Define good parts? Underpowered, underclocked without SIMD processor? G4 horsepower is the absolute minimum imo... I wouldn't spend more than $99 with something that has the horsepower of an Efika.

Look at ARM offer like BananaPi... Amazing for a sub $50 budget.

OTH, I am willing to spend $3K for a Power8 system without problem...

If it's worth it I would too. But only if the power of the POWER8 was actually used and not just "supported". I won't pay 3K for a system that needs to use Linux to be fully useful.

That said, I think supporting old PPC macs was a stroke of genius. It has prolonged the usability of MorphOS by a number of years. I hope they will focus on the OS when the SAM 460 port is done

Quote:

Even old 7447 that equipped Pegasos G4 are obscene high pricey on this site.

How could this be possible?

Octopart is the only reseller or there are others?

> http://octopart.com/partsearch#!?q=powerpc&start=0&limit=10&filter%5Bfields%5D%5Bcategory_uids%5D%5B%5D=2f57e25d1efcf490&sortby=avg_price%20desc
> Just a quick look at how much cost a suitable PowerPC CPU for desktop usage nowadays...

Apart from your search parameters being unfair with regards to proving your point as they produce a list with descending price order so that special expensive parts made for rough environments are listed on top, this "powerpc" search doesn't even find the more modern and more interesting Power Architecture parts like QorIQ T2080 or T4080.

Quote:

Well, at the end of the Pegasos 2 era, the Peg2 mobo including a 1GHz G4 CPU card was sold new at $399 (that is 326 EUR, quite close to your 299 price tag). And this was almost *a decade ago*, when its specs was actually kind of decent, and prices for hw was generally higher, and bang/buck ratio for general computers were far worse than today. Still, it was *not* considered "a deal for the entire community" back then.

Amiga One showed us what a bad idea it is to use development/evaliation boards for comsumer products. They are simply not meant for this. Bplan/genesi and Individual Computers are examples of highly skilled developers who can and will focus their development and extensive testing/bug elimination towards consumer level products, and put a long term commitment on support and (if needed) improved revisions over time. Buying a development/evaluation board won't give you this, and neither is schemes where someone pays a huge pile of money to some external entity for a one-time deal for them to develop a motherbard, if the end user product because of this arrangement will cost north of $3,000.

>> http://octopart.com/partsearch#!?q=powerpc&start=0&limit=10&filter%5Bfields%5D%5Bcategory_uids%5D%5B%5D=2f57e25d1efcf490&sortby=avg_price%20desc

> Even old 7447 that equipped Pegasos G4 are obscene high pricey on this site.

1.0 GHz 7447(A) starts at about 125 USD on that site. You find that obscenely high?

> Octopart is the only reseller or there are others?

Octopart is no reseller at all but a search engine for electronic parts. It lists the prices of the parts at the various resellers offering the parts.

I found this site:

Design your own dream motherboard:

http://www.element14.com/community/community/dreamboard?CMP=SOM-ADS-CARD-DREAMBOARD

[ Edited by Raf_MegaByte 05.01.2015 - 11:26 ]

Quote:

Done.

http://www.element14.com/community/docs/DOC-73376
I believe such T1042 based boards can be manufactured+built for less than EUR200 (including Radeon HD & case & etc).
So, for EUR299 one should get small margin to cover some costs.


Comment to previous discussions:
Also this "1250 - Freescale T1040RDB" is a full system with rack case etc., IIRC.

[ Edited by KimmoK 08.01.2015 - 10:10 ]

"Design" looks more like a painture than like a circuit board. What is the real use of this website?

@KimmoK

_IF_ you can build it for that price and MorphOS Team _can_ and _will_ port MorphOS to it, I will buy a computer for myself and chip in for some computers for the developers too.

Lot of buts and ifs here though.

hope for the future of morphos

hi

Quote:

+1

Marketing?

Quote:

Yes. I saw this fact 4 days ago. I started a motherboard PPC design to check if the site is for discussing design with expert people or if they propose to design dream motherboards for customers at a fixed price.
Still monitoring what happens. I'll make you aware of any news.

[ Edited by Raf_MegaByte 11.01.2015 - 14:01 ]

Neat idea.
I think Andreas mentioned something similar to me over a year ago.
Continuing to avoid mainstream hardware appeals to my hacking instincts.

