That's too bad, since it doesn't answer the question. I didn't ask about inner workings, I asked about actual numbers. Real-world bare facts, not theory and should-bes.
I can tell you that it's very hard to get. The price per pound doesn't reflect the price of the material. It costs far more than that.
While this is industrial retail, I buy from them. Check out the prices. wide sheets are much more per inch.
...
A little bit of an offshoot update, but my dad bought a square yard of carbon fiber cloth for $21 from his supplier that he usually gets his fiberglass materials. Fiberglass cloth of same thickness is $9/yd, it doesn't seem to be too big of a cost difference to me. I didn't ask why he got it though.
A little bit of an offshoot update, but my dad bought a square yard of carbon fiber cloth for $21 from his supplier that he usually gets his fiberglass materials. Fiberglass cloth of same thickness is $9/yd, it doesn't seem to be too big of a cost difference to me. I didn't ask why he got it though.
That's a hell of a difference in cost to a manufacturer, who measures his costs in fraction of pennies. It's almost 2 1/2 times the price. Cloth is cheaper to buy than stiff carbon fiber sheet which is usually used for airframes and other impact resistent objects. They are then heat, and vacuum formed, onto a mold. The cloth needs further processing, making it more expensive to use because of the extra materials, and processing steps.
The plastics used for computer cases, including laptops, cost no more than a fraction of a dollar for the entire case, and are injection molded, which is a much cheaper, and common, process.
The other difference is that materials such as cloths, and sheets must be trimmed, and the excess discarded, which can amount to a large fraction of the material, effectively raising the cost appreciably. Injection molded materials have little waste, other than trimming off the sprue's and, rarely, removing the tiny amounts of plastic that sometimes result from mold halves not matching exactly. Often, that waste is recycled into a later batch.
Yes it does. Don't forget, it's the North Bridge that addresses the RAM, and the current North Bridge is part of Napa, a 32 bit chipset. The 3.2 GB problem will go away with Santa Rosa.
No it doesn't. This is an architectural limitation, the chipset has nothing to do with it. If you dump the memory-mapped IO you can address full 4 GB (minus a couple of megabytes, perhaps), but then you get a huge performance hit on every basic operation.
Additionally, the 3.2 GB value is taken out of the blue, this depends on hardware configuration.
No it doesn't. This is an architectural limitation, the chipset has nothing to do with it. If you dump the memory-mapped IO you can address full 4 GB (minus a couple of megabytes, perhaps), but then you get a huge performance hit on every basic operation.
Additionally, the 3.2 GB value is taken out of the blue, this depends on hardware configuration.
With Leopard, though, the system will run in 64-bit mode where possible/applicable, therefore allowing a much larger addressing space ?*and consequently allowing the limit to be bumped up considerably.
With Leopard, though, the system will run in 64-bit mode where possible/applicable, therefore allowing a much larger addressing space ?*and consequently allowing the limit to be bumped up considerably.
But still, nothing to do with Santa Rosa.
I think it's a question of how is Mac Pro beating that barrier but the notebooks can't.
I think it's a question of how is Mac Pro beating that barrier but the notebooks can't.
Because the Mac Pro uses an x86 technique whose name eludes me right now that allows a ceiling higher than 4 GBs despite 32-bit. So, the problem is merely shifted a few notches higher.
Whereas, for the other Macs, that's not an option either; the chipset doesn't implement this, and it wouldn't be very useful as you'd need more than two slots. (Unless you want Apple to start doing soldered RAM again. E.g., 1 GB soldered, 2x2 slotted.)
Because the Mac Pro uses an x86 technique whose name eludes me right now that allows a ceiling higher than 4 GBs despite 32-bit. So, the problem is merely shifted a few notches higher.
Whereas, for the other Macs, that's not an option either; the chipset doesn't implement this, and it wouldn't be very useful as you'd need more than two slots. (Unless you want Apple to start doing soldered RAM again. E.g., 1 GB soldered, 2x2 slotted.)
Can it be pushed to the top of installed memory? Say, to the 7th GB for 8 GB installs, or 15 GB, for16GB installs?
Can it be pushed to the top of installed memory? Say, to the 7th GB for 8 GB installs, or 15 GB, for16GB installs?
I wish I could give you more details, but without knowing the name, I can't look it up (my attempts at finding it have failed so far), and I don't have sufficient personal experience.
Comments
That's not an answer.
http://news.google.com/news?q=strain...n+~performance
http://www.google.com/search?q=strai...n+~performance
http://scholar.google.com/scholar?q=...n+~performance
That's my final answer.
http://news.google.com/news?q=strain...n+~performance
http://www.google.com/search?q=strai...n+~performance
http://scholar.google.com/scholar?q=...n+~performance
That's my final answer.
That's too bad, since it doesn't answer the question. I didn't ask about inner workings, I asked about actual numbers. Real-world bare facts, not theory and should-bes.
http://news.google.com/news?q=strain...n+~performance
http://www.google.com/search?q=strai...n+~performance
http://scholar.google.com/scholar?q=...n+~performance
That's my final answer.
Maybe your final answer, but not to the questions asked!
Why should I have to find your answers?
http://xtreview.com/review139.htm
3% performance increase for the shipping x1700 model. Enjoy!
ps: I highly recommend this latest version of OmniWeb. Very snappy™!
