My point was that the bottom surface of a hot object has the worst thermal transfer coefficient, in non-circulated air. Actually moving air would be better, but this product really doesn't seem to try very hard at that either.
It's not ideal. But the fans are no doubt the largest that could fit. They also take up power. Being mechanical devices, no matter how efficient they are, the draw would be considerable if they were to be on for a large percentage of the time. The battery itself gets hot as current is being drawn.
As I said, there isn't a better place on the machine that could sink the heat other than the bottom. would you rather they sunk it on the top, through the keyboard?
As I said, there isn't a better place on the machine that could sink the heat other than the bottom. would you rather they sunk it on the top, through the keyboard?
Actually they do that too. There is a considerable amount of heat going to the keyboard. It's not as much of a problem as there is no heat trapping going on.
Actually they do that too. There is a considerable amount of heat going to the keyboard. It's not as much of a problem as there is no heat trapping going on.
But that isn't intentional. Some of the heat rises through the keyboard because it has no other place to go. Heat trapping isn't 100% efficient.
But that isn't intentional. Some of the heat rises through the keyboard because it has no other place to go. Heat trapping isn't 100% efficient.
Regardless, I've looked around, the part between the keyboard and screen exceeds 120 F and has been the subject of numerous complaints.
I've looked around, the claim that using the fan would waste too much power doesn't hold much water. I found a few small ones that are quiet and only consume 0.5 W. Still, something like this wouldn't be as good as not turning electrical power into heat in the first place, especially at idle. The amount of heat generated at idle is the reason I'm not convinced that SpeedStep is being used properly.
Regardless, I've looked around, the part between the keyboard and screen exceeds 120 F and has been the subject of numerous complaints.
I've looked around, the claim that using the fan would waste too much power doesn't hold much water. I found a few small ones that are quiet and only consume 0.5 W. Still, something like this wouldn't be as good as not turning electrical power into heat in the first place, especially at idle. The amount of heat generated at idle is the reason I'm not convinced that SpeedStep is being used properly.
It's aomething that Apple has no control over. The chip makes all of the decisions. So you have to blame Intel on that one.
Look, I'm with you on the heat thing. I think that the machines could be a bit thicker, and and as a result, have a better internal path for the heat to be withdrawn through. But, as I said, thin is in.
As far as the fans go, I don't know what models Apple uses, or the cfm. But even if they use .5 watt, there are two of them, and even that 1 watt would draw that battery down if they were on much more often. also, if you read many comments in the thread, you will see that people just HATE it when the fan comes on for long periods. So, one just can't win with this.
Don't forget to say that you want to be able to OMG UPGRADE TEH VIDEO CARD and OMG UPGRADE TEH PROCESSOR and OMG UPGRADE TEH HARD DRIVE and OMG I WANT MORE PIXELS and OMG WTF NO PCI SLOT.
It's aomething that Apple has no control over. The chip makes all of the decisions. So you have to blame Intel on that one.
Did Intel remove the software controllability of SpeedStep? I haven't used a Pentium-M or Core anything laptop, but my PIIIm (1.2GHz) only needed a simple little applet (SpeedSwitch XP for Windows XP) to force any SpeedStep capable laptop into the slow clock. That unit didn't need it so much because it doesn't run hot. That laptop was very thin for the time and light too, 1.25" thick and 5lb.
Did Intel remove the software controllability of SpeedStep? I haven't used a Pentium-M or Core anything laptop, but my PIIIm (1.2GHz) only needed a simple little applet (SpeedSwitch XP for Windows XP) to force any SpeedStep capable laptop into the slow clock. That unit didn't need it so much because it doesn't run hot. That laptop was very thin for the time and light too, 1.25" thick and 5lb.
It's not so much as removed it, as this being a new series, it was never put in. That doesn't mean that some hack doesn't exist that might allow it. But, right now, at least, Intel doesn't support it.
What speed machine do you have? The faster, the hotter, after all.
As I said, there isn't a better place on the machine that could sink the heat other than the bottom. would you rather they sunk it on the top, through the keyboard?
Of course not. MBPs should have liquid cooling, with a radiator on the back of the lid.
It's not so much as removed it, as this being a new series, it was never put in. That doesn't mean that some hack doesn't exist that might allow it. But, right now, at least, Intel doesn't support it.
What speed machine do you have? The faster, the hotter, after all.
I wasn't aware of that.
