Nvidia crushes MSI's Lucid - RETRACTED

[Utnorris]

No it's not a bad thing to want it to happen. Of course not. But if people go around making claims that it "will" happen while dissing the current solutions they should be prepared to get called on it and defend their opinion. It would also be easier if you could back up your faith with some technical reasoning. The people challenging Hydra are doing so on technical grounds, not because they hate progress or don't want a better solution.

The hardware you run is an indicator of whether you're in Lucid's target market. Multi-GPU is primarily a performance enhancing option. If you're happy with a single 8800gt then I don't see how Hydra's benefits would apply.

1 - Nobody is "dissing" SLI or CF, just that Lucid on paper is supposed to be better.

2 - Nope, not one person challenging whether or not Nvidia is blocking Hydra on a MSI board have brought up that Hydra is technically impossible or what not.

3 - Reread my post. Hydra actually would allow someone with a 8800gt to upgrade to the next generation while using the 8800gt, hence is why EVERYONE WANTS IT!!!!! Hydra would take away the limitations that SLI and CF place on users, not to mention you could use a Nvidia card with an ATI card.

Seriously, this discussion was on whether or not Charlie was correct and if so, then what? Stay on subject.

[BenchZowner]

Hydra is currently having issues combining two current high end gpus ( same model & card, not different gpus ), do you really think it can mix and match different GPUs ?

[tajoh111]

Hydra is currently having issues combining two current high end gpus ( same model & card, not different gpus ), do you really think it can mix and match different GPUs ?

I think this is the more likely reason for hydra coming out.

[Utnorris]

Hydra is currently having issues combining two current high end gpus ( same model & card, not different gpus ), do you really think it can mix and match different GPUs ?

Do you have a source or have you seen it first hand?

[BenchZowner]

I told you guys I know somebody who works for LucidLogix.
He's also a member of this very extreme forum, but he doesn't seem to want to post anything himself ( I wouldn't too, if I was working for them and was bound by NDA and risk losing my job ).

Now if you guys are so eager to hear that it was nVIDIA who stopped MSI then I guess there's nothing that can be done.
Even if an MSI rep. posted something I'm quite sure lots of people wouldn't believe his words...

Anyway... keep it up...

[Olivon]

This thread really suxxx ...

[Bodkin]

this been posted yet? http://techpowerup.com/107704/NVIDIA...cid_Hydra.html

They reliable enough for you tim?

[trinibwoy]

Hydra actually would allow someone with a 8800gt to upgrade to the next generation while using the 8800gt, hence is why EVERYONE WANTS IT!!!!! Hydra would take away the limitations that SLI and CF place on users, not to mention you could use a Nvidia card with an ATI card.

Do you believe in the easter bunny too?

Seriously, this discussion was on whether or not Charlie was correct and if so, then what? Stay on subject.

MSI already called him on his BS. Move along.

this been posted yet? http://techpowerup.com/107704/NVIDIA...cid_Hydra.html

They reliable enough for you tim?

You're kidding right? Their source is Overclock3D which got their info from SemiAccurate. See how Charlie's BS spreads and gets people all excited?

[BenchZowner]

And finally somebody who knows about news circulation

[trinibwoy]

Well if you are still hanging on to your premise that hydra is technically impossible, then it doesn't matter if Charlie's article is right or wrong because the product will never materialize anyway. Is that still the case, that you think hydra is impossible to implement?

It's kinda hard to say something is technically impossible when you don't know what that something is. But let's step back for a minute and see what people are getting so excited about. The current data path is:

Application -> DirectX -> IHV Driver -> IHV Hardware.

Now what I would like is for one of the believers to tell us where Lucid is going to inject itself in that stream in such a way that they could manipulate what commands get sent to the hardware. Note that it IS impossible for them to insert themselves between the driver and the hardware because that communication is completely proprietary.

[SKYMTL]

this been posted yet? http://techpowerup.com/107704/NVIDIA...cid_Hydra.html

They reliable enough for you tim?

No, it shouldn't be reliable enough for anyone. That is just a rehash of what Charlie posted. The same goes for what is posted over at OC3D (the original source).

Unfortunately for people who "ran" with the story, MSI came out with a statement that debunks what everyone was so eagerly lapping up like lovestruck puppies.

[Bodkin]

Damn, that's a shame, I really thought techpowerup was a good source, well its gone from my rss feeds

[Russian]

From what i've heard it had nothing to with nvidia or anything... its just that the Lucid drivers didn't pass MSI's QA... so they delayed the product until they're sure its working correctly. And from what i've seen in the past, the drivers were the biggest issue that most people complained about when they encountered this Hydra chip.

[Russian]

Do you believe in the easter bunny too?

MSI already called him on his BS. Move along.

You're kidding right? Their source is Overclock3D which got their info from SemiAccurate. See how Charlie's BS spreads and gets people all excited?

Yeah, its really annoying that people still read his site considering all of the biased crap he continues to post. And then people take his stuff, because its so sensationalist and post it around, and then he gets more and more hits. Which is ultimately achieving his goal of more traffic and bashing nvidia.

[Solus Corvus]

It's kinda hard to say something is technically impossible when you don't know what that something is. But let's step back for a minute and see what people are getting so excited about. The current data path is:

Application -> DirectX -> IHV Driver -> IHV Hardware.

