@Troels

Quote:

Why make a G5 capable motherboard now? I would offcourse buy one if it was priced right (which would be hard with a little production) but how many others would.. 1000 maybe 1500 Amigans and some Linux people?

1) Geek factors
2) Power factors
3) Price factors

1 and 2 are kind of related. The people that like the latest, greatest or the ability to use the latest greatest may find it attractive. I don't know of any announcements saying the G3 and G4 series are ending but the G5 is out and the G6, Power5 derived chip, is to be the next CPU so it seems the newer setup may be better for the future. However, no geek will look sad at a G5 that's 2Ghz vs having a 1Ghz G4. Even if AmigsOS doesn't make use of 64bit extensions the system will still be able to render faster then the G4 system.

Price -- Apple, supposedly, is buying G5 processors for cheaper then G4 processors. If this is the case and someone can make a G5 motherboard for equal to the cost of the G4 motherboard it'll save us all a bit of change. What can we do with that money? Buying more Amiga Software to use on the machine sounds like a great idea to me.

@tomazkid

Quote:

Agree that the most important thing now is to get OS4 ready, and get some good PPC-native software running on it. It's no harm though in planning a few steps ahead on the hardware-side. But what's the use of a G5 without software that can use it to it's full potencial? Except the geek-factor

I won't argue with that reasoning, and I agree to an extent. But there is no reason to not move forward with the hardware. Think about how old even the G4 is now. By the time an Amiga G5 project meets the market, Apple may well be moving to the next processor. If Amiga is going to gain serious momentum in the long run, the hardware has to catch up.... I know that OS4 is not even out officially yet, but the hardware design has to start sooner...

Think of it like this, the hardware will take a particular cycle of time to design and develop and test. Then there is additional time required to port the required parts of OS4.X to run on the new board.... then there is beta testing both parts.... Look at how long this has all taken with the current hardware (although I understand this is the initial port and that does make it take longer).... But would you rather that they wait for another year or two just to start designing the G5 based hardware, then another 6 months to a year just to get it and it's OS4 Port available, or start planning ahead now? That doesn't mean that people should wait for the G5 version before buying a next gen Amiga... but it just gives them incentive to support the current line, knowing that there is something better in the works that will make a nice upgrade sometime down the line....

@Herewegoagain

Heh, maybe Eyetech are working on it, but would rather we buy the current offerings than hold our breaths another year or two for G5.

I think there should be a push for a G5 in future Amigas. Okay, OS 4 isn't taking advantage of all the features of the PowerPC processor, but Ben Hermans did say that the blood, sweat and tears getting a working AOS on PPC hardware are over, now they can start concentrating on enhancing the Amiga OS to take advantage of such technologies that PowerPC is well known for.

The G5 means prestige, which is a good selling point in its own right. I hope that Eyetech has the PPC-970 somewhere in its sights for future boards.

_________________

@AmigaMac

Quote:

I hope that Eyetech has the PPC-970 somewhere in its sights for future boards.

Hopefully on upgrade cpu-cards fitting in the AmigaOne

_________________
Site admins are people too..pooff!

@tomazkid

The only way that would be feasible is by making a cache-controller with three busses:
G5 front side bus, 60x front side bus and memory bus for connecting L3 cache (the G5 doesn't support an L3, which is why you would need to make the support for it yourself). Given a reasonable amount of L3 cache running at the G5 front side bus clock and bus width you would get a reasonable amount of data throughput. It would not come cheap, though.

