> the cancellation of the PowerXCell8i

Err, PowerXCell 8i is there. You mean PowerXCell 32ii and PowerXCell 32iv, right?

Sorry, bad cut and paste.
Did I at least get the name of the CTO correct?

> Did I at least get the name of the CTO correct?

I think so ;-)

> Your posts contained the first speculation about Power8 featuring SPEs
> that I'd seen, but it does look like this is going to happen.

There're dissenting opinions as well:

"IBM [...] hinted that in the future its commercial-grade Power chips would incorporate many of the functions of the Cell chips. This [...] could happen with Power7+ or Power8 chips -- or, more likely, not at all. It seems far more likely that IBM will now focus on hybrid CPU-GPU combinations for certain HPC customers"
http://www.theregister.co.uk/2011/06/28/ibm_kills_qs22_blade/

@ Andreas_Wolf

Speculation. I bet we don't get a solid answer about this for a year or more.

> I bet we don't get a solid answer about this for a year or more.

First solid information on POWER7 preceded its official release by 1.5 years. Going by that, 1 year from now plus 1.5 more years would mean early 2014 for POWER8 release. But as we know that POWER8 is scheduled for 2013 (and rather early than late that year), solid information may be imminent ;-)

Quote:

It's not quite the same.

As I understand it, on ARM the problem would be intermixing big endian code (68K, PPC) with little endian code (ARM). In both cases the data is all big-endian.

On IA32 everything is little-endian, code and data.


What I had suggested is essentially 2 A-Boxes. One includes a PPC emulator to run PPC and 68K code.

The second A-Box (B-Box?) would only run natively compiled ARM code. No intermixing.


This has a big advantage for moving things forward:
If you are building a new box that you know requires at least a re-compile, you could also add other conditions. e.g. Forbid() could be either be ignored or forbidden (pardon the pun).

You could also put down plans for moving towards things like full memory protection and multi-processing.

I'm not sure how you'd go about this but you could possibly do it with proxies. The app would talk to the proxy thinking it was the OS. The Proxy then does the interaction with the OS.

Then, you build a "C-Box" with memory protection etc. In this box your app still speaks to the proxy but this time it doesn't write directly into kernel memory as the Amiga does.

I don't know how easy or difficult that would be in practice but it would give you an ordered transition that would allow you to migrate today's software into a rather more modern environment. A sudden leap is not likely to work.

There would be a cost to using proxies of course but given how much ARM's performance will shoot up over the next couple of years it's not something you're likely to even notice.

Is MorphOS going the arm route?. Whatever route it chooses, i hope it gets powerful hardware.

Quote:

No idea, but that's the rumour.
They have to go somewhere after the G5.

Quote:

I think that's a given whatever they do.

The g5 route is interesting. I tried to get a dual g5 on ebay but got overbid on the last minute, damn!.
Still looking for a good deal also on a g4 tower.
The problem, is that the arm processors i know, aren't powerful enough for desktop, only for phone/tablet use. Are out there more powerful arm cpus?

> The problem, is that the arm processors i know, aren't
> powerful enough for desktop, only for phone/tablet use.
> Are out there more powerful arm cpus?

More powerful than the ones you know? Nobody can answer this as long as you don't list the ARM CPUs that you know ;-)
In any case, the most powerful one currently still seems to be the one that I mentioned back in January in the last sentence of that posting:

https://morph.zone/modules/newbb_plus/viewtopic.php?forum=3&topic_id=7001&start=525

Latest information on that:
http://www.youtube.com/watch?v=3ibShHKLjMM
http://www.youtube.com/watch?v=9yq9TW6xQOw

At 1.6 GHz and below it's still a chip intended mainly for mobile computing including notebook/netbook, not only phone/tablet. In the first video however they also show an AIO computer prototype based on it, and in the second video they call this prototype a reference design. So from that I conclude that they also intend to target general desktop computing with this chip, presumably with the 2.0 GHz version.
That said there seems to be a little confusion regarding the 2.0 GHz version and laptops. In the first video it's said that the 2.0 GHz chip will be used in laptops because its power consumption is too much for mobile devices. Some might say that a laptop is a mobile device as well...

Quote:

The Nufront chip is fastest announced but I've no idea if it's actually shipping yet. They have shown it working a few times so i guess it'll be soon. That said, that clock and memory system should put it faster than any G4.

There will be others at a similar speed later in the year though.

Next year is where it gets really interesting with the A15 At 2.5GHz that should match any G5.

[ Edited by minator 10.07.2011 - 02:20 ]

> I've no idea if it's actually shipping yet.

According to the Nufront representative in the first video it is, including the 2.0 GHz version. The video is from early June btw.

> There will be others at a similar speed later in the year though.

Yes, but it won't be Cortex ones it seems ;-)

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=7675&forum=3&start=168

> Next year is where it gets really interesting with the A15
> At 2.5GHz that should match any G5.

