Yes different test, but we can see the difference a SSD makes, which is my whole point.
We don't know if the difference is chassis cooling properties, lack of SSD, the amount of thermal paste on the chips, or the specific binning lottery of these particular chips. More and better controlled experimentation will need to be performed to call anything more detailed than:
It appears that the rMBP cools better than the uMBP based on one sample set. Which is what I am saying.
A CMBP and a retina have the same CPU so temp should be the same there. The GPU's are the same save for memory so temps should be similar also. That right there should tell you something. Yes the RMBP has better cooling and fans. But it needs them because of less 'cooling space' within the uni-body case, than a CMBP.
The same model chip can have somewhat different thermal characteristics due to the binning lottery. Yes it should be identical, but it probably isn't, and that is the reality of modern processor yields (especially early in a fab lifespan, like we are with IVB).
A HD will add to the total internal temperature of the machine, so yes it will add significantly to the CPU and GPU temperatures. Add a SSD into the mix and watch those total temps drop.
Not buying it. The HDD is not even spinning during these operations. The default settings are for it to spin down when it isn't being used. When I try to save a screenshot, it pauses as it spins the drive up to save it. I doubt the disk is contributing anything, or at very most 1-2 degrees. Not ten.
Note that this is a 5400 RPM notebook drive, not a toasty blistering speed demon 15K Cheetah. SSDs are also not magical heat black holes, the controller does produce heat, as does the NAND switching.
http://www.anandtech.com/show/6023/the-nextgen-macbook-pro-with-retina-display-review/12
What is the biggest difference between the two. The CPU and the bottom of the two. A SSD makes all the difference in the world. Anything with moving parts that spin as fast as a HD does gives off allot of heat. And that is a understatement. Both MBP's in the test above had SSD's in them.
These devices are very thermally integrated. It is almost impossible to separate the heat generation of one part from the heat generation of another. Especially when dealing with the two largest contributors (CPU/GPU). The huge differences between Anand's two devices are the following:
- Completely different cooling design. Intakes, air guides, fans, radiators, and heat pipes.
- CPU process shrink from 32nm to 22nm.
- GPU process shrink from 40nm to 28nm, and architecture/vendor change from AMD/Whistler to Nvidia/Kepler.
That is a lot of huge differences. The contribution (or not) of the HDD vs SSD difference will be completely lost in the noise.
Subtract the thermals from the 2011 model from a 2012 model CMBP regarding the CPU (ivy bridge runs cooler), and it would be essentially be a wash.
The biggest difference in these tests is the CPU, and on a 2012 model, it would be almost the same or similar both with ivy bridge.
Taking the sum of Anand's computer's temp measurements and then finding the difference is a totally ridiculous way of determining the absolute heat handling ability of the machine. Not to mention projecting what might be if a 2011 were to contain IVB. We can take a point and say this is different by X, but we are aware that there are a lot of contributors in this equation.
My late 2011 MBP runs 10 F cooler without a HD in it. As soon as I dropped a SSD in it it ran 10 F cooler. Maybe more.
I do not believe this without data to support it. Simply an implausible statement. Maybe there were massive dust bunnies trapped in your radiators and opening the case freed them.
