AMD R9 390X Coming With Cooler Master Liquid Cooler + Estimated Performance

[KeltonDSMer]

Not at all. The entire reason silicon chips die is due to loss of mass over time. Depending on the resistance of the circuit you can easily generate more heat than the amount of power you dump into it. Go ahead and yank out your resistor circuit problem sets from your high school/college/university days. The extra energy comes from the circuit itself. There is no violation of the conservation of energy. That said, if you want the fully detailed version I'm going to have to get into quantum mechanics and multivariable calculus, something this website is not at all a good medium for.

 

Matter is conserved too. The exact amount of energy used to degrade a chip (silicon or copper solid->gas sublimation) is "recovered" as the gas condenses like 2nm away from where it was vaporized. A chip that starts with 50grams of copper and 50 grams of silicon still contains exactly that much matter after it is "dead". Maybe I am just misunderstanding your post, but "The extra energy comes from the circuit itself. There is no violation of the conservation of energy." does not make any sense to me. There is no extra energy. Energy in=energy out. A graphics card at constant temp drawing 200W is always emitting 200W of heat (and some radiation which is immediately converted to thermal anyway).

 

Would you mind clarifying for me what you are saying? I have re-read your posts and am still left confused about what your trying to convey.

 

Accelerating an object increases its momentum, but it also shaves mass in the tiniest of quantities. Also, I'm not violating the laws of conservation of energy. The circuit itself is losing mass over time, and a single atom has a ton of energy to give. Take out any circuit problem set from your first year in physics. You'll find the total power dissipated by the circuit can be greater than the electrical power pumped through it depending on the resistors you put in the way.

 

Matter is not being converted to energy nor is the opposite true. These aren't nuclear reactors! A chip's degradation is comparable to a pot of boiling water. The water on a molecular level is not changing as it becomes steam, it is simply changing phases. The amount of energy used to convert the liquid to gas is recovered when the steam condenses, there is no loss/gain of energy or matter. Same goes with a degrading circuit. Any "loss in mass" of a circuit isn't a loss. That matter is still there, just deposited in the "wrong" place.

[Lewellyn]

Matter is conserved too. The exact amount of energy used to degrade a chip (silicon or copper solid->gas sublimation) is "recovered" as the gas condenses like 2nm away from where it was vaporized. A chip that starts with 50grams of copper and 50 grams of silicon still contains exactly that much matter after it is "dead". Maybe I am just misunderstanding your post, but "The extra energy comes from the circuit itself. There is no violation of the conservation of energy." does not make any sense to me. There is no extra energy. Energy in=energy out. A graphics card at constant temp drawing 200W is always emitting 200W of heat (and some radiation which is immediately converted to thermal anyway).

 

Would you mind clarifying for me what you are saying? I have re-read your posts and am still left confused about what your trying to convey.

 

 

Matter is not being converted to energy nor is the opposite true. These aren't nuclear reactors! A chip's degradation is comparable to a pot of boiling water. The water on a molecular level is not changing as it becomes steam, it is simply changing phases. The amount of energy used to convert the liquid to gas is recovered when the steam condenses, there is no loss/gain of energy or matter. Same goes with a degrading circuit. Any "loss in mass" of a circuit isn't a loss. That matter is still there, just deposited in the "wrong" place.

 

TDP = Thermal Design Power, is the heat that needs to be dissipated in order to stay under the thermal limit of the part, or in other words the theoretical maximum of heat output. I'm getting really annoyed at people in this thread arguing about a pretty much un-standardized measurement (not you)...

[Giannhs]

OP please rename the post to TDP lessons or TDP wars  
 

*sarcasm*

[patrickjp93]

 

p=mv, decreasing an object's mass would decrease its momentum. Furthermore, reducing an object's mass would make it easier to accelerate, making it easier to increase the velocity of a massive object the closer it approached c. The usual (mis)conception is that accelerating a massive object increases its mass. In reality mass is invariant but relativistic mass is a quantity which undergoes Lorentz transformations.

 

 

 

And there was me thinking that measuring a cooler's TDP at a specified temperature, such as 20-25 degrees C to match what is usually taken as "room temperature" would be the only sensible way of doing it, and certainly not in a vacuum. Still, your radiator would actually be a radiator in that case, losing heat only through BBR

 

facepalm... E=mc^2 for any matter. This is why the sun loses mass as it gives off light, and it's the same for production of heat and infrared light. Any motion whatsoever will lose the object some of its mass due to radiation and entropy.