I bought the 1.83MHz version. I was tempted to go higher, but I am thankful I didn't. I had expected that the increased power saving of the latest SpeedStep, combined with the transistor size shrinks (.065 vs .13 on the old notebook) would allow enough efficiency improvements to offset the clock speed increase.
I did find some posts where people claimed that the 1.0.1 firmware helped so I installed that for completeness sake, but the Apple info makes no claims to that effect. That isn't to say that Apple didn't sneak in some bug fixes without saying anything, the only thing Apple claims about 1.0.1 is to support languages that read right to left.
Just increase the speed in windows and leave it alone. Apple dynamically adjusts it however they have been runnning it somewhat slower for most things.
I bought the 1.83MHz version. I was tempted to go higher, but I am thankful I didn't. I had expected that the increased power saving of the latest SpeedStep, combined with the transistor size shrinks (.065 vs .13 on the old notebook) would allow enough efficiency improvements to offset the clock speed increase.
I did find some posts where people claimed that the 1.0.1 firmware helped so I installed that for completeness sake, but the Apple info makes no claims to that effect. That isn't to say that Apple didn't sneak in some bug fixes without saying anything, the only thing Apple claims about 1.0.1 is to support languages that read right to left.
Too bad they didn't continue to offer the 1.67. You know, if the Speedstepping IS working correctly, and there isn't any reason to believe that it isn't, just think how hot it would get without it!
Apple probably stuck some very minor fixes in, but if it were anyrthing that a user, or programmer, would see, they would have mentioned it.
Did Intel remove the software controllability of SpeedStep? I haven't used a Pentium-M or Core anything laptop, but my PIIIm (1.2GHz) only needed a simple little applet (SpeedSwitch XP for Windows XP) to force any SpeedStep capable laptop into the slow clock. That unit didn't need it so much because it doesn't run hot. That laptop was very thin for the time and light too, 1.25" thick and 5lb.
into the slow clock? yeah that was P3s.
SpeedStep doesnt do that anymore. It dynamically scales its frequency from the lowest setting to the highest setting in 100mhz increments. It could be running 600MHz on second and th next at 900, or at 1000, or at 2000... it dynamically changes all the time.
Can someone check if they brought the clockspeed up on the Windows side they would hit these standards for x1600 Futuremark 3Dmark05 scores (free download from http://www.futuremark.com) ???
I feel that in Windows on the MacbookPros and iMac Intel there would be no reason for arguments, ATI has appropriate clocking tools (just check up on the PC overclocking forums) to set things to your heart's desire and get the most frame rates out of your x1600/x1600 mobility. Overclock too if you dare
I agree though there should be some way to set clock speeds on the Mac OS X side of things (the "highest" and "normal" and "reduced" settings in the system preferences), but really, if your pro apps run well and smooth why bother with clocking your GPU higher?
I agree that (and why not send feedback to Apple) one should have more control on GPU clocking on the Mac OS X side of things but it should be not too long before someone comes up with a Macintel OS X ATI overclocking tool
That said, Apple has worked out a balance of performance, heat, etc, etc, so at least you're better informed now going in to purchase a Macintel.
First of all, I'd like to say that I'm all for this. I'd rather have a laptop that runs as silently as possible, and trade some performance for longer battery life.
That being said, wouldn't it be cool if Apple offered an "overlooking" tool, like ATI does? That way, if you want to squeeze every last frame per second out of Quake 4, you could, and than clock it back to the "recommended" state.
****************************************
That's right.
Afterall, if I want to run on the power brick to do video capture / conversions etc., I should have the choice of clocking up, then back down for battery operation.
Afterall, if I want to run on the power brick to do video capture / conversions etc., I should have the choice of clocking up, then back down for battery operation.
Specifically, Core Image operations* will benefit from a higher clocked GPU right? The question then is how much Core Image operations there are in your workflow
Specifically, Core Image operations* will benefit from a higher clocked GPU right? The question then is how much Core Image operations there are in your workflow
*Core Image operations = GPU-only tasks
Video capture, and conversions are cpu tasks, not GPU tasks.
These aren't video cards, like my old TruVision boards were. Thse are graphics boards. There is a big difference.
Comments
Originally posted by JeffDM
My point was that the bottom surface of a hot object has the worst thermal transfer coefficient, in non-circulated air. Actually moving air would be better, but this product really doesn't seem to try very hard at that either.
It's not ideal. But the fans are no doubt the largest that could fit. They also take up power. Being mechanical devices, no matter how efficient they are, the draw would be considerable if they were to be on for a large percentage of the time. The battery itself gets hot as current is being drawn.
