Much has been written about the increase in heat emitted from the iPad 3 with some tests claiming that it runs as much as 10 degrees Fahrenheit hotter -- while running GPU-intensive tasks -- than the iPad 2.
Late Monday evening Engadget stoked the fire with a post claiming that the new iPad burns 10 degrees hotter than its predecessor accompanied by thermal imaging photos of the iPad 3 and 2 side-by-side:
Having been burned (pardon the pun) by CR's scathing reviews in the past (you'll recall the CR was one of the most vocal of the iPhone 4 detractors during the Antennagate fiasco with the iPhone 4 in July 2010). Apple was quick to nip this one in the bud, with Apple PR's Trudy Muller telling AllThingsD:
The new iPad delivers a stunning Retina display, A5X chip, support for 4G LTE plus 10 hours of battery life, all while operating well within our thermal specifications. If customers have any concerns they should contact AppleCare.
[Don't you love how Apple buries the response in the middle of a commercial? Geez.]
I played games on my iPad 3 for about two hours straight last night while writing a review of five games that push the iPad 3's Retina display to the limit -- and my iPad never got too hot to touch. It was definitely warmer than my iPad 2, but not unusually hot, especially for how much I was pushing the quad-core GPU and the Retina display.
For his part, Mr. Soneira thinks that the additional heat is simply due to the fact that the iPad 3 has to drive four times the amount of pixels of the iPad 2 double the amount of LEDs to light it up. In his excellent display technology shootout between the iPad 3, iPad 2 and the iPhone 4 Soneira notes that the iPad 3 has much lower display power efficiency:
The new iPad uses 2.5 times the Backlight power of the iPad 2 for the same screen Brightness. As discussed above that results from the TFT transistors in the LCD blocking much more of the light at higher ppi. On the other hand, the highest ppi iPhone 4 is the most power efficient display of all because it uses Low Temperature Poly Silicon LTPS, which is much more efficient than amorphous silicon in the iPads. All of this points to the need for the IGZO display technology discussed above, which is more efficient and lower cost than LTPS. It should be in production shortly, and is the first in a whole series of enhanced Metal Oxide semiconductors for LCD and OLED displays.
The highest ppi iPhone 4 is the most power efficient of all because it uses Low Temperature Poly Silicon LTPS (see above).
The number of Backlight LEDs has roughly doubled (from 36 to an estimated 72 to 82), so the Backlight power has approximately doubled. Since the display normally consumes about 50-60 percent of the total Tablet power, the new iPad needs at least a 50 percent larger battery. In fact, the battery increased from 25 to 42.5 watt hours, a 70 percent increase. Our measured Backlight power for the new iPad is 2.5 times the iPad 2 for the same screen brightness. In spite of the larger battery the running time at Maximum brightness in our tests was 5.8 hours, 20 percent less than the iPad 2‚s 7.2 hours. But at the Middle brightness slider setting, which is closer to typical user settings, the running time was 11.6 hours, which is almost identical to the iPad 2, indicating that Apple has used an appropriately larger battery (and confirms Apple‚s 10 hour claim).
The bottom line is that the iPad 3 backlight consumes 2.5 times as much power as the iPad 2 for the same brightness (luminance) due to the higher number of pixels per inch (ppi). The Retina display LCD has a lot lower light efficiency and therefore power efficiency.
According to Soneira, the iPad 3's LCD panel has 72-82 LEDs, which is approximately twice the amount of LEDs as found in the iPad 2. The iPad 3's LEDs give off 2.5 times as much heat as the iPad 2 as will the battery and power electronics.
He goes on to say that when the iPad 3 display is set for maximum brightness the display consumes about 65% of the devices total power, which plays a major roll in its thermal budget.
Both my new iPad and my iPad do not run excessively warm (let alone hot). As expected, I can feel that the new iPad is a bit warmer than the iPad 2, but they are both fine. Also I run mine at maximum Brightness for testing (which generates the most power and heat), whereas most people will run with a lower setting - it comes at the Middle slider setting from the factory. At the Middle slider setting the Backlight consumes only 36% as much power as at Maximum, so that is only 36% of the heat also.
It's certainly possible that CPU and GPU will heat up during extensive gaming, but the major power difference between the iPad 3 and iPad 2 is the Retina display.
Now, the issue won't be overblown if it results in massive hardware failures, but none have been reported (at least that I'm aware of) because it's just too early in the product lifecycle to know. And we're not likely to to have a definitive answer for several weeks or months to come.
Don't be surprised if someone runs a GPU hogging app continuously to see if the iPad 3 will melt or catch fire. Someone will do it and it will probably make it onto all of the cable news shows.
I would be shocked however if the additional heat was a genuine design issue. It's not out of the realm of possibility that there could be a production problem affecting a small percentage of the first batch of iPad 3s, but I doubt that it will result in a recall.
Apple announced that it sold 3 million iPad 3s on opening weekend and if the failure rate was 0.1% then that would translate to 3,000 defective iPads in the first batch -- not terrible by large-scale manufacturing standards.
I'm curious if the thermal issue will be exacerbated by putting the iPad 3 in various "full body" cases. As for me, I'm going to stick with a simple Smart Cover on my iPad 3 -- at least while it's running.