"How does the Mactel box do on floating point, 64bit and/or vector based code?"
It's using a Pentium 4 660 3.6 ghz (IIRC), which is generally faster on floating point code than G5s. Emulation of floating point code is slow because x86 and PowerPC floating point instructions do not have exactly the same behavior but native floating point code is fast.
G5s are also known for the weakness of their Altivec implementation and need a significant clock speed advantage to outperform G4s in that area. I don't know how they compare with P4s with vector code, but the gulf isn't as wide as you'd think.
The dev boxes use a chip that supports x86-64. It doesn't have 64-bit support because Apple doesn't support it at the OS level.
"The main reason for getting a G5 was to improve performance of 64bit/floating/vector code like is used in video production and scientific apps.
The main reason for the G5 was because Motorola was falling so far behind that people were switching to Windows. Same reason as the switch to Intel.
"Since Intel has always been shaky in floating point"
If by "always been really shaky" you mean "shaky for a few months in 1994", and by "shaky for a few months in 1994" you mean "shaky for a few months in 1994 before they replaced the chips for free", then yes.
"and probably doesn't really know the meaning of vector"
One of Rosetta's most important features is the ability to translate library calls such that they can go to native libraries. That speeds things up a lot, but the impact is not evenly distributed. Applications that rely on internal routines performing well will be hurt more.
Also, things like floating-point heavy code are hurt more.
Pentium Ms are known for very nice battery lives, as others have posted 7+ hours is not unheard of. They also comfortably putperform G4s, and while a G5 laptop might be able to keep up it won't be able to do it at 1.6 ghz.
The new Yonah core chips Intel will be releasing early next year also have a dual-core version in the same power envelope...
Yes, the MacOS port of Firefox is an embarrasment. The interface is responsive in the same way that steering a rocket by sending olfactory messages to slugs is responsive.
TDP is the maximum power usage, not the typical power usage. Actual power usage must be strictly less than that value, as that's what vendors use to design cooling solutions. IBM's numbers give the typical power usage.
Also, you'll note that there's a range of chips there with a TDP of 21 watts. Given the range of clock speeds, I'd be pretty surprised if none of them fell below 16 watts in typical usage.
Also, Intel sells low and ultralow voltage Pentium M chips with TDPs of 10 watts and 5 watts. 1.4 ghz at 10 watts isn't too bad IMO.
That's a more plausible alternative than reversing the switch, but IBM isn't coming up with anything that can keep up on the small server or desktop either.
Even these announced dual-core G5s won't be enough to overcome already shipping dual-core x86 chips, and I'm assuming those won't be standing still. IBM just isn't doing what it takes to stay ahead in desktops/laptops.
Lower speed Pentium Ms also consume less power. I believe 2.13 ghz is currently the fastest Pentium M.
Besides, Apple isn't going to use these things. They'd have to redesign PowerBook chipsets and motherboards for a computer that, at best, they'd be selling for less than 2 years. It's much more likely that they'll transition all current G4 computers to Pentium Ms first.
These just-announced FX chips compete with the lower end of Intel laptop chips, while the Yonah-core chips Intel is releasing early next year will improve in every way. Higher clock speeds, lower power consumption, and dual-core versions that fit within current power envolopes.
You might prefer a 1.6 ghz G5 to a 2 ghz Pentium M, but would you prefer it to a dual-core 2 ghz Pentium M that takes the same amount of power?
" Well, sure, if you're talking about 970FXs at similar clock speeds. On the other hand, we can't be sure how heavily modified these processors were modified to get them to such low power consumption - they could be even slower, or even faster."
I'm guessing it's on the slow side, but it would take a lot to make a G4 with its bus preferable.
"Hmm... perhaps Apple's differentiation between the iBooks and Powerbooks will be dual-core Yonahs in the latter, and single-core PowerPCs in the iBooks? That would be kinda cool, although I'm aware it'd also probably be a terrible way to go."
Nah. It's actually a distinction that makes sense. Single-core chips are much cheaper, and they're not really needed in most places yet.
As the number of cores on chips increases, lower numbers of cores will always be cheaper because the dies will be smaller.
IBM releasing chips that can compete with the low end of today doesn't mean much when Intel will be releasing higher speed, lower power, dual-core laptop chips early next year.
IBM is releasing laptop chips that fit into the lower end of the spectrum of current chips while Intel will be releasing the next generation early next year.
Yonah-core Pentium Ms include floating-point improvements (the Pentium M's current weakness), clock speed improvements, power improvements, and there will be dual-core versions in the same power envelope as current chips.
