Computex 2005 Early Bird Coverage

zbojnik83 writes " Today was the official start of this year's Computex, but as always we were able to get a sneak peak at the show before the floor actually opened. With the show a day barely underway, we've already seen the first AMD BTX motherboard, a number of NVIDIA G70 graphics cards, an Intel motherboard that can be switched to an Socket-939 board by just purchasing a single card and the details of ATI's new multi-GPU chipset."

Re:Does anyone know?

[Wesley Felter]

Yeah, they are showing the G70 chip and the GeForce 7800.

*sigh*

[Leroy_Brown242]

I miss the good-ol-days when everything used the same socket. It silly stuff like this a non-issue.

Re:*sigh*

[Leroy_Brown242]

HOW PARENT WAS INTENDED

"an Intel motherboard that can be switched to an Socket-939 board by just purchasing a single card"

I miss the good-ol-days when everything used the same socket. It silly stuff like this a non-issue.

Re:*sigh*

[Leroy_Brown242]

LETS GO RIDE BIKES!!!!

Re:*sigh*

[pyrrhonist]

HOW PARENT WAS INTENDED ... It silly stuff like this a non-issue.

It ironic stuff like this a +5 Funny!

Re:Not enough

[Jeff DeMaagd]

I've read all the docs and specification PDFs that were released when BTX was announced. I still contend that the BTX form factor is unnecessary.

For one, it was designed such to accomodate Intel's excessively power wasteful chips that aren't proving to be worth the power they waste.

For another, most of the changes to accomodate these wasteful chips could have been done as simple tweaks of ATX, call it ATX 3.0. Without having to contend with nearly so much of a change, ATX 3.0 cases could still have housed ATX 2.0 boards. ATX 2.0 case designs would have only needed a few changes to work. Granted, ATX 3 boards might not work so well in ATX 2 cases because the older cases wouldn't have the necessary mount points to take the stress of the extra heatsink weight.

PicoBTX is nice but given the acceptance of the small form factor ATX systems, I don't see how picoBTX is different enough to be more accepted.

Re:Not enough

[MojoStan]

PicoBTX is nice but given the acceptance of the small form factor ATX systems, I don't see how picoBTX is different enough to be more accepted.

I thought the (supposed) advantage of picoBTX was the efficient cooling system that's already part of the spec, which can be easily (and cheaply) implemented by any case maker. The components that need cooling (CPU, chipset, and GPU) are in-line with the BTX airflow and the fan/heatsink/thermal module is part of the CPU.

Quiet small form factor ATX systems tend t

Re:Not enough

[Targon]

When ATX first came out, there were a LOT of problems with the cases and motherboards of the time(Baby AT was the most common) for manufacturers. For those who don't remember the old days, we had:

All connectors were on a seperate back-plate, or you could punch out a cover and mount the serial, parallel, game port, SCSI connectors, and everything else except the keyboard connector. So building a system meant you had to connect every one of these connectors via a small ribbon cable to the motherboard. Tod

Re:Is it just me or..

[Leroy_Brown242]

Just think of the smell!

Re:Is it just me or..

[77Punker]

Since they are geeks, they probably have the latest in personal air filtration technology; this would allow them to inhale the air yet remain unharmed by the stench of sweat and the solvent escaping from freshly opened computer parts.

Re:Is it just me or..

[isny]

Since they are geeks, they probably have the latest in personal air filtration technology; this would allow them to inhale the air yet remain unharmed by the stench of sweat and the solvent escaping from freshly opened computer parts.
They do have the latest personal air filtration technology, but they pulled the batteries from them to power up a more important technology that needed the power. (laptop, PDA, cell phone).

Where is it?

[YankeeInExile]

Maybe you, like I, are wondering, Where is it?. One would find Computex 2005 in Taipei.

More information...

[BobtheGreatZeta]

For those of you interested in the cheap RamDisk tech that Gigabyte is showing off at the show, the SPCR forums have a slight bit more information... http://forums.silentpcreview.com/viewtopic.php?t=2 2448 [silentpcreview.com] Here's a quick summary: SATA Type: Unknown as of yet... Email sent to AnandTech about this topic... Maximum transfer rate of DDR200: 1600MB/sec Maximum transfer rate of PCI Bus (not used in this incarnation of the technology): 132MB/sec In addition, DDR266 and others should slow down to function at DDR200 speed without a problem.

Re:More information...

[BobtheGreatZeta]

From the information at hand, the card appears to only draw power from the PCI bus and could therefore (theoretically) be used with any motherboard...

Re:More information...

[Vo0k]

Neat idea. RAID maybe?
Too bad, that now, once it's in public domain, it can't be patented so they probably won't pick it :)

Re:More information...

[KillShill]

apparently, it's only 2x faster than a real hard drive...

i guess when i wasn't looking hd manufacturers boosted hd performance 100x...

or that because this product transfers thru the aging pci, it is limited to 133 (max theoritcal rate) MB/s.

these should have been out ages ago and been only held back by the price of ram at the time. i see no reason why this isn't just a godawful cheap hack.

it deserves no applaud. it only took about 10 years for someone to come out with a cheap gadget that could do this f

Re:More information...

[Vo0k]

No, it's probably mostly because of overblown HDD transfer benchmarks.
They love to show MAXIMUM hdd transfers, which are close to ramdisk speeds, because of cache reads, high speed of disk spinning and data density in a sector. Therefore a single, continuous big file read is really fast. But the seek times still suck - physically moving the head takes a lot of time and only the data from cache can be transferred in the meantime. Read speed of multiple small files scattered over the drive will take much long

Bad link to "ATI's new multi-GPU chipset"

[Anonymous Pundit]

Um, the "ATI's new multi-GPU chipset" link actually points to a slashdot article entitled "AMD Athlon 64 Dual Core Chips Released".

Re:Bad link to "ATI's new multi-GPU chipset"

[Toad McFrog Esq.]

I can't wait to see how ATI's new dual card solution (which they call MVP) stacks up against Nvidia's. Because the link to more info is jacked up, here [gizmodo.com] is a link to a Gizmodo entry with links to more information.

Re:The sad thing is..

[bemenaker]

Well, things have gotten fast enough now, that changes are smaller and smaller. Not a big reason too. I follow everything, but I haven't seen a product that made me say, "WOW OMFG I HAVE TO BUY THAT YESTERDAY!!!" in a few years. Well, not in the computer field at least. :D Need some new thinking in the design teams.

Re:lack of interest

[biancmb]

it must becasue it says "sneak PEAK" instead of "sneak peek"...