Lucid Hydra on Smackover 2

[Xoulz]

Im not saying its a hoax. And you now say that Lucid will test and make profiles for every game? You cant load balance something if you dont know what it is. And you need software to tag and determine it.

I guess you have your mind set on that this is a PC revolution. Where I just think its a console revolution. The Hydra 100 chip aint even PCIe 2.0.

Understood, but to be 99% efficient with certain aspect of the rendering process, the chip will probably need updates. But even without them, the Hydra Chip's design is inherently better than the SLI solution.

A "profile" may increase a certain games performance say another 7%, or remove an anomaly, but understand, we are talking about a processor solely dedicated to nothing but guiding information on a PCIe bus. A new motherboard design is required to make use of this technology!



Secondly, I understand it's limits, but that doesn't mean other products or revisions are not in testing or development. Lucid is probably releasing this chip for a revenue stream before setting their sights on the PCIe 3.0 standard.

As with anything in the hardware world, update or die..!




.

[MrHydes]

any news on this Lucid ting? lol

this seems to be just crap after all

[Aberration]

How does it seem to be crap after all?

[STEvil]

Does the chip even need to be PCI-E 2.0? All it does it sit on the bus and send certain stuff where it has to go.

As before there's nothing mystical about what is happening here.

[Macadamia]

Any news? Err... AMD's chipset group seems to be EXTREMELY interested in Hydra that they made Device IDs for it.


http://www.xtremesystems.org/forums/...d.php?t=212665

[Sr7]

Any news? Err... AMD's chipset group seems to be EXTREMELY interested in Hydra that they made Device IDs for it.


http://www.xtremesystems.org/forums/...d.php?t=212665

That site sure doesn't look very official. I'd be interested to see the official registry to see if it has something similar.

[Macadamia]

That site sure doesn't look very official. I'd be interested to see the official registry to see if it has something similar.

As if they'd show something similar BEFORE releasing the product


DeviceIDs have always been leaked through driver INFs or Linux driver info. RV770 and 740 were mentioned in Linux drivers quite a while back before they were even released/people knew what they were.

[Decami]

Also rememebred they demoed one thing in a highly controlled situtation with 2 identical cards. Plus it was in a setup that was very far from a PC design.

Funny you dont remember that thing, it was Crysis, and from what I remember reading in the review, THEY WERE PLAYING CRYSIS.

hardly controlled, hardly a time demo and hardly a game that requires no input.

to add to that,

This "very far from a PC" may have looked strange to you, but I dont even know what demo you were looking at, the only strange thing was the MB and thats because they had to build their own for the chip to be placed in, everything else was normal, the memory, the cpu, all normal, and why wouldnt it be, my god man, I suppose we never went to the moon either.

I do remember the reviewer saying soemthing about a little bit of input lag here and there, but Lucid assured him that would not be a problem in the final product.

Now listen its not like i dont have questions either, or that im confused about how certain things are done. Im not even saying the chip will be the best thing since sliced bread, or the savior of anything.

All I am saying is that this will most likely be a better solution to SLI/CF, if it wasnt, they wouldnt be getting the lazer eyes from the companies they have been lately, and Intel wouldnt be strongly backing this either, if its gonna be as much of a failure as you think it is, Intel hasnt pulled out yet, and with a backer like them, its not their in the dark, so something has got to be good about it.

[Shintai]

Any news? Err... AMD's chipset group seems to be EXTREMELY interested in Hydra that they made Device IDs for it.


http://www.xtremesystems.org/forums/...d.php?t=212665

Aha...

Name: RD890 Northbridge only dual slot (2 x 16) PCI-e_GFX Hydra part

shane.huang

2007-08-07 11:29:21

Not really an official site either...

[Sr7]

As if they'd show something similar BEFORE releasing the product


DeviceIDs have always been leaked through driver INFs or Linux driver info. RV770 and 740 were mentioned in Linux drivers quite a while back before they were even released/people knew what they were.

And similarly this random registry of device id's would have access to this info ahead of time because...?

The site looks hokey as .. just wondering why something as specific as device id's and descriptions would show up there first instead of being leaked on forums or something.

[MrHydes]

Lucid can scale with 2 GPUs close to linear
Chips in 2H 2009

We had an interesting meeting with the president and CIO of Lucid, a
company that gained a lot of attention at IDF 2008. Lucid promises a small
chip that can make your two GPU scale almost linear. After a long discussion
we learned that Linear is almost linear, and that linear is best case scenario
and that real world performance should be close to linear.

This would mean that two GPUs, let's say in SLI or Crossfire, powered by
Lucid chip and driver, could score close to 100 percent faster than a single
GPU. Let's say that if a single GPU would score 100 FPS in one game, two
would score close to 200FPS.

The chips that will find their place in both motherboards and graphics cards
will appear on the market in 2H 2009. The prototypes are already finished,
but the company didn't want to talk about any specific partners.

This leaves a lot of unclear questions to be answered, as we suspect that
both ATI and Nvidia might have something against a chip that makes SLI and
Crossfire look like child's play, and that this could lead to a driver that will
block Lucid’s ability to shine.

Source: fudzilla

[STaRGaZeR]

The end of the need of software profiles for CF/SLI? Bring it on, now!

[vietthanhpro]

Lucid Chip in 2H 2009

[HKPolice]

Too good to be true. 1 big issue: the chip only has PCIe 1.1 interfaces. It'll be splitting a single 16x PCIe 1.1 into 2x 16x PCIe 1.1. Clearly bandwidth will be a bottleneck. Latency is another big issue.