> I think Andreas mentioned something similar to me over a year ago.

If I have I must have forgotten about it :-)

Quote:

You don't remember specially pointing out those lower cost PPCs to me for possible use on a board? ;)

Don't get too humble on me friend.

First you draw everyone's attention to the processor that eventually gets used in the X5000.

Later you mentioned lower cost derivatives of that core to me.

I remember that.

>>> I think Andreas mentioned something similar to me over a year ago.

>> If I have I must have forgotten about it :-)

> You don't remember specially pointing out those lower cost PPCs to me
> for possible use on a board? ;)

I thought you were referring to Raf_MegaByte's postings #32 and #38. If you didn't, I'm not sure to which posting you were replying to. And of course I do remember reporting the announcement of the first QorIQ T1 chips here on MorphZone.

Sorry, I forgot to approach this thread in the linear fashion you do.

Yes, you were the first to mention the e5500 core here and first to bring up the T1 series (which still looks like our best low cost option).

AND, to completely diverge here, many posters here seem to underestimate the difficulty in creating designs based on such high frequency parts.

Creating a new motherboard would be a task better left to a professional.

Quote:

>>>>> I think Andreas mentioned something similar to me over a year ago.

>>>> If I have I must have forgotten about it :-)

>>> You don't remember specially pointing out those lower cost PPCs to me
>>> for possible use on a board? ;)

>> I thought you were referring to Raf_MegaByte's postings #32 and #38.
>> If you didn't, I'm not sure to which posting you were replying to.

> Sorry, I forgot to approach this thread in the linear fashion you do.

Nothing about linearity. You said I mentioned "something similar" but didn't specify similar to what, so you must have implicitly referred to another posting in this thread. I assumed it must be Raf_MegaByte's postings, but as you say it's not, I'm lost here.

Yes Andreas, it was in reference to Raf's and Kimmok's comments.

I don't remember saying it wasn't.

Only that you read these threads differently than I do.

And the second comment was directly in regard to "asking manufacturers".

Where did you get lost?

[ Edited by Jim 11.01.2015 - 19:28 ]

Quote:

Sure all available aid from professionals must be used to avoid unnecessary revisions & learning everything the hardest possible way. And I fully understand why I often see HW designs that resemble a lot some reference boards!

HW wise I see those SoC chips appealing to work with as they have so much integrated...

After I got my NDA:s sorted out, I became overwhelmed with the details of T1xxx chips. Thousands of pages reference documents .... it takes me time just to learn to filter out the relevant for configuring the chip for hobby niche product (&to imagine what kind of driver work there is).
Well, at least it's interesting stuff to learn for an engineer. 8P

In the mranwhile I still monitoring ongoing of Open Source Hardware Laptop PPC Project from an italian hacker and young engineer but nothing still in sight...

http://deliriotecnologico.blogspot.it/2014/11/linuxday-2014-grande-interesse-per-il.html?m=1

[ Edited by Raf_MegaByte 12.01.2015 - 14:39 ]

Quote:

Well, he has a website
http://www.powerpc-notebook.org/

We can hope its in design phase, target is end of 2015.

What is new is that its supposed to be multicore 64-bit PPC laptop, which is leap forward from Mac models.

Interesting, FAQ reveals idea to virtualize PPC consoles emulation in addition to use Linux, so might be a first PPC gaming laptop

Will it be a success depends on design as well as on crowfunding, I hope we are at stage 2) - Actions to spread the idea.

Will it be "Gnome pants plan" or reality, waits. However, it would be good to contact the author to see what we can do as interested group.

Have you attempted a crowd source campaign?

A crowd source campaign could be an option ( in a next stage), another option could be a group buying campaign ( we are building a work-group of studying about these topics , contact us if you would like to join)

What is the roadmap?

Build a multi-disciplinary project team
Actions to spread the idea
Crowd-funding/Group buying
Beta tester program for software optimization
To go into production before the end of 2015

I hope something does come out if it and that it can/will be ported to MorphOS. If it's any good and doesn't cost a fortune. I can also join in a crowd funding if that's the case.

Keep us posted.

Quote:

It is various months they talk of crowfunding but I still haven't seen any bounty. We will seat and see....