Convinced you of what? That even though it isn't more cost effective for you, that you should spend the money anyway? Or that you aren't a professional and should just buy a lower margin iMac because Apple thinks so little of you that they won't even supply the most basic tower level expansion when they finally get around to making a decent performer.
How does everyone know the Dual is the best selling config? If it is, then that means that the high end pros you talk about are the only people buying the machines (and that they won't bother with the 1.6 and 1.8. But the 1.6 and 1.8 are well expensive, too expensive in fact for non pros, so that kinda strands them without a market.
Apple MUST either raise the relative performance of ALL its towers, if it wants to steadily sell all three models, or it has to drop the price of the single 1.6 and 1.8 level machines. The bottom tier has to reach down to 1500 to cover a wider market, and the middle tier down to 2200, if they've no intention of adding duals.
As you all are so fond of saying, cost is no object "pros" will buy the top config anyway, so if the bottom two machines become better deals, thats no harm to sales of the top tier, it just reaches people who would NEVER buy the dual for cost reasons but who would buy the singles if they represented better value.
Likewise, since ONLY a select cast of "pros" who meet AI fanboy criteria are supposed to buy PM's, the other tact, adding more dual configs at the same prices can divulge two ways. "Pros" will continue to buy the top tier only, and lesser dual configs will not harm sales, or the top tier will sell less because even pros like to save some money if they can and 1999 duals will be "good enough."
Oh but nothing is ever "good enough" for a real pro.
Maybe,
But you've all just finished shouting how "real pros" buy new equipment every year anyway and they'll take that expense/depreciation write off over just saving the money in the first place. What none of you seem to understand is that "real pros" will certainly take the write off when they have the expense, but they've no need (and more wisdom) that to just create the expense for the sake of the write off. "Real pros" can make their write offs on a host of other expenses that before they ever get to the computers that sit on their desks. At the end of the day the goal is to spend as little as can possibly get the job done and try to figure out how to spend even less.
But lest get back to the Ai conception of a "real pro"
When the DP 867 was released it was actually a decent value for a few months. People snapped them up. Must have hurt DP1.25 sales because Apple promptly dropped the dual "good" config at the next update. "Pros" is appears, like to save money.
The only other conclusion is that NOT JUST PROS, or more precisely, not just the narrow AI conception of "real pros", buy Powermacs, we should probably call these people "AI pros" but anyway... if the PM market stretches beyond just a particular type of pro, then it makes sense either to broaden the appeal of the machine (including the drive issue) or to add options and tiers.
By popular consensus, the two bottom tiers are stranded in the current Pm line.
Which is why, going all the way back before the forced digression into storage foisted on this thread by hmurchinson's pre-emptive fanboy reaction , I mentioned storage in the first place -- differentiation.
Have your products reach more people.
to that end:
SP1.8-2 Ghz
DP 1.6
Dp 2.0
Or, you get the picture, use something fast but single in the low end, use something slower but DUAL in the middle, and use FAST and DUAL in the top. This way, there's no "Dual at 50% off!" like there was with the DP867, and the middle dual, while 500 cheaper, is not so close to the top that it convinces otherwise top end buyers to drop down a rung. A DP 1.8 is awfully close to a Dual 2Ghz for effective marketting, too close. Tweak the internal layout of all these machines (an ABSOLUTELY MINIMAL modification) without increasing the exterior dimensions (MORE THAN POSSIBLE) to accept more drives.
Then, in the place of the current PMG4 introduce a "cubish" product that takes RAM, CPU daughtercard, and AGP expansion, and has the same I/O as the PM's, but costs less and uses a G5.
1299-1599 Cube G5 (not a "cube" per se, just something in that range)
with slower singles, SP 1.6, and lacking the HDD and PCI expansion of the PMG5.
1999-2999, PM G5, Single, dual, dual. Better expansion and faster/dual configs.
Like so.
Cube G5 1.6 1299-1599
PM G5 2GhzSP 1999
PM G5 dual 1.6 2499
PM G5 dual 2.0 2999.
Nice even steps, lotsa selection, appropriate flexible options and performance for everybody and minimized "cannibalization" though I hate that term.
And don't force me to lay the smack down on you by mentioning the iMac.
The top iMac is so grossly overpriced that it must immediately drop to 1399 just to pretend to be in a consumer price range.
PS, it's obvious that ATM Apple is doing a little "demand control" by making the top end, and most expensive machine, appear to be the best value. Things should change by the time the G5 supply improves, after a quarter or two.
PS, it's obvious that ATM Apple is doing a little "demand control" by making the top end, and most expensive machine, appear to be the best value. Things should change by the time the G5 supply improves, after a quarter or two.
I don't agree with this part.
(I'll keep my mouth shut about how much of the rest of that I agree with, it is kinda disturbing )
IF Apple knew people were going to dive into the Dual 2.0 this way (Both feet first, cheering all the way), there's no way the pricing would be the way it is. Assuming that Apple has to buy 1.6's and 1.8's along with the 2.0 chips from IBM in whatever percentages they pass their tests at -> there's going to be a lot more 1.6s.
So _IF_ they knew in advance, they'd have to have some way of selling the 1.6's. Remember, these particular 1.6s will never be cool enough for a laptop -> they need to be sold nowish. Selling dual 1.6's would certainly move chips.
The only other things that make any sense would be: 1) IBM's yields are _much_ better than expected, or 2) Apple _isn't_ forced to buy the lower speed chips at a set ratio to the high speed chips.
I'd have Single 1.8, Dual 1.6, Dual 2.0 as my lineup in order of price.