Now what I would like is for one of the believers to tell us where Lucid is going to inject itself in that stream in such a way that they could manipulate what commands get sent to the hardware. Note that it IS impossible for them to insert themselves between the driver and the hardware because that communication is completely proprietary.

Sure, we need to be skeptical about news reports/rumors of unreleased products. Especially one like this where the product is so uncertain that assigning blame on Nvidia for delaying it is uncalled for without a higher level of evidence.

We don't know how it supposedly works so being skeptical is a reasonable response. I'm just saying that the other side of skepticism is that we don't know how it supposedly works so how could we possibly have enough information to come to a conclusion that it likely won't work? How hard it would be to implement a product like this depends on a lot of factors, not the least being where in the do they intend to insert themselves in the instruction stream. But we don't even know that much. So I'm not going to say such a product is likely or unlikely without a little more information.

[trinibwoy]

We do have information. We know how DirectX and GPUs work. The guys who write engines for a living are asking the same questions and showing the same skepticism (see the Lucid thread at B3D).

[Solus Corvus]

We do have information. We know how DirectX and GPUs work. The guys who write engines for a living are asking the same questions and showing the same skepticism (see the Lucid thread at B3D).

Ok, but questions about how it could work doesn't equal impossibility.

[Xoulz]

well you seem to be a big fan of hydra you have any test to show us? or just all the hype that they shoveld us the last few yrs?
i cant think that the support for this would be any good if you did have issues with it.
or just make us pay more for a mb that has this feature only to turn it off cus its broken.

Yeah, I'm highly interested in Lucidlogix's Hydra chip...

Who isn't?



Though, I'm not as cynical as most, because of the technology itself. Honestly, to think SLI/Xfire is the absolutely best way, is ignorant & laughable.

Ironically, I think THESE people thought so too.. and decided to do something about it. I would assume they know more about motherboards, PCIe and DirectX calls than any of the thread-crappers here.

It's works, unless you call the press release and demos fake?

So.. the only thing left to discuss is it's performance... not HYDRA's validity. Thats a forgone conclusion at this point. The doom & gloom is overdone at this point, the chips are real and on boards... the drivers are being tweaked.



What is it exactly that you are trying to dismiss..??

[Chumbucket843]

dont be so naive xoulz, its just a chip and so what if it provides perfect scaling. thats not nearly as exciting as the rest of hardware out there especially with the speed increases with next generations of gpu's. it makes the hydra engine pretty insignificant. if you read trinibwoy's post above you can see people who write engines are skeptical about it. thats not a good sign coming from a forum like b3d. the people there probably know more than lucid anyways. i dont understand how you can validate a product by a demo either. what if it only works with ue3?

[grimREEFER]

im gonna find it hilarious if lucid finally does come out with the chip, but it doesnt perform as well as SLI and crossfire, or introduces noticeable input lag or other problems.
then we'll have endured all this hype for nothing lol

[Solus Corvus]

dont be so naive xoulz, its just a chip and so what if it provides perfect scaling. thats not nearly as exciting as the rest of hardware out there especially with the speed increases with next generations of gpu's. it makes the hydra engine pretty insignificant. if you read trinibwoy's post above you can see people who write engines are skeptical about it. thats not a good sign coming from a forum like b3d. the people there probably know more than lucid anyways. i dont understand how you can validate a product by a demo either. what if it only works with ue3?

It's good to be skeptical. But questions about how it could possibly function don't necessarily mean that it cant. But trinibwoy, you, and some others seem to want to use this as a reason to doubt that it could function at all, ie. that it is a nonexistent product.

But inclusion in a future motherboard seems to indicate that the product is in fact real, though perhaps not functioning properly. So did Lucid trick MSI into designing a board around vaporware or do you think that they might know more about it then we do?

[003]

this been posted yet? http://techpowerup.com/107704/NVIDIA...cid_Hydra.html

They reliable enough for you tim?

No, in fact, it's what we would refer to as an Epic Fail.

TPU's source is Overclock3D, and Overclock3D states their source as being SemiAccurate!!

[To(V)bo Co(V)bo]

What we have hear is a FAiL to communicate.

I could write a story about how having a hydra chip swinging from my nutsack helps taking double headed dookies, and somebody would believe it.

[Utnorris]

[QUOTE=trinibwoy;4098376]Do you believe in the easter bunny too?QUOTE]

Actually....................................no, but my point was if it does what it is suppose to do, then it would mean the end to SLI and CF. They have done limited demo's, so it isn't vaporware. Now the real question is whether Nvidia is trying to kill it, hence the delay or was it crappy drivers that Benchzowner has stated that he got from someone he knows from an inside source as being the real reason for the delay. We don't know either way and since I do not know Charlie or Benchzowner, then I will take both with a grain of salt.


We will just have to wait till January to find out.

[Andrew LB]

this been posted yet? http://techpowerup.com/107704/NVIDIA...cid_Hydra.html

They reliable enough for you tim?

Ummm... you posted an article from "TechPowerUp" which sources it's information from "Overclock3D.net", which then in turn sourced it's information from "Semi-Accurate" aka Charlie Demerjian.... the original source of the story.


Great job at reading your articles.