Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Upgrade MacBook Air 2013 i5 to i7?
- Thread starter kuzzkuzz
- Start date
- Sort by reaction score
The article isn't very clear, but I think they did run the test on the 2013 i7 model, though it's not listed in the table.
It looks like they had previously tested the 2013 i7 model here, and in the body of the text of this article, they use that data to say that the 2013 i5 model (the standard configuration) lasted 11 mins longer than the 2013 i7 model (the 'ultimate CTO') on the movie test. However, they're not directly comparable, as they have different RAM (4GB vs 8GB) and hard-drive capacity (though I'm not sure that would make a difference?) too.
Sean Dempsey
macrumors 68000
No. It will be much faster.
CPU clock speed is largely meaningless. This isn't the 90's anymore. Your 2.4ghz core2duo will be orders of magnitude slower. Speed of a modern computer is 10% CPU and 90% the other tech.
With the i5 vs. i7 question, it's like asking if you'll need your car to drive 200mph instead of 180mph. You'll never go fast enough to hit either unless you're a serious pro, and if you were you wouldn't be asking.
As far as future proofing, you think you will need to go 200mph in a few years when the roads are still 75mph and you've only ever gone 110mph that one time to show off?
If you have the money, just buy whatever makes your anatomy enlarge. Otherwise, you don't need to buy performance limits you won't dream of hitting.
jmpage2
macrumors 68040
Sorry Shawn but your post is extremely misleading.
To some extent, raw clock speed is "misleading" in that newer CPUs typically have many enhancements beyond the raw clock rate that improve performance (things like hyper-threading, cache, instruction set extensions, etc).
However to claim that a 1.3ghz i5 is "orders of magnitude" faster than a 2.8ghz C2D is again misleading since it largely depends on what tasks are being performed.
Number of clock operations still do matter, which is why very high clocked dual core chips can outperform very low clocked advanced CPUs in some tasks.
A full performance list of CPU tests with synthetic benchmarks can be found here;
http://www.cpubenchmark.net/cpu_list.php
The closest two CPUs for comparison purposes are a 1.7ghz i5 with hyperthreading (dual core) and a 2.4ghz C2D. The performance difference in synthetic benchmarks is about 2X in favor of the i5 but obviously the clock speed difference between 1.3ghz Haswell and 2.8ghz C2D is unknown at this time. It might be 2X or so faster.
Probably not "orders of magnitude" faster... but certainly faster than the older C2D (the SSD will actually translate to most of the "wow" factor of how much faster the new MBA is compared to the old MBP).
One of the biggest advantages of these new CPUs is their extremely efficient power design, allowing for small thin enclosures and super long battery run times.
To some extent, raw clock speed is "misleading" in that newer CPUs typically have many enhancements beyond the raw clock rate that improve performance (things like hyper-threading, cache, instruction set extensions, etc).
However to claim that a 1.3ghz i5 is "orders of magnitude" faster than a 2.8ghz C2D is again misleading since it largely depends on what tasks are being performed.
Number of clock operations still do matter, which is why very high clocked dual core chips can outperform very low clocked advanced CPUs in some tasks.
A full performance list of CPU tests with synthetic benchmarks can be found here;
http://www.cpubenchmark.net/cpu_list.php
The closest two CPUs for comparison purposes are a 1.7ghz i5 with hyperthreading (dual core) and a 2.4ghz C2D. The performance difference in synthetic benchmarks is about 2X in favor of the i5 but obviously the clock speed difference between 1.3ghz Haswell and 2.8ghz C2D is unknown at this time. It might be 2X or so faster.
Probably not "orders of magnitude" faster... but certainly faster than the older C2D (the SSD will actually translate to most of the "wow" factor of how much faster the new MBA is compared to the old MBP).
One of the biggest advantages of these new CPUs is their extremely efficient power design, allowing for small thin enclosures and super long battery run times.
Sean Dempsey
macrumors 68000
He said "mostly school research and writing and photo editing."
I should clarify - the computer as a whole will be significantly faster for those tasks. It's not even a competition.