 

Matter is conserved too. The exact amount of energy used to degrade a chip (silicon or copper solid->gas sublimation) is "recovered" as the gas condenses like 2nm away from where it was vaporized. A chip that starts with 50grams of copper and 50 grams of silicon still contains exactly that much matter after it is "dead". Maybe I am just misunderstanding your post, but "The extra energy comes from the circuit itself. There is no violation of the conservation of energy." does not make any sense to me. There is no extra energy. Energy in=energy out. A graphics card at constant temp drawing 200W is always emitting 200W of heat (and some radiation which is immediately converted to thermal anyway).

 

Would you mind clarifying for me what you are saying? I have re-read your posts and am still left confused about what your trying to convey.

 

 

Matter is not being converted to energy nor is the opposite true. These aren't nuclear reactors! A chip's degradation is comparable to a pot of boiling water. The water on a molecular level is not changing as it becomes steam, it is simply changing phases. The amount of energy used to convert the liquid to gas is recovered when the steam condenses, there is no loss/gain of energy or matter. Same goes with a degrading circuit. Any "loss in mass" of a circuit isn't a loss. That matter is still there, just deposited in the "wrong" place.

 

No, no, and no. If you take the mass of a chip after running it for 6 years, even under normal circumstances, using very precise instruments, it will have lost mass due to production of light in the infrared spectrum due to the heat. Matter is constantly being converted into lower forms of energy including heat and higher forms such as light. This is invariably true. Even at absolute zero all matter continues to lose mass by actions in the Fermi energy states. Consult any university physics professor or any expert in quantum physics. No process conserves mass. If any did, the laws of thermodynamics would not hold and the heat death of the universe would not be inevitable. 

 

energy in=energy out, but you're failing to take in other sources of energy besides the electricity. Some energy is lost to light at low intensities (most of it outside the visible spectrum), some is lost to heat, and that total sum of transformed energy is greater than the electrical energy pushed in. If you don't believe that go yank out your old college physics text and read very carefully the sections on thermodynamics before going back to your circuits section. Every single motion of every particle causes loss of energy to heat, and that results in loss of mass over time. It's an inescapable fact of this universe.

[patrickjp93]

TDP = Thermal Design Power, is the heat that needs to be dissipated in order to stay under the thermal limit of the part, or in other words the theoretical maximum of heat output. I'm getting really annoyed at people in this thread arguing about a pretty much un-standardized measurement (not you)...

Each company has different ways of measuring TDP. It is by no means a theoretical maximum, as Nvidia's power draw is greater than the TDP they publish. Intel's TDP is a weighted average calculated by an algorithm where the input is the circuit design and properties of the silicon. AMD's is similar.

 

You're the one way off base here, no offense.

[othertomperson]

facepalm... E=mc^2 for any matter. This is why the sun loses mass as it gives off light, and it's the same for production of heat and infrared light. Any motion whatsoever will lose the object some of its mass due to radiation and entropy.

 

That's the energy required to create a particle of a given mass at rest.

 

The energy of a particle that's not at rest is given by

 

E²=m²c⁴+p²c²

 

You don't just take it out of the rest mass when you move a particle! That would literally change what kind of particle you would be dealing with. Notice how when you use the more general energy-mass relation you can give a particle momentum without altering its mass.

 

The sun creates energy by fusing two Hydrogen nuclei, turning one of the protons into a neutron to create deuterium. Two deuterium nuclei then fuse to form Helium. This process releases neutrinos, positrons and photons. This comes from the change in the binding energy of the nuclei involved (Helium is a lower energy system than two hydrogen atoms, so you get energy out in the form of new particles, the neutrinos and photons. If you stop to think about it, a Hydrogen plasma is just protons existing on their own not in any bound state with anything, which is why fusion involving Hydrogen yields the most energy release of any element). Binding energy is negative, so it acts to reduce the energy of the bound system below that of the constituent elementary particles, it doesn't actually change the mass of the elementary particles themselves.

 

I think what you're getting confused about is that the mass of a Helium atom, say, is less than the mass of two protons and two neutrons existing on their own. This is because of the negative binding energy, it requires you to put energy into the system (equal to the binding energy) to make up the difference, that is to separate them. (Same as in chemistry, by the way), and this energy difference is where the energy to form the new particles comes from.

[Lewellyn]

Each company has different ways of measuring TDP. It is by no means a theoretical maximum, as Nvidia's power draw is greater than the TDP they publish. Intel's TDP is a weighted average calculated by an algorithm where the input is the circuit design and properties of the silicon. AMD's is similar.

 

You're the one way off base here, no offense.

 

TDP isn't power draw and TDP will always be lower than the power draw, you can't produce heat greater than the power consumed...

