Topic: CCminer(SP-MOD) Modded NVIDIA Maxwell / Pascal kernels. - page 639. (Read 2347641 times)

It has to transfer to the backplate in the firstplace, which is where the insulation comes from. There is a extra layer here... Also known as a insulator. Air is very bad at conducting heat and electricity...

Just saying 'newtonian physics' or 'fluid dynamics' does not make you right or that you know what you're talking about.

I'm already telling you, that I OWN THE CARDS, and have experience using them. That there is absolutely NO AIRFLOW coming out of the cards. The air needs to blow through the heatsinks. If there is no airflow, then heat can't transfer or reach 'equilibrium' because the heated air does not move out into the environment and equalize.

Heatsinks work at dissipating heat from smaller surface areas to larger ones. A heatsink does not work if there is no where to remove the heat from. They could simply remove the backplate and it would skip the step of transferring heat to the plate and the card would be cooler because of it... with the exception of the pieces of ram.

If what you said worked, they could just put backplate on top of backplate on top of backplate and it would cool the card infinitely better... That's also known as layers of insulation.

"Still air is just about the best insulator there is. At the molecular level, heat transfer is easier when atoms are packed closely together.  So the actual point of most insulations is to get rid of as much solid material and trap still air inside little pockets."

https://www.quora.com/Is-air-an-insulator



1, 3, 4, and 6 are all Zotac 1070 Amp edition cards. Only reason 1 is so low is because RDP is connected and causing the card to lose hashrate. They're also set to 87c as max temp or they'd be sitting at 92, still at 100% fan usage. The coolers are trash, mainly because of the lack of airflow.

The other two cards are MSI Gaming 1070s. Notice that they're sitting at about 70% fan utilization? They'd be cooler if the Zotac cards cooled better and had more airflow around them to push heat away.

Something like that. I'm using 1.0.7 IIRC, because the latest version was not provided for Cuda 6.5 (which is the most that this old driver will support). You'll need to experiment with the different releases to find out one that works.

My 980's are somewhat overclocked (not much), and are doing about 22MH each if I'm not mistaken.

Oh thanks , download the 347.88 for win10 x64 , and with ethminer-0.9.41-genoil-1.0.8.zip daggerhashimoto for eth i have 20 mh/s ? .

THERE YOU SAID IT!--

The backplate is in contact with the airspace and conducts heat away from the air, dissapating it on the cooler back side.  The air is not trapped between the components and the backplate, thermal currents flow to the backplate, and are dissapated.   The backplate "sinks" the heat from the air.  It is not magic, it is Newtonian physics.  There are numerous vents along the back and edge sides of the backplate to allow for fan driven, or thermal driven, airflow.  I just looked.  You brought up fans, anyway.

A heatsink is called a "sink"; it is an analogy to the flow of water.  Heat moves along the path of least resistance, like water, or electricity.  Heat flowing to the backplate will be dissapated rapidly by the high thermal conductivity of the metal forming the backplate.

--scryptr

You just need an older driver version, for example 347.88 will get you over 20MH/s on the 980.

I think we're on two different wavelengths here. I said the backplates are bad in this case because they wrap the card and there is no room for exhaust (they cover the sides of the card where exhaust from the heatsink normally happens). In other words, the fans can't push air out the other side of the heatsink because it's largely blocked.

That aside, in order for a heatsink to work it has to make contact with whatever it's transferring heat from and something like thermal grease or a TIM is used to make sure there is no air between the two objects. Like when you mount a heatsink on a CPU, you don't want air bubbles as they're a insulator. Backplates generally do not have this. The only thing they make contact with is the memory chips, but cooling memory is futile as it almost never gets hot and it's not related to the ability to OC them. So you have a giant stagnant air bubble behind the card between it and the metal backplate.

Heatsinks don't just magically suck heat out of one environment into another.

I have gtx 980 with windows 10 x64 home edition get only 5 mh/s .

METAL IS NOT AN INSULATOR OF HEAT--

It sinks the heat from the air and dissipates ithe heat on the cooler side.  This creates a fluid dynamic air flow away from the hot components.  The entire metal backplate is a conductor of heat.  Most users report 1-5 degree cooler running temperatures when using a backplate.  And, the gas is not trapped as it is in windows.  I said nothing about fans, fluid dynamics accounts for the airflow in the space between hot components and the backplate, and includes fan currents.