I'm not sure this is true at all. Speed is nice, as is 'future delaying' (rather than future proofing). But if the palmrest gets hot doing basic things or battery life is, say, >10% worse, then I'd say it's not worth the upgrade for me. It's exactly those differences that we haven't yet nailed down.
If past i5 v i7 comparisons are anything to go by the i7 will run about 1-2c hotter, consume around 5% more battery life under light load, 10-15% more battery life under heavy load, spin up the fans sooner but complete high CPU tasks like video encoding 25-30% quicker.
However this was a comparison between the ivy bridge filler generation i5 and i7 chips.
The haswell architecture and power efficiency might well be better at curbing the extra power draw and heat.
Either way, I think generally in basic tasks the battery and heat difference will be hardly noticeable.
When you run demanding tasks they will finisher sooner at the expense of a little extra battery consumption.
Update
In addition to my ivy bridge summary I found this on the sandy bridge i5 v i7 : http://www.macworld.com/article/1161434/macbook_air_2011.html
As I said < 5% more battery light loading 10-15% heavy loads. The sandy bridge article reflects this too. 8% more battery life consumed on the i7 v i5 under a fair load.
I think the differences will be very similar with haswell too. Waiting for the reviews myself.
However this was a comparison between the ivy bridge filler generation i5 and i7 chips.
The haswell architecture and power efficiency might well be better at curbing the extra power draw and heat.
Either way, I think generally in basic tasks the battery and heat difference will be hardly noticeable.
When you run demanding tasks they will finisher sooner at the expense of a little extra battery consumption.
Update
In addition to my ivy bridge summary I found this on the sandy bridge i5 v i7 : http://www.macworld.com/article/1161434/macbook_air_2011.html
As I said < 5% more battery light loading 10-15% heavy loads. The sandy bridge article reflects this too. 8% more battery life consumed on the i7 v i5 under a fair load.
I think the differences will be very similar with haswell too. Waiting for the reviews myself.
Last edited:
jmpage2
macrumors 68040
With Haswell, both parts have a 15 watt total power draw. Battery life should be fairly similar between the two as should heat levels.
Personally I ordered the i7 for my wife. She does run an XP image with virtualization image and it's a bit of a resource hog. I get a discount through my employer so bumping the CPU to an i7 and going to 8GB was $1440 with free shipping out the door... which is reasonable compared to the rMBP she had been considering.
Personally I ordered the i7 for my wife. She does run an XP image with virtualization image and it's a bit of a resource hog. I get a discount through my employer so bumping the CPU to an i7 and going to 8GB was $1440 with free shipping out the door... which is reasonable compared to the rMBP she had been considering.
There's a lot of should be's and would be's and NO Concrete battery test for i7 till now...pretty sad 🙁
Lame journalists are sleeping literally...!
I think people who need an i7 KNOW they need an i7...if you are questioning, you probably don't.
Due to lack of +1 button, I'll just quote your statement. Studying the currently available information, I think this is precisely what people should expect.
I can add one comment regarding games like WoW and LoL: the proc will be pushed to the limit if there is no limitation on the fps, thus the i7 will actually have 10-15% shorter battery life. With lower graphics settings and fps limiting (which both games support), the battery will last longer.
curtoise, can you run some gaming benchmarks on both machines? would be interresting how faster the i7 with his 100MHz higher GPU-turboboost actually is.
Strategy
This comment is not directly aimed at the above, but more a general remark. I've been on a mac for more years than I can count and due to cash constraints I've always run the machines into the ground before upgrading. This means that I'm now 'working' (trying to work, that is) on a Powerbook 12" 867Mhz PowerPC.
Now as most of you will know, that is one FRUSTRATING thing to be doing and it's extremely hard to be productive (I find Word and Excel to work fine on it). I have though had many a conversation with some friends of mine who all have more or less the latest kit, and who are graphics and photo professionals and we've all agreed (without doing any calculations I might add) that what makes the most sense is to buy the value option when it comes out but to upgrade every year.
