I'm not starting yet another thread questioning if the core should or should not shut down. Not even concerned (much) on why it appears easier to cause this on the Air. What I am confused about is the why.
Let me say that for most activity I don't see this, in fact to cause a core to drop takes pushing this system to work at the level of a desktop.
I know several of you have seen my review of the Vaja leather suit. One of my observations was that the Air was more easily pushed to a core shutdown with the case installed. Because of this I wanted to understand what part of the system was overheating and driving this event.
I was creating some content for my iPod using Handbrake with the suit applied and noticed a slowdown. Checking activity monitor I noted that I had indeed lost a core. Since I had several hours of this activity queued up I decided to do a little testing and monitoring to help better understand the problem.
I looked at both iStat and Coolbook reports for system temps. The CPU was running at about 70 c and most other temps seemed within the norm from my original temp testing.
Here is the funny part, I don't believe the core getting hot is the reason for the drop. I monitored temps with and without the case running Handbrake and the core stayed about the same. Even more noteworthy all the temps in iStat didn't show much of a change regardless of the case being installed.
I really believe what I am seeing with at least the CPU temps. Coolbook, even when not active, does provide this information on the CPU. It seemed within a degree or 2 of what iStat was showing. Is it possible that one of the temps iStat is monitoring is not correct? Could they be missing something in their reporting?
Since I don't see a major difference in the temps of the CPU I don't believe the chip itself is the source of the shutdown. Could it be system firmware detecting an overall heat issue within the machine and using the core shutdown to help release some of that overheating? What temp is it monitoring that we don't see reported?
When I replaced the thermal paste on the Air it did help. Prior to that I couldn't even watch a 15 minute video with issues. Now, without the case, I can watch videos, surf the web, have several other task active without any issues. When I replaced the paste I was sure to apply paste correctly to both chips using the heatsink. I was under the impression that the second chip under the heatsink is the bridge chip. Is it or is it a GPU? I know I don't have a specific temp reading on this one but the heatsink temp would seem to indicate that it is running in the norm.
One thing I did notice during my testing was how hot the power supply felt. The temps in iStat didn't change much but when I removed the case and placed my hand under the upper left corner of the base it was almost to hot to touch.
Would really like to hear others thoughts on this.
Let me say that for most activity I don't see this, in fact to cause a core to drop takes pushing this system to work at the level of a desktop.
I know several of you have seen my review of the Vaja leather suit. One of my observations was that the Air was more easily pushed to a core shutdown with the case installed. Because of this I wanted to understand what part of the system was overheating and driving this event.
I was creating some content for my iPod using Handbrake with the suit applied and noticed a slowdown. Checking activity monitor I noted that I had indeed lost a core. Since I had several hours of this activity queued up I decided to do a little testing and monitoring to help better understand the problem.
I looked at both iStat and Coolbook reports for system temps. The CPU was running at about 70 c and most other temps seemed within the norm from my original temp testing.
Here is the funny part, I don't believe the core getting hot is the reason for the drop. I monitored temps with and without the case running Handbrake and the core stayed about the same. Even more noteworthy all the temps in iStat didn't show much of a change regardless of the case being installed.
I really believe what I am seeing with at least the CPU temps. Coolbook, even when not active, does provide this information on the CPU. It seemed within a degree or 2 of what iStat was showing. Is it possible that one of the temps iStat is monitoring is not correct? Could they be missing something in their reporting?
Since I don't see a major difference in the temps of the CPU I don't believe the chip itself is the source of the shutdown. Could it be system firmware detecting an overall heat issue within the machine and using the core shutdown to help release some of that overheating? What temp is it monitoring that we don't see reported?
When I replaced the thermal paste on the Air it did help. Prior to that I couldn't even watch a 15 minute video with issues. Now, without the case, I can watch videos, surf the web, have several other task active without any issues. When I replaced the paste I was sure to apply paste correctly to both chips using the heatsink. I was under the impression that the second chip under the heatsink is the bridge chip. Is it or is it a GPU? I know I don't have a specific temp reading on this one but the heatsink temp would seem to indicate that it is running in the norm.
One thing I did notice during my testing was how hot the power supply felt. The temps in iStat didn't change much but when I removed the case and placed my hand under the upper left corner of the base it was almost to hot to touch.
Would really like to hear others thoughts on this.