Custom Motherboards?

Druegan asks: "I've been rooting around on the net lately checking out all the latest and greatest in new PC parts, plotting out the design for my next build. I'm finding lots of neat stuff, but I can never seem to find a main board that has just the right combination of features. Therefore, I want to Ask Slashdot: Is there any way your esteemed readership knows as to how a person might get a main board custom made?" "I don't know how practical this is, BUT I'm looking for a mainboard that supports a dual processor configuration for the AMD64 FX 55 processor, built around the nVidia nForce 4 chipset. I'd like two full x16 PCI-express slots with support for the nVidia SLI, as well as room for at least 2gb of dual channel DDR, and SATA Raid support. I also am looking to be able to overclock the bejeesus out of the whole mess.

This is only a test case, but there currently is no such mainboard available. I'd like to know if there is some way to get one custom built though, even if it is ridiculously expensive.. (yes, this might fall into the 'more-money-than-brains' dept.)

I'd just like to build the system to see how it'd work."

Comments

Doy

[Apreche]

The reason that you can't get a board that meets your needs is because your needs are unmeetable. IIRC the Athlon 64 FX can't be paired in a dual processor configuration. That's what the Opterons are for.

However, if you come up with requirements for a motherboard that are possible and you want to have one made, good luck. There are many companies that can do it for you for an extremely large pile of money.

Re:Doy

[OAB_X]

It cant be paired, you need the operton 2 series. Its basically the same as an FX, just with it being able to be used in multi-cpu configs.

Re:Doy

[davez0r]

when you said "it's basically the same as the FX", i'm reminded of my dual athlon XP computer from a couple of years back. AMD said that you need athlon MPs for SMP. but the MP and the XP were essentially the same processor. the only differences were that the MP had added testing done, and the XP had some traces cut on its surface to prevent SMP capability. some conductive paint fixed that. is that the case here? or does the opteron actually have additional functionality that the FX physically lacks

Re:Doy

[xenocide2]

Well, there is the significant factor of the extra pin in the 940 slot architecture. I'm reasonably certain it's not for nothing.

Re:Doy

[scheme]

when you said "it's basically the same as the FX", i'm reminded of my dual athlon XP computer from a couple of years back. AMD said that you need athlon MPs for SMP. but the MP and the XP were essentially the same processor. the only differences were that the MP had added testing done, and the XP had some traces cut on its surface to prevent SMP capability. some conductive paint fixed that. is that the case here? or does the opteron actually have additional functionality that the FX physically lacks

Yeah, the 2xx opterons have the capability to setup a single coherent hypertransport link. You need coherent hypertransport links to get processors talking to each other without managling each other's memory.

Re:Doy

[DetrimentalFiend]

I'd hate to say it, but this Ask Slashdot is retarded. The most pci express channels that I've seen any consumer chipset support yet is 20x, and 2*16 = 32. Soon, there should be dual core FX's, but you'll need to go opteron if you want dual 64bit cpu's. That much ram is no issue and can be found on nearly every mother board, but this guy FOR SURE has more money than brains.

Re:Doy

[ameoba]

This is all that needs to be said - the discussion can stop here.

Re:Doy

[OAB_X]

>is that the case here? or does the opteron actually have additional
>functionality that the FX physically lacks

As mentioned, it does have aditional functionality, but the FX and Opertons have a thermal spreader on them, so you cant even get to the top of the chip where you can complete these traces.

Even if you do, there is no gaurntee that it would work. Its better to just get the dual operton board.

Re:Doy

[sharkey]

The most pci express channels that I've seen any consumer chipset support yet is 20x, and 2*16 = 32.

Tyan offers a dual-Opeteron board with dual x16 PCI-Express slots. http://www.tyan.com/products/html/thunderk8we.html

Re:Doy

[Anonymous Coward]

Emphasis on the "extremely large pile of money".

And yes, you need Opteron 2xx CPUs for dual setups.

Custom or semi-custom, one-off or limited run production of high performance boards is possible, but expensive. This is why hardware from Cray costs more than hardware from Dell.

Check out Tyan. I'm sure they have something that will do the job.

Re:Doy

[morcheeba]

I'll just extend on that...

The Athlon used an external bridge chip that connected directly to each processor (the 760, as used on the Thunder K7).

The Opteron processors connect directly to each other, and one (or more) of them also connect to the rest of the computer. It's glueless, like in this board. Good thing, too, because one chip that connected to 4 or 8 processors would have so many pins on it that it would be incredibly expensive.

Re:Doy

[cowbutt]

The reason that you can't get a board that meets your needs is because your needs are unmeetable. IIRC the Athlon 64 FX can't be paired in a dual processor configuration.

That said, there were a number of 486-based SMP machines around pre-1995. The 486 wasn't SMP-capable, but the custom chipset used made such usage possible.

To the original poster, the cheapest way to have your own custom motherboard is probably to buyout Giga-Byte, or Asus or ECS, or something. :-)

Re:Doy

[cavemanf16]

Indeed. In fact, he's not even concerned about the SLOWEST component of the entire system: the hard drive! Get a custom-built 20,000 RPM hard drive and the entire system moves faster. Get that 2GB of memory cached on board the CPU chip die and the system moves faster. He just wants to throw money at a poor solution to his computing "needs" to compensate for his less than l337 computer hacking skills. He should just resort to having num-chuck skills and be happy.

Re:Doy

[masklinn]

I couldn't agree more with parent.

AFAIK nF4 doesn't handle dual processors or 32 PCIe lines.

If poster drops both (and he has to unless he wants to drop nF4 and move on to an Opteron based system), i'd recommand the DFI Lanparty nF4 SLI-DR

• Up to 4Gb memory with dual channel (i'm running mine with 2Gb dual channel as 2*1Gb)
• SLI (obviously)
• Overclocking bitch (able to feed your RAM with 4V, or raise FSB into the high 300s)
• Extremely complete packaging
• 2*PATA + 4*SATA2 raid 0/1 + 4*SATA1 raid 0/1/5

It's more than likely THE best board for A64 in performances and "tunability" (overclokability if you prefer), it's got every setting you may want in the bios (including the ones from A64Tweaker for your memory), embedded Memtest86 runnable from the bios, onboard reset and on/off switch (near the SATA plugs) when you're playing the chain crash game and it's able to save up to 4 BIOS configurations (to save stable settings, or whatever you want to save).

