Intel rolls out Broadwater, says 3.2GHz Woodcrest planned - Page 3

Very impressive. A woodcrest workstation will be awesome. If Apple don't release a MacPro with woodcrest there will be some bitchin.

Whenever Apple release anything short of awesome, there will be some bitchin!

I know, I'm one of them!

Woodcrest here we come!

You wont get any arguments out of me. I've already started selling my comic book collection, some guitars, and anything else I can do away with so I can get the highest performance model available from day one. Apple better have SLI cause I'm going dual Quadro baby!

That would be nice.
In an Apple store you could go to the genius bar and ask an asian cock how to handle Apple's new wokstation or about a new recipe...

Should have said " That would be rice" .

And with that, the thread strayed back on topic.


WOOOOOo! WOODCREST MACS! I WANT ONE!

Intel's own motherboards are designed by Intel and produced by Foxconn. God knows who makes Apple's motherboards but I assume it's the companies which are the ODM's for Apple's computers.

I agree if by that you mean the G5 from last year vs Core 2 coming soon but that's not to say there wouldn't have been changes to PPC in the last year too.

The G5 has it's faults but then so does Core 2.


No, that's not really true either. The problem with the perception of the G4 is it stopped with the 7447A. Apple never used the 7448 or the 8641D. The 8641D in particular is pretty good (dual core, lower power than Core, built in dual memory controllers, 667Mhz bus, PCI-E, Gig Ethernet on the CPU..) but it shipped too late. If it shipped a year ago I think we'd not be going through the switch.

I think the switch is more a matter of timing than anything. Apple needed laptop chips desperately and PowerPC chips just weren't ready. I think it's a real pity as there's some real work going on around PPC just now in IBM and PA Semi. Maybe Apple will get back to PPC someday.

I also think there's a software plan here that's not apparent yet.

I don't think anything would have made G5/G4 development proceed as quickly as Intel's: neither IBM nor Freescale want to be in the personal computer processor business, and neither of them finds it profitable to do so just for Apple. So that's pretty much that--years of experience showed us that PPC had promise, but Apple could NOT make IBM/Freescale turn that promise into reality. Not consistently and on time and without shortages.

Reality means on time, not a year late, and it means shipping in QUANTITY, which is a huge question mark with those newer G4s.

So a nice new G4 a year too late or in small volumes isn't good enough. And if it was, then what comes after that? Another promise that is likely to be a missed target.

Meanwhile Intel has abandoned the Netburst mistake and has an excellent new direction. The choice was easy, as long as Apple could get their OS on x86. Which, thankfully, is a safety net they had ready.

PowerPC lives on, and I'm glad of that--but the new developments are in chips for game consoles, phones, servers... everything but computers. And those specialized chips just can't take the place of a personal computer processor in a desktop or (especially) laptop. Cell, for instance, is really cool but is not useful to take the place of a computer CPU.

But that's not to say there *would* have been changes. What the G5 could or could not have been is pure speculation...

...on the other hand, Intel *is* delivering in a tangible manner. And if you're gonna make me believe that IBM or Motorola/Freescale would suddenly have gotten off their duffs to actually deliver decent processors improvements, you're going to have to try much harder.

Ever since the switch, we've been bombarded with Intel chip news. They have been following their roadmap pretty well...too well even that chips are being released earlier then even Intel expected. How the fuck can you ask for more? Heck, I'm even sick and tired of hearing about Woodcrests beating all other chips by a factor of almost 2...and Merom being speedy and consuming 35 watts... (ok, I'm not really sick of hearing it, but it's a massive change to how everything was just a year ago.)

I'll tell you one thing... had Steve decided to stick with IBM another year, we'd have no news on chip dev, and we'd sure as hell not have some ass-spankin' laptops right now (despite some minor heat issues).

I'm not saying the switch to Intel wasn't a good idea. Far from it, Intel are delivering well enough and I think the Core Duo iMac is the best thing Apple have ever produced. The problem I have is with the bleak picture that gets painted about PowerPC when it really wasn't the case (bar laptops) and isn't now. Occasionally the PowerPC gave the Mac a lead and you must now realise that'll never be the case again. It's all about the software now.

