Which Mobo?

[Skrying]

If it's the best overclocking chipset, that is a huge deal to me. Though I agree we only have a few sources claiming the X38 is a great overclocker, but if true that's a big selling point.

As for PCI-E 2.0 support, it's like the old AGP 8x vs PCI-E debate. Outside of multiple GPU's, does PCI-E have a real world benefit over AGP 8x? Regardless of it being better or not in real world situations, PCI-E 2.0 will become the new standard and thus should be at least treated as a worthwhile bonus.

The best means little if you're only better by 2%, which in my opinion it will only be marginally better. The current chipset are already great overclockers, that's plenty for me and unless you're a serious OC'er then I'd imagine current chipsets are plenty and the increased price is again another factor that makes it not worth it to me. Current chipsets have been shown over and over to reach very nice overclocks.

As for PCIe 2.0 support, again unless it holds a advantage then its pointless honestly. Being the new standard or not means nothing if don't offer a real advantage. If all it brings me is not having to plug in one more cable then its not worth the wait nor increased price.

 

[willardjuice]

Being the new standard or not means nothing if don't offer a real advantage.

So why aren't we still using AGP?

 

[Skrying]

So why aren't we still using AGP?

Who knows? It doesn't matter to this conversation.

 

[willardjuice]

Who knows? It doesn't matter to this conversation.

Yes it is because PCI-E 2.0 will do to PCI-E what PCI-E did to AGP, even though there wasn't a performance increase.

 

[Blazkowicz]

well PCIe 2.0 will do to PCIe 1.0 what AGP 8x did to AGP 4x. or what SATA 2 did to SATA. many people will think it's something important but nobody should actually have to care about the difference.

 

[Skrying]

Yes it is because PCI-E 2.0 will do to PCI-E what PCI-E did to AGP, even though there wasn't a performance increase.

No its not. You're acting like just because its new I should care, yet it brings nothing to the table for me at all. So why should I care? Your only reason seems to be "because its the future", which isn't good enough for the wait and price increase.

 

[SugarCoat]

Yes it is because PCI-E 2.0 will do to PCI-E what PCI-E did to AGP, even though there wasn't a performance increase.

Yea...to echo what blaz just said, its like AGP 8x to 4x, certainly isnt some huge leap in capability that cant be done under current conditions and despite the added power through the slot, i dont know about you but i dont WANT graphics cards to suck down any more power then they already do dont even want to give AMD or nVidia the ability to be lazy. They should be getting better as the fab process gets smaller, not worse. It also doesnt matter whether the card can get its power through the slot or from the rail, its going to be out of the 12V (the power through the slot isnt free in other words) of the PSU one way or the other so that doesnt matter either.

 

[ShaidarHaran]

You guys are missing the point about PCIe 2.0. The point is not that it will bring about some monumental performance increase, but that it will supplant PCIe (for graphics anyway). Eventually all high-end video cards will only be available on PCIe 2.0.

 

[willardjuice]

You guys are missing the point about PCIe 2.0. The point is not that it will bring about some monumental performance increase, but that it will supplant PCIe (for graphics anyway). Eventually all high-end video cards will only be available on PCIe 2.0.

Exactly.

 

[SugarCoat]

You guys are missing the point about PCIe 2.0. The point is not that it will bring about some monumental performance increase, but that it will supplant PCIe (for graphics anyway). Eventually all high-end video cards will only be available on PCIe 2.0.

not true, its backward/forward compatable. Theres nothing that i've seen that says PCIE2.0 graphics cards will not work in 1.0x slots, nor should there ever be. Even the pin I/O connection is the exact same form factor. The entire system is designed specifically for the advantage of doing that to stop the exact problem of alienating customers which is exactly what you and a apparently a few others fear.

 

[willardjuice]

not true, its backward/forward compatable. Theres nothing that i've seen that says PCIE2.0 graphics cards will not work in 1.0x slots, nor should there ever be. Even the pin I/O connection is the exact same form factor. The entire system is designed specifically for the advantage of doing that to stop the exact problem of alienating customers which is exactly what you and a apparently a few others fear.

The PCIe Base 2.0 specification supports both the 2.5GT/s and 5GT/s signaling technologies. A device designed to the PCIe Base 2.0 specification may support 2.5GT/s, 5GT/s or both.

Only if the card supports both 2.5GT/s and 5GT/S signalling. I'll admit I'm guessing most video cards will, but eventually...

 

[Skrying]

You guys are missing the point about PCIe 2.0. The point is not that it will bring about some monumental performance increase, but that it will supplant PCIe (for graphics anyway). Eventually all high-end video cards will only be available on PCIe 2.0.

