I still don't see what Apple can do to update the Powerbook line at the end of this year or the beginning of next year. Mr. MacPhisto's post says that there won't be a portable G5 before next summer, and on the other hand Freescale is not going to sample the 7448 before H1 2005. So, what Apple has in the hands now for the next Powerbook update? I don't see anything, except perhaps a silent minor speed bump in the 7447A family.
I still don't see what Apple can do to update the Powerbook line at the end of this year or the beginning of next year. Mr. MacPhisto's post says that there won't be a portable G5 before next summer, and on the other hand Freescale is not going to sample the 7448 before H1 2005. So, what Apple has in the hands now for the next Powerbook update? I don't see anything, except perhaps a silent minor speed bump in the 7447A family.
I said early 2005, possibly late this year if things work without much tinkering. This is for the low voltage 970FX, so there is hope sooner rather than later. What Rhumgod unearthed in the code of OS X makes it appear that Apple is preparing a G5 PB in the nearer future. I don't see why they'd put into Panther if they didn't have some inkling of it becoming available prior to Tiger's launch early next year. So there is hope. I previously felt that FreeScale would deliver dual-core and low-voltage SoC chips before IBM. It is now appearing to not be so. Too bad. Like Amorph, I had hoped for a viable alternative to IBM to keep them on their toes. The are making progress at Fishkill though. 90nm G5s are becoming more common now, though Apple wants more. I do wish they could deliver dual-core for portables sooner; I'd love for the opportunity to replace my iBook.
I said early 2005, possibly late this year if things work without much tinkering. This is for the low voltage 970FX, so there is hope sooner rather than later.
Ah, you meant indeed the low voltage 970FX. So, in some sense, things are already determined. There is nothing else in the horizon for the Powerbooks than a lower voltage 970FX.
Please name one thing that would make the 8641 an unsuitable notebook chip.
Are there going to be any graphics chipsets available that support PCIe x8? If not, there's one reason.
Apple would need to design a RapidIO I/O ASIC for USB, Firewire, audio, SATA/ATA, WiFi, Bluetooth, etc. They could do it, but it would be money that also could be spent elsewhere.
Quote:
Secondly, the 7448 uses 10W typical @ 1.4Ghz the 7447A uses 21W at the same clock.
I would be very very careful of any advertised power consumption specs for 90 nm chips, especially from a company who hasn't produced a real 90 nm part yet.
Quote:
How is not that an improvement, aspecially concidering that the 7448 gets a 100% larger L2 cache and a ~20% faster fsb.
Not to mention that the .pdf I linked to shows that it's going to clock to "atleast" 1.5Ghz. (Looks like Freescale has leant the merit of being conservative in there preformance estimates)
It's an improvement no doubt. But they are rumored to sample in 1H of 05. June 30, 2005 is in 1H 05. Production quantities may not even come until late Q3 05. That's the problem. Freescale will ship the 7448 ~8 months too late, assuming everything will go all right with their 90 nm fab.
Again, I have no problem with Apple shipping a 1.5" thick Powerbook G5 with a 1.8 GHz 970 based chip, along with a reduced price, more mobile Powerbook G4 and low-end iBook G4.
If laptops are a bigger market for Apple, maybe they should start shipping a larger product line. They have 3 different desktop Macs, why not 3 different laptop Macs?
Freescale is not going to sample the 7448 before H1 2005. So, what Apple has in the hands now for the next Powerbook update? I don't see anything, except perhaps a silent minor speed bump in the 7447A family.
I wouldn't put any faith into what EE-Times has to say. They are about as reliable as MOSR, if that.
I wouldn't put any faith into what EE-Times has to say. They are about as reliable as MOSR, if that.
You can't be serious - eeTimes is an Electronic Engineering site that publishes news stories. I don't think I've ever seen an article based on speculation and rumor and heresay.
Except that the 8641D isn't pin compatible with current mobos (the 7448 is).
The 8641D isn't intended for computing, it's intended for embedded markets.