Nobody have any clue of the current yields of the PPC 970. It's even possible that most of the chip produced are able to reach 2 ghz, if it's the case there is no real background for this rumor.
Nobody have any clue of the current yields of the PPC 970. It's even possible that most of the chip produced are able to reach 2 ghz, if it's the case there is no real background for this rumor.
Yep, could be that the 1.6 GHz are just 2GHz run at a lower voltage and overclocking the chip is easy but FSB makes it difficult (type of memory).
Nobody have any clue of the current yields of the PPC 970. It's even possible that most of the chip produced are able to reach 2 ghz, if it's the case there is no real background for this rumor.
That is why its a rumor and not a *confirmed* rumor
Of course *confirmed* rumors are far different then "The Truth", and then there is what apple announces being the *real confirmed in writing* truth which is no longer a rumor.
Convinced you of what? That even though it isn't more cost effective for you, that you should spend the money anyway? Or that you aren't a professional and should just buy a lower margin iMac because Apple thinks so little of you that they won't even supply the most basic tower level expansion when they finally get around to making a decent performer.
I'm a prosumer and will get the most cost-effective dual G5 avaliable (which for me will probably be a DP 1.8, if Apple chooses to release it), when I have the $$$. Maybe I should add, that even though I'm not needing the power of the G5 to get the work done, to sustain economic income, I still want one because I feel this dual 1GHz G4 is way too slow. And I would be surprised if I'm the only one.
Though I definetely desire more expansion, I realize that (that's what Ed M. convinced me of), given the resistance I get on the internal storage-issue, I will get nowhere by claiming that a sufficient amount of professional customers would gain from the extra expansion, to justify the possible added engineering and production cost, without having any evidence.
But there are more than professional customers, the question is: How many, and how's their need for extra expansion? There's a 97,5% marketshare (that's a group almost 40 times bigger than the current group of mac buyers) for Apple to catch (they say they want it), and that is a big target, with lots of different needs. Someone mentioned PC-fanboy sites, that could have led me into believing we, or I need extra expansion. Those sites are read by more than just PC-fanboys, even professionals, regular consumers - which means they too can be led into believing that they need the expansion. They will look at the newest, glossiest, greenest monster-PC, which features three times the expansion. If they're in the dilemma, G5 or glossy, hairy PC - it'll affect their decision. I don't know if that's a large group or not, but that's the group Apple can't catch with an iMac, but still targeting. Their weapon is the G5, they're getting alot closer, but is it enough?
Sure, I must precise that even I'm desiring more expansion for myself, I think the G5 is the most awesome computer ever built, it's not that, but Apple has to - if they want "everything", they need something that suits everyone. A DP 1.8 is a considerable step in the right direction, more bays will make it suit more people, unless it makes the case grow noticeably, but I don't think it really will.
Bah, all this blahblahing, even though I said I would stop talking about this issue which is a very small one, IMO, but almost the only one - hence it's what I have to "whine" about, if I'm feeling the need to complain about something.
</off topic>
Oh, and Matsu, I like the idea of a SP 1.8 or 2.0 low end. According to Apples own benchmarks (well, SPEC - ugh, and add altivec), a single 1.8 or 2.0GHz G5 is very close to the processing power the competition offers - 3.2Ghz P4. One SP model to face the competetition at the SP side of things, approximately the price range of comparably specced PC towers. Duals, even lower clocked than the SP tower, would and will take care of the higher-end segment of the market. I think most pros ( aren't they the only ones buying high-end towers? ) would understand that the 1.6DP is faster than the SP one, in their daily work.
I thought it was pretty good the days when there was a SP 867 G4 mid-end, and DP 800 G4's in the top end. But it wasn't perfect, since the mid-end had a faster processor than the top-end, which meant you didn't get the most technically possible processing power when buying the top-end, that there's more potential. Always though it was too bad they didn't offered the best of the best, but it was quite okay anyways.
SP 2.0 (or 1.8 ), DP 1.6, eventual DP 1.8 and finally a DP 2.0 would be perfect in this regard. Apple would sell more than they currently do, and they'd get a much more balanced demand. Now, by making a DP 1.8 they'd be in the right direction, but we won't for sure be seeing the other stuff.
Or, you get the picture, use something fast but single in the low end, use something slower but DUAL in the middle, and use FAST and DUAL in the top. This way, there's no "Dual at 50% off!" like there was with the DP867, and the middle dual, while 500 cheaper, is not so close to the top that it convinces otherwise top end buyers to drop down a rung. A DP 1.8 is awfully close to a Dual 2Ghz for effective marketting, too close. Tweak the internal layout of all these machines (an ABSOLUTELY MINIMAL modification) without increasing the exterior dimensions (MORE THAN POSSIBLE) to accept more drives.
Then, in the place of the current PMG4 introduce a "cubish" product that takes RAM, CPU daughtercard, and AGP expansion, and has the same I/O as the PM's, but costs less and uses a G5.
1299-1599 Cube G5 (not a "cube" per se, just something in that range)
with slower singles, SP 1.6, and lacking the HDD and PCI expansion of the PMG5.
1999-2999, PM G5, Single, dual, dual. Better expansion and faster/dual configs.
Like so.
Cube G5 1.6 1299-1599
PM G5 2GhzSP 1999
PM G5 dual 1.6 2499
PM G5 dual 2.0 2999.
Nice even steps, lotsa selection, appropriate flexible options and performance for everybody and minimized "cannibalization" though I hate that term.
