Apple A4

[wombat888]

lilo's link contained valid and useful info. It didn't attempt to discuss battery life. We know what battery life is for each device, but without comparisons like this, we don't know performance.

I'm much more interested in battery life, but it's still best to have all the info available.

 

[MikhailT]

lilo's link contained valid and useful info. It didn't attempt to discuss battery life. We know what battery life is for each device, but without comparisons like this, we don't know performance.

I'm much more interested in battery life, but it's still best to have all the info available.

I updated my post to be more clear. I agree that we need comparisons but CNET was just horrible.

 

[Bentov]

The ARM cortex a-8(which is probably what the A4 is based off of, and is a single core chip) seems to be a 13 stage pipeline processor. It is basically, the same chip that is in the iphone 3gs(but that one is only running like 600 mhz I think). As others have said though, it isn't "more powerful" than a 1 ghz processor, since it is a 1 ghz processor. You can't compare chip speeds across families, only in the same family. Knowing the chip speed of the a4 is meaningless since it is the only one in it's "family"

 

[lilo777]

I updated my post to be more clear. I agree that we need comparisons but CNET was just horrible.

I agree with you but this seems to be one of very few tests available. But it does show that the raw performance of Atom might be significantly higher than that of A4. At least CNET used the same browser (Safari) to do the testing Very often when iPad is compared to netbooks the testers not only use different OSes (iPhone vs, say, Windows 7) but also different browsers. As I understand, real testing of iPad is problematic due to the closed nature of the platform (no benchmark apps in App Store ).

 

[HiRez]

You really can't have a meaningful comparison of mobile chip performance without factoring battery life (as well as the related weight and heat production) into the equation since that's the whole point. Anyone could make some tablet and shoehorn a Core2 Duo or something into it, but it would weigh 3+ lbs., get scorching hot, have short battery life, and would probably require more RAM to make use of it. The Atom is not that bad, but it's in-between somewhere. That's where the A4 beats the pants off of Atom, especially with Apple iPhone OS being so optimized for low power consumption. Apple supposedly used their Intrinsity assets to strip out all non-essential instructions from the ARM8. Other companies using off the shelf chips are at a disadvantage since those chips have to cover a wide range of general use possibilities, whereas Apple knows exactly what they need in them because it runs their software, and only their software, on their hardware (accessory chips, graphics package, battery, radios, etc.)

 

[MikhailT]

The ARM cortex a-8(which is probably what the A4 is based off of, and is a single core chip) seems to be a 13 stage pipeline processor. It is basically, the same chip that is in the iphone 3gs(but that one is only running like 600 mhz I think). As others have said though, it isn't "more powerful" than a 1 ghz processor, since it is a 1 ghz processor. You can't compare chip speeds across families, only in the same family. Knowing the chip speed of the a4 is meaningless since it is the only one in it's "family"

You don't consider Cortex A8 chips all in the same family? I do. There are many Cortex A8 designs, they are all not the same. Each ARM designer can improve it in their own ways that's better than the reference.

Look at this from Anandtech's review lilo linked,

If we take the network out of the equation, the A4 in the iPad has a 37.6% performance advantage over the Qualcomm QSD8250. This actually supports some of the larger performance differences we saw earlier. If Apple can manage to deliver this sort of performance in its smartphone version of the A4, we're in for a treat.

The why is much more difficult to ascertain. It could be as simple as the the iPad OS being better optimized than Android, a definite possibility given how much longer Apple has been working on it compared to Google. The advantage could also be hardware. The A4 may boast higher IPC than Qualcomm's Snapdragon thanks to better core architecture, larger caches or a faster memory bus. The likely case is somewhere in between, where the iPad's advantage comes from a combination of hardware and software.

Snagdragon is a 1GHz processor based off ARM Cortex A8 and yet it's a bit slower than A4.

 

[prvt.donut]

I updated my post to be more clear. I agree that we need comparisons but CNET was just horrible.

CNET is usually horrible. Their reviews are nigh on worthless.

 

[Bentov]

You don't consider Cortex A8 chips all in the same family? I do. There are many Cortex A8 designs, they are all not the same. Each ARM designer can improve it in their own ways that's better than the reference.

Look at this from Anandtech's review lilo linked,


Snagdragon is a 1GHz processor based off ARM Cortex A8 and yet it's a bit slower than A4.

I do consider all Cortex A8's to be in the same family to a point, but I don't believe that all Apple did was take a dremel to the chips, removing the A8 and stamp an A4 on it. I think there comes a point where the architecture of the chip is changed enough for it to be considered almost a different chip.