|||

[ Edited by Velcro_SP 29.06.2011 - 12:14 ]

IMHO, such a bounty would only make sense if some of the bounty-money would go back in to a price-cut on the licence-kwy for those who contributed, ideally into a general price-cut as it kind of sucks when the OS costs allmost as much as the HW...

@Velco-SP

We have nearly 20 CherryPal/LimePC units as we did a significant amount of software development around the 5121e. This work was funded by Freescale over months.

We tried to explain a number of times here and on PowerDeveloper why the 5121e is a waste of time for use with an OS like MorphOS. Please don't waste you time and money on this proposed bounty. If a MorphOS-Team developer wants one of these systems we would be happy to send one to them. This is not the issue.

2603053373_0424f96061.jpg

Here is one next to an open Open Client earlier this year on a table in the Office.

Here are some more reasons:

1. The Freescale Team that handled the business and marketing for this part were let go by Freescale last month. The roadmap that was planned for the 5121e, including the 5123 has been significantly altered. It is doubtful that the 5123 will ever be produced. Neither the 5125 or the 5130 will have an MBX or AXE core. The MBX and the AXE and the load balancing work to be done in conjunction with the e300 core would be only useful for the 5121e. The 5121e is now fully positioned as an embedded part focused on embedded markets. Freescale's consumer focus is to be around the i.Mx family (ARM).

2. CherryPal lost their funding partner because they continuously misrepresented themselves to the market and to their partners. CherryPal has absolutely no intellectually property and no internal technical skills. They are a marketing organization with no money.

3. The mctera/LimePC management have absolutely no regard whatsoever for commonly accepted business practices. All the significant work done on the hardware was outsourced. Most of the companies and people that provided services to mctera have still not been paid. We heard as recently as last week that even component suppliers have not been paid -- this includes Freescale themselves.

If you are looking for more specific information, please write us.

Porting MorphOS to the 5121e is a waste of time. It is too bad things worked out like they did. We had high hopes for this chip.

R&B

so is there actually another potential morphos hardware besides the used and old mac systems? what is morphos's roadmap for the next few years? anything on that genesi will participate and contribute to? maybe a new pegasos3 on a 8610 basis?

to me it looks like, morphos will end with the completion of the mac mini (ibook) port. so what to expect?

Quote:

TerraSoft PowerStation :))) Bounty? ;)

|||

[ Edited by Velcro_SP 29.06.2011 - 12:13 ]

What's the benefit for the MorphOSteam compared to their Efika port ? Just another underpowered machine that runs MorphOS. I know it's all good to fight global warming, but does that fight really have to start with people that want a nextgen Amiga?

I actually wonder how many people actually registered MorphOS for Efika and how many (did) have an Efika and resold it ( or want to )because they don't feel comfortable with it's specs and spending a 150 Euro for the OS.

A bounty to port to something faster ... or maybe a macbook ... much better. And running MorphOS on PPC macs is recycling ... also good for the environment.

|||

[ Edited by Velcro_SP 29.06.2011 - 12:10 ]

@Velcro_SP

Please don't be so hardheaded I disagree with you. Genesi know what they are talking about with this hardware because they have been working with it for several years now. Also Genesi have always backed things like bounties and MorphOS.

I'm not sure what the future holds for Genesi, because Freescale seem to be going downhill. The group who bought Freescale for Billions should know what is going on!

In these tough economic times, Freescale can't afford to lose all their "mission critical" customers because they couldn't deliver a fully functional chip.

If a company had spent 2 years and millions of dollars developing for this chip they would not be able to deliver a "mission critical" product based on it. What would you want them to do? Spend another 2 years re-developing the software and millions more? Please understand, in these tough economic times.. they don't have that option! Even the auto companies are asking the Gov for loans of billions of dollars!

You could make the argument that Linux coders will work for FREE and could put together a functioning linux for this device.. but that's not acceptable for most commercial "mission critical" customers. It just ends up looking like another Articia chip..

@ Velcro_SP

I read the 8610 specs slightly different: For the 800MHz and 667MHz versions it should be possible to operate them w/o the need of active cooling.
The 800 MHz version is rated @ 9,5 W for worst conditions (max temp, all subsys 100% active) which is a bit critical for a passive heat sink, but typical max load power consumption is als low as 5.8W.
On 667 MHz the TDP drops to 8.5 W for worst conditions and to 5 W for typical max load. The 667 MHz version should work with only passive cooling. And a 667MHz 8610 with the e600 core offers quite a lot more computing power than a 400 MHz 603 core.
The lack of 3d hw acceration may get compensated by Altivec, or for higher spec applications, through the addition of dedicated gfx chip.

The 8610 is cheap *and* scaleable enough to make it for very low end (667 MHz version) to low end (800-1000 MHZ version) to mid range (>1Ghz plus optional gfx card): One basic design, many applications. I guess that's the only way to cut down investments and to actually get the chance to see a ROI and keeping end prices relatively low.

I loved the 5121 idea (super duper cheap and ultra lowest wattage) a lot, but I think it's time to move on. Plus the cherry pal (as it is offered now) is not what I would call super duper cheap.

[ Edited by Zylesea on 2008/12/8 15:47 ]

Quote:

Similarities is perhaps more obvious in this picture:
https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=5822&forum=11

Quote:

Yeah, I want to see a 8640 based desktop machine in MicroATX form factor!

(Doubt it would happen though)

Quote:

AFAIK, Genesi was part of *creating* this vision, unlike CherryPal. Together with THTF. CherryPal isn't making hardware, they are marketing a rebranded product that is what *should have been* an "NG Efika"/"MyPal"/LimePC/Call it what you like (and of which MorphOS could have been part of). Things obviously didn't play out that way.

