Possible New G3 Chip Coming?
This is wild speculation and may make a junk discussion topic, but I think it is possible. Look at the situation with PowerMacs. Why isn't Apple selling one with really competitive performance? We all know the answer. They will when the chip is ready, the IBM 970. Chips take time, unfortunately.
We all know about the 970, but we are asking the same kind of question for the low end. Why isn't Apple selling a Mac with a really competitive price? Could it be that the chip is not ready? Quite some time ago there was development work on a G3 processor that had other on-board circuitry for many of the motherboard functions. The idea was to make a very low cost motherboard for the low end market. Interesting idea but it faded away and hasn't been discussed in years.
The lack of a low end Mac is such a glaring deficiency that we cannot help but ask why? Maybe there is something in the works for the low end? With the space saved by not having a SIMD engine, and the use of smaller processes, IBM could likely pack a lot into the processor chip. Another possibility is a separate chip to take on many motherboard functions, rather than putting them into the processor. Either way lowers manufacturing cost significantly. As I said, it is wild speculation.
We all know about the 970, but we are asking the same kind of question for the low end. Why isn't Apple selling a Mac with a really competitive price? Could it be that the chip is not ready? Quite some time ago there was development work on a G3 processor that had other on-board circuitry for many of the motherboard functions. The idea was to make a very low cost motherboard for the low end market. Interesting idea but it faded away and hasn't been discussed in years.
The lack of a low end Mac is such a glaring deficiency that we cannot help but ask why? Maybe there is something in the works for the low end? With the space saved by not having a SIMD engine, and the use of smaller processes, IBM could likely pack a lot into the processor chip. Another possibility is a separate chip to take on many motherboard functions, rather than putting them into the processor. Either way lowers manufacturing cost significantly. As I said, it is wild speculation.
Comments
The real reason is because Apple does not have a price competitive "tower" of any kind. The only single CPU desktop they sell is an all in one. Bad move imo.
I cannot see why it would cost more than, say, $700-800 for 800 MHz G3, 256MB RAM, 80GB HD, CD-RW, GeForce 4.
Lusers bought iMacs because they were/are simple.
Power users bough Power Macs because they wanted the expandibility.
Prosumer is a myth.
Barto
PS it seems that what people want is something for nothing, ei a sub $1000 Power Mac
[ 01-22-2003: Message edited by: Barto ]</p>
<strong>it seems that what people want is something for nothing</strong><hr></blockquote>
Undeniable.
<strong>Putting a G3 in it would be instant death to a resurrected cube. But I think you can get a low power G4 (7457) running at a reasonable speed and fit an AGP and expansion slot in there. Get rid of a DIMM slot; 2 is enough. A 1GHz 7457 will not need such a big heat sink, and for God's sakes put a fan in there! And I don't think a $999 PowerMac is unreasonable. Especially in Cube disguise. The plastics are incredibly simple and with a fan much of the central heat column can be easily displaced making it easier to pack things in more tightly since heat is actively being removed from the case.</strong><hr></blockquote>
put a 1Ghz ppc7447 in it, instead of a ppc7457 and you have a headless imac
It should also be noted that what you can now buy for 999USD [sans monitor] is not in any way a low end machine. Such a machine includes at least 64MB of graphics memory; at least 2 ATA-100 channels, often ATA133, and soon SATA150; at least the equivalent of a combo drive (usually in the guise of one CDRW and one DVD-ROM accross TWO full size external bays); either 10/100 ethernet or 56K modem, or both; comes with 256-512MB of RAM and expands to at least 1.5GB but takes as much as 3-4Gb depending on the system. It is solidly in the mid range in terms of power and features.
999 is hardly bargain basement, what Apple offers for 999USD on the desktop side is utterly pathetic. Some manufacturers even manage to throw in a display in addition to the features I described, no one here is even asking Apple to do that much.
I would hardly call it price whining, I would call it reality and Apple better wake up to it!
<strong>The other 97.3 % of the computing world buys 999 and under SP towers, NOT A-I-O's, Apple ONLY sells as many AIO's as it does because they lack a decently priced expandable model. Expansion/upgradeability and a sub K price. That's what people want, if Apple really wants market share that's what they'll start selling. It doesn't have to be a "tower" but I have to be able to attach my own display, it has to have a full size AGP slot, one PCI slot open, user accessible INTERNAL drive bays, and a REPLACEABLE CPU card/slot/whatever.
</strong><hr></blockquote>
You are projecting what you want onto the rest of the market. Most of the non-geeks I know don't care about expansion, upgradability, drive bays, multiple attachable displays, etc, etc, etc. They want a machine that is small and unobtrusive (not to mention stylish). And guess what... most of the world is filled with non-geeks. They just don't hang out on boards like this to bemoan how Apple doesn't have low end towers -- they tend to have a life.