So I'm jumping in right now, buying the 1.3/8mb/128 machine, with a view to upgrade it in one year's time.
As an example:
Two months ago I started looking at a base model MBP13 on Ebay. A second hand one, released in October 2012, would sell for about £750. A brand new one sold at the same time on Amazon for £849. A 2011 model would be north of £600 depending on nick, and a 2010 model was between £500 and £550.
So the point is that in order to be able to do what you need to do, for the least amount of money over time, it seems to be a better option to go for the 'budget' version and upgrade, rather than 'maxing it out' in order to future proof it for the next three+ years.
A brand new base model will most likely outperform the previous year's maxed out model, so you'd get sub-par performance for Y1 but for Y2 and 3 you'd get better performance than one who maxes out and keeps for three years.
Now, if only that pesky MBA could come a little faster!!!!!!!!!! 😀
This comment is not directly aimed at the above, but more a general remark. I've been on a mac for more years than I can count and due to cash constraints I've always run the machines into the ground before upgrading. This means that I'm now 'working' (trying to work, that is) on a Powerbook 12" 867Mhz PowerPC.
Now as most of you will know, that is one FRUSTRATING thing to be doing and it's extremely hard to be productive (I find Word and Excel to work fine on it). I have though had many a conversation with some friends of mine who all have more or less the latest kit, and who are graphics and photo professionals and we've all agreed (without doing any calculations I might add) that what makes the most sense is to buy the value option when it comes out but to upgrade every year.
So I'm jumping in right now, buying the 1.3/8mb/128 machine, with a view to upgrade it in one year's time.
As an example:
Two months ago I started looking at a base model MBP13 on Ebay. A second hand one, released in October 2012, would sell for about £750. A brand new one sold at the same time on Amazon for £849. A 2011 model would be north of £600 depending on nick, and a 2010 model was between £500 and £550.
So the point is that in order to be able to do what you need to do, for the least amount of money over time, it seems to be a better option to go for the 'budget' version and upgrade, rather than 'maxing it out' in order to future proof it for the next three+ years.
A brand new base model will most likely outperform the previous year's maxed out model, so you'd get sub-par performance for Y1 but for Y2 and 3 you'd get better performance than one who maxes out and keeps for three years.
Now, if only that pesky MBA could come a little faster!!!!!!!!!! 😀
I think you may well be right, but there is something fundamentally wrong with a world where this is the case! I just cannot stomach getting through something as valuable and as environmentally-unfriendly as a laptop every year, even if economically, on a personal level, it makes the most sense.
The amount I've spent to keep my current laptop usable from its original specs (upgrading RAM, harddrive, and OS), I probably would have saved money buying a new laptop outright - but I'm perversely quite proud to have kept this old soldier going as long as it has.
Agreed. This also ignores the value of the time and effort spent churning into and out of a new machine every year, plus the risk factor of all those transactions.
I've always kept my computers 3-5 years. As I await my first ever Mac (just left Shanghai this morning), I am on a deathly slow older Dell laptop. Embarrassingly slow. But I am, like you, "perversely proud" to have hung in there with the old dog. It also makes the leap to the new system all that more pleasurable whereas an annual incremental bump has no real emotional or functional pleasure.
Although I don't disagree with you in one sense, I think you're missing that by upgrading annually there's little to suggest (as far as I can see) that this behaviour would contribute to an increase in the over all production of laptops.
What it would do though (as I see it) is to democratise the market, drive down prices of second hand machines and allow more people to have a good mac.
I've frequently borrowed a mate's mbp 2011 and that is a machine that's doing everything I'd ever want it to, so if I would have had one of those I'd been happy.
What I don't want to do is to pay 70% of the price of a new machine when the new machine is 2 generations ahead, which is the case at the moment.
Good point - but I guess that if you do a migration from time machine it shouldn't be such a hassle right? But I do get the value of a clean install, tend to do it already so I probably would have gone down this route myself..
Jazwire
macrumors 6502a
Was getting tired of waiting on reviews myself, called both Apple Stores in my area, but now they are all sold out of the Ultimate versions. *sigh*
My wife has a base model so I could test the difference between the two.