... Things should change by the time the G5 supply improves, after a quarter or two.
I don't usually agree with Matsu, but your logic seems right on as far as these offerings. However, this logic is not based on any real information we have on costs. For instance, maybe it's very expensive to have memory systems at 1GHz, and it's much cheaper to have one at 800MHz. Maybe the dual processor motherboard is much more expensive than a SP. It could be that the cost of a DP 1.6 and a DP 2.0 is almost the same. It may be that the cost spread between the current SP 1.6 and SP 1.8 models is almost entirely due to the PCI difference and the memory bus speed difference, and not due to processor cost differences. Apple's margins on these may all be relatively low, and Apple may be betting on component costs going down for long term profits.
Next spring we'll have a better idea about costs - Apple will undoubtedly adjust price as all the lines migrate to the G5.
As far as storage, give it a rest. There's no one perfect solution, plus Apple has made mistakes before. Remember DVD drives but no CD-RW drives? It will work itself out. This isn't a make or break issue. Computers are more and more becoming commodities - buy what you need today and for next week. When you need that 3rd drive, won't you also want the DP G6?
Someone mentioned PC-fanboy sites, that could have led me into believing we, or I need extra expansion. Those sites are read by more than just PC-fanboys, even professionals, regular consumers - which means they too can be led into believing that they need the expansion. They will look at the newest, glossiest, greenest monster-PC, which features three times the expansion.
Expansion is not just hardrives and opticals. There is more than enough room for extra cards and RAM. If the lack of internal options for drives stops you from purchasing a G5 then you don't need the processing power.
Apple had four choices:
1. Apple releases new G5 in modified el cap enclosure. Very noisy. People complain about noise.
2. Apple releases new enclosure with many fans, designed for quiet operation but with limited expansion of drives. People complain about expansion.
3. Apple releases enclosure with many fans, designed for quiet operation and same expansion options as el cap. People complain about size.
4. Apple releases enclosure with many fans, designed for quiet operation and expansion options. But the enclosure will not work with newer faaster and hotter chips to be released in January. People complain that Apple is a bunch of short-sighted, stupid, talentless fscks who can't design an enclosure.
Pick your poison. BTW people, for the most part==mac forum members.
Zap, I think that for now the ideal lineup would move the 1.8 into an SP config and offer a dual middle tier of DP1.6. DP1.8 is wayyyy too close to DP1.8, even DP1.6 is getting close. WE will have to wait for the top end to stretch a little further. Soon, according to Apple.
Also, by "demand control" I wouldn't mean to say they're trying to limit pre-orders of expensive machines, just that they know of certain "pro" markets that have been craving such a machine, and for at least a quarter they want to make as much cash as they can on a limited supply. Give IBM and Apple a bit of break, they just started making these things, however badly you may want one, the wait is proof enough that supply is still limited. The fab still has a move to .09u to make, and speed increases within 12 months as stated by Apple. It leads one to believe that this first run into the PM's while fully workable isn't really what fishkill/G5 is all about and they will realize the true output of their new facility/chip in about 6-12 months. By then, prices and speeds might change dramatically, as well as configs.
You shouldn't care about the resistance of a few Apple apologists. They have no more right to dictate what you should want than I do, or Apple for that matter. IDE drives, and lots of them, are by far the most cost effective simple and reliable solution for you, YOU HAVE A RIGHT, as a customer, to make demands of the products you buy and the companies you buy from. It's the only way companies will learn to respect you.
If you buy the machine because the performance of your previous setup is so woeful, but you really wan the extra space, then you'll have to spend still more to get it externally, I know your money means nothing to Apple, but it damn well means something to you, doesn't it?
So you have to buy the G5 because, frankly, drives or not, you NEED it. That's fair of course.
That's what's so damned comtemptible about this move. Apple knows, or feels, that they have you by the pubes. So they'll sell you what they damned well please because you have a software investment in them and they can milk it. Then, in 6-12 months, they can "fix" the drive limitation and sell a whole new round of machines to people like you -- people that were pretty much "forced" to buy the first G5 they could get. Sure you could by external storage, but with the good resale value of PM's, it might be better just to sell and get a machine with the storage you wanted all along.
Mebbe that's even smart for them, short term, but it still treats you like an a$$hole, just for the privledge of giving them your business. Contemptible on their part, though a great many pros might be similarly constrained.
jwdwaso,
About bus speeds. IT IS NOT written in stone that the memory system has to function at 1:2 processor speed. Yes, for now the FSB is locked that way, but Apple is making their own custom controller, or getting IBM to make it, that means that they can get the memory to controller to CPU ratio wherever they need it to be. And they will have to. In twelve months they'll have a 3Ghz chip. Will they use a 1.5Ghz FSB/memory system? What happens when they get to 3.5, and then 4 or 5 Ghz ???
WHile the FSB is HUGE, it's virtual certainty that the ratio cannot remain locked at 2:1 forever, or even for long into the forseeable future. The will need higher and more flexible multipliers soon.
So, while it may not be as I say NOW, it will be possible in future iterations to have keep the FSB manageable on lower end machines even as clock speeds increase.
This is a V1 product afterall, things will improve in the future.
I wasn't arguing about Apple's costs at all, but the market reaction to machines that appear to offer similar performance. The cost difference in the 1.6 and 1.8 is probably as you say, down to Mobo differences and not the chips themselves. But that only makes my argument more plausible. Stick a faster single in the low end (assuming FSB ratios can be sorted) and put duals higher up. Slightly slower duals will be slightly cheaper, but the pair will cost more. In the end, ideally, you can get a nice consistent margin for your machines and a good reaction from the market, ie that the lower machines don't pummel the upper machines, sales wise.