 

"The thermal design power (TDP), sometimes called thermal design point, is the maximum amount of heat generated by the CPU that the cooling system in a computer is required to dissipate in typical operation."

[KeltonDSMer]

Each company has different ways of measuring TDP.

 

That statement is exactly why I feel so strongly about TDP being a useless number in these situations.

 

 

 

 

No, no, and no. If you take the mass of a chip after running it for 6 years, even under normal circumstances, using very precise instruments, it will have lost mass due to production of light in the infrared spectrum due to the heat. Matter is constantly being converted into lower forms of energy including heat and higher forms such as light. This is invariably true. Even at absolute zero all matter continues to lose mass by actions in the Fermi energy states. Consult any university physics professor or any expert in quantum physics. No process conserves mass. If any did, the laws of thermodynamics would not hold and the heat death of the universe would not be inevitable. 

 

energy in=energy out, but you're failing to take in other sources of energy besides the electricity. Some energy is lost to light at low intensities (most of it outside the visible spectrum), some is lost to heat, and that total sum of transformed energy is greater than the electrical energy pushed in. If you don't believe that go yank out your old college physics text and read very carefully the sections on thermodynamics before going back to your circuits section. Every single motion of every particle causes loss of energy to heat, and that results in loss of mass over time. It's an inescapable fact of this universe.

 

Okay that is interesting, but atoms emitting radiation from an excited state aren't losing mass. Sure the emitted energy can be considered as mass, but this energy came from outside the system to begin with so the net gain/loss is zero. What are you saying is happening on a molecular level as described in the first paragraph quoted? All materials emit radiation as they reach excited states, but they atomically do not change as they emit radiation. Lasers would last all of 2 nano seconds if the gases inside the case changed their atomic structure as they excite/relax. An incandescent light bulb does not have its tungsten filament magically turn to lead throughout its lifespan.

 

If we could crack atoms in a controlled manner inside our computers to release "extra" energy, the US DOE would be writing us fat checks for our new found solution to the world's energy crisis. Sadly, our GPUs are not sustaining a controlled fission-like reaction.

 

Feel free to PM me to discuss further if you want to, I am genuinely interested in what you have to say.

 

Sorry for muddying up the 3xx series thread, I will not post any more off topic here.

[Sakkura]

Not at all. The entire reason silicon chips die is due to loss of mass over time. Depending on the resistance of the circuit you can easily generate more heat than the amount of power you dump into it. Go ahead and yank out your resistor circuit problem sets from your high school/college/university days. The extra energy comes from the circuit itself. There is no violation of the conservation of energy. That said, if you want the fully detailed version I'm going to have to get into quantum mechanics and multivariable calculus, something this website is not at all a good medium for.

 

In the absence of a chemical or nuclear reaction, what you are suggesting is a clear and direct contradiction of conservation of energy. You can't just handwave it away with quantum mechanics, it applies just as much there.

 

No, no, and no. If you take the mass of a chip after running it for 6 years, even under normal circumstances, using very precise instruments, it will have lost mass due to production of light in the infrared spectrum due to the heat. Matter is constantly being converted into lower forms of energy including heat and higher forms such as light. This is invariably true.

 

No, that is invariably wrong. When the chip is not running, it is in thermal equilibrium with its surroundings, receiving as much radiation as it emits. Clearly no net mass or energy transfer. When it is running, it is receiving energy from the electrical grid. That energy raises the temperature of the chip, which it mostly conducts into the cooler, though a tiny fraction could be lost radiatively. However, that does not affect the mass of the chip, since it's just throwing out the energy that was poured in.

 

 

Even at absolute zero all matter continues to lose mass by actions in the Fermi energy states. Consult any university physics professor or any expert in quantum physics. No process conserves mass. If any did, the laws of thermodynamics would not hold and the heat death of the universe would not be inevitable. 

 

energy in=energy out, but you're failing to take in other sources of energy besides the electricity. Some energy is lost to light at low intensities (most of it outside the visible spectrum), some is lost to heat, and that total sum of transformed energy is greater than the electrical energy pushed in. If you don't believe that go yank out your old college physics text and read very carefully the sections on thermodynamics before going back to your circuits section. Every single motion of every particle causes loss of energy to heat, and that results in loss of mass over time. It's an inescapable fact of this universe.

 

No process conserves mass? Funny. There's something called conservation of energy, and there's something called mass-energy equivalence. The obvious conclusion is that there must also be conservation of mass, and as it turns out that's another fundamental fact of physics, including in special relativity. The heat death of the universe is irrelevant, because it will not destroy any energy or mass, it will simply be distributed differently.