Hey guys the whole world is a market for embedded devices, lets knock it off with respect to intended markets. Much of what we are likely to get at anyone time is the result of supply. That is if Apple had a low power G5 that meant all of their requirements you can be sure that it would end up in a portable of some type soon after it came into existance.
The bigger issue that I think some here are missing is that there is currently a market for both processors. The G5 can go inot laptops for people that often use rubber rulers and the G4 products can go into people that need real portable performance.
Quote:
And the 8641D won't sample until the 2nd half '05.
Actually I hope this is a sign that Freescales is maturing quickly and not publicly promising more than they can deliver. It is refreshing to see that some of the e500 devices have started shipping. One can always hope.
Quote:
And typical Moto... more than a year since the 1.5GHz 7447 and they've bumped...
100 MHz! Woot!!!
Frankly that does suck. On the other hand the high integration devices do appear to be customed mad for Apples need. PCI-Express and a built in memory controller would have been fantastic for a portable, but the new chips have even more goodies. Frankly Apple could build one hell of an iBook with one of these.
The question is can IBM deliver similar products at similar power / performance points. Right now they can't but it will be very interesting to see what is happening in the new year.
Are there going to be any graphics chipsets available that support PCIe x8? If not, there's one reason.
*groan* You could hook up a PCI-E x16 interface to a x1 line if you wanted and it'd still work. Infact I'd say that it's better then a full x16 interface because, 1) it'll drain less energy, 2) it do is enough bandwidth, people have trouble seeing difrences between AGP x4 and x8, PCI-E has got, full duplex, roughly the double bandwidth of AGP x8. 3) Being on chip ought to give the GFX interface a lower latency access to the CPU wich ought to give it a slight preformance boost.
Quote:
Apple would need to design a RapidIO I/O ASIC for USB, Firewire, audio, SATA/ATA, WiFi, Bluetooth, etc. They could do it, but it would be money that also could be spent elsewhere.
With That is what they'd have to build an ASIC for, there is no etc. The DMA and interrupt controller is on the CPU, as is the GFX interface (and of course the memory controller. Oh and if they wanted, they could make it a PCI-E interface instead.
Quote:
I would be very very careful of any advertised power consumption specs for 90 nm chips, especially from a company who hasn't produced a real 90 nm part yet.
Corolles is producing 90nm parts and is sampling 65nm parts though.
[edited]Don't post early in the morning (like I did)...[/edited]
I'm going to order me a Powerbook soon, but I was planning on waiting to the next Rev. But hell, I'm not waiting THAT long.
I might as well order my Powerbook today..
If you can wait, I would recommend waiting until the end of October to see if there is an update of any kind. Even a 100MHZ bump would make the current PBs a little bit cheaper so you could save some money on them. However, if you need a notebook and your are set on an Apple (there are reasons outside of processor speed to buy them) then I'd recommend plopping down the cash.
There is a possibility of low-voltage G5 availability before the year is out, though I wouldn't hold my breath.
You could hook up a PCI-E x16 interface to a x1 line if you wanted and it'd still work. ...
That is what they'd have to build an ASIC for, there is no etc. The DMA and interrupt controller is on the CPU, as is the GFX interface (and of course the memory controller. Oh and if they wanted, they could make it a PCI-E interface instead.
It could work. I agree with you. I was promoting this idea 4 years ago myself. But I'm not getting a warm fuzzy with Apple using a SoC like this yet.
Quote:
Corolles is producing 90nm parts and is sampling 65nm parts though.
Which 90 nm parts is Crolles producing? I know about the 90 nm Philips ARM part in sampling, with expected volume production in Q4 04, but what else? Freescale is saying that the PowerQuicc III 90 nm parts won't be sampling until Q2 05, so I think they have a ways to go with the fab.
Freescale itself (slide 41 of SNDF 04 presentation) says that they will sample the 7448 in 1H 05 and have volume production in 2H 05. The 8641D will sample in the 2H 05 and have volume production in 1H 06. That's a lot time for IBM to come out with low voltage 970fx, 970mp and 970 w/on-die mem controller parts.
