing" 3GHz machines by Summer 2004 - and consequentially looking at sales of only 60,000 PowerMacs.

So you're saying that 40,000 of the 100,000 pre-orders they had by the time the G5's started shipping late last summer were withdrawn?
Or did you, perhaps, mean 600,000?

 

[dongmin]

Did you not read the earlier posts in the thread? If you didn't, I made a post about this very issue - that the planned replacement for the PowerPC 970 was the PowerPC 980, not the 975.

confirmed:
970
970fx
power 5

unconfirmed:
975
976
980
990
etc.

It's true that the rumors were calling the power 5 derivative the 980 before. But at this point, the 980 is no more real than the 975. They're both pure guesswork.

 

[SeaFox]

You're right about the price drops with the new models, I just wish they would lower prices as the models aged, and before the other models were released, so there would be some incentive to buying a 9 month old computer *before* the new models come out.

- reaper

There is. It's all those rebates/bundled offers with iPods and printers and Cinema Displays that appear in the month or two prior to an update.

 

[soosy]

Apple never discloses what its going to introduce. They've never done that, and never will do that. If they do then people will know that new products are coming out and nobody will buy the currently models. This makes it very hard for Apple and Apple Resellers to get rid of their current inventory of the old machines to make room for the new ones.

That may be true but you have to wonder about other companies who are more open about what is coming up around the corner. Do they suffer this problem? I'm wondering about the Dells/Gateways/etc. There may be a factor of scale that makes the difference, but maybe not.

A LOT of people (possibly the great majority) don't follow the rumor sites to see what is coming up. And even if Apple pre-announced products or gave roadmap indications, I'd have to think those same people still wouldn't be aware and just buy machines when they needed them. I guess the question is, how big of a user segment follows Apple so closely and demands only the latest and greatest?

Another angle is how many people would have bought machines during months and months of waiting for new models, if they just new that nothing was coming out for a while?

On the other hand, when Apple does "preannounce" a new product that won't be shipping for a few months, there is a segment of people who complain to no end... and I suppose those who do keep up on Apple would definitely wait for the updates.

 

[AidenShaw]

Just ask Intel - roadmap is on the main page

That may be true but you have to wonder about other companies who are more open about what is coming up around the corner. Do they suffer this problem? I'm wondering about the Dells/Gateways/etc.

Don't ask Dell or Gateway, Intel puts the roadmap on their website.

Go to:

http://www.intel.com/products/desktop/processors/index.htm?iid=ipp_desk+proc&

and click on the Processor Roadmap link.... You'll see the planned roadmap for the next 6-12 months.
---
IMO, when people see that something is 6 months or more away, they'll buy today if they need something today. If the upgrade is one month away, they'll be more likely to wait.

With Apple, there's always a concern that it's only one month away, so they wait....

 

[wms121]

..I still want a Moto 7540

http://e-www.motorola.com/files/netcomm/doc/fact_sheet/MPC8540FACT.pdf



...but if you guys find that IBM page.

ww ; )

 

[wrldwzrd89]

confirmed:
970
970fx
power 5

unconfirmed:
975
976
980
990
etc.

It's true that the rumors were calling the power 5 derivative the 980 before. But at this point, the 980 is no more real than the 975. They're both pure guesswork.

OK - point taken. I still claim the validity of my point, which is that the PPC 980 was rumored to replace the PPC 970 before the rumored PPC 975 came into the picture. (keyword: rumored)

 

[Sun Baked]

OK - point taken. I still claim the validity of my point, which is that the PPC 980 was rumored to replace the PPC 970 before the rumored PPC 975 came into the picture. (keyword: rumored)

But it's a little hard for me to check back on one of the publishers of that roadmap rumor, since MacOSRumors isn't responding.

 

[justytylor]

The little dirty secret of corporate computer purchases is that you don't wait for the Next Great Thing to buy - you buy when you need new machines, no sooner and no later.