Oh, and even though it uses a Realtek chip the DFI guys designed a module called Karajan that allows the chip to use like a third of the CPU resources it usually uses.

Really hard...

[GreyWolf3000]

Most processors now have so many pinouts that you need multi-layer boards just to house all of them. Not even double sided boards would work. So any 'custom motherboard' would have to be made in a really expensive factory cranking them out by the thousands. However, you can sodder off the connections you don't use (like the built-in audio) and pretend you don't have them :P.

Re:Really hard...

[Anonymous Coward]

"Not even double sided boards would work"

You just made my day! That's one of the funniest examples of "n00b miscalculation of complexity" I've seen in a while.

Re:Really hard...

[amigabill]

The number of pins per device or layers oer board isn't the hard part. Current technology has some really fast stuff happening, and the timing and impedance matching will be the truely hard things to cope with.

I majored in computer engineering in college but we didn't take any EM fields classes. I would love to know this stuff today, but textbooks on the subject aren't easy recreational reads... Does anyone have recommendations for good books or web pages or courses that are good for those who are interested particularly in learning high-speed PCB layout? I've bought the two "Black Magic" books by Johnson and Graham but haven't had much time to study them yet. I'd love any other recommendations and advice I can find on the subject.

Re:Really hard...

[Anonymous Coward]

Toss the Johnson books, esp the second one. It's all easy stuff, covered elsewhere. You want Maxwell stuff.

Re:Really hard...

[Anonymous Coward]

Wow. You mean you can't just buy a resist pen and blank copper-clad at Radio Shack and do your own motherboard? Who would have guessed! I guess you could go reeelly high-tech and wire-wrap it. Just need to pind a CPU thats in a DIP package.

Sorry kids. This is only funny if you're over 30. Just humor me.

Re:Really hard...

[anderm7]

You can get cheap 10 layer boards from places on the net. Then you can build a solder reflow oven (a toaster+microcontroler) to mount the BGAs and those surface mounted resistors and caps. But the real problem is designing the damn thing. You have multi-inch traces that need to go 100+ Mhz. At 50 Ohms/50pF, that is one pain in the ass, even if you have an EE degree.

I don't think that will work

[Jjeff1]

Someone check me on this, but the FX series of chips don't support dual CPU.

And the Nforce 4 isn't a dual CPU chipset. So you'll never find what you're looking for.

Re:I don't think that will work

[OAB_X]

You would need a nForce Pro 4 chipset ofor dual CPU.

Re:I don't think that will work

[MerlynEmrys67]

Actually - the PCI-Express Chipset for AMD (The NForce) IS just a HyperTransport - Device chip... It will work for either a single, a dual, a quad, or for that matter an 8 way motherboard.

You can have two of them in your dual proc system (or 4 of them for that matter) and get your multi 16x.

Don't look for any manufacturer to produce such a thing, and frankly - doing a custom motherboard is probably out of your reach unless your first name is Bill and you don't mind throwing down a few million for a custom job

Cool.

[TsukasaZero]

Enjoy your new, non-realistic, motherboard.

What's the production run?

[Anonymous Coward]

Unless we're talking about thousands or tens of thousands, it's not economically viable. This is not a system-on-chip design, so it requires considerable debugging. Prototype manufacturing costs are always high as well.

My advice as an electronic engineer: give it up, your idea doesn't make sense.

Re:What's the production run?

[Anonymous Coward]

You're an electronic engineer? Wow, all this time I've been using the old-fashioned biological kind.

Re:What's the production run?

[harrkev]

What the parent says is correct, but let me elaborate. Let's assume that you DID have the knowledge and education to undertake this. Here is what you would need...

1) A complete schematic capture system with a really good board layout tool. My company provides one for me, so I do not know how much they paid. But I would guess that for one license, you would have to spend $10K or so. Certainly more than $5K. Note that there are budget packages out there ($2K or so), but you get what you pay for.

2) You will need to get the board manufacturered. Expect at least $2K or so for a handfull of boards (probably closer to $5K). Since you will not get it right on the first try, you will have to get multiple revisions made.

3) Buy parts. Some items have a minimum buy quantity. You might need to get entire tape-n-reel packs.

4) Get the board built. This is not something that you can do yourself, unless you have $100K equipment lying around. Contract this out. My best guess is between $1000 and $5000 for small quantities. I am not sure about this part, though.

5) Debug. You do have test equipment, I hope. Minimum equipment will be a good logic analyzer and oscope. The minimum that I would recommend would be an Agilent frame with a good analyzer card, pattern generator card, and o-scope card (tektronics make awesome scopes, but I do not like their logic analyzers). Expect to drop at least $30K on this. You can rent, though, if money is tight ;) Now, if you want to make your job easier, you might want to get some specialized equipment. Special analyzers exist which analyze one type of interface. You should probably have one for HyperTransport (assuming AMD), one to analyze the DRAM channels, one to analyze PCI, one to analyze USB, etc. You get the idea. This stuff will be used a lot less, so you should rent it.

6) Let's assume that the hardware works. Did you want a BIOS with that? License it. I have absolutely no idea how much this costs. You will likely have to costomize the BIOS for your board. You might be able to do this with open-source tools. Expect to spend some time on this, though.

7) You want that done this year? While you hold down a day job? Expect to spend $10/day on no-doze and Jolt cola.

As you can see, there is a LOT involved. The only reason that you can get a motherboard for $100 or so is that they make a LOT of them. The first motherboard is incredibly expensive. The second one is dirt cheap.

Re:What's the production run?

[thegrassyknowl]

And don't forget that it needs cup holders, wings and bubbles. What good is a motherboard without all of those things? I mean, really!?!?!?!?!

I'm your man.

[CarnivoreMan]

Go ahead and pay me $10,000 to my PayPal account and I'll get started right away. It'll be in the mail real soon! I can even through in a nucular mobo battery and wire it with a turbo button.

Re:I'm your man.

[foos_guy]

I'll have the $10k to you as soon as possible, once I get my millions of dollars from my Nigerian friend, we'll be in business!!!

I believe that the Tyan ....

[the eric conspiracy]

has something very close to what you want in their Thunder K8WE. You need Opterons though because the AMD64 doesn't support dual CPU, and the chipset is the NVIDIA nForce(TM) Professional 2200 and 2050 series, which is needed for dual CPU and PCI-X.