Comments
Originally posted by PB
I still don't see what Apple can do to update the Powerbook line at the end of this year or the beginning of next year. Mr. MacPhisto's post says that there won't be a portable G5 before next summer, and on the other hand Freescale is not going to sample the 7448 before H1 2005. So, what Apple has in the hands now for the next Powerbook update? I don't see anything, except perhaps a silent minor speed bump in the 7447A family.
I said early 2005, possibly late this year if things work without much tinkering. This is for the low voltage 970FX, so there is hope sooner rather than later. What Rhumgod unearthed in the code of OS X makes it appear that Apple is preparing a G5 PB in the nearer future. I don't see why they'd put into Panther if they didn't have some inkling of it becoming available prior to Tiger's launch early next year. So there is hope. I previously felt that FreeScale would deliver dual-core and low-voltage SoC chips before IBM. It is now appearing to not be so. Too bad. Like Amorph, I had hoped for a viable alternative to IBM to keep them on their toes. The are making progress at Fishkill though. 90nm G5s are becoming more common now, though Apple wants more. I do wish they could deliver dual-core for portables sooner; I'd love for the opportunity to replace my iBook.
Originally posted by Mr. MacPhisto
I said early 2005, possibly late this year if things work without much tinkering. This is for the low voltage 970FX, so there is hope sooner rather than later.
Ah, you meant indeed the low voltage 970FX. So, in some sense, things are already determined. There is nothing else in the horizon for the Powerbooks than a lower voltage 970FX.
Originally posted by Eric_Z
Please name one thing that would make the 8641 an unsuitable notebook chip.
Are there going to be any graphics chipsets available that support PCIe x8? If not, there's one reason.
Apple would need to design a RapidIO I/O ASIC for USB, Firewire, audio, SATA/ATA, WiFi, Bluetooth, etc. They could do it, but it would be money that also could be spent elsewhere.
Secondly, the 7448 uses 10W typical @ 1.4Ghz the 7447A uses 21W at the same clock.
I would be very very careful of any advertised power consumption specs for 90 nm chips, especially from a company who hasn't produced a real 90 nm part yet.
How is not that an improvement, aspecially concidering that the 7448 gets a 100% larger L2 cache and a ~20% faster fsb.
Not to mention that the .pdf I linked to shows that it's going to clock to "atleast" 1.5Ghz. (Looks like Freescale has leant the merit of being conservative in there preformance estimates)
It's an improvement no doubt. But they are rumored to sample in 1H of 05. June 30, 2005 is in 1H 05. Production quantities may not even come until late Q3 05. That's the problem. Freescale will ship the 7448 ~8 months too late, assuming everything will go all right with their 90 nm fab.
If laptops are a bigger market for Apple, maybe they should start shipping a larger product line. They have 3 different desktop Macs, why not 3 different laptop Macs?
Originally posted by PB
Freescale is not going to sample the 7448 before H1 2005. So, what Apple has in the hands now for the next Powerbook update? I don't see anything, except perhaps a silent minor speed bump in the 7447A family.
I wouldn't put any faith into what EE-Times has to say. They are about as reliable as MOSR, if that.
Originally posted by TWinbrook46636
I wouldn't put any faith into what EE-Times has to say. They are about as reliable as MOSR, if that.
You can't be serious - eeTimes is an Electronic Engineering site that publishes news stories. I don't think I've ever seen an article based on speculation and rumor and heresay.
Originally posted by curiousuburb
Except that the 8641D isn't pin compatible with current mobos (the 7448 is).
The 8641D isn't intended for computing, it's intended for embedded markets.
Hey guys the whole world is a market for embedded devices, lets knock it off with respect to intended markets. Much of what we are likely to get at anyone time is the result of supply. That is if Apple had a low power G5 that meant all of their requirements you can be sure that it would end up in a portable of some type soon after it came into existance.
The bigger issue that I think some here are missing is that there is currently a market for both processors. The G5 can go inot laptops for people that often use rubber rulers and the G4 products can go into people that need real portable performance.
And the 8641D won't sample until the 2nd half '05.
Actually I hope this is a sign that Freescales is maturing quickly and not publicly promising more than they can deliver. It is refreshing to see that some of the e500 devices have started shipping. One can always hope.