Plus everyone likes to apportion blame and usually it's not Apple that's the bad guy. That's not what I've heard from IBM engineers.

I just find it hilariously funny that we've gone from dissing the Intel gang but as soon as we've moved gangs it's now fashionable to bash our old once cutting edge friends.

Fully agree. I can't understand this yearning for ppc chips. Intel is keeping their end of the bargan so far. They are ahead of schedule in bringing chips to market. What else do they have to do to get a little respect around here? And if Intel begins to falter, AMD is only a phone call away.

Actually, there are lots of people that were bashing "our cutting edge friends" when Intel wasn't in the picture at all. I repeatedly bashed Motorola during the G4 days. I repeatedly bashed IBM during the G5 days. This isn't some magical and recent phenomenon. People were unhappy with Motorola/Freescale and IBM before any of the switches (be it to G5s or be it to Core Duo.)

And you're not going to hear the unbiased story from an IBM engineer...just like you won't hear it from Apple employees.

But, in my wise and humble opinion, I've seen Apple work its ass off offering cutesy designs to make up for the sorry fact that they to use some mediocre to average chips by Freescale and IBM (respectively). If they packed computers with value-added gizmos such as Bluetooth and Airport and made nice enclosures, it was to make customers not concentrate on the processors.

The fault may have lied in part on Apple...but I'm pretty sure the majority of the fault lied on the suppliers dicking around.

In the pre-Intel days, I heard a lot of bashing of Netburst/Pentium, but no bashing of Core or Core 2 (which obviously didn't exist back then).

I'm still willing to bash Netburst. And I'm in good company: Apple and Intel have both rejected Netburst.

So I don't see any "fashion change" of Apple changing their minds about Netburst, nor Mac users. I see only one party changing their minds about Netburst: Intel. I'm glad they did.

As for bashing Freescale and IBM, I think all Mac users appreciate that PPC had potential, and that for a while that potential was actually realized. But Mac users badmouthing IBM and Freescale is nothing new at all. It began with the delay of the first iMac G5s and with the Pentium M vs. G4 comparisons. (Pentium M is not Netburst--it's a predecessor of Core and was the one execllent chip Intel had for a while there.)

A few points:

1. TomsHardware and AnandTech have demonstrated 20% or so increase in performance in their benchmarks with Conroe over AthlonX2.

2. I agree iMac goes Conroe, Mac Pro goes pure Woodcrest (dual and quad models). Redesigns to accomodate Mac Pro Conroe is too messy this year. That means NO mid-range tower from Apple. NO MID RANGE HEADLESS CONROE TOWER IMO. Too messy, too much for Apple to handle this year.

3. I agree MacBookPro goes Merom, sticks with ATI mobility graphics. MacBook remains faster Yonah, maybe better integrated graphics. MacBook maybe goes Merom towards end of the year.

4. MacBook Pro and MacBook HAVE HEAT ISSUES. Some of you seem to be thrilled about the Intel transition. I am too, but there is no denying that the heat issues with the brand new laptops are really not something Steve wanted with the whole performance-per-watt faster, cooler, better deal. THESE HEAT ISSUES NEED TO BE ADDRESSED and I'm sure that Apple is working on it.

5. There will be no Jonathan Ives redesign until 2007 IMO. Apple is too hard at work with the internals and dealing with the fast rate of Intel chips that major physical appearance stuff will have to wait until 2007.

6. Netburst is not dead for PC-land. Intel is pushing out Pentium 4s, Pentium Ds, and Pentium Extremes at low, low prices. Conroe availability is expected at 30% or less in the overall chip availability from Intel.

Intel Pentium D 820 (2.80GHz/1Mx2) \t$133
Intel Pentium D 805 (2.66GHz/1Mx2) \t$93

For about a hundred plus you can get a dualcore machine, be it Netburst. But Netburst WILL still be around this year. A pity for the PC users but with the low prices they become very competitive with Athlons.


Just some thoughts and predictions......
.......................................

Originally posted by smalM
That would be nice. In an Apple store you could go to the genius bar and ask an asian cock how to handle Apple's new wokstation or about a new recipe...

Umm, typo...

Ever since the original iMac, Apple has focused on high margins and slow growth for the Mac market.

