Oh my god Andreas, sometimes I seriously wonder if you are for real...

> Oh my god Andreas

Some time ago someone used to nonsensically call me his "prophet" here on MorphZone. Now I am your "god". What's next? :-)

> sometimes I seriously wonder if you are for real...

I am, and so are my explanations to you that you tend to ignore.

My God, Andreas.
You are a prophet, but you will never be a deity.

Although, at times, I wonder if you are "for real" or just a really good artificial intelligence program.

The kind that WILL eventually exist, that will make us all feel...smaller.

> http://www.arm.com/about/newsroom/arm-launches-cortex-a50-series-the-worlds-most-energy-efficient-64-bit-processors.php

Finally :-) Seems my assumption about Calxeda's Lago being their own core design was wrong. So in terms of 64-bit ARMv8 (AArch64) cores, there is announced so far:

- Cortex-A53 and Cortex-A57 from ARM Ltd.
- Denver from nVidia
- X-Gene from Applied Micro
- Thunder from Cavium

> You are a prophet

And here you had me hoping we were over this... ;-)

Hey, at least I never claimed you were God.

Chromebook test:
http://www.anandtech.com/show/6422/samsung-chromebook-xe303-review-testing-arms-cortex-a15

More on the A15:
http://www.extremetech.com/computing/139393-arm-cortex-a15-explained-intels-atom-is-down-but-not-out

...and the A57
http://www.eetimes.com/electronics-news/4400351/ARM-A57-core-detailed-in-technical-session

On a topical note, I'm sure you'll work out a Spec2000 number for the A15 from the slides (hint: about 1000). That puts it just shy of a 2.2GHz G5 and about the same as where the X1000 should be.

> I'm sure you'll work out a Spec2000 number for the A15 from the slides (hint:
> about 1000). That puts it [...] about the same as where the X1000 should be.

...at 2.0 GHz, which PA6T chips available on the market never really ran at. So it's only about 900 SPECint2000 for the 1.8 GHz PA6T.

Andreas_Wolf,
Quote:

I suspect a few special customers might have got some.

Quote:
990 ...the estimate was 1100 at 2GHz.

That said I still suspect it's not set up correctly in the X1000 and you wont get that score if you ran it.
OTOH, I did once read a comment that newer firmware versions improved performance so maybe results are better now.

Quote:

I remember the Articia-S controller. Rated to be AGP 2x in speed and marketed as such. In real tests (that was conducted by various community members many, many years ago) it turned out that the real-life data bandwidth was about AGP 0.5x in practice, due to a buggy/unfinished design (they were only made for development boards, like the Terons, that someone resold as AmigaOne's.) It's not always about "settings" to get something right, some times a design can just have been released too early, too underdeveloped, too untested, and wasn't really meant for end-user markets anyway.

>> it's only about 900 SPECint2000 for the 1.8 GHz PA6T.

> 990 ...the estimate was 1100 at 2GHz.

Ah okay, thanks. A quick search found me only the 1000 figure, which is probably rounded.

> Articia-S [...] were only made for development boards, like the Terons [...].
> [...] some times a design [...] wasn't really meant for end-user markets anyway.

https://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=7675&start=336

And from the horse's mouth:

"Articia S is the first ready-to-market chipset of the Articia family of chipsets. It is a highly integrated, low-cost and high-performance system IC, ideal for network computers, web servers and thin servers. [...] Articia S is the leading system controller that enables AGP support for RISC processor based systems. It meets the increasing demands for today's embedded systems, such as video/3D graphics oriented, multi-tasking and real-time applications."
http://web.archive.org/web/20060514160730/http://www.mai.com/products/articia_s.html

No doubt that the ArticiaS wasn't ready for any commercial product, but that doesn't mean that MAI didn't try to make and sell it for exactly that or that MAI didn't claim it to be designed for end-user markets.
Regarding PA6T and end-user markets, you should keep in mind that it was originally designed as a chip for Apple laptops.

takemehomegrandma,
Quote:

MAI logic were incompetent, they didn't test the chip properly and didn't even know it had basic faults.
Some faults can be very difficult to find - but not the basic ones.

The PA6T was designed by people who knew what they were doing. Some of them have quite a record, Alpha, StrongARM, Opteron, and now Apple's A6.

There could be a number of other reasons though, it's designed to run code for a G5, it may simply be bad at running code designed for a G4. Maybe the 32 bit mode sucks.

But did they really, *really* reach 100% all the way to a 100% tested and 100% functional *end-user level* product before Apple pulled the plug? Albeit seen and used in some product at early stage, had it really finished sampling and all necessary revision tapeouts to become final? Is there some web page in the archive with a general, public press release saying "We are finished and PA6T is now open for business and we will be happy to take your orders" kind of thing?

I've no idea about the actual details but it was sufficiently finished for US military suppliers to use it. When Apple bought the company the US government even forced Apple to keep making them for some time afterwards.