We will see if there'll really be 2.5 GHz Cortex-A15 based chips as early as 2012. OMAP5 at least is only announced to go as fast as 2.0 GHz for 2012:

http://www.ti.com/ww/en/omap/omap5/omap5-OMAP5430.html
http://www.ti.com/ww/en/omap/omap5/omap5-OMAP5432.html
http://www.youtube.com/watch?v=o-QzNpN77go

Quote:

I meant the ~1.8 GHz chips meant for later in the / year early 2010.

> I meant the ~1.8 GHz chips meant for later in the / year early 2010.

I guess that should read "later in the year / early 2012" ;-)
If you're now talking about early 2012 as well instead of just later this year like you did before then yes, there should be 1.8 GHz OMAP4470 in Q1/2012.

So I have to wait till next year for an ARM processor that runs as fast as my G4?
I think I'd prefer the G5.
Then, later, we can consider ARM (as more powerful CPUs come to market).

> So I have to wait till next year for an ARM processor that
> runs as fast as my G4?

As I said, Nufront claimed in early June that their 2.0 GHz Cortex-A9 chip was available. Is their claim true? I don't know.

> I think I'd prefer the G5.

Yes, at least there're already desktop systems built around that :-)

> Then, later, we can consider ARM

You mean "we" as in "the MorphOS Team"? ;-)

@Jim

ARM can't compete with x86 on desktop. IMHO choosing ARM would be a suicide. It doesn't fix any problem, it doesn't bring any advantage to customers, it's yet another hard to find hardware platform. If MorphOS was going to migrate to other cpu it should be x86 because you can buy x86 machines everywhere, there's enough variety. I'm not interested in vapour cpus with inflated benchmarks. ARMs performance sucks. 1Ghz ARM cpus are slower than all PPC equivalents. There's no ARM machine that can compete with a QuadG5 machine in price/performance right now.

If we switched to other architecture it should be x86 (it's cheap, it has a wide variety of models from low consumption to high performance, it's easy to find, it won't dissapear)

So BAF could also mean Blind ARM Follower? :-)

> It doesn't fix any problem, it doesn't bring any advantage to customers

It would at least make hardware available that you can buy unused. If you see that as an advantage would depend on whether you see the non-availability of unused hardware as a problem, though.

> 1Ghz ARM cpus are slower than all PPC equivalents.

I'm not sure that's a true statement. A comparison for instance between a 1 GHz PPC440 based chip (e.g. PPC460EX) and one core of a 1 GHz Cortex-A9 based chip would prove interesting I bet. We have NBench figures for 1 GHz Cortex-A9 for example. Now we'd only need someone with a Sam460ex to run this benchmark for us to get an idea of the trueness of your statement. Unless you think of NBench as an "inflated benchmark" that is ;-)

> There's no ARM machine that can compete with a QuadG5 machine in
> price/performance right now.

...and not in performance alone either :-)

> So BAF could also mean Blind ARM Follower? :-)

A MorphOS Team member making fun of ARM fans doesn't sound too good for an (allegedly rumoured) ARM port of MorphOS ;-)

Quote:

True but ARM doesn't try to compete on the desktop - at the moment.
They're moving up very rapidly though.
Currently, the highest end stuff is based on phone chips but that's changing as tablet chips become available. It'll get really interesting when server chips appear.

Quote:

Since when has cheap, available, new hardware been a disadvantage?

Quote:

Actually, at 6 billion a year they're pretty hard to avoid.

They're not exactly well represented in the desktop space (in the west) but there are some and there's more due at the end of the year.

If you just want a cheap ($149 - $199) dev board there's a few here:
http://www.linaro.org/low-cost-development-boards

Quote:

Can't argue with that.

Quote:

Good. Neither am I.

Quote:

You base this on what exactly?

Quote:

Quote:



Of course you can't be referring to me because this is one area where I can't be described as blind...

>> There's no ARM machine that can compete with a QuadG5
>> machine in price/performance right now.

> That'll change next year. The year after, the opposite will be true.

Don't be so sure about that. While it's true that ARM performance will increase over time and actually make a jump next year with the advent of Cortex-A15 based chips, leading also to an improvement in price-performance ratio, used quad-G5 Macs are becoming even cheaper, improving their price-performance ratio as well, all the time. Currently, the average price that a (fully functional) quad-G5 Mac is sold for on eBay Germany is about 530 EUR, whereas 16 months ago it was still 910 EUR, which is a reduction by 42%. If that trend continues (and I guess it does) new machines with future ARM chips will definitely have a rough ride beating used quad-G5 Macs in terms of price-performance ratio for some time to come.

Addendum:

> Page 12 of FTF11_NET_F1176 document mentions "Improved load /store
> to ease bit alignment" and "New instructions for video analytics".

Found another interesting comment on the new AltiVec in the QorIQ AMP:

"An important note regarding AltiVec is that some users have previously found the ISA difficult to use. AMP marks its relaunch in a form that allows a far easier definition of parameters."
http://www.techdesignforums.com/embedded/embedded-topics/freescale%E2%80%99s-qoriq-amp-series-illustrates-growing-demand-on-the-control-and-data-plane/