So what's preventing you from making corporate orders of Apple products then? There's obviously something new "always" on the horizon (sometimes the horizon is further away). Get what you need: while the criticisms of the 970 above may be valid, the G5s are very good machines, particularly when you get into the dualies. I use a 1.6 GHz G5 at work and I'm very pleased with it: it chugs through an hour of digital video in 4 hours (dual pass); my 667 MHz G4 would take at least 24 hours, and I'm sure my G5 would outperform my new 1.2GHz Aluminum Powerbook, so I don't even bother encoding on that.

 

[kenaustus]

What ever is announced at the WWDC is not going to be a last minute decision on Apple's part. They would have been working with IBM to make a powerful (and dependable) announcement at the WWDC - with discussions STARTING big time when the 970fx problems surfaced in Dec/Jan. I also believe that when both Steve and IBM said 3 gigs in a year they were fairly well along with prototypes.

The interesting issue now is what IBM is going to do now and over the next 6 to 12 months. Apple's going to get some good pricing to make up for this year's lost sales and probably some very interesting offers to bring the lower end models to a "G5" level.

The word from on high at IBM will be to get the job done, even without Steve's prodding - and I'll bet that Steve has been prodding just a little bit . . .

 

[Mord]

I have been skeptical about the existence of the 975 since the first rumours came up. IBM should solve the yield issue with the 970FX before developing an entirely new processor. But of the course the question remains which cpu will be put in the new PM revision.

they have already desighned the ppc 980 it's not about the desighn it's about the size of the die and the heat and the yeild

 

[melgross]

IBM info

This is from IBM's site. I wish it had held the formatting. I tried to manually lay it out. If you have enough horiz. space it might work. You can read about the speeds, again-beyond 2.0GHz-not explicitly stated as 3.0GHz. Also of interest further down in the specs is the junction temps-to 85C for the 970, but 105C for the FX. That might tell us something about top achievable speeds. So far, all of the analysis done by the engineers at Microprocessor Report and others have agreed that the 970FX won't get to 3.0GHz. Take it from there. Also, I checked out IBM's entire site. No mention of the 975, 976, or 980.

PowerPC 970 & 970FX Processors Highlights • Based on the award-winning POWER4TM architecture • Manufactured in IBM’s 90 nm (970FX) and in 0.13 ?m (970) copper process technologies with lower power and small die sizes • Products can range from1.4 to beyond 2.0GHz for high performance computing, networking, storage, workstation and entry-level server systems • High-superscalar 64-bit design, with dual fixed point, dual floating point, and dual load/store units with 64-bit data paths, along with AltiVec* SIMD units with 128-bit data paths • 32-bit compatibility allows continued use of legacy operating systems and applications and offers the ability to migrate to 64-bit at your own pace • Provides 512KB of internal L2 cache with ECC • Achieves high system performance with up to 1.1GHz processor bus, with dedicated read and write data paths, to avoid performance bottlenecks
PowerPC 970 & 970FX Processors
970 970FX Frequency
1.4 – 2.0GHz 1.4 – 2.0+GHz
Performance 1.8GHz 2.0GHz
SPECint2000 828 (est) 890 (est)
SPECfp2000 1036 (est) 1100 (est)
Dhrystone 2.1 MIPS 5800 7584
Typical power 51W@1.8GHz 12.3W@1.4GHz
24.5W@2.0GHz
Voltage (logic/ I/O) 1.3V / 1.3V 1.0V / 1.3V
Junction temp. range 0o to 85oC 0o to 105oC
L1 Caches 64KB instr. w/par 64KB instr. w/par
32KB data w/par 32KB data w/par
L2 Cache Internal 512KB w/ECC Internal 512KB w/ECC
Bus Interface 1.0GHz 1.1GHz
Dual unidirectional Dual unidirectional
32-bit logical busses 32-bit logical busses
w/ECC w/ECC
Package (mm x mm) 576-CBGA (25x25) 576-CBGA (25x25)

 

[ingenious]

Check this out:

http://www.theregister.co.uk/2004/05/26/ibm_ppc_975/

They're saying yes, the 975 and the 976 are real.