Not for awhile, not for a very long while. I'll put out two years at least, possibly 2.5.

 

[Blazkowicz]

let's imagine a 10 gig ethernet controller on a motherboard, that could be a PCIe 2.0 device supporting only 5GT/s.

I feel confident. sometimes compatibility really works, to the point you can put a 360KB 5"25 floppy drive along with your core2quad or plug a 750GB disk in a 486 (assuming there's linux taking care of BIOS limitations, or you use a PCI controller card)

 

[D3v0ur3r]

That's great about PCI-e 2.0 but honestly, its going to be pretty useless for what another year or 2 before there is a significant difference that would warrant the need for it? By that time the 2 mobo's I have here in this thread to choose from will long be out of date as will the CPU, memory, my PSU & the current line of V-cards if I want to stay in the ballpark of up to date hardware.

I do not want to pay for things that I can't use just to stretch my e-penis.

 

[ShaidarHaran]

not true, its backward/forward compatable.

Link? I can't find anything to support this statement.

 

[ShaidarHaran]

As for a use for PCIe 2.0, one need look no further than GPU physics which, if it ever materializes as planned would allow for more than just "eye candy physics". Multi-GPU configs beyond 2 GPUs can also benefit from PCIe 2.0.

 

[SugarCoat]

Link? I can't find anything to support this statement.

Really? You couldnt look for yourself but just assumed they'd be incompatible? Comes across as a little lazy on your part but okay...

http://www.pcisig.com/news_room/faqs/pcie2.0_faq/

Q5: Then PCIe 2.0 must be backward compatible with PCIe 1.1 and 1.0?
A5: Yes. The PCIe Base 2.0 specification supports both the 2.5GT/s and 5GT/s signaling technologies. A device designed to the PCIe Base 2.0 specification may support 2.5GT/s, 5GT/s or both. However, a device designed to operate specifically at 5GT/s must also support 2.5GT/s signaling. The PCIe Base specification covers chip-to-chip topologies on the system board. For I/O extensibility across PCIe connectors, the Card Electromechanical (CEM) and ExpressModule™ specifications will also need to be updated, but this work will not impact mechanical compatibility of the slots, cards or modules. Currently, the PCI-SIG is defining the PCIe CEM 2.0 specification which has been released to members for review at v0.5. There are currently no plans to adapt the PCIe Mini CEM specification for the faster bit rate as the market need has not yet materialized.

As for a use for PCIe 2.0, one need look no further than GPU physics which, if it ever materializes as planned would allow for more than just "eye candy physics". Multi-GPU configs beyond 2 GPUs can also benefit from PCIe 2.0.

When your best defense is the 2% of gamers who can afford to spend thousands on just the graphic hardware alone, i think you've only proven what a few of us have been saying. Theres hardly a need to bring it up as a great reason to wait another 6 months in every hardware suggestion thread.

 

[ShaidarHaran]

Really? You couldnt look for yourself but just assumed they'd be incompatible? Comes across as a little lazy on your part but okay...

http://www.pcisig.com/news_room/faqs/pcie2.0_faq/

PCI SIG was the first place I looked, actually. I didn't see an FAQ for PCIe 2.0, and didn't think they'd bother to put technical details in the press section.

When your best defense is the 2% of gamers who can afford to spend thousands on just the graphic hardware alone, i think you've only proven what a few of us have been saying. Theres hardly a need to bring it up as a great reason to wait another 6 months in every hardware suggestion thread.

GPU Physics is intended to be a mainstream product. 2% isn't mainstream.

 

[D3v0ur3r]

GPU Physics is intended to be a mainstream product. 2% isn't mainstream.

But how soon would mainstream be & mainstream would suggest to me not extremely costly. I would say that's at least 2 years away at best on a GPU. Its great to have the slot available to develop the hardware for it but its not going to be mainstream any time soon.

Wouldn't it seem more practical to use multi cored CPU's for physics? Like Quad cores especially which would be mainstream sooner then a GPU solution wouldn't it?

 

[bystander]

Going to do an upgrade pretty soon (finally) & was curious which mobo the pro's here thought would be better.

This nForce chip board. http://www.newegg.com/product/product.asp?item=N82E16813128044

or this Intel chip board. http://www.newegg.com/product/product.asp?item=N82E16813128050

The only big difference I can see is the nForce board uses DDR2 667 while the Intel board uses DDR2 800 or DDR2 1066.

I plan on buying a Intel e6850.

If you have any questions about the Gigabyte P35C DS3R, I'm using one in conjunction with a E6420 OC'd to 2.84Ghz (FSB is at 355Mhz without any voltage adjustments). I'm certain you can use DDR2 667 on this board as well.