A single 1.6 ghz G5 might be welcome on PowerBooks (particularly since it replaces the archaic bus), but it's not going to stack up very well against a dual-core 2 ghz Pentium M. IBM simply is not willing to put in the kind of R&D it takes to keep up with Intel on laptop chips. IBM might be not-quite-so-behind now, but brief periods where their chips are not a crippling weakness for Apple are not enough to change the decision.
Pentium Ms perform a bit better than G5 chips at similar clock speeds, and Pentium Ms scale to much higher clock speeds than these lower power FX chips.
The new FX chips would probably be a welcome replacement for G4s if only to replace the archaic bus (though I doubt Apple will bother), but they're not good enough to replace Intel's current laptop chips, much less the future chips.
Intel is releasing Yonah-core Pentium Ms early next year. They're going to address the Pentium M's floating point weaknesses, increase clock speed, lower power usage, and there will be dual-core versions within the same power budget as current chips. I don't think a single G5 at 1.6 ghz would do too well against a dual-core 2 ghz Pentium M.
Basically, IBM is releasing chips that fit into the lower end of current laptop chips a few months before Intel releases the next generation.
"Speaking of Roadmaps... Why not would Apple not switch to AMD? AMD's chips run with less power consumption and way less number of transistors. When comparing the Dual core chips from AMD and Intel, AMD wins on power consumption. But I thought Jobs said Intel had the best Performance per watt? ADA4800DAA6CD (AMD Dual core 64-bit): 110W Intel® Pentium® Processor Extreme Edition: 130W These are the latest and greatest from Intel and AMD right?"
This has been discussed about a million times on any site that posted any news about the switch.
You're wrong in two ways. First, you don't understand what Apple cares about. Second, you don't understand the situation in the area that Apple cares about.
Apple cares more about laptops. Intel wins easily in this area. They beat every current or planned PowerPC laptop chip, and they beat every AMD laptop chip. There's basically no serious competition at this point (AMD is trying but they're not yet serious competition).
AMD wins on power consumption on the desktop right now, but Apple cares more about laptops and also Intel is going to be moving their laptop chips into the desktop because the P4s have dead-ended. In the 2006-2007 timeframe, Intel is going to have very powerful multi-core low power chips on laptops and desktops.
Intel supplies chipsets as well, and their chipsets are pretty nice. They're not always the best, but they're usually close and they're almost always better in laptops. Having the chipset provided by Intel cuts down on engineering costs as well, which is important for Apple. Their volumes are small by most OEM and motherboard maker standards.
"Is the real reason we backed away from manned exploration in the 70's because the "right" people weren't making a profit off of it? It could well be that, until there is a consistent and projectable profit to be had from the manned exporation of space, we will be stuck back here on Earth. I memorized every announced space launch, manned or otherwise, when I was a kid. I looked forward to life in space, or at least something better from the space program than satellite TV and phone service. A terribly sad situation all around."
I'm not sure about exploration, but I think there will be a profit to be made from space-based solar power in the next few decades, say before 2050 or so. Long before commercial fusion power happens.
If that happens, the capability to launch large exploration missions will be commercially available.
"But you say "it's trivial". Um, maybe for you. But how many typical Chinese users know how to do this, or even that this could be done, or even that they need to do this?"
I imagine there's people that know how over there, but I can imagine that they wouldn't want to go around telling everyone.
Something I do for wireless networks is set up an SSH tunnel to a proxy server at home, and browse through that. If China lets people SSH out, it's trivial to bounce connections off of an outside computer.
Doesn't that make circumventing the firewall pretty easy? People could just make an SSH tunnel to a proxy server. I do that on public networks and they're not trying to stop me from getting out.
"You know, because it would be horrible to have this as an emmissions-free source of energy. Incredible."
You're missing the point. Even if ITER works, we're still many decades away from commercial fusion power. More improvement would result from spending that money in optimizing what we already have.
Also, IIRC Greenpeace grudgingly supports nuclear technology because it's the lesser evil.
Slashdot Mirror
The altivec implementation on G5s is weak compared to that on G4s, so the advantage isn't as great.
No, the rationalization is that it's a recent development.
"How does the Mactel box do on floating point, 64bit and/or vector based code?"
It's using a Pentium 4 660 3.6 ghz (IIRC), which is generally faster on floating point code than G5s. Emulation of floating point code is slow because x86 and PowerPC floating point instructions do not have exactly the same behavior but native floating point code is fast.