Isn't Google wonderful?

 

[Sun Baked]

Check this out:

http://www.theregister.co.uk/2004/05/26/ibm_ppc_975/

They're saying yes, the 975 and the 976 are real.

Isn't Google wonderful?

Nope...

Try reading down a bit where they say the IBM document comes from a local paper.

Or do they? We found a reference to the CPU on IBM Taiwan's web site, but closer inspection reveals it to have been taken from a local publication, IT Home.

So there is no proof, at least from IBM.

 

[LauShu]

Taiwanese & Chinese

I was directed to this forum and thread by an American friend of mine who is very interested in Chinese culture. I am from Taiwan. I am also a dedicated Mac users. I know MacRumors for long, but this is my first time posting because the discussion has become unbelievable.

So here is the deal.

First of all there are three locations for discussion: China, Hong Kong, and Taiwan.

The language Chinese does have two prominent forms, one called Mandarin, and one called Cantonese. As someone pointed out, the writing is the same, but the speaking is different. To be more accurate Cantonese is a dialect, just like there is a dialect in Taiwan called, Taiwanese. There are hundreds of dialects all over China. What's different about Cantonese though, is that they use the Chinese characters to read and write, unlike most other dialects, they are for speaking purpose only. But, Mandarin IS the language that uses the Chinese characters to read write and speak. The catch is that, as time go on, people try to write Taiwanese, so there are words invented, or phrases written to immitate the sounds of Taiwanese, just like there are characters that people from Hong Kong write is not in the real Mandarin's character set. One thing to note is the writing. There are two types of writing, Traditionally and Simplified. In China, Singapore, and Malaysia, they all use Simplified. In Taiwan, we use traditional, and I believe so do people in Hong Kong. But this is just the sake of the writing of the characters, the language is the still the same.

So Who does what?
Almost all Chinese from China speak Mandarin.
All Chinese from Hong Kong speak Cantonese, and some speak Mandarin.
All Chinese from Taiwan speak Mandarin, and some speak Taiwanese.
We all speak Chinese, the language.

Cantonese is also spoken at the south region of China, as you may see from a map, that's where Hong Kong is near at.

There is nothing to do with Thai the language and Thailand. It's a totally different race though some of them are Chinese heritage which might speak a little Chinese.

Many Chinese heritage from Malaysia and Singapore also speak Chinese, though the official language in Malaysia is Malay, and English for Singapore.

 

[jakemikey]

let's stick to the topic please

As interesting as that last post about all the ins and outs of the dialects and languages of the Asian continent was, I think it shows that this thread is quickly running out of steam. Come on, people!

 

[Rower_CPU]

As interesting as that last post about all the ins and outs of the dialects and languages of the Asian continent was, I think it shows that this thread is quickly running out of steam. Come on, people!

Agreed. Let's get back to and stay on the topic, folks.

 

[makkystyle]

Did you not read the earlier posts in the thread? If you didn't, I made a post about this very issue - that the planned replacement for the PowerPC 970 was the PowerPC 980, not the 975.

I did read them and my point was an agreement with yours and to provide a reference to material that substantiated this.

 

[makkystyle]

Here is an interesting link. It leads to a PDF in french that mentions IBM using "PowerPC 970 at 3Ghz" in their Blade servers. If someone out there reads french and can please give us a rundown on what this says that would be great. I can't vouch for the validity of this article, because I have no idea where it came from (the whole not being able to speak french thing). Also it doesn't lend itself to Sherlock translation very well because of the format. Hope someone can shed some light:

http://www.irisa.fr/orap/Publications/Bi-orap/biorap-35.pdf

 

[Telomar]

Did you not read the earlier posts in the thread? If you didn't, I made a post about this very issue - that the planned replacement for the PowerPC 970 was the PowerPC 980, not the 975.

The 980 was always an invention of the rumour boards for what would follow the 970. The 975 comes more out of IBM roadmaps that showed a 97x processor I suspect.