But, PowerPC chips were so cheap, IBM/Moto saw almost none of the huge profits Apple was taking home. And to make matters worse, Apple played the ultra-segementation game meaning most of their volume was extremely low-profit low-end chips (think how long the G3 hung around).

If Apple had grown their marketshare to 5% and put higher-end chips in their consumer systems, you can bet that those suppliers would have stop dicking around and got Apple some better chips. As it was they were simply a lousy customer.

With Intel, Apple wanted started with a bang -- using only the highest-end, best performing laptop chips. This is a total reversal from their PPC product planning, and frankly I wouldn't expect it to last.

By next year, count on Apple's consumer systems lagging behind with Intel's cheaper, slower chips. Sure there will be a couple speed-queen models in the line, but for the most part Apple doesn't need to use Mhz to make large margins, preferring instead to rely on model segmentation and form-factors.

That's probably a business decision, not a resource decision. The circuit design boys being busy doesn't necessarily mean that the fluffy-shirt type design boys are off designing circuits to fill the need manpower.


Later in the game, Netburst did get a little power inefficient but they do the job and are reliable. It did hold its edge for half its life, and in media handling for a while longer. They got a little carried away with Prescott, Nacona and later, I never had those, but the Northwood and Prestonia has served me without issue.

You forget all the things IBM and Freescale promised or anticipated that never happened. IBM didn't say "3 Ghz in a year IF you put high-end chips in your low-end systems." They said "3 Ghz in a year."

And what higher-end chips were you thinking would go in the iMac (which was G5) and the iBook (which was G4)? I can only assume you are criticizing the Mac Mini for not having a G5--but that was both a cost and space issue, and even if it had used a G5, I don't see how that would magically have made IBM and Freescale "stop dicking around."

Apple wasn't a "lousy customer," their needs simply didn't match what IBM and Freescale want to do, which is embedded, server, and console chips, not computer chips. Apple was the wrong customer for them, and the right customer for Intel who actually WANTS to make computer chips.

>...ass-spankin' laptops... despite some minor heat issues...

Not issues - features.
The heat is actually what makes the MacBook a WOKstation (as previously mentioned in this thread).

It's quite simple. The PPC was and in some respects still is a better design and if you tie your ship up with Intel in particular (AMD less so) there's no way the Mac is ever going to be faster or more architecturally interesting than a PC. It's lost one of it's differentiating factors.

It's all about software now. The hardware wars are over. PowerPC lost on the desktop.

The iMac G5 delay wasn't IBM's fault but that's who got the blame from the Apple fanboys.

Badmouthing Freescale started way way before they were even called Freescale, back as far as the G4 introduction and all the way through the cancelled Moto G5 project. Actually, probably as far back as the 68060!

You can blame IBM or Freescale all you like but ultimate responsibility for not delivering lies with Apple. They can't just sit back and hope IBM or Freescale produces a useful chip to beat Intel/AMD, they have to make sure it happens. There's a lot of history between Apple, IBM and Motorola and a lot of it bad and caused by Apple. I'm not surprised IBM or Freescales attitude to Apple starts with 'Show me the money'.

Going Intel sidesteps the chip wars by throwing in the towel. I don't mean that in a bad way. Everyone has to know when to quit and when to pick a new fight and that's what Apple have done. That doesn't make losing a good, unique architecture any less sad.

Actually, IBM didn't say that. Steve Jobs did.

Originally posted by aegisdesign
It's quite simple. The PPC was and in some respects still is a better design and if you tie your ship up with Intel in particular (AMD less so) there's no way the Mac is ever going to be faster or more architecturally interesting than a PC. It's lost one of it's differentiating factors...It's all about software now. The hardware wars are over. PowerPC lost on the desktop.


Ah, true words. However, for 99% of the marketing that drove Apple's profits and revenues in the PowerPC era, it wasn't because of "PowerPC inside". We'd like to think that the Pentium-on-a-snail ads and toasting-bunny-suit-people were really effective, but that would be a fantasy IMO.

For the average Mac buyer, PPC was just part of the Mac and they accepted it as it was, most of the time, regardless of how much they knew about PPC chips compared to x86. I don't think they were like OMFG PPC w0000t!!!