That's the effect they're going for now, but the offering are somewhat lopsided ATm, iThink owing to the newness of the design. Things should settle down in revs 2 and 3.
I don't think the supply and demand argument is entirely valid here. The supply of ALL configs at this moment is effectively zero so should they bump the price of all the configs to help control demand? I think that the only given here is that the one system with dual processors has proven most popular.
I'd like to see what sort of demand there would be if each system offered dual processors right now with all other options left alone. I am positive you'd see strengthened demand for all models.
The only thing bumping the price of the sole dual machine would do it spike short term profit while inspiring longer term ill will. It will hurt more in the long run. The machine was priced with appropriate profit figured in from the start.
I am speculating based on the MacBidouelle report of 65K? preorders for the dual 2 gig versus 37K? preorders for the single 1.8 gig.
As for generating ill will, Apple has been doing that for years and it is all forgotten as soon as the new generation of faster machines is introduced. If the dual 2 gig is heavily backloged, that by itself will generate a lot of ill will.
Kickaha and Amorph couldn't moderate themselves out of a paper bag. Abdicate responsibility and succumb to idiocy. Two years of letting a member make personal attacks against others, then stepping aside when someone won't put up with it. Not only that but go ahead and shut down my posting priviledges but not the one making the attacks. Not even the common decency to abide by their warning (afer three days of absorbing personal attacks with no mods in sight), just shut my posting down and then say it might happen later if a certian line is crossed. Bullshit flag is flying, I won't abide by lying and coddling of liars who go off-site, create accounts differing in a single letter from my handle with the express purpose to decieve and then claim here that I did it. Everyone be warned, kim kap sol is a lying, deceitful poster.
Now I guess they should have banned me rather than just shut off posting priviledges, because kickaha and Amorph definitely aren't going to like being called to task when they thought they had it all ignored *cough* *cough* I mean under control. Just a couple o' tools.
Don't worry, as soon as my work resetting my posts is done I'll disappear forever.
First, could we drop the storage argument or take it elsewhere? I'm not saying it's not relevant, but it's a tangent that's hijacking the thread and taking it nowhere.
Now, with that out of the way:
It came out in the quarter ending last January that the average selling price of the MDD PowerMacs was exactly $200 [edit: I originally said $100] over the entry price for the line. Considering that sales of PowerMacs dropped even further, it's a fair bet that the average this past spring was no better, even though Apple gimped the low end model (although I haven't had time to dig up the documents and confirm this hunch yet).
That's pathetic.
So I really can't blame Apple for hypercorrecting with the G5, and going all out to make the fastest (and most profitable) PowerMac model as attractive as possible. Although Apple consistently identified Quarklessness and a crumbling economy as contributing to poor PM sales in part to avoid mentioning the elephant in the room, those were significant factors, and they haven't improved (Quark 6 is out, yes, but enough of the initial reactions from the field are so negative that it might as well not be - from what I've heard, DTP professionals have long since learned not to trust point-zero releases from Quark). But I digress. Apple bent over backward to upsell after several quarters where people overwhelmingly bought the bottom of the line machine, and judging from the reaction (the 2x2GHz machine alone among PowerMacs in the Apple Store's bestseller list) they were too successful. If my line of speculation is on target, then this recalibration of the line would make sense. If the 2x2GHz machine is flying out the door this fast, then the take-home message is that for the first time in a long time Apple doesn't have to gimp the lower line machines in order to upsell. This is good.
My analysis might be biased by the fact that I firmly believe that with OS X two brains are better than one, and I'm definitely in the camp that would stuff dual processors into everything that could take them. OS X is so much more fluid and responsive on two CPUs, and it seems to me that this benefit would be particularly felt in the consumer range, where people are more likely to run several lightweight applications than one big heavy one. But, again, I digress.
My analysis might be biased by the fact that I firmly believe that with OS X two brains are better than one, and I'm definitely in the camp that would stuff dual processors into everything that could take them. OS X is so much more fluid and responsive on two CPUs, and it seems to me that this benefit would be particularly felt in the consumer range, where people are more likely to run several lightweight applications than one big heavy one. But, again, I digress.
I have no doubt that the dual G5 will rock
However nobody has an idear how OS X will perform in term of fluidity and responsability under a single G5. And i speak here for Panther, because Apple have alreday announced that under Jaguar, the speed of the OS will be in the same range than the one of the G4 towers.
Apple bent over backward to upsell after several quarters where people overwhelmingly bought the bottom of the line machine, and judging from the reaction (the 2x2GHz machine alone among PowerMacs in the Apple Store's bestseller list) they were too successful.
I agree completely. I think the last few years created a corporate mindset at Apple that made them despise their own high end offerings. The Dual 2.0 G5 is the pendulum swinging the other way, fast.
Whether there will be a dual 1.8 rejigger to the line will depend on the contract with IBM. If Apple must buy 1.6 and 1.8 chips in certain quantities, then they'll make changes. If not, then why not just let this "problem" continue? If they sell twice as many high end machines and it doesn't get them into trouble with their supplier, so be it. Such sales figures could certainly be spun very positively in the press.
However nobody has an idear how OS X will perform in term of fluidity and responsability under a single G5. And i speak here for Panther, because Apple have alreday announced that under Jaguar, the speed of the OS will be in the same range than the one of the G4 towers.
Where did this myth come from? Apple has announced no such thing. Apple hasn't made any performance announcements other than the now famous benchmarks.