At some point this might happen for the the G4 Macs, might be something to keep an eye out for.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@Troels
Simply because to atract new and old users back to the platofrm we need a fast machine, the current A1's are expensive and slow, even compared to the budget G4 macs. They use old processors (The 1.4ghz modules aren't available to the public so they don't count) they use old, slow, rare and expensive RAM. They don't have modern conectivity options (AFAIK, I can't actually find any detailed specs regarding IDE, USB etc. on Eyetech's website). Mai haven't released the Articia P or Sa chipset so we will be stuck with the slow RAM and PCI and AGP until they do.
Also we need to keep up, Apple have G5's, who is going to but a slow and old G4 system twhen you can go and get a fast G5 with many more expandability options. Yes OS4 is a very good reason, but I doubt it is good enough for a lot of people to spend the sort of moeny we're talking about on.
Another aim of this project is to cut prices, IBM are ehavily backing 3rd party PPC970 motherboards and I hope to get help from them, also if KMOS are interested. The money that we raise from such companies would be used to pay for inital production costs to do relativly high volume runs (compared to the current A1's), end users would then be able to purchase 1 board per year (possibly per 6 monthes) at a discount which means they only pay cost, there is no money made. FUrther boards from teh same customer or large quanitity orders by companies would be cahrged the full price, this money would then be used for developing and producing new/more boards.

@AmigaMac
I'm not sure if I understand you, what do you mean what does it lack?
It has everything taht a normal motherboard would have, it uses the IBM CPC925 northbridge which gives it 8x AGP, 64bit PCI-X slots, DDR400 Dual channel 128bit RAM (A huge boost over PC100/PC133 single channel 64bit SDRAM) and up to 2x1ghz FSB PPC970 connections. The southbridge is an AMD one that connect via a hypertransport bus, this would give the same connectivity that AMD 64 motherboard have as they alsouse the 8111 southbridge.

@all
I hate to say this but can we please get back on topic, this is about getting a the money to make our own PPC970 board, not about Eyetech and their future plans.

@digitaldisaster

Hi,
I agree with your idea that AmigaOne is a developtment motherboard and is too old (Ram, USb 1.1, Hd, etc ) to consider a good item to buy now and to give for three / four years .... I am Amiga fan but I'm not so happy to buy a old hardware (AmigaOne) to have a new software (OS4).

Could you give me an idea of how many money minimum needs your idea please ??

@quartex

Right, just for you I got off my arse and set up a Pay Pal account, in total we need $6000 to purchase an evaluation board, once that is done I will work as fast as I can to create the PCB layout for the initial A1 board and send it off to be manufactuerd and assembled, this will cost a bit (A few hundred maybe) so we may need more payments in the future, basically any money you donate will go towards building an ATX G5 system. once that is done I will work on a microATX then mini-ITX system (In that order) which shouldn't take as long to do and will have minimal development costs (manufacturing and assembling of testing boards)

I am not expecting one inidivual to pay the ammount of money required, it would have to come from the whole community, think of it as $10 from 600 people and we're there!

This (should) be a button that links to a page for donating to the PayPal accoun I just set up, it depends on how much this forum can deal with HTML, otherwise transfer money to amiga@bezilla.rog (temporary address unitl either amingaworld.net or amiga.org or someone lese offers me hosting and an email address) Also, once I get home from my hols I will start work on the web site which I will host at bezilla.org untill I can find somewhere else for it.





Edit - OK that didn't work! The code's stil there but you can't see the button (fat lot of use that is!) For now can you just use PayPal to transfer money to amiga@bezilla.org until I set up the website

@digitaldisaster

I'm not sure if I understand you, what do you mean what does it lack?
It has everything taht a normal motherboard would have, it uses the IBM CPC925 northbridge which gives it 8x AGP, 64bit PCI-X slots, DDR400 Dual channel 128bit RAM (A huge boost over PC100/PC133 single channel 64bit SDRAM) and up to 2x1ghz FSB PPC970 connections. The southbridge is an AMD one that connect via a hypertransport bus, this would give the same connectivity that AMD 64 motherboard have as they alsouse the 8111 southbridge.

Everyone here seem to be discussing what was needed to make it complete or something (at least that was the impression I was getting). Anyways I did go look at the specifications myself a little further and it looks like a complete solution to me, just as you say.

I guess all is needed is for Hyperion to add this baby to the support list

_________________

@digitaldisaster

Sorry to sound negative but surely you must be aware of the history of the platform and the bad experiences of people that pre-paid for a future dream amiga machine.