I know for email, browsing, word, etc the i5 is fine.
I will game occasionally on the machine, playing LoL which is CPU intensive my wife's i5 does hit 100% cpu usage), I will occasionally do video editing, i will occasionally do coding/compiling.
I am willing to get the i7, the little extra money for it bothers me not at all, however I don't want to cook my lap either, and I dont want to have 1 hour less battery life over my wife's i5 either under normal usage.
I fully understand if i am making the processor go to 100% of course the battery is going to last shorter, and its going to get warmer on the i7. Because it working harder than the i5....
But when I am browsing the internet, checking email I dont want the i7 to be noticeably warmer / battery noticeably shorter during those times.
Also like to see tests with the exact same things be ran on them at the same time.
All I would need to make my choice is
An i5 ran for x (1,2,5 hours) amount of time with x application (Netflix or a Game or Video/Movie) running.
an i7 ran for same "x"amount of time with same "x" application running.
Give me the temps. (Doesn't even have to be exact, barely warmer, considerably warmer will suffice.)
Give me the battery life. (Started at 100% i5 is at "x" and Started at 100% and i7 is at "x" after running the same app(s).
But no one on the entire planet apparently has done or is willing to do this.
I was going to do it myself but dang Apple store sold out of i7's. 😡
My wife has a base model so I could test the difference between the two.
I know for email, browsing, word, etc the i5 is fine.
I will game occasionally on the machine, playing LoL which is CPU intensive my wife's i5 does hit 100% cpu usage), I will occasionally do video editing, i will occasionally do coding/compiling.
I am willing to get the i7, the little extra money for it bothers me not at all, however I don't want to cook my lap either, and I dont want to have 1 hour less battery life over my wife's i5 either under normal usage.
I fully understand if i am making the processor go to 100% of course the battery is going to last shorter, and its going to get warmer on the i7. Because it working harder than the i5....
But when I am browsing the internet, checking email I dont want the i7 to be noticeably warmer / battery noticeably shorter during those times.
Also like to see tests with the exact same things be ran on them at the same time.
All I would need to make my choice is
An i5 ran for x (1,2,5 hours) amount of time with x application (Netflix or a Game or Video/Movie) running.
an i7 ran for same "x"amount of time with same "x" application running.
Give me the temps. (Doesn't even have to be exact, barely warmer, considerably warmer will suffice.)
Give me the battery life. (Started at 100% i5 is at "x" and Started at 100% and i7 is at "x" after running the same app(s).
But no one on the entire planet apparently has done or is willing to do this.
I was going to do it myself but dang Apple store sold out of i7's. 😡
I second this!! Will someone with an i5 and an i7 be so generous/awesome/kind to run a test like this for just a couple hours?
Curtuoise? You have both, no? 🙂 🙂 Pretty please?
I have 3 now 😱 but two are nicely packed and ready to be returned at the nearest Apple store.
Booji
macrumors 6502a
Exactly...I'd be willing to go to the i7 but I'm buying the Air for awesome battery life and the value of the upgrade depends on the power/stamina tradeoff. If I loose an hour with normal use, its a no-go.
I'm getting the i7 because I'll be using Final Cut Pro X and windows-based statistical modeling programs for business school, and I believe I'll benefit a little by having a little additional power. However, most of the time when I'm using those CPU-intensive programs, I'll be plugged in, using an external monitor for video editing, or when sitting around a conference room table at school.
Most of the time when I'm on battery I'll just be doing some light work in Office, web surfing, etc., and I won't "need" the additional power of the i7, so it'll throttle down and battery life shouldn't take much of a (if any) hit.
That's my theory anyway.
Most of the time when I'm on battery I'll just be doing some light work in Office, web surfing, etc., and I won't "need" the additional power of the i7, so it'll throttle down and battery life shouldn't take much of a (if any) hit.
That's my theory anyway.
Good choice. My i7 lasts 12 to 13 hours. When i do CPU intensive work, its almost always plugged in.