As I said, there isn't a better place on the machine that could sink the heat other than the bottom. would you rather they sunk it on the top, through the keyboard?
Originally posted by melgross
As I said, there isn't a better place on the machine that could sink the heat other than the bottom. would you rather they sunk it on the top, through the keyboard?
Actually they do that too. There is a considerable amount of heat going to the keyboard. It's not as much of a problem as there is no heat trapping going on.
Originally posted by JeffDM
Actually they do that too. There is a considerable amount of heat going to the keyboard. It's not as much of a problem as there is no heat trapping going on.
But that isn't intentional. Some of the heat rises through the keyboard because it has no other place to go. Heat trapping isn't 100% efficient.
Originally posted by melgross
But that isn't intentional. Some of the heat rises through the keyboard because it has no other place to go. Heat trapping isn't 100% efficient.
Regardless, I've looked around, the part between the keyboard and screen exceeds 120 F and has been the subject of numerous complaints.
I've looked around, the claim that using the fan would waste too much power doesn't hold much water. I found a few small ones that are quiet and only consume 0.5 W. Still, something like this wouldn't be as good as not turning electrical power into heat in the first place, especially at idle. The amount of heat generated at idle is the reason I'm not convinced that SpeedStep is being used properly.
Originally posted by JeffDM
Regardless, I've looked around, the part between the keyboard and screen exceeds 120 F and has been the subject of numerous complaints.
I've looked around, the claim that using the fan would waste too much power doesn't hold much water. I found a few small ones that are quiet and only consume 0.5 W. Still, something like this wouldn't be as good as not turning electrical power into heat in the first place, especially at idle. The amount of heat generated at idle is the reason I'm not convinced that SpeedStep is being used properly.
It's aomething that Apple has no control over. The chip makes all of the decisions. So you have to blame Intel on that one.
Look, I'm with you on the heat thing. I think that the machines could be a bit thicker, and and as a result, have a better internal path for the heat to be withdrawn through. But, as I said, thin is in.
As far as the fans go, I don't know what models Apple uses, or the cfm. But even if they use .5 watt, there are two of them, and even that 1 watt would draw that battery down if they were on much more often. also, if you read many comments in the thread, you will see that people just HATE it when the fan comes on for long periods. So, one just can't win with this.
Originally posted by lundy
Don't forget to say that you want to be able to OMG UPGRADE TEH VIDEO CARD and OMG UPGRADE TEH PROCESSOR and OMG UPGRADE TEH HARD DRIVE and OMG I WANT MORE PIXELS and OMG WTF NO PCI SLOT.
This is....TEH 13373ST P0sT oF TEH y34R!
Originally posted by melgross
It's aomething that Apple has no control over. The chip makes all of the decisions. So you have to blame Intel on that one.
Did Intel remove the software controllability of SpeedStep? I haven't used a Pentium-M or Core anything laptop, but my PIIIm (1.2GHz) only needed a simple little applet (SpeedSwitch XP for Windows XP) to force any SpeedStep capable laptop into the slow clock. That unit didn't need it so much because it doesn't run hot. That laptop was very thin for the time and light too, 1.25" thick and 5lb.
Originally posted by JeffDM
Did Intel remove the software controllability of SpeedStep? I haven't used a Pentium-M or Core anything laptop, but my PIIIm (1.2GHz) only needed a simple little applet (SpeedSwitch XP for Windows XP) to force any SpeedStep capable laptop into the slow clock. That unit didn't need it so much because it doesn't run hot. That laptop was very thin for the time and light too, 1.25" thick and 5lb.
It's not so much as removed it, as this being a new series, it was never put in. That doesn't mean that some hack doesn't exist that might allow it. But, right now, at least, Intel doesn't support it.
What speed machine do you have? The faster, the hotter, after all.
Originally posted by gugulino81
I'd like to know, how other Laptop-Manufacturer handled this problem? It seems, that the ATI graphic-card is a bit too hot for Notebooks!
Making a laptop that's thicker than a fucking inch?
Originally posted by melgross
As I said, there isn't a better place on the machine that could sink the heat other than the bottom. would you rather they sunk it on the top, through the keyboard?
Of course not. MBPs should have liquid cooling, with a radiator on the back of the lid.
Originally posted by melgross
It's not so much as removed it, as this being a new series, it was never put in. That doesn't mean that some hack doesn't exist that might allow it. But, right now, at least, Intel doesn't support it.
