Core i7 vs. Core i5: Understanding the Power Story

Between generations Apple constantly struggles between squeezing every last ounce of max performance out of silicon and reducing system temperatures. I believe Apple's philosophy here is that most of the time your CPU should be running at relatively low utilization and as a result offering the full dynamic range of CPU performance is preferred to clamping max performance in order to preserve lower thermals. The problem is that in some cases, lazy background task management (e.g. keeping too many Safari windows open with Flash active) can drive CPU usage and thermals up even if you're actively doing nothing on the machine. This scenario coupled with Haswell ULT's excellent idle power consumption I believe are primary motivators for Mavericks' App Nap and occluded window slumber features.

 

 

To understand the impact on thermals (and battery life) of the Core i7-4650U on the 13-inch MacBook Air you need to understand what's going on under the hood. To hit higher frequencies, the i7-4650U generally requires a higher voltage. Power consumption (and thus thermal dissipation) can scale linearly with frequency, but it scales quadratically with voltage. The combination of the two is quite possibly the worst case scenario from a power consumption standpoint. This is why it's generally always best to increase performance via process shrinks or architectural enhancements vs. simply scaling frequency. In the case of the i7-4650U we're not talking about huge frequency/voltage scaling here, but rather a tradeoff between added performance and increased power consumption. In the table below I noted typical CPU core voltages for a couple of different operating modes on my i5-4250U and i7-4650U samples. Several years ago Intel introduced voltage binning even at a given frequency, so the voltages you see in the table below are only applicable to my parts (or other similar parts) - you could see a range of acceptable voltages in other binned parts even carrying the same model number. The values in parantheses indicate the CPU frequency (or frequencies) observed during the workload.

13-inch MacBook Air (Mid 2013) CPU Comparison - Observed Voltages
  Idle Cinebench 11.5 (1 thread) Cinebench 11.5 (4 threads)
Intel Core i5-4250U 0.665V
(800MHz)
0.852V - 0.904V
(2.3GHz - 2.6GHz*)
0.842V
(2.3GHz)
Intel Core i7-4650U 0.655V
(800MHz)
0.949V - 1.041V
(2.9GHz - 3.3GHz*)
0.786V - 0.949V
(2.8GHz - 2.9GHz*)

There are a bunch of observations here. First off, the two parts are very comparable at idle - this is how Apple can quote all implementations of the MacBook Air as being capable of up to 12 hours of battery life. At idle large parts of the silicon are clock gated if not fully power gated. Idle voltages are extremely low (even compared to what you find in modern smartphones) and both parts run at the same 800MHz frequency at idle, so power consumption is comparable between the two at idle.

Using Cinebench 11.5, I ramped up a FP intensive single threaded workload. FP workloads tend to force a bunch of large units into switching making this a great test for voltage scaling. Here we see that the i5-4250U is capable of hitting its max turbo frequency but for the most part it hangs out around 2.3GHz. The same is true for the i7-4650U, 3.3GHz is possible but most of the time it's sitting down at 2.9GHz. The i7-4650U needs higher voltages all around to hit these higher frequencies.

Next, I cranked up the number of threads. First you'll notice a reduction in clock speeds and voltages. This is where multithreading can actually be good for power consumption. Running more cores at a lower voltage for a shorter period of time can reduce total energy consumed while performing a task. The i5-4250U has no issues running at its max DC turbo frequency (2.3GHz), while the i7-4650U mostly sticks to 2.8GHz with occasional bursts up to 2.9GHz. Note that the 4650U's min voltage at 2.8GHz is actually lower than the 4250U's here. In order to hit these higher frequencies within the same TDP, Intel does have to bin for parts that do a bit better at higher frequencies whereas to make the cut for a 4250U the leakage requirements aren't as severe.

There shouldn't be any surprises thus far, but this data should give us an indication of what we can expect in terms of battery life and thermals. Where the i7 vs i5 comparison becomes tricky is if you look at workloads that can complete quick enough due to the faster performance in order to offset any additional power consumption.

CPU Performance Battery Life & Thermals
POST A COMMENT

127 Comments

View All Comments

  • FwFred - Thursday, July 04, 2013 - link

    Probably the benefits matter much more in non-optimal cases. I think Anand's experience mirrors my own. One bad browser tab can kill the battery life due to javascript or flash. If Mavericks is able to prevent this somehow (not really sure how it could do it within Chrome between tabs, but maybe Safari), I would be very pleased. Reply
  • solmaker - Thursday, July 04, 2013 - link