Not a good idea

[photon317]

Aside from what others have already pointed out, namely that your chosen CPU and Chipset don't appear to support dual processors to begin with, the idea of having a custom motherboard made is silly.

Motherboards are extremely complex peices of equipment. An enormous amount of work goes into getting them production-quality, it's a lot more than just wiring the rights pins of the right chips and sockets together. There's all the EMF and heat effects to consider, trace lengths and their effect on signal propogation, etc. Then ocne you have a baord that's even capable of functioning reliably, you have to make a BIOS for it and get all the right parameters tweaked correctly to initialize the board the right way - there's a lot of values tuned by the vendor for the board in question that you never see in your little BIOS setup screen.

Even among commercial boards, as we've seen on review sites, there are varying levels of success at building a rock-solid stable board. It requires an enormous amount of engineering man-hours to go through the design and testing process, and sometimes they still can't get it quite right, and half the boards are a little "flaky" under the wrong conditions.

So even if you wanted to drop some enormous sums of money (very enormous, I would imagine, orders of magnitude more than the cost of any custom built PC), it would be unsupported by other vendors (drivers, etc), and likely be plagued by little one-off problems like so many new boards are. Usually the vendor can see the trends in the problems based on numerous end-user bug reports, and fix it in the BIOS - but with just one user, good luck.

Chances are that if you actually made a competent choice about what motherboard features and components really suit your needs, you'd find they already make it anyways.

Re:Not a good idea

[ka9dgx]

Amen!

Computer design is not an exact science. The current hardware and software development processes use the end user as a test fixture, and rely on the sheer quantity of end users to help fix bugs. A one-off bespoke solution is not an option.

Accept the things you can not change, This is one of them.

--Mike--

Re:Not a good idea

[forkazoo]

Uhh, This is one of the hundred or so "me too posts." The Asker has no idea what he is doing.

That said... When I glanced at the headline, I assumed it was about somebody doing a custom 8 bit or 16 bit system, and was looking for some resources to make his tricked out Zilog box look more professional.

So, I'd love to hear from some experts about just what level of custom board would be doable for a hobbyist? How about a dual 8080, or maybe (joygasm) a dual 386 custom board? What free/cheap tools are available, and who does cheap low-volume PCB runs? Let's all pretend that a useful question was asked, and answer that!

Re:Not a good idea

[booch]

Let's all pretend that a useful question was asked, and answer that!

Damn, where's my mod points when I need them? Parent: +1 Insightful

Re:Not a good idea

[HeyLaughingBoy]

So, I'd love to hear from some experts about just what level of custom board would be doable for a hobbyist?

Depends on your definition of "hobbyist."
I'm an electrical engineer who also does hardware/software design & development both as a hobby and for pay as a side job. You can build pretty sophisticated hardware these days: it's never been easier.

I use the free version Eagle CAD to do PC boards under Linux, and there are plenty of shops that will build prototypes for under US$100. Most of the important parts manufacturers will either give you free samples, or sell small quantities off their website. My compiler of choice is gcc, also free.

My biggest problem is usually coming up with a cool project that's (a) worth doing, (b) doesn't cost an arm and a leg, and (c) can be done by one person who's married with a small child in less than a lifetime (i.e., 6 months or so :-)

As a result, my most interesting projects tend to be from the people who pay me to design something they need, but have no idea how to do.: generally people with plenty of expertise in fields other than electrical and software engineering.

The stuff I come up with on my own tends to not get finished; OTOH, knowing someone's waiting on me to deliver (and is paying) is a huge motivator to get done on time.

HTH :-)

Re:Not a good idea

[Short Circuit]

Nice post. You only left out one part...your contact information.

;-)

Tyan Thunder K8WE

[DeadBugs]

http://www.tyan.com/products/html/thunderk8we.html

- Dual AMD Opteron(TM) 200 series processors
- DUAL PCIe x16 with FULL SPEED x16 lanes
- (8) DIMMs for Reg'd DDR400 memory
- U320 SCSI and SATA-II with NVRAID(TM)
- Dual Gigabit Ethernet with ActiveArmor(TM)
- FireWire and USB 2.0 ports

Re:Tyan Thunder K8WE

[wh173b0y]

DUAL PCIe x16 with FULL SPEED x16 lanes... no, those are PCI-X slots, not PCIe.

Re:Tyan Thunder K8WE

[MerlynEmrys67]

DUAL PCIe x16 with FULL SPEED x16 lanes... no, those are PCI-X slots, not PCIe.

Try again. Those are dual PCI-Express x16 slots. You will notice 3 chips on the motherboard (well, you didn't since you obviously didn't go to the website where these claims are made in plain english - with a picture that shows the slots as well) - two chips are HT-PCI express chipsets, the 3rd is a HT-legacy chipset giving you an additional set of 2 PCI-X 100 and 1 PCI-X 133 slots

Thank you for playing - come again

Re:Tyan Thunder K8WE

[wh173b0y]

i looked right past them. an excuseless blunder.

Re:Tyan Thunder K8WE

[superpulpsicle]

The dream motherboard would have 10 PCI slots. That thing only have 3 PCI slots.

Most I have ever owned was on an Abit board with 5 PCI, 1 ISA. I swear, PCI slots mean so much to users, yet 90% of the board out there have 4 max.

Re:Tyan Thunder K8WE

[pipingguy]

Can I put two dual-core Opterons on that and have it work to my advantage? I'm waiting for 64-bit Windows and more multi-threaded applications before replacing my recently upgraded AMD Athlon SMP machine.

Re:Tyan Thunder K8WE

[Short Circuit]

If you want more than five PCI slots, you need a PCI-PCI bridge on the board. PCI cards can be electrically noisy on the bus.

Or so I've read, anyway. Studying electronic technologies is my hobby, not my job. So consider the source.

Re:Tyan Thunder K8WE

[Nutria]

If you want more than five PCI slots, you need a PCI-PCI bridge on the board. PCI cards can be electrically noisy on the bus.

Not only that, but you need a big-ass motherboard.

The only ones I know of that have so many PCI slots are expensive rack-mounted "enterprise" systems.

Re:Tyan Thunder K8WE

[runderwo]

I have a MSI BX-Master which has 6 PCI master slots with a custom PCI arbiter chip. All slots are filled (net, sound, mpeg2 out, scsi, video in/out, modem) and I've never had any problems.

Re:Tyan Thunder K8WE

[Short Circuit]

Found my source. See the second paragraph.

