Techpowerup
The cooler cumulatively has 20 heat pipes, a surface area of 2 square meters across 280 aluminum fins. The second module is optional, and is only recommended if the operating environment is warm, or if the end user is overclocking. The card is said to use a complex VRM design that draws power from two 8-pin PCIe power connectors, though we suspect the idea behind that to be heat spreading. The design guarantees reliable silent operation at reference clock speeds or mild overclocking. Display outputs include one each of dual-link DVI, HDMI, and DisplayPort. Colorful is still assessing the marketability of the design. Watch this space for more.

 

 
Wow, this thing is just massive!
 
Source: http://www.techpowerup.co...680-Graphics-Card.html

I'm amazed they would dare create such a thing.  With the extra heatsink, thats a lot of weight on the gfx card.

crocodolljr


I'm amazed they would dare create such a thing.  With the extra heatsink, thats a lot of weight on the gfx card.

Yeah, imagine someone using this card with the Nocuta DH-14 cooler, the motherboard would probably break off.

What a monstrosity! No, thank you.  My 680 card from EVGA with a fan is way quieter than my 460 leaf blowers and runs cooler than those at 100% usage.  I'll stick with what I got.

I can't imagine much of a market for that card. 

Exodai

Techpowerup
The cooler cumulatively has 20 heat pipes, a surface area of 2 square meters across 280 aluminum fins. The second module is optional, and is only recommended if the operating environment is warm, or if the end user is overclocking. The card is said to use a complex VRM design that draws power from two 8-pin PCIe power connectors, though we suspect the idea behind that to be heat spreading. The design guarantees reliable silent operation at reference clock speeds or mild overclocking. Display outputs include one each of dual-link DVI, HDMI, and DisplayPort. Colorful is still assessing the marketability of the design. Watch this space for more.

Wow, this thing is just massive!

Source: http://www.techpowerup.co...680-Graphics-Card.html

That's what she said.  At first I thought this was 2 cards.

Imagine trying to SLI 2 or 3 of them. 

I would get it for the possibility for the absence of the loud stock nvidia fans...

dbe425

I can't imagine much of a market for that card. 

 
I agree. I would rather companies devoted resources to making a single slot version or a quieter dual slot.

Does seem like a fairly pointless exercise.  Takes up a lot of motherboard realestate, & offers no real advantage.  A card with a fan on it will not be any louder than the system fan(s) when operating at stock clocks.
 
Scratching my head on this one...

I think it is brilliant, and if I could afford to get it, it would be a perfect fit for me (except it doesn't look like it is mini-ITX compatible, heatsink would overhang the CPU socket). But it is nothing a flexible PCIe riser wouldn't fix.
 
That's called modding: Making something work (even if it doesn't seem to).
 

dbe425

I can't imagine much of a market for that card.

Unless this is only meant as a proof-of-concept, no company invests their money on a hunch something will sell. They all need gaurantees of return-on-investment prior (meaning Colorful already identified the very same maket you can't imagine) in order to doing R&D, design, fabrication and marketing afterwards to get it sold.  Give them credit for doing their homework.
 

crocodolljr

dbe425

I can't imagine much of a market for that card. 

I agree. I would rather companies devoted resources to making a single slot version or a quieter dual slot.

There is a single-slot version by Galaxy, no ETA.  Besides, the fact that this card's heatsink is so large tells us one thing: They couldn't make it smaller with the resources and technology available to them.   Hint: Lean Manufacturing.
 
Summary of the link: No wasteful spending, if you could design something to be perfect it would take forever and never make it to market, so you got to make a call.  Most importantly, it must satisfy the demand's requirements.  Colorful's market research identified a demand for fanless GTX680, so they went out and found a way to do it.
 
Yet some criticise that decision; why don't you guys design your own heatsink?  There is a reason the examples we see are the way they are, any better makes them expensive or physically bigger or both.  Can't have something for nothing in the engineering world, you have to sacrifice something.  If anything, this GTX680 and possibly the Galaxy single-slot, aren't aimed at people that intend on more than one graphics card.
 
Fact is the market for those getting just one GTX680 is bigger than those getting more than one for SLI, at least I'm betting Colorful's market for a silent 680 is bigger than you guys think.

The backside is only for overclocking.   You don't actually need it for stock clocks.

I give it a 9 to 10 pts on creativity, but if you OC it, and it looks like it's going to cover 3 to 4 slots with that 2nd heatsink, I wouldn't even touch it.  Plus, is this a dual gpu 680 like the Asus Mars III, or a single gpu card?  I hope it's not like the 680 HC with a voltage lock on it--no offense EVGA.  680 + fancy, creative-looking heatsinks + limited or voltage locks = epic fails?  Can't wait to see the benches...

