Projects are approved when they are approved; it isn't about which hardware will be available first :)

The current program will run based on Freescale's board. When we have our own board, we will swap to ours. The whole point is to improve the environment for the release of the consumer level product.

Otherwise we will be trailing by six months waiting for someone to pick up dual core optimizations and PCI Express graphics driver bugs etc. :)

This way we can "seed" the market, the best developers will get early access to the chip. Remember, it is still not a giveaway. We will be looking at applications which can SPECIFICALLY take advantage of the hardware, rather than just throwing boards at anyone who thinks four ethernet ports will save them buying another ethernet card.

Personally I am looking forward to projects about multithreading improvements, AltiVec optimization of existing and non-existing software (feel free to create some AROUND this chipset).

It really is sort of tiring to see 200+ project ideas which are "I want to compile something already in Linux and just run it". It is very easy for us to look at a project idea and tell how long it will take you to do if you are competant; and if you are competant, and it will take a few days, is that worth $4000 to us? Is there a benefit, a market opportunity opened, that we could not have just done in-house?

I'm more interested to know when the project might start and not when exactly it is approved. I think that's the whole point of projects here anyway, so sorry for not so clear formulation above. :-)

You are talking about multithreading improvements. The problem is that I already submited such project for OSW, since it's better to have 4 cores than 2 for tuning concurrency models inside the software. Do you think I should also submit the same project to 8641D projects besides my existing project in this cathegory?

Thanks,
Karel

I've improved the description of my project, giving the steps of the developement, which I think, might be interesting for Genesi, because it would uses multithreading, and Altivec optimizations for heavy computations, and would provide a good 3D engine to the platform, with a new 3D application.

>>> Open source 3D real time engine, optimized for dual core processor with Altivec <<<

Of course, for this project I also need good 3D drivers using PCI-Express slot (this explaining my first question to which czp did answer).

Will the $1500 system include an EIZO TFT?

Not likely!

EIZO TFT Prices

R&B :D

@bbrv:

Better source ;-)

Feel free, and feel free to improve the project description while you do it. The better it is...

They're the same prices just converted to Euro :)

$1500 does not include a monitor, but it may include a nice keyboard and mouse if you go through a local reseller (we don't have and don't expect to have the capability to buy keyboards of 25 different keymaps unfortunately).

I hope we will define THE defacto standard configuration in cooperation with resellers though. The idea would be that everyone can sell a Pegasos 8641D Workstation - and as long as you're all using the same parts, things get easier. It worked for Apple, with a limited range of standard configurations, so why not us?

>They're the same prices just converted to Euro :)

If you don't calculate with an old Pentium 1 ;-) you get 439 USD out of 344 EUR which is different to what the URL to EIZO said.

With today's competitive landscape (Mac Mini, iMac) I recommend bundling such a quality monitor if you really shoot for a $1500 G4 system.
The customer receives a complete PPC turnkey solution in a well rounded package that delivers true value.

That is not a bad idea, but perhaps we make it thus:

1500 USD Pegasos8641D
1500 Euro Pegasos8641D (RoHS compliant)

This is still a problem. The availability and the prices have not adjusted in the market.

R&B :)

bbrv:
You should make it RoHS compliant from the start. Isn't USA
or China going to regulate similar to RoHS any time soon?

RoHS was a major pain for Genesi ( http://www.powerdeveloper.org/forums/vi ... =5466#5466 ), and finding RoHS compliant parts for everything is still difficult and more expensive. As someone with a diploma in environmental protection, I would prefere to have RoHS compilant machines all over the world. On the other hand, it's easier (cheaper) to compete with non RoHS compliant machines outside Europe.

I think you misunderstood COMPLETELY how we would design it.

RoHS is about reducing the amount of hazardous materials in products. We can design around components which are RoHS compliant. This is easy enough if they are available (sometimes they are not, and you need to do some serious searching and even redesigns around new ones). But now that everyone is moving across the cost is neglibile, and production is easy.

However the RoHS compliance guidelines are not simply about buying the RIGHT components; you also have to pay the government agency to certify that your systems are compliant and tell them that you promise to destroy the systems mindfully when they are done with.

That costs some several (potentially several hundred) thousand dollars in fees, don't get me started on the TIME this takes.

Therefore it is much cheaper to build systems in the USA or China which are not certified as RoHS compliant, for the time being, to ship to those countries which do not require it. It may not sound like we are being totally environmentally friendly of course, but the basic designs will ALL be RoHS compliant, we just can avoid the costly certification process (which we are forced to do in Europe, so the systems will have to be compliant) if we ship to the USA (exept California, they have RoHS-alike rules), Asia etc.