It's actually meant for things like 10Gbit ethernet, Infiniband controllers and potentially storage arrays (like 3Ware/AMCC did).

The actual applicability to a desktop system is about zero, and the timeframe? 4Q 2010 won't be met, these deadlines never are.

_________________
Matt Sealey

Impossible.

http://www.haiku-os.org/blog/nielx/2009 ... week_later

32k downloads the FIRST week. Now it's more than 100k. Quite possible.

I seriously doubt there's so much public interest for Haiku. If there was, it would be mainstream.
It even has the same limiting factor of MorphOS: Haiku also lacks the "official" trademark ("BeOS") that sparks people's curiosity, just like MorphOS, that lacks the "Amiga" badge, and solely for that is almost unknown to people. People remember brands, not technologies.
Heck, if you google for "morphos", there's a hell of a lot more results than for "haiku"!

Of course, I could be wrong. But I don't want to deviate attention in this thread. How's that PowerPC computer going, Konstantinos? Have you taken a look at COM Express modules?

Haiku is something that people can try on new and easily found hardware. There is a LOT of excitement about the project, there are new developers *every* day joining the project and helping here and there. From a developer's point of view -like myself- Haiku is just beautiful, which is not something that I can say for MorphOS -I used to code for the Amiga OS back in the 90s and compared to today's OSes it was just nightmare. Both OSes are fast, offer quick boots, etc, but Haiku is much more probable to get good software ported than MorphOS or AmigaOS, since it's POSIX compatible and it has modern compilers like gcc 4.3 (and now 4.4). Heck, in just a few months Qt4 and Arora have both been ported. And it's OPEN.
MorphOS: 263,000
Haiku: 8,440,000 (but that's incorrect, haiku is just 'poem' in Japanese)
Haiku OS: 655,000
I don't see it happening. I'll post details on my blog tomorrow as I promised, but so far response has been very small. I can't even convince myself to spend money on it with such numbers, let alone convince an investor.

I was/am interested in a new powerpc board. I didnt respond because i am quite low on cash right now and sadly I had doubts about this really happening. I was interested in the ppc efika or sam440, but hesitant about anything less than 1 Ghz. Just googled to see if the sam440 are still around and all I get are dead links. Seems the only option for powerpc now is an IMB workstation; not cheap and probably a really hot toaster too.

The ps3 could have been a decent option if sony made any effort (no hardware accelerated graphics and no support on the new slim).

Right now I hoping my 2 ibooks keep working :)

matt

Konstantinos : I am sad but I understand your decision ... I saw your project as a great opportunity !

matt : An email didn't mean "I will buy the board, I have the money for that", it was for an evaluation of interested people (which board, wished cost, ...).
You don't find Sam440 to sell ???
http://www.acube-systems.biz/index.php?page=hardware
And you also have to click on the page "resellers".

About the number of Haiku downloads, I have doubts too but even if it's true, we have to say that people downloaded just to try for example like me with VMplayer. But there is no software to run MorphOS or OS4.

yes it's sad but yet it's a way to face the truth about the opportunity for a ppc desktop market, even small.
i suppose it arrives a bit too late.

i downloaded Haiku myself a couple of times as i used it for a VMware workstation and esx server training i had to give last week, i wanted the training members to install something else than windows for a change. So i agree that it doesn't reflects the userbase accuratly.

I think your little "market research" online has been kind of pointless, and you can't use the results for any kind of decision making IMHO. There are several flaws, but most of all you never provided a clear option on which people could decide upon. You were really going to some *investor* with this fuzzy idea?

But OK, here are my (rather late) thoughts: The 8610 is the most interesting option IMHO. Its built video and the way it works (renders the screen directly from system mem, 1080 HD, etc) makes this chip the most interesting IMHO.

Make a modular board using the 8610 with a very small form factor (i.e. could fit into a box like this one). It should have better sound than the AC97, and it should have all the usual Audio/Video connectors as modern HiFi consumer electronics has (like TV's, STB's, etc). It should have a single gigabit ethernet controller, a 2x SATA controller, and a 4x USB controller on board (at least one for "internal" or "front side" use). Beyond that it should have an option to add daughter cards or even bus expansion boards.

Then you would have a board that could be used for a numerous things:

Put it in the case similar to that Efika MX Open Client one, and you will have a similar product but with better performance and better screen resolutions.

Put it in a case similar to the Efika MX Netbook case, and you will have a Netbook with very interesting specs.