I can

[Anonymous Coward]

Give me lots of money and time. I will need to learn all about the processor's power requirements, the connection to all the other chips, their power requirements, their datasheets, until I know everything about every part that's gonna be needed.

Then you need to pay a licence for a schematic-capture to PCB suite with simulation and auto-routing. I think a few 100K$ for the Cadence suite should do it. Oh, you thought that someone who thinks two-layer boards are a challenge and uses Eagle could do it? Nope.

You'll need someone to procure small quantities of the parts you want, that is, get to know all the reps in your area and squeeze them for samples. In between the badgering phone calls, you still have to order all the other parts, while trying to find a local PCB assembler to handle your parts. Don't forget to supply the pick-and-place file and keep in mind the constraints when building your parts library. Oh, you thought you can just buy parts and the software automatically draws them for you with the IBIS model already connected? Nope.

Advise the fab of your PCB ahead of time of the layer count and size of the PCB so they can at least get the materials and open up a time slot for you. When you start the layout, you should be able to estimate a completion time, so it's not a problem, right? You did think of the stack up BEFORE laying out, right? You didn't just use any thickness you felt like, right? You know the difference between pre-preg and core, right?

Anyways, once you've done all the placement, assigned all the properties to the nets (unless you did all that on the schematic and configured the packager to forward the properties to the board), set up the DRC rules and routing areas and keepouts, and defined your via technology, and routed the critical paths, you can unleash the auto-router.

While Specctra is churning away, you can go back to the BOM. Are all your parts going to ship at the same time? Will you receive them at the same time? Will you ship them to your assembler or will you want to look at them first? Anyways, you should be able to tell them when to open up a slot for assembly so they can schedule it.

After the auto-router is done, it's clean up time. Check the artwork carefully. Create the assembly drawings, mechanical drawings, drill files, pick-and-place and IPC files. You *will* want to electrically test the PCB before tossing it to your assembler, right?

So the layout is done, you can start sending gerbers to your fabricator for DFM checking, and you can send the pick and place and solder stencil files to the assembler. While the DFM checking is done, you can extract the layout into Signal Explorer and run some simulations to see if there are any signal integrity issues. What's that? You want to run those tests BEFORE actually getting the board built??

Well, it's already been a year since the project started, and no one can wait anymore....

Dude, what you're asking for is the single most ridiculous thing I've ever heard on /. Even more outlandish than the "can I use a laptop LCD on a desktop?"-type questions.

Re:I can

[voisine]

I agree that the request is rediculous, but all that stuff you mentioned is why he wants to hire a company build the thing for him. It's not like he asked how to build one himself.

Re:I can

[Anonymous Coward]

The submitted obviously doesn't know anything about the complexity of a motherboard. So the grandparent response here should be very enlightening.

Personally, I get the feeling that the submitter is a 15 year old spoiled rich kid.

Re:I can

[John_Booty]

"Dude, what you're asking for is the single most ridiculous thing I've ever heard on /."
-----

I agree completely.

I've been posting on Slashdot for around 7 years and I think this is my first "me too"-style post. Please forgive me. It's just that this question was so. fucking. ridiculous. Literally the stupidest thing I've ever seen posted on Slashdot.

Re:I can

[amigabill]

>Dude, what you're asking for is the single most ridiculous thing I've ever heard on /.

I'd like an Amiga laptop computer. Now that I just made this guy look totally sane, go get to work and build his totally sane motherboard for him... :)

Re:I can

[Anonymous Coward]

Read some of his other posts - he's wondering why he's having problems breaking into IT. Maybe it's because he's so utterly clueless. Maybe.

Re:I can

[zero-one]

Once I've done all that, could you give me some tips on designing my own processor? I like those Intel and AMD ones but what I really wanted was a four core chip with a built in USB controller.

Re:I can

[harrkev]

Oh. His request CAN be done. Advance payment of $100K required. Total bill is likely to be in the %500K range.

You can do almost anything with enough money. Man went to the moon in 1969. The only reason that we don't do it today is that it is terribly expensive.

Re:I can

[Anonymous Coward]

Buy this software IC stuff

Learn it.

Go to TSMC, they will build it for you.

Happy?

Re:I can

[Jennifer E. Elaan]

You are absolutely right. I could probably manage the request, even, but it would take years and require a lot of equipment that I don't have (and can't afford). The task isn't so much difficult (in the sense of complexity) as it is massive and expensive.

$100k up front, $500k by the end sounds about right.

Re:I can

[Anonymous Coward]

Well... That sounds more feasible and cooler than what the other guy asked for.

Re:I can

[suraklin]

Ask and ye shall recieve

http://amiga.emugaming.com/paws.jpg

Buy some stock.

[jamey.v]

The only way I could see this happening is if you bought a controlling interest of a MB manufacturer stock. Think of it as an investment.

What's the production run?-Realty-Check.

[Anonymous Coward]

"My advice as an electronic engineer: give it up, your idea doesn't make sense."

Now maybe people will realize what software people have to put up with.

ridiculously expensive

[j-turkey]

even if it is ridiculously expensive

You're asking for quite a bit. A few people here have pointed out that FX 55 does not support a multi-cpu configuration (so you're already asking for a custom CPU/chipset), however, taking it a step further, the development process that goes into hardware development (especially for mainboards) is astounding. From reference chipset design (as well as testing and manufacture), to OEM's implementing that design (again, testing and manufacture). It's a big deal, and the process costs millions from end-to-end.

If you can afford this kind of solution, and are willing to take it on for personal use, I think that you're right that the discussion is quickly exceeding the more-money-than-brains department and entering the more-money-than-god department. Hell, if you can justify the cost of not only one, but two FX55's for a personal machine, you may already be in the more-money-than-brains department.

Check out the AMD roadmaps at Anandtech.com. You may just want to wait for the next FX chip release with dual cores, and an NF4 SLI nForce chipset. The dual core chip is as close as you'll get to true SMP, SLI will provide your 2 x16 PCI-e slots, the NF4 will provide support for dual-channel memory. These systems also come with SATA raid support, but IMO, an external controller is best (especially if you can find one with a battery backed cache). You'll have plenty of cash leftover for a badass liquid cooling system to overclock the crap out of that sucker. You will lose the exclusivity of having a super-custom system that nobody else can get, but hey -- it'll still be badass, and you will save $millions over custom hardware development.