Your final answer? Here's a lifeline:
http://xtreview.com/review139.htm
3% performance increase for the shipping x1700 model. Enjoy!
ps: I highly recommend this latest version of OmniWeb. Very snappy?!
Not too impressive, is it?
Clock the memory up more than the performance numbers.
I can tell you that it's very hard to get. The price per pound doesn't reflect the price of the material. It costs far more than that.
While this is industrial retail, I buy from them. Check out the prices. wide sheets are much more per inch.
...
A little bit of an offshoot update, but my dad bought a square yard of carbon fiber cloth for $21 from his supplier that he usually gets his fiberglass materials. Fiberglass cloth of same thickness is $9/yd, it doesn't seem to be too big of a cost difference to me. I didn't ask why he got it though.
A little bit of an offshoot update, but my dad bought a square yard of carbon fiber cloth for $21 from his supplier that he usually gets his fiberglass materials. Fiberglass cloth of same thickness is $9/yd, it doesn't seem to be too big of a cost difference to me. I didn't ask why he got it though.
That's a hell of a difference in cost to a manufacturer, who measures his costs in fraction of pennies. It's almost 2 1/2 times the price. Cloth is cheaper to buy than stiff carbon fiber sheet which is usually used for airframes and other impact resistent objects. They are then heat, and vacuum formed, onto a mold. The cloth needs further processing, making it more expensive to use because of the extra materials, and processing steps.
The plastics used for computer cases, including laptops, cost no more than a fraction of a dollar for the entire case, and are injection molded, which is a much cheaper, and common, process.
The other difference is that materials such as cloths, and sheets must be trimmed, and the excess discarded, which can amount to a large fraction of the material, effectively raising the cost appreciably. Injection molded materials have little waste, other than trimming off the sprue's and, rarely, removing the tiny amounts of plastic that sometimes result from mold halves not matching exactly. Often, that waste is recycled into a later batch.
I'm not real familar with carbon fiber material. What would the advantages be of using it as the exterior case material?
Weight (lighter) and strength (won't dent).
Yes it does. Don't forget, it's the North Bridge that addresses the RAM, and the current North Bridge is part of Napa, a 32 bit chipset. The 3.2 GB problem will go away with Santa Rosa.
No it doesn't. This is an architectural limitation, the chipset has nothing to do with it. If you dump the memory-mapped IO you can address full 4 GB (minus a couple of megabytes, perhaps), but then you get a huge performance hit on every basic operation.
Additionally, the 3.2 GB value is taken out of the blue, this depends on hardware configuration.
No it doesn't. This is an architectural limitation, the chipset has nothing to do with it. If you dump the memory-mapped IO you can address full 4 GB (minus a couple of megabytes, perhaps), but then you get a huge performance hit on every basic operation.
Additionally, the 3.2 GB value is taken out of the blue, this depends on hardware configuration.
With Leopard, though, the system will run in 64-bit mode where possible/applicable, therefore allowing a much larger addressing space ?*and consequently allowing the limit to be bumped up considerably.
But still, nothing to do with Santa Rosa.
I'm not real familar with carbon fiber material. What would the advantages be of using it as the exterior case material?
It's stronger, and has more torsional resistance (twisting, remember the titanium?). It also weighs less than equal material.
With Leopard, though, the system will run in 64-bit mode where possible/applicable, therefore allowing a much larger addressing space ?*and consequently allowing the limit to be bumped up considerably.
But still, nothing to do with Santa Rosa.
I think it's a question of how is Mac Pro beating that barrier but the notebooks can't.
Weight (lighter) and strength (won't dent).
But, it does crack at a sharp point of impact, which polycarbonate won't.
I think it's a question of how is Mac Pro beating that barrier but the notebooks can't.
Does it, though?
The region above 3.2 GB and 4Gb is fine.
I think it's a question of how is Mac Pro beating that barrier but the notebooks can't.
Because the Mac Pro uses an x86 technique whose name eludes me right now that allows a ceiling higher than 4 GBs despite 32-bit. So, the problem is merely shifted a few notches higher.
Whereas, for the other Macs, that's not an option either; the chipset doesn't implement this, and it wouldn't be very useful as you'd need more than two slots. (Unless you want Apple to start doing soldered RAM again. E.g., 1 GB soldered, 2x2 slotted.)
Does it, though?
The region above 3.2 GB and 4Gb is fine.
So a person that buys 4GB with a Mac Pro is going to lose nearly a gig?
Because the Mac Pro uses an x86 technique whose name eludes me right now that allows a ceiling higher than 4 GBs despite 32-bit. So, the problem is merely shifted a few notches higher.
Whereas, for the other Macs, that's not an option either; the chipset doesn't implement this, and it wouldn't be very useful as you'd need more than two slots. (Unless you want Apple to start doing soldered RAM again. E.g., 1 GB soldered, 2x2 slotted.)
Can it be pushed to the top of installed memory? Say, to the 7th GB for 8 GB installs, or 15 GB, for16GB installs?
So a person that buys 4GB with a Mac Pro is going to lose nearly a gig?
I don't know, it was just an observation.
Can it be pushed to the top of installed memory? Say, to the 7th GB for 8 GB installs, or 15 GB, for16GB installs?
I wish I could give you more details, but without knowing the name, I can't look it up (my attempts at finding it have failed so far), and I don't have sufficient personal experience.