<strong>Increase in market share encourages development for the platform that in turn drives more sales.</strong><hr></blockquote>
Yes. That's why it may make sense to sell some models of Macs at zero profit with the only goal to increase their quantity. Not going to happen. <img src="graemlins/hmmm.gif" border="0" alt="[Hmmm]" />
<strong>
You are projecting what you want onto the rest of the market. Most of the non-geeks I know don't care about expansion, upgradability, drive bays, multiple attachable displays, etc, etc, etc. They want a machine that is small and unobtrusive (not to mention stylish). And guess what... most of the world is filled with non-geeks. They just don't hang out on boards like this to bemoan how Apple doesn't have low end towers -- they tend to have a life.
True, but not exactly. Name on PC AIO that hasn't flopped? They flop, not only because they tend to look like crap but also because they same manufcaturer typically offers a better deal in the form of a tower or desktop. Consumers will buy laptops in droves before they ever make a whole-scale migration to AIO's. I don't consider myself a power user or geek, just a guy who works fast and tends to have a lot of free work time (it helps to work from home most days) Still I have a perfectly usable 6 year old tower (AMD300) for office internet and some CD burning (4X) that would not have survived as long as it did if it were an AIO. When the monitor went last year, I bought a new one for 275 Canadian, a very good one. WHEN something goes wrong with your eMac or iMac display a few months out of the ridiculously expensive AppleCare, just fixing the machine is going to cost a lot more than that. If you open up your iMac to swap the HDD, I'm pretty sure that violates the warrantee agreement. When I ran out of space on this machine, a cheap INTERNAL ATA drive was a swap away. And even if I couldn't do it myself, installation was a piddling 30 Canadian.
Laptops or towers, AIO's really don't make sense unless they aquire a greater degree of modularity. That is drives, RAM, CPU nd GPU all standard sized and user accessible.
<strong>
You are projecting what you want onto the rest of the market. Most of the non-geeks I know don't care about expansion, upgradability, drive bays, multiple attachable displays, etc, etc, etc. They want a machine that is small and unobtrusive (not to mention stylish). And guess what... most of the world is filled with non-geeks. They just don't hang out on boards like this to bemoan how Apple doesn't have low end towers -- they tend to have a life.
The problem with that is that most of the salesmen are PC Geeks who dont know enough about Macs to effectively sell them. They do however, have a tendency to talk people out of them in favor of a computer that they do know enough about to sound intelligent.
<strong>
Yes. That's why it may make sense to sell some models of Macs at zero profit with the only goal to increase their quantity. Not going to happen. <img src="graemlins/hmmm.gif" border="0" alt="[Hmmm]" /> </strong><hr></blockquote>
No need to sell at zero profit. They sell iBooks and eMacs for around $999 or just over a grand. You telling me they can't take an eMac specwise, cut out the monitor and stick it in a small case, and NOT make a profit at $999?
DurH! Of course it didn't.
Is the iMac2 working much better? A little. It's not the runaway success the first one was...
In the absence of a decent and cheap edu' box, the iSlab/iCube headless would suffice for many. (Guess, I'm 'projecting' though...)
Any guarantee that a cheaper version of the Cube wouldn't sell? At least much a guarantee that price affected it's ability to sell in the first place? People like a bargain. They like cheap. Rip off the iBook monitor and sell the slab even cheaper. Who'd buy..? Try it and find out Apple. Unless you try entering vacated desktop market segments...then you'll never have the answer to why you don't have 10% let alone 97%.
What we know is that Apple is still selling 60K of the 'crap and cheap' G3 iMac. Plenty of eMacs... So, I guess there was a market for a Mac under a K after all...
Programmer, Matsu may be 'projecting'. That's what people on rumour boards do. Apple has a laptop range that scales handsomely.
The same cannot be said of the desktop. No single cpu, expandable machine for less than? Oh. They don't have one. Massive holes in single cpu performance.
Expandable? Don't have one of those for less than £1,395.
A cheap white box with integrated crap in it to sell to skinflints. Don't have them either. (But they're willing to sell the 'crap' iMac G3 with a couple year old spec and the same inflated price...)
A tower for around a £1k?
Gee, if Apple's desktop range wasn't littered with more holes than Swiss cheese than I'd probably quit 'whining' about it. When you're ignoring 97% per cent of what PC buyers want it's hardly surprised Apple's desktop sales are pants.
And they want 10%?
Er. Try harder with your desktops Apple.
Which, I'm sure they'll update. Y'know, after 'figuring out' what the problem was with the iMac 2... Of course, they will. Logically, they will. But will they 'try harder?' Will they do the 'best with what they've got'? Let's see.
Lemon Bon Bon
[quote]They just don't hang out on boards like this to bemoan how Apple doesn't have low end towers -- they tend to have a life. <hr></blockquote>
The sound of breaking glass, Programmer...
Still, most of those 'real people' have a life with a 'low end' Wintel tower which they got wayyyyyy more affordably than a £1,395 (starting at folks!!!) Apple tower. Is there a market for a cheaper Apple tower. I have no idea...
[ 01-22-2003: Message edited by: Lemon Bon Bon ]</p>
<strong>
True, but not exactly. Name on PC AIO that hasn't flopped?</strong><hr></blockquote>
Name one PC AIO that hasn't sucked.
Most of them have any of: Crap LCDs, "integrated" graphics, loud fans, low RAM ceilings, and aggressively ugly, user-hostile designs. I love Apple's AIOs, but that's because they know how to design them. If I had to choose between a PC AIO and a PC tower, I'd get the tower too. Not because towers are better than AIOs, but because every PC AIO I've ever seen was a bad joke.