As an alternative, if you're after exclusivity, you can start a hardware review website, gather a large readership, and then ask manufacturers for pre-releases of the latest greatest hardware for testing purposes. You'll get better-than-consumer support, and will have a machine that none of your friends will be able to buy for at least a few months.

Re:ridiculously expensive

[wgaryhas]

One nitpick: in SLI mode you have 2 x8 PCIe slots, even if they physically are 2 x16 they share a single x16 channel unless you go with the Tyan Thunder K8WE mentioned in a previous post that has 2 nforce 4 Pro chipsets on it.

Nice Troll :-)

[Anonymous Coward]

Running an Athlon FX (or two Opterons) on a MB while using the built-in sata raid. Next thing, you'll be asking for a high quality AC'97 codec to go with it...

dude

[Anonymous Coward]

you're a retard. custom motherboards..? what do you want next? a girlfriend?

Re:dude

[AndroidCat]

You might be able to get customized. (Careful with the overclocking!)

As easy as Legos

[Anonymous Coward]

I don't know what this guy is thinking but he seems to think that creating motherboards are as easy as piecing legos together. If only it were that simple. Maybe 20 years down the line we'll see something like this.

easy answer

[jenkin sear]

The easy answer- just buy ABIT. Not the board. The company.

Re:easy answer

[jasonmicron]

I'm more of an ASUS kind of guy. :p

Re:easy answer

[littlerubberfeet]

well, one could buy the board as well...member by member.

Dear Slashdot

[justforaday]

Dear Slashdot,

I'm looking for a videogame system that will play every game that's ever been made out there. I don't care how much it costs. I want it to be able to play every imaginable format, all through a single universal slot. Oh, and it can take any sort of controller ever made too. And while you're at it, make sure it can play every type of audio and video format (I'm planning on using it for my 1337 home theater setup). And don't forget to make sure that it'll be able to play everything that comes out in the future, too. I realize I may be asking a bit much, but I'm pretty sure it's doable...

Re:Dear Slashdot

[Ubertech]

Just give the MAME developers enough time 8-)

Re:Dear Slashdot

[justforaday]

No, I want to be able to use the original media! And remember, it has to be able to play wax cylinders, 8-tracks, vinyl, cassette, MD, etc all through the same single slot that it uses to play games...

Re:Dear Slashdot

[meltedeyes]

Mr. Compatibility by Tom Payne from the tape Vince Emery Presents The Funniest Computer Songs Would you put this system together for me some I ordered by mail and parts I got free. The pieces all work, except two or three. Please, Mr. Compatibility. Now with my Amiga, all I want to do is running MacDraw underneath OS/2, to write all my data on ROM-type CD, and display EGA on projection TV. I'd like to plug hypercard into my VAX, run my parallel printer on standard coax; Use it all under MS-DOS version 3.2, and a VCR on RS-232. My Cray running friendly interfaces with ease, to movie film shown on a 13-foot screen; My plotter and laser should speak Postscript well, networked on Appletalk under Novell. Autocad my Atari, VGA TTL; tied to my Apple ][ running Excell. Lay optic fiber, shaped in a token ring with a scanner and tablet - I forget anything? Put all this together - take an hour or two; hide those ugly cables, and when you are thru write up a contract saying you'd guarantee it'll all work forever, or you'd fix it for free. Would you put this system together for me some I ordered by mail and parts I got free. The pieces all work, except two or three. Please, Mr. Compatibility.

Re:Dear Slashdot

[justforaday]

Yes! That's exactly what I want. Except in a game system... : /

Three easy steps

[n1ywb]

Step 1: Pay major mobo mfg about $1,000,000

Step 2: ...

Step 3: Profit!

Re:Three easy steps

[PhraudulentOne]

You messed that up...

Step 1: Pay major mobo mfg about $1,000,000

Step 2: Wait a long time for it to be built

Step 3: Debt!

DIY

[amigabill]

Well, you could get licenses from the chipset vendors and hire up a layout guy to do the dirty work. Realize that one vendor refusing to NDA can toast the whole idea though, and Nvidia doesn't have a reputation for signing with just anyone that applies. A few years ago I asked them for an NDA that would allow my employer to develop proprietary (not open-source) drivers for their graphics chips, and they never even bothered to write back with a polite "no" answer. Good luck...

And this may not be a totally insane idea. FX support for muliprocessor issues aside, if you have an idea that's worth a profound amount of money for you yourself to have one, you might as well build a few extras and sell to other people too, and get at least some portion of your costs back... There's a few big companies that had their beginnings this way.

Yeah, I'm sure you can afford it...

[fmaxwell]

The same person who wants to buy a one-of-a-kind custom motherboard has to use a free e-mail address from Yahoo: Druegan2001@yahoo.com. Now that's funny!

Get an 8 way iwill system instead

[scheme]

Why not just get Iwill's 8 way opteron system, load it with 8 dual core optersons, put in two nvidia 6800 pci-e cards in sli mode, and put in a terabyte in scsi raid 5 storage as well as a well one or two scsi raid 1 arrays for the system and scratch disks. You can do this and still come out about $10 million ahead of your custom motherboard with dual fx-55 support. Besides why use a dual cpu solution when you can have a 16 cpu monster ready to do your bidding.

Re:Get an 8 way iwill system instead

[justforaday]

Because that would take way too long to overclock. Sheesh, don't you have any perception of reality? : p

What in the hell.

[Anonymous Coward]

Earlier this morning I posted a very serious "Ask Slashdot" regarding getting advice on requirements for a remote "store and forward" system I'm trying to design for physicians in very remote areas of third world countries. It's good to see that my topic gets ignored while ass-drippings such as this get air-time.

Re:What in the hell.

[rindeee]

But hey, what are the sick and dying in a 3rd world country in comparison to some knucklehead who needs dual Athlon FX's? You're so selfish.

Gee, let's ask Slashdot!

[c0d3h4x0r]

Posted by Cliff on Friday April 22, @10:30AM
from the we'll-publish-any-fucking-question-from-any-retard -we-can-find dept.

Fucktard asks: "I've been rooting around on the net lately checking out all the latest and greatest in new PC parts, plotting out the design for my next "Ask Slashdot" post. I'm finding lots of neat stuff, but I'm a whiner and I want something twenty times more technically sophisticated than anything that has ever been made by mankind. Therefore, I want to Ask Slashdot: WTF n00b? Why can't I have, like, anything I want?"

