Nvidia crapping their pants over Lucid?

[Acanthus]

It seems to me that this is good news for the consumer and bad news for nvidia who wanted to sell a physically worthless chip to mobo makers to enable SLI.

Hot off the press:

http://www.anandtech.com/cpuch...el/showdoc.aspx?i=3395

 

[AmberClad]

There's sort of an ongoing discussion about this over here. (I have no idea why the person posted it in CPUs, instead of here or even Motherboards.)

 

[Acanthus]

That is a strange place for the discussion... i didnt even look there.

I just wonder if the lucid demonstration not long ago is what made nvidia finally give up the sinking ship of forcing the hardware unlock of SLI...

Especially since this new development seems to have hastily presented itself... They may have realized that they wont be able to corner the SLI market with the Hydra 100 on the way.

 

[BLaber]

I have to say Nvidia ===> Compromised. LOL

 

[Sylvanas]

Now theres even less reason to buy an Nvidia motherboard.... Bring on Lucid!

 

[chizow]

I don't think Lucid had anything to do with it, their solution also relies on an add-in chip for mult-GPU functionality. From the various blurbs, it was probably due to OEMs not planning to use NF200 on X58 boards at all, which basically forced NV's hand into removing the artificial limitation on X58. I saw this coming down months ago, even with them keeping the limitation intact on older boards and chipsets.

 

[taltamir]

with AMD widening the gap with CF scaling compared to SLI scaling, lucid actually helps nvidia alot by evening the playing field.

But the link you posted doesn't have anything to do with the hydra. It is about nvidia allowing INTEL (not lucid) to integrate SLI support into the X58.
And requires mobo makers to pay nvidia for the privilage of doing so.

Once cracked drivers hit. It will mean you could run nvidia SLI on intel boards (more nvidia cards sold), and nvidia could still extort a premium from mobo makers for the SLI certification.

 

[Wreckage]

This has nothing to do with Lucid and everything to do with Intel with holding QuickPath.

 

[nonameo]

Originally posted by: Wreckage
This has nothing to do with Lucid and everything to do with Intel with holding QuickPath.

I wonder if there's a way to make intel CPUs work with hypertransport?

 

[taltamir]

whatfor? without the license its not legal to sell.

This is why nvidia doesn't make x86 CPUs.

 

[Borealis7]

Originally posted by: taltamir
But the link you posted doesn't have anything to do with the hydra. It is about nvidia allowing INTEL (not lucid) to integrate SLI support into the X58.
And requires mobo makers to pay nvidia for the privilage of doing so.

correct me if im wrong here, but i remember Lucid being an Israeli company funded by Intel (or competely owned by Intel). i'll go check my Israel Intel connections...

 

[taltamir]

As you said, i'ts an israeli company, funded by intel, making a chip called hydra that will allow 100% caling multi GPU with both AMD and nVidia on the same hydra chip.

And it has nothing to do with this announcement. This announcement is that the X58, an intel chipset, will be able to run xfire and SLI natively.
When hydra would be ready, it could be used with X58, X48, Xwhatever, AMD chipsets, and nVidia chipsets. And it will not be either SLI or CF, it will be hydra multi GPU.

 

[SunnyD]

Originally posted by: Wreckage
This has nothing to do with Lucid and everything to do with Intel with holding QuickPath.

And has nothing to do with Lucid making the NVIDIA 200 chip pointless right? My bet is that the Hydra chip would be made available to Intel and OEM's far cheaper than NVIDIA's chip, and NVIDIA knows this, so it's not a cash cow for them all of a sudden (assuming integrators wanted to put it on their boards). Coupled with NVIDIA's lack of a QPI/CSI license, they stand to have EVERYTHING to lose especially since they no longer have the highest performing graphics parts to boot.

So you're absolutely right, I'm sure is has nothing to do with Lucid at all. :roll:

 

[rgallant]

Originally posted by: Wreckage
This has nothing to do with Lucid and everything to do with Intel with holding QuickPath.