G5s are also known for the weakness of their Altivec implementation and need a significant clock speed advantage to outperform G4s in that area. I don't know how they compare with P4s with vector code, but the gulf isn't as wide as you'd think.
The dev boxes use a chip that supports x86-64. It doesn't have 64-bit support because Apple doesn't support it at the OS level.
"The main reason for getting a G5 was to improve performance of 64bit/floating/vector code like is used in video production and scientific apps.
The main reason for the G5 was because Motorola was falling so far behind that people were switching to Windows. Same reason as the switch to Intel.
"Since Intel has always been shaky in floating point"
If by "always been really shaky" you mean "shaky for a few months in 1994", and by "shaky for a few months in 1994" you mean "shaky for a few months in 1994 before they replaced the chips for free", then yes.
"and probably doesn't really know the meaning of vector"
SSE is what, 6 years old now?
The two are not necessarily contradictory.
One of Rosetta's most important features is the ability to translate library calls such that they can go to native libraries. That speeds things up a lot, but the impact is not evenly distributed. Applications that rely on internal routines performing well will be hurt more.
Also, things like floating-point heavy code are hurt more.
Pentium Ms are known for very nice battery lives, as others have posted 7+ hours is not unheard of. They also comfortably putperform G4s, and while a G5 laptop might be able to keep up it won't be able to do it at 1.6 ghz.
The new Yonah core chips Intel will be releasing early next year also have a dual-core version in the same power envelope...
Yes, the MacOS port of Firefox is an embarrasment. The interface is responsive in the same way that steering a rocket by sending olfactory messages to slugs is responsive.
"apple cares about laptops?"
They'd be stupid not to. Laptops now dominate new computer sales.
"all that fuss about imacs and what not"
They focus on iMacs because their laptops are behind by a factor of three in performance terms, and a shorter battery life.
TDP is the maximum power usage, not the typical power usage. Actual power usage must be strictly less than that value, as that's what vendors use to design cooling solutions. IBM's numbers give the typical power usage.
Also, you'll note that there's a range of chips there with a TDP of 21 watts. Given the range of clock speeds, I'd be pretty surprised if none of them fell below 16 watts in typical usage.
Also, Intel sells low and ultralow voltage Pentium M chips with TDPs of 10 watts and 5 watts. 1.4 ghz at 10 watts isn't too bad IMO.
That's a more plausible alternative than reversing the switch, but IBM isn't coming up with anything that can keep up on the small server or desktop either.
Even these announced dual-core G5s won't be enough to overcome already shipping dual-core x86 chips, and I'm assuming those won't be standing still. IBM just isn't doing what it takes to stay ahead in desktops/laptops.
Lower speed Pentium Ms also consume less power. I believe 2.13 ghz is currently the fastest Pentium M.
Besides, Apple isn't going to use these things. They'd have to redesign PowerBook chipsets and motherboards for a computer that, at best, they'd be selling for less than 2 years. It's much more likely that they'll transition all current G4 computers to Pentium Ms first.
These just-announced FX chips compete with the lower end of Intel laptop chips, while the Yonah-core chips Intel is releasing early next year will improve in every way. Higher clock speeds, lower power consumption, and dual-core versions that fit within current power envolopes.
You might prefer a 1.6 ghz G5 to a 2 ghz Pentium M, but would you prefer it to a dual-core 2 ghz Pentium M that takes the same amount of power?
" Well, sure, if you're talking about 970FXs at similar clock speeds. On the other hand, we can't be sure how heavily modified these processors were modified to get them to such low power consumption - they could be even slower, or even faster."
I'm guessing it's on the slow side, but it would take a lot to make a G4 with its bus preferable.
"Hmm... perhaps Apple's differentiation between the iBooks and Powerbooks will be dual-core Yonahs in the latter, and single-core PowerPCs in the iBooks? That would be kinda cool, although I'm aware it'd also probably be a terrible way to go."
Nah. It's actually a distinction that makes sense. Single-core chips are much cheaper, and they're not really needed in most places yet.
As the number of cores on chips increases, lower numbers of cores will always be cheaper because the dies will be smaller.
Doesn't matter.
IBM releasing chips that can compete with the low end of today doesn't mean much when Intel will be releasing higher speed, lower power, dual-core laptop chips early next year.
They're not going to reconsider.
IBM is releasing laptop chips that fit into the lower end of the spectrum of current chips while Intel will be releasing the next generation early next year.
Yonah-core Pentium Ms include floating-point improvements (the Pentium M's current weakness), clock speed improvements, power improvements, and there will be dual-core versions in the same power envelope as current chips.