Put it in a traditional HiFi CD/DVD player kind of case, with a Blu-Ray unit, an internal multi memory card reader (from USB?), and a HDD as an option, and you would have a high end, networked media player.

Or skip the traditional HiFi rack cases, and make something new of it instead, with a "biggish" touch screen, on which you could interact in a way á la iPhone?

Make a DTV-T/C/S tuner as a daughter card, and you would have a high end STB that thanks to the "open OS option" and ethernet connector will be very interesting for the "tinkerers".

Make a daughter card with *two* (or more) DTV-T/C/S tuners on it, add a HDD, add a Blu-Ray R/W unit, and you would have a very high end PVR unit on which you can watch one show and record others at the same time, and thanks to the "open OS option", ethernet, etc it will have specs that few regular consumer products of this kind has.

The two products above could of course be built directly into a TV screen, for a different kind of product.

Or of course, you could design a bus expansion board (á la Zorro expansion boards for A1200, etc), that makes you mount the whole shebang into a standard desktop case, and use it as such.

I think it would be difficult to *not* sell more than 120 units of a flexible and modular board like that. In fact, I even think it would be *easy* to add a few zeros after that figure. But of course, your little "market research" can't possibly hint anything of that, and that's why it was kind of pointless IMHO, and can't be used as decision making materials, neither for you nor for any investor...

Hi.
I think it has not been pointless at all. I think it has shown quite clearly how little interest there is in a new PowerPC based platform - except for a few die-hard fans here and there.

I think we have to face the truth of the situation here: the PowerPC platform, for all its advantages and plus points, is dead in the form most of us would like to see.

The cost and time investment needed to build something up again just does not sound realistic. The industry is focussing on one thing right now: ARM - let's take the opportunity to ride the wave and not look back or focus on what are basically impossible dreams in the current market.

Besides all of this - the hardware platform really doesn't matter that much. Focussing on ARM+NEON will have some real world benefits today - not in the distant future. Jobs in this market are required today. The combination of Linux on ARM and Symbian kernel on ARM, with Qt on the GUI side is where Nokia is heading. There are a lot of other players in this market. It is a focus - let's be part of that and get noticed by the rest of the world.


Johan.

_________________
Johan Dams, Genesi USA Inc.
Director, Software Engineering

Yep, I have a blog... PurpleAlienPlanet

If the sam440 is still going then there is hope :)

Except that it *doesn't* show that, for a numerous reasons.

And BTW, I have two NAS systems from Synology, a DS107 and a DS107+. They look like this:



Not until recently, when I was about to flash their firmwares, I noticed that the first one mentioned was actually based on PPC (while the other isn't). Did I care? No. Did it affect my purchase decision? Absolutely not. I wanted some technology that had a certain set of specifications. That was why I bought it, not because of some "interest in PPC hardware".

I believe you could create a wide set of very interesting products based on pretty much the same 8610 motherboard, if you are clever enough when outlining its design, and open up for modularity.

Maybe this is more what an investor would like to hear, and not some vague poll on Amiga forums, where people didn't even have clear options to vote for?
I know the PPC roadmap isn't of much interest for any of us anymore, and I too am very interested in what's happening in the ARM arena at the moment, but that doesn't take away the fact that the *8610* is still a very interesting CPU, that can be used to create *very* interesting products.

The Sam price/performance ratio makes it utterly useless, and it's certainly *not* an option for... well, *anything*...

The CPU has never mattered for the broad audience - only for die-hard
fans. Even Windows-users couldn't care less.

People want a complete package that works.

My ZyXEL NSA-2401 NAS runs a PowerQUICC CPU. Most of ZyXEL's
new USG-firewalls all run on PowerQUICC.. Nobody cares and I
wouldn't have known it unless I opened them up and checked
the CPU myself.. (and when I attached the serial cable and
saw the Linux/PPC kernel booting..)

Hi.

That's my whole point. The CPU doesn't matter. No one cares. No matter how nice the 8610 is, where is the future in this line?

It's not enough to have a one-off product, there should be a successor to the product, and a successor to that. This is just so much easier to develop when the CPU has a roadmap, and thus you don't have to change the entire design migrating to a new CPU architecture for the next product in line.

This is what investors will look at.

ARM provides just that - a clear roadmap ahead. Where is e600 going at Freescale?


Johan.

_________________
Johan Dams, Genesi USA Inc.
Director, Software Engineering

Yep, I have a blog... PurpleAlienPlanet