I wonder too if Intel , when they summit theirs boards and pay the fee , if they have to hand over a lot of restricted tech doc's not offered to those that do not need to know , and now nv needs to know ?

 

[aka1nas]

Originally posted by: SunnyD

Originally posted by: Wreckage
This has nothing to do with Lucid and everything to do with Intel with holding QuickPath.

And has nothing to do with Lucid making the NVIDIA 200 chip pointless right? My bet is that the Hydra chip would be made available to Intel and OEM's far cheaper than NVIDIA's chip, and NVIDIA knows this, so it's not a cash cow for them all of a sudden (assuming integrators wanted to put it on their boards). Coupled with NVIDIA's lack of a QPI/CSI license, they stand to have EVERYTHING to lose especially since they no longer have the highest performing graphics parts to boot.

So you're absolutely right, I'm sure is has nothing to do with Lucid at all. :roll:

If it had anything to do with the Hydra chip, they wouldn't be only offering this for X58 boards, which 99% of the market isn't going to purchase.

 

[AmberClad]

Timing seems a little tight, what with Fud speculating that the i7 launch date will be around the first couple weeks of November. Hopefully this won't result in half-baked drivers.

 

[NicePants42]

Originally posted by: chizow
I don't think Lucid had anything to do with it, their solution also relies on an add-in chip for mult-GPU functionality. From the various blurbs, it was probably due to OEMs not planning to use NF200 on X58 boards at all, which basically forced NV's hand into removing the artificial limitation on X58.

Wait, so you're saying that this decision by nVidia is based on real information concerning the size of their potential market in the next chipset refresh, and not speculation based on a closed tech demo and two vague AT articles?

*facepalm* Brilliant!

 

[SunnyD]

Originally posted by: aka1nas

Originally posted by: SunnyD

Originally posted by: Wreckage
This has nothing to do with Lucid and everything to do with Intel with holding QuickPath.

And has nothing to do with Lucid making the NVIDIA 200 chip pointless right? My bet is that the Hydra chip would be made available to Intel and OEM's far cheaper than NVIDIA's chip, and NVIDIA knows this, so it's not a cash cow for them all of a sudden (assuming integrators wanted to put it on their boards). Coupled with NVIDIA's lack of a QPI/CSI license, they stand to have EVERYTHING to lose especially since they no longer have the highest performing graphics parts to boot.

So you're absolutely right, I'm sure is has nothing to do with Lucid at all. :roll:

If it had anything to do with the Hydra chip, they wouldn't be only offering this for X58 boards, which 99% of the market isn't going to purchase.

Gotta start somewhere. Plus there's the who "SLI Validation" crap too, which I doubt they can retroactively go back and test every single board that has been release to date, let alone possibly cannibalize any existing sales potential for boards based on NVIDIA chipsets that still are being sold.

Face it, it's all about dollars and cents to them. People weren't going to drop their chip on an X58 board for the added cost. NVIDIA won't have an alternative chipset for Nehealm because they don't have a bus license. The only thing they have left to sell in terms of Nehealm are graphics cards... of which their potential market is halved because Nehealm boards wouldn't be running SLI. Add that to the fact that Intel boards natively run CrossFire, and the fact that AMD has the faster net graphics solution, and if we ignore Hydra all together - NVIDIA is screwed. Their revenue stream went down the shitter with Nehealm.

Come to think of it, mayhaps Lucid would be beneficial to NVIDIA in this case...

 

[NicePants42]

Originally posted by: SunnyD
Face it, it's all about dollars and cents to them. People weren't going to drop their chip on an X58 board for the added cost. NVIDIA won't have an alternative chipset for Nehealm because they don't have a bus license. The only thing they have left to sell in terms of Nehealm are graphics cards... of which their potential market is halved because Nehealm boards wouldn't be running SLI. Add that to the fact that Intel boards natively run CrossFire, and the fact that AMD has the faster net graphics solution, and if we ignore Hydra all together - NVIDIA is screwed. Their revenue stream went down the shitter with Nehealm.

It's good that you understand what aka1nas and Wreckage were trying to tell you.

