Update:

> I think Applied Micro (X-Gene) or even Cavium (Project Thunder) might
> arrive in high-end ARM land before nVidia (Project Denver/Boulder).

Calxeda revealed as new ARMv8 contender with "Lago":

http://www.calxeda.com/calxeda-lays-out-a-vision-for-the-hyper-efficient-datacenter/


Edit: Seems like Calxeda won't design its own ARMv8 core but will use a ready-made core from ARM Ltd. instead:

"Announced licensees of the new processor series include AMD, Broadcom, Calxeda, HiSilicon, Samsung and STMicroelectronics."
http://www.arm.com/about/newsroom/arm-launches-cortex-a50-series-the-worlds-most-energy-efficient-64-bit-processors.php

[ Edited by Andreas_Wolf 30.10.2012 - 23:44 ]

>> ARM Cortex-A15: 3.5

> Talking of which...
> http://www.theverge.com/2012/10/18/3521492/google-new-samsung-chromebook-249-monday

Nice :-) Samsung made its promise true and beat TI.

Hits the shelf on Monday?
OMG! Guess I need to figure out what $250 expense has to be foregone.

minator,
Quote:

Well, being the first(?) Cortex-A15 in a consumer product is kind of cool, and so is the price, but the display seems a bit tacky judging from the article (a poor display is usually a deal-breaker IMHO, no matter the price), and it's a bit worrying if it can't properly play HD videos on true HD displays? Maybe that CPU was really tailored for a future Galaxy S IV smartphone, and less suitable in a device like this? And the lack of both Ethernet and 3G/4G wireless makes it even less usable, and puts it even further down my wishlist...

> lack of [...] 3G/4G wireless

http://www.samsung.com/us/computer/chrome-os-devices/XE303C12-H01US-specs

Indeed, but then it's $330 and with a plan with a phone company I don't have access to where I live.

I'll just keep looking, better deals are bound to happen a bit further down the road...

> it's [...] with a plan with a phone company I don't have
> access to where I live.

Yes, obviously, as that's the US release, which is the initial one. Releases for other parts of the world will follow I guess.
I just wanted to give the correct information that there is a version with hardware for 3G connectivity included, rather than imply that this version was currently suited for where you live.

Since all I want is wifi (for the casino I work in and at home) I couldn't care less about the 3G or 4G features.
Gotta have one.
And then a hosted version of AROS.

Quote:

The previous chips could do it so I'm guessing it's a software problem.

Quote:

Possible, or it's an issue of premature/rushed HW...

Update:

> I doubt the A6 will clock higher than 1.2 GHz in the iPhone.
> Maybe later in the next iPad, as the tablet form factor
> allows for a bit more power consumption.

As we know, the A6 chip in the new iPhone clocks at 1.3 GHz. The A6X chip in the new iPad announced yesterday is said to run at 1.5 GHz and thus faster than A6, as expected:

http://arstechnica.com/apple/2012/10/deducing-details-about-apples-a6x-processor/

Quote:

Rushing a 32nm SoC would be a very bad idea given the exorbitant cost of even taping it out.

@minator

As far as I know, some problems is only/can only be spotted *after* you have it in real silicon, and AFAIK it's not very uncommon that you have to taping out a few revisions before reaching a level that can be judged good enough for end-users. For example: Freescale did at least 3 tapeouts of the i.MX515, and Genesi used the TO2 in early board designs/targeted at early developers. Only the TO3 was used in the final, end-user products:

Quote:

This is what I meant with "being rushed"; maybe the chip hasn't reached its final tapeout revision yet, but Samsung/Google is eager to push a first version of their Cortex-A15 smarttop to the market? Some things can be worked around in drivers, other things can't and will need an updated chip. I guess we'll see.

You can find out for yourself now with an Arndaleboard.

$250, same chip as the Chromebook.

minator,
Quote:

And only another $250 for the LCD screen, $160 for the SSD, $120 for the connectivity and sound boards, $60 for the camera, and $60 for the accesory package (including cables, pwr supply, and micro SD card).

$900 anyone?

I'll take the Chromebook.

Jim,
Quote:

The basic board includes Sound and I/O cards (the $120 is for different versions). I don't think you really need any of the rest as they are pretty much standard parts.

Quote:

I might get one myself, might try some OpenCL on it.

[ Edited by minator 26.10.2012 - 23:30 ]

Quote:

OTOH if you were say developing something, I don't know, say, an OS. the Arndale board might be a better bet.

There's certainly a lot of information on that site, including fairly comprehensive information on the processor.

minator,
Quote:


Yes, say as a basis for a Ubuntu system, the board probably would be the better choice.

Some benchmarks:

http://www.androidauthority.com/exynos-5-dual-benchmarks-125134/

Granted Javascript benchmarks aren't very good as a CPU test but they are very nice scores. Quite a bit better than I expected.

Yep, pretty good.
But I think we were all expecting that.
And they choose to compare it to some pretty lame processors.

>> Quite a bit better than I expected.

> I think we were all expecting that.

Well, minator said he was not, in the posting you just replied to ;-) And I'm siding with him here in that the results of the Cortex-A15 vs. Krait and Swift are better than expected (by me).

> they choose to compare it to some pretty lame processors.

They compared it to what is relevant for this site to compare, i.e. the top-end chips used in mobile Android and iOS devices.

>They compared it to what is relevant for this site to compare, i.e. the top-end chips used in mobile Android and iOS devices.

Personally, since the first device I've seen with an A15 is the Chromebook, I would have prefered comparisons with other netbook and low end laptop CPUs.
We were fairly certain that the A15 woulld outperform Apple's offering and I've always thought the predictions for Krait were over blown.

> since the first device I've seen with an A15 is the Chromebook, I would have
> prefered comparisons with other netbook and low end laptop CPUs.

I'm sure those will come (if they're not out there somewhere already), but on other sites than androidauthority.com.

> We were fairly certain that the A15 woulld outperform Apple's offering

Yes, after it became known that the Swift core in the iPhone achieved its surprisingly high Geekbench score at 1.3 GHz instead of the 1.0 GHz that was reported before, the expectations regarding Swift's per-clock performance decreased under those of Cortex-A15.
What minator and me didn't expect though was *how much* the Cortex-A15 would outperform Swift (and Krait).

> I've always thought the predictions for Krait were over blown.

The predictions for Krait were that it would be roughly one third faster per clock than Cortex-A9, and I think that's what it has delivered for real.

Andreas_Wolf,
Quote:

You are better informed then me, I wasn't aware that the anticipated performance was that low.
If they sucessfully clock the A15 higher then 1.7 GHz or manage to double the cores without a significant speed decrease, then this core is likely to dominate for quite awhile.

> You are better informed then me, I wasn't aware that the anticipated
> performance was that low.

Yeah, I've been mentioning the anticipated Krait performance for the last 1.5 years here on MorphZone :-)

> If they sucessfully clock the A15 higher then 1.7 GHz or manage to double the cores
> without a significant speed decrease, then this core is likely to dominate for quite awhile.

Yes, absolutely. I doubt Apple will go above 1.5 GHz dual-Swift anytime soon, but 1.7 GHz quad-Krait is already there in the APQ8064 (and soon to come MPQ8064).