However I will venture that the difference now with Intel is Intel becomes a key selling point. It helps bridge the gap between "What is this weird Mac thing" to "Oh okay, it's like a PC but no viruses and looks cool...". Intel inside helps reduce the alienation of the potential buyer.

I mourn the PowerPC on the desktop but from a marketing standpoint, PPC-inside contributed very little and now Intel-inside contributes something.

Originally posted by aegisdesign
...That doesn't make losing a good, unique architecture any less sad...


I guess that's a personal thing*. There are a lot of other architectures out there doing lots of cool stuff - Pentium M (it was/is decent), Core, Core 2, Athlon, Opteron, PS2, PS3, Xbox, Xbox360.

*For me owning my very own G5 to produce trance music (Reason, Rewire, Ableton Live on software, M-Audio hardware) gave me a lot of joy. The people that make Reason (http://www.propellerheads.se/) coded a fantastic music application that was really quite complex but remarkably stable. My single 1.8ghz g5 blew away the dual 1.something g4 I had previously when handling synth tracks, sound processing, etc.

Reason is now a Universal application. Will the iMac Core Duo serve musicians now just as the single or dual G5 did? For studios that really want the power would they be on the dualcore G5s or quad G5s and waiting for Conroe 2 to catch up?

I guess I'm starting to ramble, maybe the point I'm trying to make is that all the tech sentimental attachments aside, for a lot of the public, Mac users are oblivious to the massive efforts put into the hardware and software engineering. And the creative professionals, for them, they will use what is best, affordable, and comfortable for the work they produce.

G3, G4, G5, Cell(???), Core -- all will just be footnotes in history as we move on to the end of this decade.

Actually they both said it. We had this debate a year after, and there was a link posted to the original broadcast, and many of us watched it. They both said it.

More details on the Broadwater chipset...

http://www.reghardware.co.uk/2006/06..._hd_with_g965/

That's one of the areas where the PowerPC will be missed. AltiVec is head and shoulders better than Intel's SSE and that difference pushed the Mac into areas PCs couldn't reach.

Back when the G4 came out I was on a mailing list for BeOS and we had audio engineers saying they were getting 10 times speed ups in their code because of AltiVec. Then again, we had ex-Moto engineers saying how screwed up Austin was.

Core 2 however should be quite capable still as it's now got 128bit SSE. It's the one thing that really cheered me up about the switch - finally a decent vector unit. Took them long enough.

Yes, I'm quite sure the Altivec of the G5 was very well utilised by Propellerhead's Reason. Made handling all those synth tracks really amazing.

(Ahh..! With the whole music side of my explorations 2001-2004 just using a G4/G5 Mac and Reason was a breath of fresh air from the late 90's trying to pull off stuff using Creative's AWE32 http://en.wikipedia.org/wiki/Sound_Blaster_AWE32)

Not if we're talking about a chicken stir fry!

Ohhhh yes, of course...Apple had a change of heart when they got a change of processors. Do you really believe that horseshit?

You really think Apple was like "yeah, let's keep the G3 in the low-end for just another year?"

Time will tell. Frankly if they attempt that strategy sales will stagnate severely IMO. Apple already has a premium price on many of their products, using stale components in their models would be very detrimental to sales. When Apple switched to Intel, they bought into the Intel product cycle. This cycle moves rapidly and Apple can keep up or fall behind. With the margins Apple has they can afford to keep up.

The problems with PPC development started shortly after the G3 was released and Apple pulled the plug on licensing the OS. Before that IBM and Moto were working on building a new platform which could run multiple OS's, the Mac OS being just one but there was also a version of Windows NT for the PP platform which was pulled at about the same time. IBM who was the chief backer of the CHRP saw broad support in their efforts in this development fall apart. This was followed up by IBM and Moto splitting on the direction to move with the next generation chip, the G4, which IBM wanted to eventually take into a dual core chip and Moto wanted to add Altivec circutry. The chip couldn't handle both and with the "Death" of the CHRP there wasn't a prospective large enough market for two chip designs. This brought an end to the joint development of the PPC and the speed of that development that was enjoyed in it's earlier days slowed to a crawl. I think that this would have been different if there was a larger market for the PPC in desktop computers and servers but there wasn't a viable OS other than Apple's, and Apple didn't have enough market share to sustain the development costs of both Motorolla and IBM that were needed to keep the PPC desktop chip development up with Intel and AMD's competing pace for their chips.