What makes you think that people will donate towards such a project in the current climate.

Don't take this wrong but have you thought about how this will be percieved? How do you show your credibility?

_________________
Trust me. I'm a doctor.

@GadgetMaster

Yes I am fully aware of the history of the Amiga platofrm having been a part of it since 1992. I can't really proove anything, but I am not promising miracles etc. what I am primising is quite realistic, a reference board allready exists. It is a modififed version of this board which would be used. This is no"dream machine" it is very real. It is quite similar to what Eytech did with their A1 which is based on Mai's Teron board.
Also, I am not asking for large sums of money, as I said before $10 from 600 people would make this possible, I will of course put my hands into my own pockets as much as I can, this will most likely be to pay for manufacturing costs of developemnt/testing boards.
The reason that I had not tried to raise support for such a project earlier was exactly this reason, how would I be perceived.
Mind you, a lot of people donated to Amizilla with no guarentee of return. This is not the £500+ for an A1 and a promise of AOS4 that a lot of people were willing to pay for but $10 or so for the realisitic promise of a G5 A1 with AOS4 port.
Also this is not a give me money and eventually you get a machine, you aren't actaully buying anything, you are paying for some of the development costs.

@digitaldisaster

What's the paypal account?

@digitaldisaster

Fair enough.

How about a website outlining the plans and letting us know exactly how far you are and what needs to be done. How many people are involved, progress reports etc. etc.

In aid of legitimacy, the donations need to be audited to make sure that the money is indeed going where the donaters want it to go.

I would like to see you succeed thats why I would like to see everything done properly. The more transparent the plan the better.

_________________
Trust me. I'm a doctor.

@BrianK
amiga@bezilla.org

@GadgetMaster
Like I said, I am on holiday at the moment so I can't, as soon as I get back (Sunday, though I may sleep all of the day!) I will do a simple (graphically - anyone that wants to help is welcome to) website. Also I need to decide on the name of the project so that AW.net can set up a hosting account for me and email address. So far the suggested names are:
AmigaOne Point Five
AmigaTwo
AmigaFive
Amiga64

Other name suggestions are, of course, welcome. As are logos.

@digitaldisaster

PM me when you get back and want to progress. I'd be glad to discuss what I can do to create/host website and work to provide updates to contributors.

@digitaldisaster

I would be happy to donate but like most folks here need a bit more 'comfort factor'.

My initial thoughts are on the raw production cost of a small run of boards. Do you have any idea how much it will cost? Rough breakdown of materials/labour/sub-contracting....

On top of that there isn't much point building this board unless there is some sort of guarantee we can get support in software:
- U-boot
- AOS 'dongle' code
- OS drivers for chipset(s) - third party support?

Ducks need lining up, eh?


Thanks for running with this...

@digitaldisaster

A much lower priced G4+CPU combination would be much better than a G5 mobo.
We need cheaper A1s, that's my opinion.

I've spoken with many friends, over the net, and personally, and most of them are old amiga lovers, they would buy an A1 but the price keeps them to do so.

Before starting donating money to you digital, i'd wait for an official word from Hyperion.

@digitaldisaster

As Amiga is a registred trademark I would suggest to call the project something else for now.

Btw. a name for the motherboard isn't really important, the fact that AOS4 and Linux is fully supported is.

Guess the project stil needs a name :)
How about calling it AG5, A-G5 or something, atleast the A would indicate it was an Amiga (like A-1200) but not brake any trademarks (I think).

_________________

@Troels

As this is not a commercial effort (the idea of this project is a board made for the community by the community), nor is it a nmae of a product we can use the name Amiga to refer to the project in the same way that AmigaWorld.net can use it as their name. If I get any complaints from KMOS/Amiga Inc then I will of course change it.
No names have been decided for the motherboard but as it is to be an AmigaOne that is what it will be called with some form of product number/letter to desinguish it from Eyetech's A1-SE, A1-XE and uA1.
I have chosen AmigaFive as the name as it is nice and short and relates both to the project and to the AmigaOne.