What speed machine do you have? The faster, the hotter, after all.
I wasn't aware of that.
I bought the 1.83MHz version. I was tempted to go higher, but I am thankful I didn't. I had expected that the increased power saving of the latest SpeedStep, combined with the transistor size shrinks (.065 vs .13 on the old notebook) would allow enough efficiency improvements to offset the clock speed increase.
I did find some posts where people claimed that the 1.0.1 firmware helped so I installed that for completeness sake, but the Apple info makes no claims to that effect. That isn't to say that Apple didn't sneak in some bug fixes without saying anything, the only thing Apple claims about 1.0.1 is to support languages that read right to left.
Originally posted by JeffDM
I wasn't aware of that.
I bought the 1.83MHz version. I was tempted to go higher, but I am thankful I didn't. I had expected that the increased power saving of the latest SpeedStep, combined with the transistor size shrinks (.065 vs .13 on the old notebook) would allow enough efficiency improvements to offset the clock speed increase.
I did find some posts where people claimed that the 1.0.1 firmware helped so I installed that for completeness sake, but the Apple info makes no claims to that effect. That isn't to say that Apple didn't sneak in some bug fixes without saying anything, the only thing Apple claims about 1.0.1 is to support languages that read right to left.
Too bad they didn't continue to offer the 1.67. You know, if the Speedstepping IS working correctly, and there isn't any reason to believe that it isn't, just think how hot it would get without it!
Apple probably stuck some very minor fixes in, but if it were anyrthing that a user, or programmer, would see, they would have mentioned it.
Originally posted by JeffDM
Did Intel remove the software controllability of SpeedStep? I haven't used a Pentium-M or Core anything laptop, but my PIIIm (1.2GHz) only needed a simple little applet (SpeedSwitch XP for Windows XP) to force any SpeedStep capable laptop into the slow clock. That unit didn't need it so much because it doesn't run hot. That laptop was very thin for the time and light too, 1.25" thick and 5lb.
into the slow clock? yeah that was P3s.
SpeedStep doesnt do that anymore. It dynamically scales its frequency from the lowest setting to the highest setting in 100mhz increments. It could be running 600MHz on second and th next at 900, or at 1000, or at 2000... it dynamically changes all the time.
source: http://service.futuremark.com/orb/index.jsp
(yes I love to cross-post my pretty pictures )
I feel that in Windows on the MacbookPros and iMac Intel there would be no reason for arguments, ATI has appropriate clocking tools (just check up on the PC overclocking forums) to set things to your heart's desire and get the most frame rates out of your x1600/x1600 mobility. Overclock too if you dare
I agree though there should be some way to set clock speeds on the Mac OS X side of things (the "highest" and "normal" and "reduced" settings in the system preferences), but really, if your pro apps run well and smooth why bother with clocking your GPU higher?
I agree that (and why not send feedback to Apple) one should have more control on GPU clocking on the Mac OS X side of things but it should be not too long before someone comes up with a Macintel OS X ATI overclocking tool
That said, Apple has worked out a balance of performance, heat, etc, etc, so at least you're better informed now going in to purchase a Macintel.
Originally posted by Robin Hood
First of all, I'd like to say that I'm all for this. I'd rather have a laptop that runs as silently as possible, and trade some performance for longer battery life.
That being said, wouldn't it be cool if Apple offered an "overlooking" tool, like ATI does? That way, if you want to squeeze every last frame per second out of Quake 4, you could, and than clock it back to the "recommended" state.
****************************************
That's right.
Afterall, if I want to run on the power brick to do video capture / conversions etc., I should have the choice of clocking up, then back down for battery operation.
Could this be a simple SW solution ???
Originally posted by gooddog
****************************************
That's right.
Afterall, if I want to run on the power brick to do video capture / conversions etc., I should have the choice of clocking up, then back down for battery operation.
Could this be a simple SW solution ???
Revving the GPU up won't help that.
Originally posted by melgross
Revving the GPU up won't help that.
Specifically, Core Image operations* will benefit from a higher clocked GPU right? The question then is how much Core Image operations there are in your workflow
*Core Image operations = GPU-only tasks
Originally posted by sunilraman
Specifically, Core Image operations* will benefit from a higher clocked GPU right? The question then is how much Core Image operations there are in your workflow
*Core Image operations = GPU-only tasks
Video capture, and conversions are cpu tasks, not GPU tasks.
These aren't video cards, like my old TruVision boards were. Thse are graphics boards. There is a big difference.