[Face]

Lucid's Hydra 100 - Closer Than Ever

ELSA Japan, a popular AIB (Add-In-Board) company has announced the first product to incorporate Lucid Logix's Hydra 100 series ASIC. The product is aimed at HPC (High performance computing), the same space targetted by Nvidia's Tesla™ and AMD's Stream™ computing products.



Lucid claims that this solution could be aimed at two scenarios; at the chipset level (ex: nForce SLI chipset) and at the board level (ex: 4870X2). In either of those scenarios, the ASIC may come as a big boon for gamers. Typically the (current) vendor offered scaling for multi-gpu setups tops out at 250% in a quad-gpu setup. It could theoritically play to AMD's advantage since they have taken the path of developing mid-size GPUs and try to scale that performance up by means of a multi-gpu solution (X2). Reading more into the Hydra's specs, it currently is compatible with DirectX®9 only with DirectX®10 support coming very soon. Power consumption of the ASIC is rated at 6W, slightly higher compared to the PEX8647 bridge chip used in the Radeon HD 4870X2, which is rated at 3.8W.

Pairing four AMD FireStream™ 9270 with ELSA-Lucid's new product could rival Nvidia's Tesla™ S1070 server, which also makes use of four GT200 GPUs. That is assuming the promised linear scaling by the Hydra 100. This hypothetical pairing would be just shy of a 1 TFlop of double-precision (peak) performance.

Source:
http://gpucafe.com/2009/02/lucids-hy...ser-than-ever/

[clayton]

If this works, then all I can say is "1 miracle at a time". Quadfire/sli would be smoking.

[Shintai]

Gflops scaling? On HPC apps that doesnt require any form of SLi/CF/Hydra? Seems like PR shift...

[Xoulz]

Gflops scaling? On HPC apps that doesnt require any form of SLi/CF/Hydra? Seems like PR shift...

Again.. until they get an actual product out, THEN use those revenues to further develop the DirectX 10.1 Hydra chip and eliminate the needs for SLI/Crossfire completly...!

Forever!

[grimREEFER]

i feel like if i played around with the metrics, i could say the same about sli/crossfire scaling.
Nearly 100%!
it gets pretty close on some of the really gpu limited stuff with good profiles.

[MrHydes]

ELSA Japan and LucidLogix to Introduce High Performance Computing Products to Japanese Market

The companies have teamed up to transform high performance computing in
the Japanese marketplace. For the first time, a product based on Lucid’s
HYDRA technology will be used in a new line of ELSA Japan high performance
systems for the HPC, broadcast and medical markets.


End of Mar 2009, The new solution will allow ELSA to provide a cost affective
solution based on multi GPUs from any vendor. First products will feature dual
and quad GPU configurations

source: elsa-jp

[clonez]

what a monster, wall of cards
funny kind of connectors..
and an ethernet-port?!

[Tonucci]

Lucid can scale with 2 GPUs close to linear
Chips in 2H 2009

IM sorry but anything from FUD should not even be posted here imo.... We need an TMZ'ish Rumor sub forum for the likes of him.

Much less make a statement in big letters based on something from him

[gumballguy]

ELSA Japan and LucidLogix to Introduce High Performance Computing Products to Japanese Market

News via elsa's website is enough to excite me

[nr4]

I read most of the posts in this thread and i like Xoulz's, del_fuego's, Decami's and others opinions.

I'm not a guru or something like that, but i've experienced systems with feedback concept (very, very many electronic products have something similar implemented) and in my opinion Hydra is very real and can do what is said. I'm speaking about the scaling between lets say 2 different Nvidia video cards.

Let me explain the feedback concept:
The input is controlled by the ouput.
So i have a detector that analyzes the output and sets the input so the output is the desired one.
This is done in real time.
So the Hydra engine can analyze the outputs of the 2 video cards an rebalance the inputs for the 2 video cards.

You dont have to know what the cards can do, or their specifications, because you are testing them in every moment in real time, and you are adjusting the input data of the work in every moment so the system can be balanced in load.

This is my guess how could Hydra work, but i am not saying this is the real way.
For example every power supply unit has implemented this feedback system (just for record).

[MrHydes]

LucidLogix’s Hydra Renders Objects, Not Frames

The guys over at Fudzilla have been talking to the people over at Lucidlogix.

LucidLogix, if you did not already know, is the maker of the very highly
anticipated Hydra Chip. They asked questions about ELSA Japan’s recent
usage of the Hydra Chip and were told that ELSA is not using it for graphic
performance but as more of a PCI-Control Chip for GPGPU functions.

In Graphics Performance Mode the Hydra changes the traditional way that
multiple GPUs deal with rendering. In traditional SLI or CrossfireX setups the
GPUs share the rendering burden by rendering alternate items; this is usually
broken up in to frames, (AFR) Tiles, and in older SLI cases lines. With Hydra
the Hydra Chip controls the rendering, it is capable of sending commands to
the GPUs telling them which objects to render in a give frame and then it
assembles the distributed objects and send that data to the monitor.


This is a significantly more efficient method. The problem is that no one has
seen a truly working demo of this. Perhaps by CeBit we will know more.

source: tweaktown

More details about Lucidlogix Hydra revealed

Another interesting aspect is that its possible to cascade multiple Hydra
chips, although at the moment this is limited to two chips, but this would
allow for more cards to be used. This could at least in theory allow for four
GPU graphics cards, but Nir suggested that in the future up to 8 GPU's per
Hydra chip could be possible, in theory at least.

For those that are worried that this will be another nForce 200 chip, there's
no need to worry as Nir assured us that the Hydra has a TDP of a mere 4W
and as such it shouldn't even need a heatsink, although we expect
motherboard makers to add one just in case.

Source: fudzilla