"I don't know how practical this is, because I'm a retarded jackass, BUT I'm looking for a car that can go 20,000 MPH and gets 800 miles to the gallon. I'd also like it to talk to me like that car from Night Rider, because I'm too socially inept to deal with real people. Oh, and while we're at it, could I get some of those awesome free-spinning chrome rims so that everyone will know how l33t I am? I also am looking to drag race the fucker through residential areas and risk hitting small children."

"This is only a test case. Had it been a real Ask Slashdot, the question would have been even more asinine. I'd like to know if there is some way someone could please track me down and kick my ass, even if it is ridiculously expensive.. (yes, this might fall into the 'more-money-than-brains' dept.)

I'd just like to poke the hornet's nest to see how it'd work."

has to?

[Anonymous Coward]

why pay for something you can have free? got something to prove?

Re:has to?

[fmaxwell]

why pay for something you can have free?

With e-mail accounts:

1. Privacy. If it's on Yahoo!'s servers, you have no idea who can see it.

2. Spam filtering. Yahoo!'s spam filtering is practically non-existent.

3. Backup capability. With everything stored on Yahoo!'s servers, you have no real ability to backup your e-mail.

4. Contract. Because you have no contract with Yahoo!, they can decide next month to discontinue their mail service, deleting all of your stored e-mail in the process, and there's nothing that you can do about it.

5. Capabilities. Yahoo!'s e-mail cannot be used with standard SMTP/IMAP mail clients, meaning that you only have access to their limited feature set.

6. Attachment size. With a Yahoo! e-mail account, one is limited as to the size of e-mail attachments.

7. Storage. The total storage for e-mail is limited to 250MB.

got something to prove?

Yes -- that I have the technical savvy to run my domain and e-mail server. My mail does not sit on some ISP's server. I can use any spam filtering methods and lists that I choose. I can use standard mail clients and backup my e-mail so that I don't lose valuable correspondence. If I want someone to send me a 650MB attachment, I can enable it. If I want to store all of the e-mail that I receive, I can (and do).

How are we supposed to take a question about cost-is-no-object custom motherboards seriously when it comes from someone who uses a free Yahoo! e-mail address, with all of it's inherent limitations, because he's either cheap or lacks the technical skills to run a domain?

Re:has to?

[amigabill]

> 1. Privacy. If it's on Yahoo!'s servers, you have no idea who can see it.

And if webcrawlers harvest my email address posted here, do you think I want it to find my hotmail address, or do you think I want it to find my "real" email address? I think that using free email accounts in relation to posting in forums is a good idea.

If my hotmail account gets overburdened with spam, I abandon it for another one. I ain't gonna do that with my real address... Stuff that's actually important gets told about my real address where I have more space, privacy, larger attachments, etc.

I'd say that him using a yahoo address for you guys to post on the front comments page is a good idea, as it help him maintain the privacy of his real address...

Re:has to?

[fmaxwell]

And if webcrawlers harvest my email address posted here, do you think I want it to find my hotmail address, or do you think I want it to find my "real" email address? I think that using free email accounts in relation to posting in forums is a good idea.

I create and delete e-mail addreses and even subdomains on an as-needed basis. An attempt by a spammer to send to an older, harrvested e-mail address of mine will fail and it may even be directed to a non-existent subdomain.

If my hotmail account gets overburdened with spam, I abandon it for another one. I ain't gonna do that with my real address... Stuff that's actually important gets told about my real address where I have more space, privacy, larger attachments, etc.

That's what confuses me: Why have a single "real" e-mail address? I have given out dozens of addresses, usually personalized for the sender (guess who got "_yahoo@{mydomain}" when I placed an order?). That way, I can kill an individual e-mail address given to a single sender without losing correspondence from every other sender. That lets me pick up, organize, store, backup, etc. all of my e-mail with normal SMTP e-mail clients and tools.

easy answer

[Anonymous Coward]

Seeing as how this guy has a degree in religion, the answer is totally obvious. If he simply prays hard enough, God will give it to him.

Cool, great idea

[nurb432]

Id like to build a custom handheld computing device, with.......

While anything is possible, i dont think you realize the costs/time involved. And if you have to ask here, i guarantee you dont have the funds or the expirence.

That is what PCI slots are for.

[Perdo]

Turn off the features you don't want.

Insert PCI cards with features you do want.

Looks like cost is no object, so try the K8WE, a pair of Opteron 275s, Dual 6800 ultras, and 8 GB of ram.

That will give you a quad with dual video cards for 4 monitors or one monitor in SLI mode.

firewire? check.
dual gigabit ethernet? check.
Any raid across 4 sata II disks? check.
SCSI320? check.
PCI-X, PCI-Express? Check.

The absolute fastest workstation on the planet right now.

http://www.tyan.com/products/html/thunderk8we.ht ml

Want it all in SFF?

http://www.iwill.net/product_2.asp?p_id=36

ZMAXdp

http://www.iwill.net/product_2.asp?p_id=36

try Iwill.

Motherboards ARE custom pieces

[mnmn]

Motherboards are supposed to only be cpu socket, chipset, and other sockets and connectors. You're supposed to add/remove cards and devices in a custom fashion.

Apart from the obvious non-dual nature of Athlon FX55, you'll find motherboards that meet your demand.

I personally am looking for a simple nforce3 or nforce4 motherboard with large number of memory and pci slots. I mean more than 5 slots, if 10 slots are possible in that space, great, else a pci brige chip/riser card could be used.

And would be awesome, and with the price tag of current boards too.

For the price of a custom mainboard...

[NoMoreNicksLeft]

He could easily afford a x8 or x16 cpu machine from a big vendor. One that has been thoroughly tested, rather than an engineer's first prototype. With his pick of actual SMP-capable cpus, no less.

So how about a less dumb motherboard question?

Q) I need as many PCI slots as possible, with at least a few being 64bit. 4-6 slots isn't nearly enough, I'm a guy that could fill 10+ easily. And a few (read:2-3) ISA slots would be nice also.

I'm not so picky on other things, but a wishlist in order of priority is as follows:

#(10+) PCI slots
dual cpus
64bit cpus
amd cpus
dual onboard gigabit
#(1-3) isa slots
dual onboard serial ports

I think that a passive backplane is the answer I'm looking for. Things like the Magma PCI expansion system (where 7 pci slots sit in their own rackmount case) aren't quite what I need. I understand enough about backplanes now, to know that I need a PICMG single board computer. Is it the right answer for what I want?