Comments
Originally posted by Zapchud
Ed M.: Very good post! Convinced me.
Convinced you of what? That even though it isn't more cost effective for you, that you should spend the money anyway? Or that you aren't a professional and should just buy a lower margin iMac because Apple thinks so little of you that they won't even supply the most basic tower level expansion when they finally get around to making a decent performer.
How does everyone know the Dual is the best selling config? If it is, then that means that the high end pros you talk about are the only people buying the machines (and that they won't bother with the 1.6 and 1.8. But the 1.6 and 1.8 are well expensive, too expensive in fact for non pros, so that kinda strands them without a market.
Apple MUST either raise the relative performance of ALL its towers, if it wants to steadily sell all three models, or it has to drop the price of the single 1.6 and 1.8 level machines. The bottom tier has to reach down to 1500 to cover a wider market, and the middle tier down to 2200, if they've no intention of adding duals.
As you all are so fond of saying, cost is no object "pros" will buy the top config anyway, so if the bottom two machines become better deals, thats no harm to sales of the top tier, it just reaches people who would NEVER buy the dual for cost reasons but who would buy the singles if they represented better value.
Likewise, since ONLY a select cast of "pros" who meet AI fanboy criteria
Oh but nothing is ever "good enough" for a real pro.
Maybe,
But you've all just finished shouting how "real pros" buy new equipment every year anyway and they'll take that expense/depreciation write off over just saving the money in the first place. What none of you seem to understand is that "real pros" will certainly take the write off when they have the expense, but they've no need (and more wisdom) that to just create the expense for the sake of the write off. "Real pros" can make their write offs on a host of other expenses that before they ever get to the computers that sit on their desks. At the end of the day the goal is to spend as little as can possibly get the job done and try to figure out how to spend even less.
But lest get back to the Ai conception of a "real pro"
When the DP 867 was released it was actually a decent value for a few months. People snapped them up. Must have hurt DP1.25 sales because Apple promptly dropped the dual "good" config at the next update. "Pros" is appears, like to save money.
The only other conclusion is that NOT JUST PROS, or more precisely, not just the narrow AI conception of "real pros", buy Powermacs, we should probably call these people "AI pros" but anyway... if the PM market stretches beyond just a particular type of pro, then it makes sense either to broaden the appeal of the machine (including the drive issue) or to add options and tiers.
By popular consensus, the two bottom tiers are stranded in the current Pm line.
Which is why, going all the way back before the forced digression into storage foisted on this thread by hmurchinson's pre-emptive fanboy reaction
Have your products reach more people.
to that end:
SP1.8-2 Ghz
DP 1.6
Dp 2.0
Or, you get the picture, use something fast but single in the low end, use something slower but DUAL in the middle, and use FAST and DUAL in the top. This way, there's no "Dual at 50% off!" like there was with the DP867, and the middle dual, while 500 cheaper, is not so close to the top that it convinces otherwise top end buyers to drop down a rung. A DP 1.8 is awfully close to a Dual 2Ghz for effective marketting, too close. Tweak the internal layout of all these machines (an ABSOLUTELY MINIMAL modification) without increasing the exterior dimensions (MORE THAN POSSIBLE) to accept more drives.
Then, in the place of the current PMG4 introduce a "cubish" product that takes RAM, CPU daughtercard, and AGP expansion, and has the same I/O as the PM's, but costs less and uses a G5.
1299-1599 Cube G5 (not a "cube" per se, just something in that range)
with slower singles, SP 1.6, and lacking the HDD and PCI expansion of the PMG5.
1999-2999, PM G5, Single, dual, dual. Better expansion and faster/dual configs.
Like so.
Cube G5 1.6 1299-1599
PM G5 2GhzSP 1999
PM G5 dual 1.6 2499
PM G5 dual 2.0 2999.
Nice even steps, lotsa selection, appropriate flexible options and performance for everybody and minimized "cannibalization" though I hate that term.
And don't force me to lay the smack down on you by mentioning the iMac.
The top iMac is so grossly overpriced that it must immediately drop to 1399 just to pretend to be in a consumer price range.
PS, it's obvious that ATM Apple is doing a little "demand control" by making the top end, and most expensive machine, appear to be the best value. Things should change by the time the G5 supply improves, after a quarter or two.
Originally posted by Matsu
PS, it's obvious that ATM Apple is doing a little "demand control" by making the top end, and most expensive machine, appear to be the best value. Things should change by the time the G5 supply improves, after a quarter or two.
I don't agree with this part.
(I'll keep my mouth shut about how much of the rest of that I agree with, it is kinda disturbing
IF Apple knew people were going to dive into the Dual 2.0 this way (Both feet first, cheering all the way), there's no way the pricing would be the way it is. Assuming that Apple has to buy 1.6's and 1.8's along with the 2.0 chips from IBM in whatever percentages they pass their tests at -> there's going to be a lot more 1.6s.
So _IF_ they knew in advance, they'd have to have some way of selling the 1.6's. Remember, these particular 1.6s will never be cool enough for a laptop -> they need to be sold nowish. Selling dual 1.6's would certainly move chips.
The only other things that make any sense would be: 1) IBM's yields are _much_ better than expected, or 2) Apple _isn't_ forced to buy the lower speed chips at a set ratio to the high speed chips.
I'd have Single 1.8, Dual 1.6, Dual 2.0 as my lineup in order of price.
Originally posted by Powerdoc
Nobody have any clue of the current yields of the PPC 970. It's even possible that most of the chip produced are able to reach 2 ghz, if it's the case there is no real background for this rumor.