> 28nm Cortex A-15 (Eagle) i.MX73 is proposed. [...]
> i.MX Roadmap

Seems this may take longer than anticipated:

http://semiaccurate.com/2012/11/01/freescale-layoffs-target-one-particular-product/
http://www.eetimes.com/electronics-blogs/the-engineering-life-around-the-web/4401885/Freescale-s-i-MX-7-canned-as-firm-refocuses-
http://www.linleygroup.com/newsletters/newsletter_detail.php?num=4917

Edit: added some more links

[ Edited by Andreas_Wolf 18.12.2012 - 00:44 ]

Did this board ever register amoung the ARM proponents here? http://apc.io/

It's billed as a $49 Android device made by Via.

(pinched from wikipedia)
ModelAPC 8750
Software Android 2.3 (PC System)
Chip VIA 800MHz Processor
Memory DDR3 512MB Memory, 2GB NAND Flash
Graphics Built-in 2D/3D Graphic, Resolution up to 720p
Input and Output HDMI, VGA, USB 2.0 (x4), Audio out / Mic in, microSD Slot
Network 10/100 Ethernet
Size 170 x 85mm (W x H), Neo-ITX Standard

ebay India

RS Components

> Did this board ever register amoung the ARM proponents here? http://apc.io/

You and me briefly talked about it about half a year ago in this thread.

https://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=7675&start=269

http://www.macrumors.com/2012/11/06/apples-potential-shift-from-intel-to-arm-for-macs-not-implausible-but-faces-hurdles/
Dont know if this already been mentioned but i suppose it has some relevance here, if Mac goes ARM the progress of it is probably gonna accelerate a lot.

Update:

> New list of DMIPS per MHz and core figures for ARM cores, taking the Swift core
> of the Apple A6 SoC into account (only with an estimation unfortunately):
> [...]
> Apple Swift: 3.3...3.5 (estimated)
> [...]

Figure for Swift substantiated and figures for Cortex-A53 and Cortex-A57 added, according to these sources:
http://en.wikipedia.org/wiki/List_of_ARM_microprocessor_cores#Designed_by_ARM
http://en.wikipedia.org/wiki/List_of_ARM_microprocessor_cores#Designed_by_third_parties
http://www.arm.com/products/processors/cortex-a50/cortex-a53-processor.php?tab=Performance

ARM Cortex-A7: 1.9
ARM Cortex-A8: 2.0
Qualcomm Scorpion: 2.1
ARM Cortex-A53: 2.3
Marvell Sheeva PJ4: 2.4
ARM Cortex-A9: 2.5
Marvell Sheeva PJ4B: 2.6
Qualcomm Krait: 3.3
(Broadcom Brahma15: 3.5 (unsure whether or not this core is just ARM Cortex-A15))
Apple Swift: 3.5
ARM Cortex-A15: 3.5
ARM Cortex-A57: 4.1...4.8 (depending on implementation)


Edit: Added Broadcom Brahma15 core.

[ Edited by Andreas_Wolf 26.12.2013 - 21:28 ]

Some more benchmarks for the Chromebook:

http://openbenchmarking.org/result/1211246-RA-EXYNOS5DU63

and some to compare it to:
http://openbenchmarking.org/result/1202030-AR-MERGE774596

A bit confusing but it comes in close to a Core Duo.

minator,
Quote:

I appreciate that you're a strong advocate of ARM cpus, but if you're going to make a comparison like that could you give us a bench that shows the two cpus together?

Edit - Andreas, can you come up with some figures that would allow us to compare Cortex A15 with e6500 core cpus (dmips, etc)?

[ Edited by Jim 25.11.2012 - 03:26 ]

> Andreas, can you come up with some figures that would allow us to
> compare Cortex A15 with e6500 core cpus (dmips, etc)?

DMIPS figures for both those cores to compare are already on MorphZone for 3.5 months. The figure for Cortex-A15 can be found in posting #420 of this very thread, and the figure for e6500 can be found there:

https://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=7001&start=743

Sorry, I should have just searched.
Surprisingly high for the A15.
But then I remembered the e6500 figures as being pretty good too.
Is that quote on the A15 for one, two, or four cores?
And is it fair to assume that an e6500 based product with far more cores would produce even better results?

Basically, would 12 cores dual threaded equal 72 (or close to that)?

> Surprisingly high for the A15.

The DMIPS figure for the Cortex-A15 has been known for about 2 years already :-)

> Is that quote on the A15 for one, two, or four cores?

Quote from that posting: "DMIPS per MHz and core" :-)

> is it fair to assume that an e6500 based product with far more cores
> would produce even better results?

Yes, it is, as the Dhrystone benchmark scales quite linearly with the amount of cores.

> Basically, would 12 cores dual threaded equal 72 (or close to that)?

Yes, 12 dual-threaded e6500 cores would deliver 72 DMIPS per MHz.