One of my friends tells me that new iBooks are on the way. It's common knowledge that IBM has a 1Ghz G3 available. Why not stick these in the new iBook? Or hell, even in the iMac or eMac. It would get better performance than the 800mhz G4 in most apps, give off less heat, give Motorola less business
The problem is the Schylla and Charbydis of the Megahertz myth, hitting Apple from both sides. Since a G4 is basically a G3 with Altivec unit, a G3 iBook running at 1 Ghz could potentially be faster than a Dual G4 867 at some tasks! It would be quite embarrassing to Apple.
That's why we probably won't ever see another Apple machine with running the G3 processor. Sad, because IBM could probably ramp up the Mhz to 2Ghz without much effort (this probably wouldn't improve the CPU too much, but it would help the Mac remain feasable in the Mhz argument.)
If they could get this "G3+" running at high Mhz, they could just write off the G4 (and Motorola) completetly once the PPC970 hits. Better performance for general purpose computing, lower power consumption, and Altivec? Who needs it
One of my friends tells me that new iBooks are on the way. It's common knowledge that IBM has a 1Ghz G3 available. Why not stick these in the new iBook? ... Since a G4 is basically a G3 with Altivec unit, a G3 iBook running at 1 Ghz could potentially be faster than a Dual G4 867 at some tasks! It would be quite embarrassing to Apple.</strong>
You have answered yourself. The main problem with the G4 is that Motorola has yet to come out with a 130 nm G4 chip, the mythical 7457, while IBM shipped a 130 nm G3 chip last Spring. This should have come out at the latest 3Q 02 (which would have been late already), but maybe Moto will ship in Q2 03. After that, Apple's entire line of Macs can be speedbumped: iBooks, iMacs and eMacs at 1 GHz; Powerbooks at 1.25 GHz; and Power Macs at dual 1.5 GHz.
<strong>Sad, because IBM could probably ramp up the Mhz to 2Ghz without much effort (this probably wouldn't improve the CPU too much, but it would help the Mac remain feasable in the Mhz argument.) </strong>
Much effort is about $2G, that's 2 billion dollars, to get a 90 nm fab working because that's the only way the current 750 design will get to 2 GHz.
Apple's problem is that Motorola hasn't produced a 130 nm G4 yet, which would be the thing that kicks the G4 up to 1.5 to 1.8 GHz. The current IBM PPC 750fx G3 chips in the iBooks are already built on IBM's 130 nm fab and are pretty much at their limit until the next 90 nm fab ramps up in 2004.
<strong>If they could get this "G3+" running at high Mhz, they could just write off the G4 (and Motorola) completetly once the PPC970 hits. Better performance for general purpose computing, lower power consumption, and Altivec? Who needs it
If IBM can extend the pipeline of the 750fx to 7 or 8 and use the PPC 970 AltiVec unit, yeah, it would be a nice replacement for the G4 and Apple can abandon Motorola. The G4 has a 50% advantage in clock rates compared to the G3. That's hard to abandon, unless IBM comes up with something equivalent. Too bad they haven't, because it Apple's CPU woes could be alleviated within a year from now.
However, the G3 - even more than the G4 - is not designed for clockspeed. It cannot and will not magically eclipse the G4 in clock rate if IBM flips a switch somewhere. The G3 is small, cool, simple and cheap, and an excellent performer given those goals. Subsequent G3s will remain true to those goals. IBM already has a desktop processor in the works (the 970), and they sensibly scaled it down from a big iron CPU, rather than trying to scale up an embedded CPU.
I can see the G3 hanging on in the iBook, which it's just about ideal for, and perhaps moving into DLDs once it gets even smaller.
[ 01-22-2003: Message edited by: Amorph ]</p>
<strong>
Name one PC AIO that hasn't sucked.
Most of them have any of: Crap LCDs, "integrated" graphics, loud fans, low RAM ceilings, and aggressively ugly, user-hostile designs. I love Apple's AIOs, but that's because they know how to design them. If I had to choose between a PC AIO and a PC tower, I'd get the tower too. Not because towers are better than AIOs, but because every PC AIO I've ever seen was a bad joke.</strong><hr></blockquote>
Excellent points, but regardless the vast pool of potential buyers has been conditioned, and they won't be broken by pretty plastic and ergonomics. (though they ought to think about ergos) Certainly, they won't be swayed by pretty plastic at inflated prices, though we'll leave that aside for now.
The AIO is the model that will eventually be squeezed out. A laptop is an AIO and over the next 5 years they will get fast enough and cheap enough to sway more consumers than any AIO. When that day comes a desktop can only really be something that expands/upgrades easily, otherwise, what's the point? People want to choose their own displays and drives, and they want a measure of security in being able to easily upgrade to meet new connectivity standards, optical drive formats, and add INTERNAL storage capacity. Not to mention the significant number of people who do upgrade thei CPU's and GPU's (white box builders are upto 30% of the market in some areas, most of their customers like to upgrade these components, that's a big pool of customers)