Of industry, pricing, loss, and media interference

[Daniel ~]

So guys, if you want an AOA motherboard, SUBSCRIBE! ":O}

[wild_andy_c]

Well a lot easier said than done. We have to think about every aspect of the business in making that kind of board. R & D would be stretched for a product that would be very very low volume, so the R & D cost per unit sold would be huge.

Add to this the amount that it would cost for a short run of 6 layer PCB's (if anyone would entertain doing them in Hong Kong - I doubt it - or you'd have to wait until the chipset was obsolete it'd take that long).

Add to this the components that you'd use and the fact that you'd buy these in short qty and would therefore have to pay for single reels of components.

It starts to add up that you'd only do a project like this by subscription - and then at the end of it if it weren't the best implementation the users would be short for their money.

Theres no option for a one off world beater (and you can't call DFI's boards world beaters unless you are looking very short ahead of your nose in the context of this thread) when you look at it in perspective. The excitement factor of the young or naive will always try to argue past this point.

I started to get mocked by a young child with a dance mat yesterday. Erk!!!

[Favu]

Spoil sport

[dsio]

Building a low volume custom motherboard may be out of the question, but is taking an existing motherboard, and pushing it in a custom direction possible?

Is picking the best motherboard possible for modification, and selectively uprating, removing, or replacing stock components going to get you half way there?

Take the board closest to what you think a top end board should be, and go nuts with it.

[Gizmo]

Originally Posted by dsio

Take the board closest to what you think a top end board should be, and go nuts with it.

The best thing to do there would be to gather a collection of modifications, detail how they are done for anyone who wants to do the mods themselves, and then offer a modding service to those who don't want to do it themselves, I think.

Still, it's an interesting idea......

[cadaveca]

One i take upon myself. There was a post a bit back by "Big Pope", offering free capacitors for people to put on thier mainboards to replace the factory ones. I have seen some great successes from this, although i am kinda wondering if part of it may be due to the caps that are on the board being a bit "dry". Anyway, those that have undertaken this project HAVE seen higher clocks...which adds, to me, even more weight to the "ultimate" board being far too expensive for any company to undertake.

On the otherhand, tho, i have been loooking at getting a list of "parts" together that the "hardcore enthusiast" might want to have for this sort of mod, but i have found that without some sort of documentation about the layout of the board, it's like playing "pin the tail on the donkey".

I'd really like to see more interaction between enthusiast board makers, and those that use thier products. Although i understand the difficulties in catering to such a small and varied audience, the same things seem to be wanted by all...but if all the boards were teh same, there'd be no competition either.

[Daniel ~]

Originally Posted by dsio

Building a low volume custom motherboard may be out of the question, but is taking an existing motherboard, and pushing it in a custom direction possible?

Is picking the best motherboard possible for modification, and selectively uprating, removing, or replacing stock components going to get you half way there?

Take the board closest to what you think a top end board should be, and go nuts with it.

Good fall back point dsio! I like your thinking here! I wanted to say that before reading on to find out why this excellent won't work.":O}

[Daniel ~]

How far out our we from a Board maker that can custom build individual boards much in the way Saturn makes cars to order. ( I realize the simile has severe limitations, but needed a quick example!":O}

[cadaveca]

about 3mil.

[Daniel ~]

Is that your preliminary or final figure there? ":O}

[cadaveca]

It's far too complex. With each company holding info on thier products, it's impossible for some like, say, me, to design a board. There's several thousand components involved, for a start...and even changing, say, a sata controller, from one to another, could be a nightmare.


I hate to say it, but it would only ever be a feasible idea if there was a complete monoply of the market, for all hardware, by one single company. ANd that ain't EVER gonna happen.

But then again, if Micheal Dell can do it, why can't I?


Like Andy says...volume.

[danrok]

I like the customised board idea. Sounds more feasible. But, I suspect a lot of the work would have to be done by hand = too expensive.

[Daniel ~]

Does the job become any easier when we look at Video cards, power supply's and other custom O/C gear?? Or is this a dead end for us?

[cadaveca]

not really. less components, but seemingly lower tolerances...hence most high-end cards only coming from a few companies.

[Gizmo]

Certain levels of customization have to be possible. For example, the computers my company uses utilize single-board computers in an active backplane. We worked closely with Radisys to get the BIOS tuned exactly the way we wanted it. Now, obviously, there is a lot of difference between making changes to the BIOS and making changes to the hardware (plus, we are talking about a $900 board here), but Radisys indicated that they were willing to make certain changes to the hardware.

Someone with enough bucks to be able to seriously talk about 1,000 unit quantites that don't involve huge changes to the board might be able to work with a mobo manufacturer. If, as Andy says, they are loosing money, then an opportunity to build something that would actually MAKE money for them would certainly get their attention, I would think.

[cadaveca]

Originally Posted by gizmo

that don't involve huge changes to the board might be able to work

What do you think Dell et al do? Got thier cash?

[Gizmo]

Originally Posted by cadaveca

What do you think Dell et al do? Got thier cash?

True enough, but Dell et. al. are also fighting for the bottom dollar in terms of price. In our case we would be looking at building a design with cost as a secondary consideration, not a primary one, and we would be looking at $150/board, not $35. However, that aside, the cash issue is still significant. As already mentioned, at $100/board for a 1,000 unit run, you are looking $1,000,000.

[Daniel ~]

Wouldn't that be a 100,000 run?

(Petty has it's privileges!":O}