The system seeks thermal equilibrium.  Thermal currents flow away from the hot components, aided by the thermal conductivity of the metal backplate.  If the backplate were made of carbon fiber or particle board, your arguments would be more true.     --scryptr

what gpu? 1070 can do 33 in oc on windows 10, plug and play

Any news for ethereum on win 10 ? New driver ? .

No, this doesn't work. Air acts as a insulator. Have you ever heard of double pane windows? Trapping a gas between two sheets of anything is a good way to create a insulator.

Moving heat between anything counts as a insulator anyway as the heat has to move through it. The backplates do not count as heatsinks either, the area the backplate uses to conduct heat is minimal as well compared to a finned heatsink. Heat dissipation is basically proportional to area (not including airflow).

They obviously didn't intend for this effect either or they wouldn't have had fans on the front of the heatsink. How well do you think heat transfer would work with no fans as you're implying?

Putting aside theory, I've tested these and they're pretty bad.

FLUID DYNAMICS--

The backplate conducts heat away from the system faster than the air in circulation around it.  As the air ( a fluid) contacts the backplate, the heat is conducted away from the chipset more rapidly.  This creates a fluid flow towards the backplate, increasing the system's thermal conduction.  In effect, the backplate draws heat, then conducts it away faster than air currents (fluid currrents, a dynamic phenomenon) alone.  The backplate draws heat like a "sink", and conducts the heat away more efficiently than air alone.

Haven't tried it yet, but I am going to buy another one.       --scryptr

Extreme also has the retarded backplate. There is more room for exhaust, but if you notice the backplate wraps around the sides and covers most of the side exhaust severely limiting airflow.

Quite funny how you can have really nice bad ass heatsinks and completely ruin them by trying to make the card look pretty. Right now the cards are hugely flawed, I actually decided to immediately sell mine and try to recoup costs rather then dealing with a dead card or dead fans 3-4 months down the road from now... The fans run at 100% all the time. I see there being almost no OCing headroom on these simply because the cooling sucks so much. Right now I have a boxfan right on top of the cards to keep them at a reasonable temp (~80s).

Really a shame, they look like they could've been good cards too... 2x8pin power connectors, high end circuitry, very beefy and heavy (which is a good sign) heatsink, decent looking fans, and TDP of 120.


3 slot coolers aren't worth it. Usually they cool just as well as a double slot cooler, only take up more space. I've had a few of these. They're generally overkill. Blowers are actually something I'm considering right now as they vacate heat from the rig area, such as with the founders edition.

Yeah, I agree, but that is the opening salvo.

Now, you know why people where opposed to Hard Fork on Ethereum Network!

ASIC hardware could be toast, if the code was changed in ASIC unfriendly manner.

12 x 1070, can do 360MH(windows and 440MH linux but consumption is probably higher) and consume 1200w, so lol at those specs, not even worth it at all

but they cost more i must admit

Thanks for the information. I didn't know of the slot occupation. Luckily not a problem for me as I will run it in an open "case".

Remains to be seen if a rendering becomes a real and working product :-)

Did find this interesting item - out of stock!

But, still, here they are:

http://asicminermarket.com/?product=am440-eth-miner

Main advantage on this offering is electricity works out to be 40% lower than running 14 R9 390's.

However, at this price you'd rather stick with GPU's

TRIPLE SLOT CARD--

The AMP Extreme is Zotacc's top of the line.  It is factory clocked higher than any other GTX of the same "rank" (e.g. 1080, 980ti).  The GTX 1080 AMP Extreme is spec'd at "2.5 slots" and has a boost clock (core) of 1900Hz + .  The triple-fan cooling system has a massive heat block that accounts for why the card is so wide.  The GTX 980ti AMP Extreme that I bought won't fit in the case, period.  I am going to build a multi-card open rig around it, instead of replacing my GTX 960 as originally intended.  A "2.5 slot"  card means that the two adjacent slots on the cooler side are blocked, effectively making the Zotac AMP Extreme card a triple slot card.       --scryptr

I'm not entirely certain who you're even replying to here. I never said anything contradictory to this, nor did anyone here. However, ASICs are still, once again, much more resource intensive then prototyping on FPGAs or mining with GPUs and generally require quite a bit of funding to start out, which was mentioned earlier.