Quote:

Maybe it isn't CherryPal that deserves the applause?

Quote:

The most interesting things would have come from *the other* LimePC devices IMHO:

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=5638&forum=2 (scroll down a bit)
https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=5820&forum=11

Quote:

...and it is built on a CPU that lacks coherent memory, and possibly other things that was announced at first.

Quote:

AFAIK it also lacks 2D acceleration (which is very nice to have for any kind of GUI), or am I wrong there? And dedicated graphics ram? And it has a limit of SXGA (1280 x 1024 or corresponding wide screen format).

Quote:

Exactly! And since we just removed the main point with using the 8610 in the first place by adding this, then we should go all the way and choose a 8640 + a SB600 southbridge instead that will offer so much more!

Quote:

I must say I agree!

Quote:

Because of cost and versatility. the 8610 based device *can* work w/o additional gfx schip, an 8640 not. Basically the same device design could be used for low cost low end to mid range.
Plus I don't know about the prices of the 8640, but the 8610 @ 667 MHz starts for less than 85 US $ (official Retailer price list quote).
I'd say the 8610 is most versatile one.

Quote:

I care about protection of the environment, ... but since nextgen Amiga's are already years behind what speed is concerning... having the Efika running MorphOS is already green enough for me. We need a powerfull machine, and that's where I would donate money fore when there is a bounty. Not for a port to hardware with questionable quality and specs.

[ Edited by Oepabakkes on 2008/12/8 19:06 ]

@ VelcroSP

Quote:

The power consumption cited in CherryPal's advertizing materials is nothing short of a bold lie.

Also, there are actually various existing thin clients that consume similar amounts of power. (The AMD Geode GX thin client reference designs, for example.)


Quote:

Manufacturing and shipping a product across the globe tends to consume a lot of power. Any company that is serious about environmental sustainability has to find a balance between minimized operational power consumption and a maximized period of product usefulness (read: performance / modern features).

Statistically, consumers replace low-end computers far sooner than mid-level to high-end models. Buying and operating three energy-efficient low-end computers over the course of a decade can easily harm the environment more than buying and operating two less energy-efficient, but mid-level computers.

In summary, the operational power consumption is only half the story at best. A faster 86x0-based computer might well be a "greener" solution in the long run if it provides a noticably extended period of product usefulness compared to a CherryPal system using the 5121e.

(On a related note, I hope that I will never have to see MorphOS 2.x and above run on any graphics chip that is on par with the 5121e's PowerVR core, which is targetted at mobile / embedded usage and inadequate for modern desktop computing.)

|||

[ Edited by Velcro_SP 09.04.2011 - 10:17 ]

|||

[ Edited by Velcro_SP 09.04.2011 - 10:18 ]

anyway, period.

the CherryPal is based on a buggy design and should be forgotten here. The MorphOS team isn't interested in it at all. They focus on the mac port. I don't think they are interested anymore in anything less poiwerful than a pegasos2.

"MorphOS on CherryPal" sucks anyway, what a lousy name for a computer


[ Edited by SoundSquare on 2008/12/8 22:56 ]

[ Edited by SoundSquare on 2008/12/8 22:58 ]

|||

[ Edited by Velcro_SP 09.04.2011 - 10:19 ]

Quote:

In the same respect, MorphOS shouldn't run on anything less than an e700/G5 level processor for 'modern desktop usage'.

Quote:

We've enjoyed computers without cache coherency since ages. Cache coherency only makes sense in multiprocessor architectures. It's not an excuse for leaving out a CPU, it only means that you have to write drivers carefully. For Genesi, the lack of cache coherency meant that they couldn't reuse a single software in different computers, so it hurted their projects.

But it's true that the MPC5121e is flawed. But more for commercial reasons than for technical. And I'm not saying that technical aren't important, because writing an operating system for this little triple core beast requires very good programmers. Which, precisely, is the kind of programmers that MorphOS enjoys in its ranks.

But anyway, the future of MPC5121e based PRODUCTS is not clear, if it has a future at all. The key people for this chip (and its successors) are out of freescale. One can expect that supply is shut down anytime freescale remembers its not making any money out of it.

Finally, all this crap about ecology makes me nuts. People, it's been always the same thing, they just want to sell more and more.
If the trend now is to make smaller and cheaper computers, it's because they want everyone to buy five computers instead of one. It's not about saving the planet, but saving sales.
As technology has evolved, in a way that computers can be consumed as commodities, it's completely obvious that the age has come for cheap commodity computers. It's not that we are all ecologic now (which we pretend to), it's that now it's possible, and marketing just makes use of it, like everything else.
It's not about saving the planet. Although it does sound right in the commercials. If we want to save the planet, we should consume as little as possible, not the opposite.
We, with our aging old computers that last for years in our caring hands, are more ecologic that those changing computer every two years, saying that each new one is more envoronmentally friendly.

The problem with CherryPal is that it's underpowered. Yeah, 400Mhz is not enough to watch 7MegaPixel images in a fast way. AXE is useless if nobody creates AXE specific code. Things would be different if CherryPal had a laptop.

There's no need or space for another Efika-like MorphOS machine. Mac Mini port is far more interesting than CheryPal port.

Couldn't we simply jump to an Altivec compatible cpu?

Quote:

But MorphOS is a perfect match for this kind of computers. it's only us that want a desktop computer. It's a real shame that the MPC5121e can't be used, because it badly needs MorphOS to shine.

Quote:

That's where the MPC8610 comes into play. Genesi keeps on progressing with freescale's development board. And that's not the only route, read all about it.