What price range are we talking, working up through modest configurations, up to the ones that meet all of my wishlist items?

Am I overlooking some other (presumably lesser known) options that would meet most of these needs?

Are there any pitfalls in installing and using linux on such a system?

I Will Design+Build it for $20 Million USD

[Anonymous Coward]

The price includes parts procurement, several board respins (I want to make sure I get the DDR interface correct and reliable), NDA and licensing agreements with various vendors, and my time (I estimate a few months to a year max)

You can have as many PCI, PCI-X, and PCI-Express slots as you like. SATA, SCSI, SAS, Fibre-Channel, etc are no problem. I don't think it will be necessary but my fee will cover the cost to design and fabricate up to one custom ASIC if required (likely needed to glue 2 Athlon64 FX parts, glue not needed for Opterons).

I accept Paypal.

While we're at it...

[perfectlynormalbeast]

... I could use a custom CPU. Since I'm left-handed, I'd like something like an athlon64, but the leftshift instruction should shift right and the rightshift instruction should shift left. I'd be willing to pay an extra $50 for that!

Go to Taiwan

[Matt_Bennett]

Seriously- if you wanted to build your own custom motherboard, do what the big guys do- go to one of the motherboard makers in Taiwan. You'll need them because they already have relationships with all the CPU and chipset companies. You (or the company that makes the motherboard) would very likely have to use something already released, or enter into some *very* restrictive NDAs to get pre-release silicon, and that would be only if you can convince the silicon manufacturers that they should use up some of their very limited pre-release silicon on you and your project.

It will cost millions and take 6 months to a year to get a good, stable system. Making a motherboard is far from the "just plug it in" stage, particularly when you get to new-ish technology (say, stuff released in the last 3 years). Everybody is still learning how to connect things up right and view the secrets of the development and testing as proprietary.

On the other hand, there is a big problem in what you're asking for- as far as I am aware, The chipset you specify doesn't support 2 x16 PCI-E slots, and opening it up to other chipsets, there are none that offer capability for 2 x16 slots (available now)

Re:Go to Taiwan

[Holi]

The chipset you specify doesn't support 2 x16 PCI-E slots, and opening it up to other chipsets, there are none that offer capability for 2 x16 slots

So add more:

From Game PC:

"Tyan's Thunder K8WE utilizes a combination of three individual chips onboard to allow for so much onboard connectivity, the nVidia nForce4 Pro 2200 primary chip, the nForce4 Pro 2050 secondary chip, and the AMD 8131 PCI-X controller hub.
The nForce4 Professional 2200 is the "primary" chip of the nForce4 chipset, as this chip supports 16 lanes of PCI Express connectivity, Gigabit LAN, and four SATA-II/300 ports, along with all the extras which go along with the nForce4 design (USB, PCI, IDE, etc). Alone, the nForce4 Pro 2200 is similar in design to the nVidia's own nForce4 Ultra for the AMD Athlon64 platform.

The nForce4 Pro 2200 can be paired with an additional chip to supplement its abilities, which is the nForce4 Pro 2050. The nForce4 Pro 2050 is very similar in design to the 2200, but is neutered to help keep costs down. The nForce4 Pro 2050 can supply an additional 16 lanes of PCI Express bandwidth, an additional four SATA-II/300 ports, and an additional Gigabit Ethernet port. The 2050 however, cannot support USB 2.0, PCI, or IDE devices. When the 2050 chip is added to a motherboard, you basically double your PCI Express bandwidth, SATA RAID port capabilities, and GigE port capabilities compared to a single chip nForce4 Pro 2200 setup. "

Re:Go to Taiwan

[Matt_Bennett]

I wasn't aware of the K8WE- but it is still not a true 2nd x16 slot off of the chipset- it is a second root-hub/north-bridge that hooks up to a 2nd processor- the only way that the primary processor has access to it is through the second processor. I'd like to see how it actually performs- from my non-SMB expert view- it looks like you still have a major bottleneck- the benchmarks aren't that stupendous.

I'm a bit leery of nVidia's chipsets- I'm working with some people that have a couple of the Tyan dual x8 PCIE boards- they've been pretty bad in terms of stability. Very fast, yes, just when they are running. They've got great potential, I just think they need more time and development for stability before I'm comfortable trusting them for any type of server application.

Re:Go to Taiwan

[Holi]

No it is actually another Chipset that hangs off the nVidia 2200. It does not hook up to the 2nd processor as far as I know and even if it does, i believe the HT bus would help alieviate that problem. Not only that but it is the only board I have seen that provides 2 FULL x16 slot, so you saying it is not a true x16 slot makes me wonder if you even checked out the board.

Re:Go to Taiwan

[Matt_Bennett]

Umm... I think you're the one that is mistaken, I did look at the datasheet on the board- I looked at the block block diagram that Tyan has on their website- The 2050 is off of the secondary CPU, and the 2nd x16 and 1 Gb Lan port (page 2 of the pdf)- everything else is off of the the primary northbridge (2200). Which kinda messes up my idea of what SLI is- I thought the PCI-E cards could talk to each other directly- actually, they have to make hops over 3 different hypertransport links (PCI-E x16 to 2200 to primimary cpu over hypertransport, hypertransport to secondary cpu, hypertransport to 2050, 20 secondary x16 PCI-E.

Re:Go to Taiwan

[Holi]

It is quite possible for me to be mistaken, and thanks for the link to the PDF.

Re:Go to Taiwan

[Holi]

Ok after checking that out I have to admit your right. I wonder how much latency that introduces or if their is a more efficient way of creating this connection. Agreed it does change how I thought SLI actually worked.

There are at least 3 boards that meet those reqs

[zipcube]

The tyan s2895 (thunder k8we)
http://www.tyan.com/products/html/thunderk8we.html

(just preordered one myself),

The Iwill DK8EW and DK8ES
http://www.iwillusa.com/product_2.asp?p_id=91&sp=Y

However, the Athlon-FX does not have enough coherent HT links for multiple cpus, so you will have to use the opteron 2xx series. As for dual 16x vs dual 8x, there is no real world performance diff.

Also, there are number of other mobo makers with similar boards on the way based on the nvidia 2200, 2050, and amd 8131/8132 chips.

what really kills me

[tolldog]

what really gets me with this request is that he is skimping on memory and HD...