A single 1.6 ghz G5 might be welcome on PowerBooks (particularly since it replaces the archaic bus), but it's not going to stack up very well against a dual-core 2 ghz Pentium M. IBM simply is not willing to put in the kind of R&D it takes to keep up with Intel on laptop chips. IBM might be not-quite-so-behind now, but brief periods where their chips are not a crippling weakness for Apple are not enough to change the decision.
Pentium Ms perform a bit better than G5 chips at similar clock speeds, and Pentium Ms scale to much higher clock speeds than these lower power FX chips.
The new FX chips would probably be a welcome replacement for G4s if only to replace the archaic bus (though I doubt Apple will bother), but they're not good enough to replace Intel's current laptop chips, much less the future chips.
Intel is releasing Yonah-core Pentium Ms early next year. They're going to address the Pentium M's floating point weaknesses, increase clock speed, lower power usage, and there will be dual-core versions within the same power budget as current chips. I don't think a single G5 at 1.6 ghz would do too well against a dual-core 2 ghz Pentium M.
Basically, IBM is releasing chips that fit into the lower end of current laptop chips a few months before Intel releases the next generation.
"Speaking of Roadmaps... Why not would Apple not switch to AMD? AMD's chips run with less power consumption and way less number of transistors. When comparing the Dual core chips from AMD and Intel, AMD wins on power consumption. But I thought Jobs said Intel had the best Performance per watt? ADA4800DAA6CD (AMD Dual core 64-bit): 110W Intel® Pentium® Processor Extreme Edition: 130W These are the latest and greatest from Intel and AMD right?"
This has been discussed about a million times on any site that posted any news about the switch.
You're wrong in two ways. First, you don't understand what Apple cares about. Second, you don't understand the situation in the area that Apple cares about.
Apple cares more about laptops. Intel wins easily in this area. They beat every current or planned PowerPC laptop chip, and they beat every AMD laptop chip. There's basically no serious competition at this point (AMD is trying but they're not yet serious competition).
AMD wins on power consumption on the desktop right now, but Apple cares more about laptops and also Intel is going to be moving their laptop chips into the desktop because the P4s have dead-ended. In the 2006-2007 timeframe, Intel is going to have very powerful multi-core low power chips on laptops and desktops.
Intel supplies chipsets as well, and their chipsets are pretty nice. They're not always the best, but they're usually close and they're almost always better in laptops. Having the chipset provided by Intel cuts down on engineering costs as well, which is important for Apple. Their volumes are small by most OEM and motherboard maker standards.
Apple isn't going to transition PowerMacs for a long time. These chips will probably make one more buying surge in the PowerMac line possible.
Unfortunately, as a niche browser Safari is always going to be behind Firefox in support from 3rd parties.
"Is the real reason we backed away from manned exploration in the 70's because the "right" people weren't making a profit off of it? It could well be that, until there is a consistent and projectable profit to be had from the manned exporation of space, we will be stuck back here on Earth. I memorized every announced space launch, manned or otherwise, when I was a kid. I looked forward to life in space, or at least something better from the space program than satellite TV and phone service. A terribly sad situation all around."
I'm not sure about exploration, but I think there will be a profit to be made from space-based solar power in the next few decades, say before 2050 or so. Long before commercial fusion power happens.
If that happens, the capability to launch large exploration missions will be commercially available.
"But you say "it's trivial". Um, maybe for you. But how many typical Chinese users know how to do this, or even that this could be done, or even that they need to do this?"
I imagine there's people that know how over there, but I can imagine that they wouldn't want to go around telling everyone.
Doesn't that render the entire thing futile?
Something I do for wireless networks is set up an SSH tunnel to a proxy server at home, and browse through that. If China lets people SSH out, it's trivial to bounce connections off of an outside computer.
Doesn't that make circumventing the firewall pretty easy? People could just make an SSH tunnel to a proxy server. I do that on public networks and they're not trying to stop me from getting out.
I imagine they stop you from SSHing out?
I'm giving my understanding on Greenpeace's stance, not knocking ITER. I don't personally think it's a waste.
I agree with you, but Greenpeace has made a different take.
My take is that ITER doesn't live in the same world as the other technologies. It's a big budget science project like a space program.
"You know, because it would be horrible to have this as an emmissions-free source of energy. Incredible."
You're missing the point. Even if ITER works, we're still many decades away from commercial fusion power. More improvement would result from spending that money in optimizing what we already have.
Also, IIRC Greenpeace grudgingly supports nuclear technology because it's the lesser evil.