I would love to grab two of those since I'm not an avid overclocker. My cpu will be on water sooner or later, so the system would literally be silent!

uh, that's a monster ... I personally wouldn't even consider this ...
 
... but, depending on the price it might a perfect solution for silent single-gpu builds without water cooling on gpu (depending on which one turns out cheaper).
 
+1 on creativeness and filling the niche on the market
-1 from multi-gpu standpoint however ... that thing is GIANT!

thats insane looking

That'd be excellent for a passive HTPC! :)

lehpron


I think it is brilliant, and if I could afford to get it, it would be a perfect fit for me (except it doesn't look like it is mini-ITX compatible, heatsink would overhang the CPU socket). But it is nothing a flexible PCIe riser wouldn't fix.

That's called modding: Making something work (even if it doesn't seem to).

dbe425

I can't imagine much of a market for that card.

Unless this is only meant as a proof-of-concept, no company invests their money on a hunch something will sell. They all need gaurantees of return-on-investment prior (meaning Colorful already identified the very same maket you can't imagine) in order to doing R&D, design, fabrication and marketing afterwards to get it sold.  Give them credit for doing their homework.

 
While I always love your point of view on any given topic, I must respectfully disagree on the above statements.  Companies and products go down the toilet every day because people don't do their homework or their homework is all wrong.  Just because people have the intelligence to create something like this card doesn't mean they have equal business sense.  A good example might be Zune.  I'm sure MS did their homework but that's never a guarantee. 

Do they include a forklift with the purchase of the heat sink?

There's certainly a market for passive cooled and if there isn't much competition in it then why not make them?

There is and will always be a market for silent graphics cards. I could see this being used in a post production / foley studio. There are times when you can't have any background noise in your computer, for the tasks at hand. Chances are this prototype card was designed to be installed in an horizontial fashion such as a rack case. For a workstation.

Some people use high end cards who do not play games. And do not need sli.

Why not just use soundproofing material on the doors? I do not see the point of these cards.

NOPE.

anarchy998


Why not just use soundproofing material on the doors? I do not see the point of these cards.

 
Sound proofing will only go so far. Anywhere there are obstructions in a fan's airflow you create turbulence. It is that turbulence that creates the noise. When the fan speed goes past 50-60% you will hear the noise out the heat sink outlet.
 
One of the best advancements to gpu fans has been the use of the squirrel cage fan. It is able to move a massive volume of air with relatively low rpm speeds. The main problem is on graphics cards the individual vanes on the fans used are square edged and it's this edge where most of the noise is generated. Your house A/C unit uses the same type of fan but has significantly more aerodynamic vanes. This makes for a near silent fan that can move 1200-1400  average cfm at 1075 rpm. If they adopted this same tapered edge as well on the gpu fans the noise level would probably be cut in half or more and actually increase airflow as well.

anarchy998


Why not just use soundproofing material on the doors? I do not see the point of these cards.

 
Because sound waves travel through air, which goes through all the fan openings anyway. Unless you use very low RPM on good sized fans, and then all the sound dampening, that's not a solution in and of itself.

Seems like wagging the dog by its tail in order to cool it off rather than just dropping the pooch in some water.

dbe425

Companies and products go down the toilet every day because people don't do their homework or their homework is all wrong.  Just because people have the intelligence to create something like this card doesn't mean they have equal business sense.  A good example might be Zune.  I'm sure MS did their homework but that's never a guarantee.

You're forgetting the caveat of time:  People change, thus demand changes. 
 
If you do your homework for a particular demand and determine a type of product, if by the time you do the R&D and produce it and try selling it, the market changes and suddenly no one wants it, who's fault is that?  Technically no one, it is not like you deliberately designed something with no demand; but if you're too far gone, you might as well get something out of it.
 
Best example is Concorde.  Back in the late 1950s, supersonic travel was hailed as the future for the masses, yet during the fifteen years it took to produce the aircraft and bring it to service, people changed and became more economically and environmentally aware of the damage SST's would have caused that back in the 50's nobody cared about.  Yet for some reason, everyone blames the airplane, as if such ideas are showstoppers without a way around it (sonic boom, fuel consumption, noise, etc).  No, they are just design challenges that require more thinking than doing so the final action is the best it can be with minimal risk and waste.  The current airliner market just likes to 'do what works' because of the minimal risk and decades of proven ideas that cannot all apply to a faster airliner.
 
It is a very similar situation with this silent GTX680, it definitely was a design challenge and not a showstopper or something that should not have been tried or has no point.  We won't know until, or if, Colorful makes this a retail product if the homework they did at the early preliminary stages still holds up.  Demand must remain unchanged and constant if not growing for companies to risk an investment.
 
All companies do their homework, they just can't control time.

Very true.  To be honest I love to see new tech come out even if there is no real market.  How many useful inventions came about because someone decided to try a seemingly useless idea.   

Neat.