And typical Moto... more than a year since the 1.5GHz 7447 and they've bumped...
100 MHz! Woot!!!
Frankly that does suck. On the other hand the high integration devices do appear to be customed mad for Apples need. PCI-Express and a built in memory controller would have been fantastic for a portable, but the new chips have even more goodies. Frankly Apple could build one hell of an iBook with one of these.
The question is can IBM deliver similar products at similar power / performance points. Right now they can't but it will be very interesting to see what is happening in the new year.
Welcome to the 90's again! [/B]
Originally posted by THT
Are there going to be any graphics chipsets available that support PCIe x8? If not, there's one reason.
*groan* You could hook up a PCI-E x16 interface to a x1 line if you wanted and it'd still work. Infact I'd say that it's better then a full x16 interface because, 1) it'll drain less energy, 2) it do is enough bandwidth, people have trouble seeing difrences between AGP x4 and x8, PCI-E has got, full duplex, roughly the double bandwidth of AGP x8. 3) Being on chip ought to give the GFX interface a lower latency access to the CPU wich ought to give it a slight preformance boost.
Apple would need to design a RapidIO I/O ASIC for USB, Firewire, audio, SATA/ATA, WiFi, Bluetooth, etc. They could do it, but it would be money that also could be spent elsewhere.
With That is what they'd have to build an ASIC for, there is no etc. The DMA and interrupt controller is on the CPU, as is the GFX interface (and of course the memory controller. Oh and if they wanted, they could make it a PCI-E interface instead.
I would be very very careful of any advertised power consumption specs for 90 nm chips, especially from a company who hasn't produced a real 90 nm part yet.
Corolles is producing 90nm parts and is sampling 65nm parts though.
[edited]Don't post early in the morning (like I did)...[/edited]
I'm going to order me a Powerbook soon, but I was planning on waiting to the next Rev. But hell, I'm not waiting THAT long.
I might as well order my Powerbook today..
Originally posted by nooon
Argh!
I'm going to order me a Powerbook soon, but I was planning on waiting to the next Rev. But hell, I'm not waiting THAT long.
I might as well order my Powerbook today..
Ahh, a sense of reason. As we've said before, need always counts more than want, even if the want is a G5 PowerBook
Originally posted by nooon
Argh!
I'm going to order me a Powerbook soon, but I was planning on waiting to the next Rev. But hell, I'm not waiting THAT long.
I might as well order my Powerbook today..
If you can wait, I would recommend waiting until the end of October to see if there is an update of any kind. Even a 100MHZ bump would make the current PBs a little bit cheaper so you could save some money on them. However, if you need a notebook and your are set on an Apple (there are reasons outside of processor speed to buy them) then I'd recommend plopping down the cash.
There is a possibility of low-voltage G5 availability before the year is out, though I wouldn't hold my breath.
Originally posted by Eric_Z
*groan*
Hehe. You asked. I gave it my best shot.
You could hook up a PCI-E x16 interface to a x1 line if you wanted and it'd still work. ...
That is what they'd have to build an ASIC for, there is no etc. The DMA and interrupt controller is on the CPU, as is the GFX interface (and of course the memory controller. Oh and if they wanted, they could make it a PCI-E interface instead.
It could work. I agree with you. I was promoting this idea 4 years ago myself. But I'm not getting a warm fuzzy with Apple using a SoC like this yet.
Corolles is producing 90nm parts and is sampling 65nm parts though.
Which 90 nm parts is Crolles producing? I know about the 90 nm Philips ARM part in sampling, with expected volume production in Q4 04, but what else? Freescale is saying that the PowerQuicc III 90 nm parts won't be sampling until Q2 05, so I think they have a ways to go with the fab.
Freescale itself (slide 41 of SNDF 04 presentation) says that they will sample the 7448 in 1H 05 and have volume production in 2H 05. The 8641D will sample in the 2H 05 and have volume production in 1H 06. That's a lot time for IBM to come out with low voltage 970fx, 970mp and 970 w/on-die mem controller parts.