I agree. I would add that I think Apple will adopt new harware technologies quicker than pc vendors but this will not make a huge difference in the hardware between macs and pcs. But honestly, do you think Apple hardware was significantly better, with respect to cpus, for a significant period of time? Seems to me the advantages were few and fleeting.

Apple can't really fall behind...when Intel stops manufacturing a chip, it really does stop*...this is very different from Freescale and IBM which can afford to keep manufacturing the same chip for years (different types of customers in the the server, game console, embedded business.)

*to some extent.

Originally posted by kim kap sol
Apple can't really fall behind...when Intel stops manufacturing a chip, it really does stop*...this is very different from Freescale and IBM which can afford to keep manufacturing the same chip for years (different types of customers in the the server, game console, embedded business.) *to some extent.


I think Apple has pretty much secured via contracts whatever G4s? and G5s it needs to complete the transition. They can't afford to not have contracts in place to secure those chips. While Freescale and IBM might still make those chips, Freescale has moved on to releasing newer models and IBM too........

Yeah, I think Apple will keep pace with Intel. However there is some possibility they might hold back slightly from the rest of the PC world once ALL apps especially macro/adobe/office go Universal... That way they get extra profit boosts off Intel's tendency to deeply discount CPUs on their way out... (look at the prices for pentium Ds and you'll see what I mean).

Apple's got a lot of room to play with now with Intel's offerings. Just FIX THE BLOODY PORTABLE FRYING PANS* first!

*MacBook Pro, MacBook

Yes, they certainly were.

2001, Apple's laptops were thin, cool and the batteries lasted 4+ hours. Nothing to touch them until the Pentium M. You could get faster laptops with Intel but not thin, light ones. Speed weenies are blind to the principal reason by laptops often though - that they're portable.

About the same time the G4 was faster than the Pentium for a while and Apple were pushing dual processors in desktops too which shows in apps on OSX compared to Windows where developers don't have the same mindset.

When the G5 came out it was faster than anything by Intel/AMD for the money and the Quad is still faster for most apps than a dual Opteron/Xeon for about the same money. It had huge bandwidth then and killer FP performance.

Sure, you can build overclocked AMD monsters that are faster and cheaper but generally from the Tier 1 manufacturers, Apple has done ok on launch of new lines, only to be caught up later. The problem is generally that they don't update often enough so keep getting caught up and passed.

That's my point.

Apple has a strategy to take advantage of this, it goes under the product names Mac Mini, MacBook, and eMac. The MacBook Pro, iMac, and MacPro's need to keep pace with the rest of the Intel market or Apple will "loose Face" which will not look good when they are trying to gain market share. In fact, I would be willing to bet that Steve will be trying to get Intel to announce one of their chips on stage with him at WWDC or in SanFrancisco, or at least be on stage when a recently released chip is introduced in a Mac. Wouldn't it be great if the timing were right for Intel to join Jobs on stage to formally announce that Woodrest (or Morem or Conroe) is shipping and the MacPro (replace with appropriate model) is the first publicly announced computer to use the chip and will be shipping within X time frame?

The iMac followed the PowerMac by more than a year -- practically an enternity relative to the PC world -- and even then it only had the slowest, cheapest G5 CPUs in them. How long were G3s used simply for segmentation purposes?

My point is really that Apple will likely revert to this style of behavior after the Intel transition is complete. Their model segmentation gaps are so large that they really have no incentive to push newer, faster technology down to the consumer lines very quickly.


Nice that you deftly ignored the argument, but having higher volumes is not "magic", and if Apple had higher volumes, there would have been more interest in making PowerPC chips for them. Seems pretty non-controversial to me.


That's what they did, didn't they? So they must have thought it.


True, but there's usually a roughly equivilant replacement. Three years ago I had a brand new 1.5Ghz Pentium-M laptop, and similar speed 1.5Ghz Core Solo is still being sold in the Mini.

@homenow -- your description is right on. Truth is that PowerPC was a deadman walking as a personal computer chip after the first generation, and with hindsight, it was a mistake for Apple to adopt it. They should have gone right to Intel in 1994.