    Thanks for the sage observations, K-Knight. If you're right that Mavericks moves heavier workloads down toward the "light" category, that would tend to decrease the i5 vs i7 battery difference, since they're very close when "light". But I'd think even the "light" case would be helped by Timer Coalescing (batching periodic timer wake-ups) and AppNap (by nailing unused background tasks). My fantasizing about 12 hour MBA 11" battery life was based on anecdotal forum reports of 13+ hour 11" life using the Mavericks beta, but that hasn't been verified. My own usage often has an idle VMware Fusion WinXP session in the background, so I'd get great benefits if AppNap could zap that. Reply
  • KitsuneKnight - Friday, July 05, 2013 - link

    Thinking about it more, the light load would also receive some benefit as well. Before when I wrote 'light', I was more so thinking about a system with no apps running, effectively just sitting idle... which is obviously not a 'light' workload, but a no-workload. Although with the current low energy usage of OS X the gains might not be massive (maybe even within the margin of error).

    I still think the best cases will be with poorly programmed apps with lots of useless timers keeping the system awake, which Timer Coalescing will help a bit with when the app is in the foreground, and AppNap much more so when it's in the background.

    I'm not sure how Fusion is architectured, but I /think/ how it works (based on how Workstation was many years ago... on Windows) it won't automagically be able to benefit from AppNap, and VMware will have to put a bit of effort into making it opt into AppNap when appropriate. Parallels Desktop already has a feature where (under the right circumstances), idle VMs will be paused, although it's not as aggressive as it could be.

    It'll be interesting to see how developers make their apps work with AppNap (most won't have to do anything beyond linking against the 10.9 frameworks). I imagine at the start many developers will just try to disable AppNap for their app and write blog posts about how 'it's so pointless and doesn't everyone just leave their computer plugged in', before everyone starts embracing it... but maybe I'm jaded from how new technologies get adopted in Windows / Linux land.
    Reply
  • solmaker - Saturday, July 06, 2013 - link

    Thanks again, K-K. I wonder whether AppNap with be "Opt In" or "Opt Out"? If the latter, then VMware Fusion may automagically AppNap when hidden. Reply
  • fokka - Thursday, July 04, 2013 - link

    thanks for the comparison, anand! these are exactly the questions i'm wondering about when researching for a mobile pc: is there a benefit in battery life when chosing a low end cpu, or is the faster cpu just quicker and can even save power with longer idling times?

    for me it would be clear to chose the i7 here, since it would be be my main computer and battery is similar to the i5 in scenarios most likely with my usage: light workload. but then, if i decide to load up lightroom or want to play a quick game, i can rest assured that i'm eeking out quite a couple percent more performance than on the stock cpu.

    now the only thing i'm waiting for is a proper zenbook-refresh :)
    Reply
  • ananduser - Thursday, July 04, 2013 - link

    I feel you Anand; too bad that other companies don't offer such "one size fits all" lineups like Apple. The exaggerate range of choices from other companies simply adds confusion.

    I like the detail and effort that went into reviewing different SKUs of the same unit. Hope you do the same for some upcoming items from Lenovo, Asus, Samsung, Sony etc.

    Since I mentioned Sony, I want to ask you if there is a chance you could get your hands on the new Vaio Pro and Vaio Duo that sport that "fancy" triluminous display. None of the reviews available has actually gone the length to see what's with that new "quantum dot" tech. Only Anandtech™ can shed light here. What ? Sony hardly sends you Vaios for review ? I dunno, ask them harder ?
    Reply
  • Awful - Thursday, July 04, 2013 - link

    Thanks Anand, very useful!

    Any chance of including some GPU/Gaming comparisons too? It would be interesting to see the results of the increased thermal constraints of the i7 on GPU perf (could even be worse I imagine under certain loads?)
    Reply
  • ciparis - Thursday, July 04, 2013 - link

    If Firefox really takes 17-37 minutes to build, all I can say is wow, what a %$#@! pig. Reply
  • Mackan - Friday, July 05, 2013 - link

    Can you also do a battery life test in Windows, via Boot Camp, using pure EFI install which seems to be supported now. Reply
  • US Fiscal Situation - Friday, July 05, 2013 - link

    Yeah.. no one else can make thinner and lighter laptops, erm... except for the firms that can and do...

    Take the Vaio Pro 11, it weighs less (870g), it's thinner and yet has the same battery life.

    The Vaio manages to cram in a much higher resolution (full HD) TOUCHSCREEN, NFC and SD card reader. It even has Ethernet support built into the power supply which creates a WIFI hotspot

    Both machines are comparable performance/processor wise but one costs £60 less (i7,4GB RAM, 256GB SSD)- No prizes for guessing which.

    BTW apropos business use- it should be noted that the lack of boot camp drivers for W8 means that you can't run all of the programs that you might need.

    US Fiscal Situation http://bit.ly/16SsEFt
    Reply

Log in

Don't have an account? Sign up now