 

As for energy being lost to heat, resulting in a loss of mass, that is wrong because that heat contains all the supposedly lost energy, and thus there can also be no overall mass loss either - merely a transfer or transformation.

[Sakkura]

Ugh, so many people trying to lecture me about TDP, and getting it wrong at that.

 

Either way, the entire point I was making, is that you cannot criticize AMD's powerusage or TDP, without taking it's performance into account. A 390x having 80% of a Titan Z' TDP is not a bad thing, if the perfomance is at least 80% of the Titan Z.

 

It's all speculation atm anyways, so let's get back to it, when this card is actually reviewed.

 

Well, Titan Z is Kepler-based. AMD cards do need to be compared with Maxwell-based cards too. But except for small form factor systems where cooling is difficult, performance is a lot more important than efficiency. Besides, people tend to overestimate the efficiency increase with Maxwell, and may be underestimating the efficiency improvements AMD has made with GCN 1.2/1.3 as well as HBM. If they move to 20nm with the 390X, that should help as well.

[mr moose]

Again, no! You are all forgetting the other source of energy in: the mass of the circuit. Any time you push electricity through a non-superconducting circuit, you shave off extremely small amounts of mass. It's just entropy at work. That's why TDP can exceed power usage quite easily. This is entry-level laws of thermodynamics and consequences of resistance in a circuit. This is second semester physics at any decent college/university in the US if you didn't take AP Physics C in high school.

 

GPU's are not like nuclear reactors, and even if what you were saying made any sense, its effects on the energy use and energy dissipation would be extremely negligible in this discussion. 

 

You find me a chip that produces more heat heat energy than it consumes in electrical energy and I'll make sure you get a nobel prize for discovering zero point energy.

[Memphaestus]

facepalm... E=mc^2 for any matter. This is why the sun loses mass as it gives off light, and it's the same for production of heat and infrared light. Any motion whatsoever will lose the object some of its mass due to radiation and entropy.

 

 

No, no, and no. If you take the mass of a chip after running it for 6 years, even under normal circumstances, using very precise instruments, it will have lost mass due to production of light in the infrared spectrum due to the heat. Matter is constantly being converted into lower forms of energy including heat and higher forms such as light. This is invariably true. Even at absolute zero all matter continues to lose mass by actions in the Fermi energy states. Consult any university physics professor or any expert in quantum physics. No process conserves mass. If any did, the laws of thermodynamics would not hold and the heat death of the universe would not be inevitable. 

 

energy in=energy out, but you're failing to take in other sources of energy besides the electricity. Some energy is lost to light at low intensities (most of it outside the visible spectrum), some is lost to heat, and that total sum of transformed energy is greater than the electrical energy pushed in. If you don't believe that go yank out your old college physics text and read very carefully the sections on thermodynamics before going back to your circuits section. Every single motion of every particle causes loss of energy to heat, and that results in loss of mass over time. It's an inescapable fact of this universe.

Why don't you link a credible source proving what you are saying.  Instead of spewing baseless statements and hoping that they are fact, maybe you could be the one to actually provide the proof for a change.  

 

That statement is exactly why I feel so strongly about TDP being a useless number in these situations.

It is the same thing with the tire industry for UTQG tread wear grade.  Every company uses a different measurement, so comparing brands using that specification is totally useless.  Now comparing them within the same company is okay.  For instance the 290x has a 295w TDP, while the 270 has 150w TDP, letting you know that the 290x produces almost twice as much heat. But comparing to anything in Nvidia or Intel is pointless.  I wish people would understand this simple fact.

 

GPU's are not like nuclear reactors, and even if what you were saying made any sense, its effects on the energy use and energy dissipation would be extremely negligible in this discussion. 

 

You find me a chip that produces more heat heat energy than it consumes in electrical energy and I'll make sure you get a nobel prize for discovering zero point energy.

I like how he compared the nuclear chemical reaction in the sun to the flow of electricity in a circuit.  Now if he was talking about the reaction in a battery, I could get on board, because that is an actual chemical reaction and not the flow of charged particles in a system, ie electricity.

[mr moose]

Why don't you link a credible source proving what you are saying.  Instead of spewing baseless statements and hoping that they are fact, maybe you could be the one to actually provide the proof for a change.  

 

It is the same thing with the tire industry for UTQG tread wear grade.  Every company uses a different measurement, so comparing brands using that specification is totally useless.  Now comparing them within the same company is okay.  For instance the 290x has a 295w TDP, while the 270 has 150w TDP, letting you know that the 290x produces almost twice as much heat. But comparing to anything in Nvidia or Intel is pointless.  I wish people would understand this simple fact.