 

[chizow]

Originally posted by: NicePants42

Originally posted by: chizow
I don't think Lucid had anything to do with it, their solution also relies on an add-in chip for mult-GPU functionality. From the various blurbs, it was probably due to OEMs not planning to use NF200 on X58 boards at all, which basically forced NV's hand into removing the artificial limitation on X58.

Wait, so you're saying that this decision by nVidia is based on real information concerning the size of their potential market in the next chipset refresh, and not speculation based on a closed tech demo and two vague AT articles?

*facepalm* Brilliant!

Yes it was so obvious that the OP and numerous others came to the conclusion SLI support on X58 was induced by Lucid. :roll:

Also, the bit about OEMs not planning to use NF200 at all is a new development as previous reports indicated board-makers would be implementing SLI via NF200 on X58 at some point.

 

[taltamir]

um, the hydra chip DOES NOT EXIST YET! it is in development, and it is not ready.
It does not compete with with the NF200 chip, and it will not compete with it this generation, and it certainly has nothing to do with THIS ARTICLE you posted.

This article is about nvidia opening SLI to the X58 natively due to mobo makers not buying thier NF200 chips. Got nothing to do with lucid developing a product that is years away if ever. (although, since intel is funding them, i doubt it is vaporware)

 

[Acanthus]

Originally posted by: taltamir
with AMD widening the gap with CF scaling compared to SLI scaling, lucid actually helps nvidia alot by evening the playing field.

But the link you posted doesn't have anything to do with the hydra. It is about nvidia allowing INTEL (not lucid) to integrate SLI support into the X58.
And requires mobo makers to pay nvidia for the privilage of doing so.

Once cracked drivers hit. It will mean you could run nvidia SLI on intel boards (more nvidia cards sold), and nvidia could still extort a premium from mobo makers for the SLI certification.

Intel is the primary investor and primary interest in lucid (and hence hydra).

 

[sourthings]

This is half-decent news. When I still used an 8800GTX, I never went SLI, because of nvidia's freaking awful motherboards. This would of made it possible.

Let's just hope introducing SLI to the intel chipset, doesn't make it horrid like 7X0 boards.

 

[Acanthus]

Originally posted by: taltamir
um, the hydra chip DOES NOT EXIST YET! it is in development, and it is not ready.
It does not compete with with the NF200 chip, and it will not compete with it this generation, and it certainly has nothing to do with THIS ARTICLE you posted.

This article is about nvidia opening SLI to the X58 natively due to mobo makers not buying thier NF200 chips. Got nothing to do with lucid developing a product that is years away if ever. (although, since intel is funding them, i doubt it is vaporware)

Its about a year off accoding people inside lucid.

But if you have to start supporting SLI via software to all platforms... because you find out that there is a better scaling design from an OEM that you have no fiscal stake in... and no one is buying into your "needing an extra chip to enable SLI" crap... and you dont have a QPI license to cram exepensive proprietary hardware anymore...

It seems to all add up to Nvidia jumping ship on hardware SLI and giving us the software SLI we needed years ago.

Now if the 200 chip ACTUALLY DID SOMETHING like the Hydra 100 (load balancing), it would be a totally different story, and nvidia would be singing a different tune about SLI imho.

 

[taltamir]

well... the PCIe v2 speeds are as following:

X58:
1. 16x
2. 16x, 16x
3. 16x, 8x, 8x,
4. 8x, 8x, 8x, 8x

X58 + NF200 (premium SLI):
1. 16x
2. 16x, 16x
3*. 16x, 16x, 16x
4**. 16x, 16x, 16x, 16x

* uses one NF200 chip.
** uses two seperate NF200 chips.

I wonder if the extra PCIe bandwidth will actually mean anything, or if it will end up just being a gimmic. Since the nVidia chart shows each NF200 chip splitting a 16 lane PCIe V2 into two 16 lanes... which seems, um, suspicious. (i guess, if it works like an ethernet hub, and the cards tend to address each other more often then addressing the CPU, then it could be done)