Yep, could be that the 1.6 GHz are just 2GHz run at a lower voltage and overclocking the chip is easy but FSB makes it difficult (type of memory).
Hopefully by the end of September we will know.
Originally posted by Powerdoc
Nobody have any clue of the current yields of the PPC 970. It's even possible that most of the chip produced are able to reach 2 ghz, if it's the case there is no real background for this rumor.
That is why its a rumor and not a *confirmed* rumor
Of course *confirmed* rumors are far different then "The Truth", and then there is what apple announces being the *real confirmed in writing* truth which is no longer a rumor.
Originally posted by Matsu
Convinced you of what? That even though it isn't more cost effective for you, that you should spend the money anyway? Or that you aren't a professional and should just buy a lower margin iMac because Apple thinks so little of you that they won't even supply the most basic tower level expansion when they finally get around to making a decent performer.
I'm a prosumer and will get the most cost-effective dual G5 avaliable (which for me will probably be a DP 1.8, if Apple chooses to release it), when I have the $$$. Maybe I should add, that even though I'm not needing the power of the G5 to get the work done, to sustain economic income, I still want one because I feel this dual 1GHz G4 is way too slow. And I would be surprised if I'm the only one.
Though I definetely desire more expansion, I realize that (that's what Ed M. convinced me of), given the resistance I get on the internal storage-issue, I will get nowhere by claiming that a sufficient amount of professional customers would gain from the extra expansion, to justify the possible added engineering and production cost, without having any evidence.
But there are more than professional customers, the question is: How many, and how's their need for extra expansion? There's a 97,5% marketshare (that's a group almost 40 times bigger than the current group of mac buyers) for Apple to catch (they say they want it), and that is a big target, with lots of different needs. Someone mentioned PC-fanboy sites, that could have led me into believing we, or I need extra expansion. Those sites are read by more than just PC-fanboys, even professionals, regular consumers - which means they too can be led into believing that they need the expansion. They will look at the newest, glossiest, greenest monster-PC, which features three times the expansion. If they're in the dilemma, G5 or glossy, hairy PC - it'll affect their decision. I don't know if that's a large group or not, but that's the group Apple can't catch with an iMac, but still targeting. Their weapon is the G5, they're getting alot closer, but is it enough?
Sure, I must precise that even I'm desiring more expansion for myself, I think the G5 is the most awesome computer ever built, it's not that, but Apple has to - if they want "everything", they need something that suits everyone. A DP 1.8 is a considerable step in the right direction, more bays will make it suit more people, unless it makes the case grow noticeably, but I don't think it really will.
Bah, all this blahblahing, even though I said I would stop talking about this issue which is a very small one, IMO, but almost the only one - hence it's what I have to "whine" about, if I'm feeling the need to complain about something.
</off topic>
Oh, and Matsu, I like the idea of a SP 1.8 or 2.0 low end. According to Apples own benchmarks (well, SPEC - ugh, and add altivec), a single 1.8 or 2.0GHz G5 is very close to the processing power the competition offers - 3.2Ghz P4. One SP model to face the competetition at the SP side of things, approximately the price range of comparably specced PC towers. Duals, even lower clocked than the SP tower, would and will take care of the higher-end segment of the market. I think most pros ( aren't they the only ones buying high-end towers?
I thought it was pretty good the days when there was a SP 867 G4 mid-end, and DP 800 G4's in the top end. But it wasn't perfect, since the mid-end had a faster processor than the top-end, which meant you didn't get the most technically possible processing power when buying the top-end, that there's more potential. Always though it was too bad they didn't offered the best of the best, but it was quite okay anyways.
SP 2.0 (or 1.8 ), DP 1.6, eventual DP 1.8 and finally a DP 2.0 would be perfect in this regard. Apple would sell more than they currently do, and they'd get a much more balanced demand. Now, by making a DP 1.8 they'd be in the right direction, but we won't for sure be seeing the other stuff.
Originally posted by Matsu
...
SP1.8-2 Ghz
DP 1.6
Dp 2.0
Or, you get the picture, use something fast but single in the low end, use something slower but DUAL in the middle, and use FAST and DUAL in the top. This way, there's no "Dual at 50% off!" like there was with the DP867, and the middle dual, while 500 cheaper, is not so close to the top that it convinces otherwise top end buyers to drop down a rung. A DP 1.8 is awfully close to a Dual 2Ghz for effective marketting, too close. Tweak the internal layout of all these machines (an ABSOLUTELY MINIMAL modification) without increasing the exterior dimensions (MORE THAN POSSIBLE) to accept more drives.
Then, in the place of the current PMG4 introduce a "cubish" product that takes RAM, CPU daughtercard, and AGP expansion, and has the same I/O as the PM's, but costs less and uses a G5.
1299-1599 Cube G5 (not a "cube" per se, just something in that range)
with slower singles, SP 1.6, and lacking the HDD and PCI expansion of the PMG5.
1999-2999, PM G5, Single, dual, dual. Better expansion and faster/dual configs.
Like so.
Cube G5 1.6 1299-1599
PM G5 2GhzSP 1999
PM G5 dual 1.6 2499
PM G5 dual 2.0 2999.
Nice even steps, lotsa selection, appropriate flexible options and performance for everybody and minimized "cannibalization" though I hate that term.
... Things should change by the time the G5 supply improves, after a quarter or two.
