The best P4 mobo

[HammerBot]

Choosing a mobo for the P4 seems to be pretty difficult. At least for me it was a nighmare. So many choises. I guess Ive read all reviews about chipsets and mobos worth reading and Im still having a hard time choosing. It appears that just the right mobo isnt available yet. But its coming I know for sure. It even has a name: Soyo P4S648 Ultra Dragon. The main specs are:

*) Intel Socket-478 Pentium 4 CPU with 400/533 MHz FSB
*) Sis648 Northbridge, Sis962 Southbridge
*) 1*AGP Pro/ 6*PCI (8xAGP / PCI V2.2)
*) 3 * DDR DIMMs (up to 3GB of DDR400 memory support)
*) Support Ultra DMA 66/100/133 IDE device
*) On board HPT372 ATA133 IDE RAID chip, (RAID 0,1,0+1)
*) On board C-media 8738 H/W Audio chip, Support 6 channels
*) On board USB2.0 solution
*) Support 10/100 Mbps Ethernet
*) Smart Card Reader connector

The problem just is that its not available yet :-( At Cebit in march the board was on display and predicted availability was end of April. Well I guess I just have to wait a little longer.

If anyone read this far, I appreciate a little input. It appears that very few or no reviewers are conserned about support on the reviewed product. Consider Gigabyte. Great boards tons of awards, but apperently they dont know what support is. Does anybody have a meaning about Soyo/Sis?

Thanks

 

[wintek]

Is the Sis chips not performing 20% slower than the new Intel chips?

 

[HammerBot]

Actually no. The Sis chipset has been getting quite good review and performs superbly. Look here:
<http://www17.tomshardware.com/mainboard/02q1/020311/index.html>
Although the tested chipset is 645DX/961, but I dont expect the 648/962 to be worse.

 

[AMD_Man]

Where are people getting the idea that the SiS 645DX is 20% slower?

<b><i>"A penny saved is a penny earned!"</i></b>

 

[HammerBot]

Well I guess its a bad rumour going around to discredit SISs success. In the past products from ALi, SiS and VIA did have some critiziable aspects, but not today. The SIS645DX really is a cool chipset. In Toms test it outperforms i850 in 10 out of 16 tests!

 

[bob_dn]

645DX outperformed i850 with 800Mhz RDRAM. i850e with 1066 RDRAM outperforms them all. 333 DDR Ram or 400 DDR. Tom's review on i850e Gigabyte GA-81HXP and Asus P4T-533C says it all. Bandwith is the name of the game with the P4 and DDR 333 can't match RDRAM 1066 4.2Gig bandwith.

 

[AMD_Man]

645DX outperformed i850 with 800Mhz RDRAM. i850e with 1066 RDRAM outperforms them all. 333 DDR Ram or 400 DDR. Tom's review on i850e Gigabyte GA-81HXP and Asus P4T-533C says it all. Bandwith is the name of the game with the P4 and DDR 333 can't match RDRAM 1066 4.2Gig bandwith.

True, but it's not 20% slower.

<b><i>"A penny saved is a penny earned!"</i></b>

 

[HammerBot]

Yes, PC1066 RDRAM currently holds the record for RAM bandwidth. But the point about Sis645 vs. 850 is that P4 does NOT require a 1:1 BW match between FSB and RAM. Toms has made this clear several times. Yes, for some tests if the CPU exclusively access the RAM for execution of code and retrieval/storage of data in large DBs, RDRAM PC1066 does outperform DDR. But for typical usage a significant amount of the FSB bandwidth is used for AGP and PCI and then the reduced BW of DDR has NO significant impact. Furhter, RDRAM is NOT an option for me. It is simply too expensive compared to the small (and insignificant IMO) performance gain.

Anyway, enough about performance. What Im really interested in is how Soyo and Sis is viewed from a support point of view. This may not be an issue for many, but for me it is. Actually thats why I dont go for a Gigabyte mobo. Apparently a lot of people are very dissatified about GAs support. I can live without that. What can be said about Soyo/Sis when it comes down to support, drivers etc.?

 

[Crashman]

Chipset support, except for drivers, is mostly covered by the board manufacturer.

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>

 

[Crashman]

Actually, I heard that, after the VIA extorsion issue surrounding the 735 chipset and board manufacturers, they decided to formulate a new game plan, and several employees have been hitting several forums with this lie in order to make it spread!

<font color=blue>At least half of all problems are caused by an insufficient power supply!</font color=blue>