Why go for the gold with a motherboard and limit it to 2gb ram and only sata raid? Lets get some real memory in it, at least 8 gb. And why not ultra scsi raid. Enough space so you can have a raid 5 setup.

I mean, if you are going to dream, dream big.

How about something more realistic/managable

[mebob]

For instance, a board where certain components are replaceable and upgradeable.
Starting even at a basic level like some of the boards in mass produced systems, where the basic board might only have 2 PCI slots, but you can slap on an additional card to extend the PCI bus. Or add-in legacy devices, such as serial, printer or PS/2 port so they would need to be on the base board. Most of this could probably be accomplished with PCI-E x1. Perhaps, their needs to be new standard or new extension of ATX for this. But in the end, this will interfere with any goals to make things smaller, at least in the short-term. I think the point is, that unless something was mass-produce to allow for customization, the chance of a truly custom system at the board level ceased to exist long before most people even knew what a PC was.
It however would be interesting to see a board manufacturer create a manufacturing process where they could build to spec. Something like people used to order from jeep, a roof, seats, doors rugs etc., were optional. Or a more recent example would be how Mini does things. None-aftermarket customization defiantly has a future in manufacturing but it will likely be a while before we go from an engraving on you IPOD to a completely custom board.

Here is how you do it.

[LWATCDR]

1. Contact AMD and have them fab you a pair of FXMP chips.
2. Contact Nvidia and have them make a custom nForce4 chipset that supports MP and 32+ channels of PCIe
3. Contact Asus and have them fab you the actual board.
4. Contact Award for the custom Bios you need.
5. Enjoy your custom MB and you claim to the title of King of the mods.
Total Cost... about 1 billion dollars. Okay maybe not but many million.

Good grief how did this make it on Slashdot?

Re:Here is how you do it.

[Anonymous Coward]

> Good grief how did this make it on Slashdot?

The real question would be why did it take so long to get posted on slashdot.

Re:Here is how you do it.

[NOPteron]

What he's asking-after, is essentially an ABIT workstation-board, which seems to be not yet on sale in RL. . .

The INTERESTING Abit 'boards.

The Workstation subsection of Abit 'boards.

The Twin-Opteron Workstation 'board that many drool after

Re:Here is how you do it.

[LWATCDR]

Except close is not not good enough for this guy.
He was very specific what he wanted.
" I'm looking for a mainboard that supports a dual processor configuration for the AMD64 FX 55 processor, "
Not Opteron but FX55s.

"This is only a test case, but there currently is no such mainboard available. I'd like to know if there is some way to get one custom built though, even if it is ridiculously expensive.. (yes, this might fall into the 'more-money-than-brains' dept.) "
I have to love this part. With out a doubt if he did build a motherboard to his specs it would be fall into the 'more-money-than-brains dept'. The very idea of a custom one off mother-board is just dumb.

All the necessary features.

[Anonymous Coward]

Make sure it has enough cup holders. Not for a little cup either, but one of those super-slurpers at the Kwik-E-Mart.

Some things are so snazzy they never go out of style!
Like tail fins... And bubble domes... And shag carpeting...

Wasn't there an article

[gotr00t]

about how the Athlon XP, which costed significantly less than the MP, could be modified (with a pencil, I beileve) slightly to achieve multi-processor capability?

I don't know if its possible in this case, but has anyone looked into modifying the FX to do the same thing?

What's the [cost of your] production run?

[Anonymous Coward]

"As you can see, there is a LOT involved. The only reason that you can get a motherboard for $100 or so is that they make a LOT of them. The first motherboard is incredibly expensive. The second one is dirt cheap."

Translation: Production costs are way more than distribution costs, and "economics of scale" are important. Now if we could only get the "information wants to be free" crowd to read your post?

Re:What's the [cost of your] production run?

[runderwo]

Production costs are way more than distribution costs, and "economics of scale" are important.

I hope you're not equating scale economics with scarcity.

Now if we could only get the "information wants to be free" crowd to read your post?

Information has a development cost. So does a motherboard (as its schematic is information). Information does not have a marginal cost of production, as its existence in unit form is not tied to scarce resources (unless you consider a human brain a scarce resource). A motherboard's existence is inextricably tied to scarce resources, so a marginal cost exists. "Stealing" information by making copies only dilutes a potential market. Stealing motherboards and reselling them not only dilutes a potential market, but also defrauds the producer who already paid for the components that went into the production of each motherboard. I hope you realize the absurdity of the comparison you have drawn.

What in the hell-Simulating an Answer.

[Anonymous Coward]

"Earlier this morning I posted a very serious "Ask Slashdot" regarding getting advice on requirements for a remote "store and forward" system I'm trying to design for physicians in very remote areas of third world countries."

And yours get's modded "funny" while someone else with the same sentiment gets a "troll". Consistency isn't a slashdot strong point. Anyway grab a book on queue theory. The rest relates to physical environment, the nature of the communications links, as well as what's being communicated.

The short version

[fm6]

Put it another way: it costs almost as much to build one motherboard from a particular design as it costs to build the thousands (anybody know a more precise figure?) as it costs to build just one.

It's sort of ironic: most of us depend for our livelihoods on the fact that computers are cheap. Computers would not be cheap without economies of scale. Yet few Slashdotters seem to grasp the concept. They're always complaining that nobody bothers to port their favorite game to Linux or Mac, or that off-the-shelf hardware never has the precise feature set they want. Or that they can't save money by ordering hardware that lacks commmon features they don't want.

Econonomies of scale explain why Sun is in trouble, and Bill Gates is the richest person in human history. It even explains why Enterprise was cancelled! People really need to understand the concept.

Chewbacca

[booch]

Wow, that sounds amazingly close to the Chewbacca defense.

Start taking basic electronics classes ...

[bergeron76]

After a few years, you might be able to build an embedded system using an ARM processor or something. From there, you'll couple your USB, etc. chips.

In about 10 years you'll be able to create your perfect board, however, technology will have advanced so far underneath you that you won't be interested in such a board anymore.

Basically, for all the work involved, it's not worth it. Buy 2 computers and get a KVM switch; or share the resources on a network, etc.

But for heaven's sake, don't try to create your own desktop mobo - it's just not practical.

I CAN

[nilbog]

Sooo....you're saying I can't use a laptop LCD on my desktop?

the hot board

[Anonymous Coward]

it would be cheaper to have a shop like Xi just build the whole box with your options