I don't usually agree with Matsu, but your logic seems right on as far as these offerings. However, this logic is not based on any real information we have on costs. For instance, maybe it's very expensive to have memory systems at 1GHz, and it's much cheaper to have one at 800MHz. Maybe the dual processor motherboard is much more expensive than a SP. It could be that the cost of a DP 1.6 and a DP 2.0 is almost the same. It may be that the cost spread between the current SP 1.6 and SP 1.8 models is almost entirely due to the PCI difference and the memory bus speed difference, and not due to processor cost differences. Apple's margins on these may all be relatively low, and Apple may be betting on component costs going down for long term profits.
Next spring we'll have a better idea about costs - Apple will undoubtedly adjust price as all the lines migrate to the G5.
As far as storage, give it a rest. There's no one perfect solution, plus Apple has made mistakes before. Remember DVD drives but no CD-RW drives? It will work itself out. This isn't a make or break issue. Computers are more and more becoming commodities - buy what you need today and for next week. When you need that 3rd drive, won't you also want the DP G6?
Someone mentioned PC-fanboy sites, that could have led me into believing we, or I need extra expansion. Those sites are read by more than just PC-fanboys, even professionals, regular consumers - which means they too can be led into believing that they need the expansion. They will look at the newest, glossiest, greenest monster-PC, which features three times the expansion.
Expansion is not just hardrives and opticals. There is more than enough room for extra cards and RAM. If the lack of internal options for drives stops you from purchasing a G5 then you don't need the processing power.
Apple had four choices:
1. Apple releases new G5 in modified el cap enclosure. Very noisy. People complain about noise.
2. Apple releases new enclosure with many fans, designed for quiet operation but with limited expansion of drives. People complain about expansion.
3. Apple releases enclosure with many fans, designed for quiet operation and same expansion options as el cap. People complain about size.
4. Apple releases enclosure with many fans, designed for quiet operation and expansion options. But the enclosure will not work with newer faaster and hotter chips to be released in January. People complain that Apple is a bunch of short-sighted, stupid, talentless fscks who can't design an enclosure.
Pick your poison. BTW people, for the most part==mac forum members.
Zap, I think that for now the ideal lineup would move the 1.8 into an SP config and offer a dual middle tier of DP1.6. DP1.8 is wayyyy too close to DP1.8, even DP1.6 is getting close. WE will have to wait for the top end to stretch a little further. Soon, according to Apple.
Also, by "demand control" I wouldn't mean to say they're trying to limit pre-orders of expensive machines, just that they know of certain "pro" markets that have been craving such a machine, and for at least a quarter they want to make as much cash as they can on a limited supply. Give IBM and Apple a bit of break, they just started making these things, however badly you may want one, the wait is proof enough that supply is still limited. The fab still has a move to .09u to make, and speed increases within 12 months as stated by Apple. It leads one to believe that this first run into the PM's while fully workable isn't really what fishkill/G5 is all about and they will realize the true output of their new facility/chip in about 6-12 months. By then, prices and speeds might change dramatically, as well as configs.
You shouldn't care about the resistance of a few Apple apologists. They have no more right to dictate what you should want than I do, or Apple for that matter. IDE drives, and lots of them, are by far the most cost effective simple and reliable solution for you, YOU HAVE A RIGHT, as a customer, to make demands of the products you buy and the companies you buy from. It's the only way companies will learn to respect you.
If you buy the machine because the performance of your previous setup is so woeful, but you really wan the extra space, then you'll have to spend still more to get it externally, I know your money means nothing to Apple, but it damn well means something to you, doesn't it?
So you have to buy the G5 because, frankly, drives or not, you NEED it. That's fair of course.
That's what's so damned comtemptible about this move. Apple knows, or feels, that they have you by the pubes. So they'll sell you what they damned well please because you have a software investment in them and they can milk it. Then, in 6-12 months, they can "fix" the drive limitation and sell a whole new round of machines to people like you -- people that were pretty much "forced" to buy the first G5 they could get. Sure you could by external storage, but with the good resale value of PM's, it might be better just to sell and get a machine with the storage you wanted all along.
Mebbe that's even smart for them, short term, but it still treats you like an a$$hole, just for the privledge of giving them your business. Contemptible on their part, though a great many pros might be similarly constrained.
jwdwaso,
About bus speeds. IT IS NOT written in stone that the memory system has to function at 1:2 processor speed. Yes, for now the FSB is locked that way, but Apple is making their own custom controller, or getting IBM to make it, that means that they can get the memory to controller to CPU ratio wherever they need it to be. And they will have to. In twelve months they'll have a 3Ghz chip. Will they use a 1.5Ghz FSB/memory system? What happens when they get to 3.5, and then 4 or 5 Ghz ???
WHile the FSB is HUGE, it's virtual certainty that the ratio cannot remain locked at 2:1 forever, or even for long into the forseeable future. The will need higher and more flexible multipliers soon.
So, while it may not be as I say NOW, it will be possible in future iterations to have keep the FSB manageable on lower end machines even as clock speeds increase.
This is a V1 product afterall, things will improve in the future.
I wasn't arguing about Apple's costs at all, but the market reaction to machines that appear to offer similar performance. The cost difference in the 1.6 and 1.8 is probably as you say, down to Mobo differences and not the chips themselves. But that only makes my argument more plausible. Stick a faster single in the low end (assuming FSB ratios can be sorted) and put duals higher up. Slightly slower duals will be slightly cheaper, but the pair will cost more. In the end, ideally, you can get a nice consistent margin for your machines and a good reaction from the market, ie that the lower machines don't pummel the upper machines, sales wise.
That's the effect they're going for now, but the offering are somewhat lopsided ATm, iThink owing to the newness of the design. Things should settle down in revs 2 and 3.
Originally posted by Matsu
About bus speeds. IT IS NOT written in stone that the memory system has to function at 1:2 processor speed.
In fact, IBM is quoted as saying 1:2, 1:3, and 1:6 FSB's were supported by the 970.
Originally posted by Bancho
I don't think the supply and demand argument is entirely valid here. The supply of ALL configs at this moment is effectively zero so should they bump the price of all the configs to help control demand? I think that the only given here is that the one system with dual processors has proven most popular.
I'd like to see what sort of demand there would be if each system offered dual processors right now with all other options left alone. I am positive you'd see strengthened demand for all models.
The only thing bumping the price of the sole dual machine would do it spike short term profit while inspiring longer term ill will. It will hurt more in the long run. The machine was priced with appropriate profit figured in from the start.
I am speculating based on the MacBidouelle report of 65K? preorders for the dual 2 gig versus 37K? preorders for the single 1.8 gig.
As for generating ill will, Apple has been doing that for years and it is all forgotten as soon as the new generation of faster machines is introduced. If the dual 2 gig is heavily backloged, that by itself will generate a lot of ill will.
Originally posted by Nevyn
In fact, IBM is quoted as saying 1:2, 1:3, and 1:6 FSB's were supported by the 970.
In fact, IBM said 1:2, 1:3, and 1:6 FSBs are supported beside 1:4 - which is used by Apple now. Don't confuse marketing and reality
Now I guess they should have banned me rather than just shut off posting priviledges, because kickaha and Amorph definitely aren't going to like being called to task when they thought they had it all ignored *cough* *cough* I mean under control. Just a couple o' tools.
Don't worry, as soon as my work resetting my posts is done I'll disappear forever.
First, could we drop the storage argument or take it elsewhere? I'm not saying it's not relevant, but it's a tangent that's hijacking the thread and taking it nowhere.
Now, with that out of the way:
It came out in the quarter ending last January that the average selling price of the MDD PowerMacs was exactly $200 [edit: I originally said $100] over the entry price for the line. Considering that sales of PowerMacs dropped even further, it's a fair bet that the average this past spring was no better, even though Apple gimped the low end model (although I haven't had time to dig up the documents and confirm this hunch yet).
That's pathetic.
So I really can't blame Apple for hypercorrecting with the G5, and going all out to make the fastest (and most profitable) PowerMac model as attractive as possible. Although Apple consistently identified Quarklessness and a crumbling economy as contributing to poor PM sales in part to avoid mentioning the elephant in the room, those were significant factors, and they haven't improved (Quark 6 is out, yes, but enough of the initial reactions from the field are so negative that it might as well not be - from what I've heard, DTP professionals have long since learned not to trust point-zero releases from Quark). But I digress. Apple bent over backward to upsell after several quarters where people overwhelmingly bought the bottom of the line machine, and judging from the reaction (the 2x2GHz machine alone among PowerMacs in the Apple Store's bestseller list) they were too successful. If my line of speculation is on target, then this recalibration of the line would make sense. If the 2x2GHz machine is flying out the door this fast, then the take-home message is that for the first time in a long time Apple doesn't have to gimp the lower line machines in order to upsell. This is good.
My analysis might be biased by the fact that I firmly believe that with OS X two brains are better than one, and I'm definitely in the camp that would stuff dual processors into everything that could take them. OS X is so much more fluid and responsive on two CPUs, and it seems to me that this benefit would be particularly felt in the consumer range, where people are more likely to run several lightweight applications than one big heavy one. But, again, I digress.
Originally posted by Amorph
Quarklessness
My preferred state.
Originally posted by Amorph
OK, let's see here.
My analysis might be biased by the fact that I firmly believe that with OS X two brains are better than one, and I'm definitely in the camp that would stuff dual processors into everything that could take them. OS X is so much more fluid and responsive on two CPUs, and it seems to me that this benefit would be particularly felt in the consumer range, where people are more likely to run several lightweight applications than one big heavy one. But, again, I digress.
I have no doubt that the dual G5 will rock
However nobody has an idear how OS X will perform in term of fluidity and responsability under a single G5. And i speak here for Panther, because Apple have alreday announced that under Jaguar, the speed of the OS will be in the same range than the one of the G4 towers.
Originally posted by Amorph
Apple bent over backward to upsell after several quarters where people overwhelmingly bought the bottom of the line machine, and judging from the reaction (the 2x2GHz machine alone among PowerMacs in the Apple Store's bestseller list) they were too successful.
I agree completely. I think the last few years created a corporate mindset at Apple that made them despise their own high end offerings. The Dual 2.0 G5 is the pendulum swinging the other way, fast.
Whether there will be a dual 1.8 rejigger to the line will depend on the contract with IBM. If Apple must buy 1.6 and 1.8 chips in certain quantities, then they'll make changes. If not, then why not just let this "problem" continue? If they sell twice as many high end machines and it doesn't get them into trouble with their supplier, so be it. Such sales figures could certainly be spun very positively in the press.
Originally posted by Powerdoc
I have no doubt that the dual G5 will rock
However nobody has an idear how OS X will perform in term of fluidity and responsability under a single G5. And i speak here for Panther, because Apple have alreday announced that under Jaguar, the speed of the OS will be in the same range than the one of the G4 towers.
Where did this myth come from? Apple has announced no such thing. Apple hasn't made any performance announcements other than the now famous benchmarks.