 

I like how he compared the nuclear chemical reaction in the sun to the flow of electricity in a circuit.  Now if he was talking about the reaction in a battery, I could get on board, because that is an actual chemical reaction and not the flow of charged particles in a system, ie electricity.

 

Its like he's done a quick google search and grabbed random equations and concepts then thrown them all in the mix in some sort of hope that we would be confused with the terms and give up. 

 

Last time I looked most SS devices didn't absorb energy or convert mass to energy on any scale measurable with domestic tools.

[s3ns3]

wtf nuclear ,sun ,chemical reactions, mass energy, particles , radiation WTF???? 

[Lewellyn]

wtf nuclear ,sun ,chemical reactions, mass energy, particles , radiation WTF???? 

 

If you're high enough I guess it could make great comedy.

[RexinOridle]

But will it play 4K Crysis?

[Hatsune Miku 「 」]

How does a GPU have the same reaction processes such as those in a nuclear mass such as the sun?

E=mc^2 is irrelevant to electrical resistance. At the very least who ever used this has an argument has no idea what he is taking about.

And electrical resistance produces heat because more voltage is needed to be applied to force a flow of electrons, NUCLEAR REACTIONS ARE NOT A THING INSIDE A COMPUTER COMPONENT.

and Electrical Resistance != TDP or heat output. If that was true we would have 100 degree heaters and even hotter electrical ovens. Thus I conclude that the heat output measured being released from a GPU CANNOT be greater than the energy drawn from the PSU if the amount energy drawn by the GPU is converted to heat, otherwise you're creating heat (HEAT IS A FORM OF ENERGY) out of no where, which is impossible due to the law of the conservation of energy.

[D3MON]

I'm putting money on the reference card being air cooled but with an improved fan design which is the main culprit of bad cooling not the chip.
also i bet the 390x2 being dual gpu with an AiO.

for those that havent realised yet TDP=Thermal Design Power is heat output and NOTHING to do with power draw, mixing the 2 is as bad as saying that a 600w power supply will run hotter than a 500w...

also gtx480 ran hot due to-1) too much factory voltage, 2) so many heatpipes it killed all heatsink airflow, 3) dinky little fan that moved NO air, stock clocks with a gtx285 fan and 580 heatsink will reduce temps allot!

also "nearly the same tdp as an nvidia dual gpu titan-z, thats terrible", wtih 300w vs 250w TDP ill take a 20% higher heat output for 40% more performance otherwise we'd all be running cute 750ti's.
oh and as for temps of a 300wTDP card these twin fan coolers from asus/xfx/msi etc are very simple really, fins tacks are hard to make sure but 2 cheap fans over a bunch of fins isnt rocket science and could easily done as factory reference (see 7990) though usually arn't done because non-heatpiped blower types are slightly cheaper when you're putting out tens of thousands of cards and you want to leave something for ITX users.

[othertomperson]

I paid £9000 for a physics degree and all it got me was an argument on the Internet

[MrSuperb]

for those that havent realised yet TDP=Thermal Design Power is heat output and NOTHING to do with power draw, mixing the 2 is as bad as saying that a 600w power supply will run hotter than a 500w...

 

of course ... it has everything to do with power draw ...

[Lewellyn]

I paid £9000 for a physics degree and all it got me was an argument on the Internet

 

Could of had a v8

[patrickjp93]

How does a GPU have the same reaction processes such as those in a nuclear mass such as the sun?

E=mc^2 is irrelevant to electrical resistance. At the very least who ever used this has an argument has no idea what he is taking about.

And electrical resistance produces heat because more voltage is needed to be applied to force a flow of electrons, NUCLEAR REACTIONS ARE NOT A THING INSIDE A COMPUTER COMPONENT.

and Electrical Resistance != TDP or heat output. If that was true we would have 100 degree heaters and even hotter electrical ovens. Thus I conclude that the heat output measured being released from a GPU CANNOT be greater than the energy drawn from the PSU if the amount energy drawn by the GPU is converted to heat, otherwise you're creating heat (HEAT IS A FORM OF ENERGY) out of no where, which is impossible due to the law of the conservation of energy.

It's not reactions in the sense you are thinking. Absolutely every process in this universe increases entropy, moves energy from higher forms to lower towards heat. All the mass and energy in the universe will one day exist as heat. Every time you pass electricity through a circuit, you generate heat which shaves mass on the tiniest of scales. It does not have to be a nuclear reaction. It's just basic thermodynamics. Did none of you take physics in college and get even this far?

 

I did not say resistance = TDP. Resistance and TDP are directly related, and Power dissipated by a circuit is the sum of the current^2 running through each resistive element multiplied by the resistance of said element. It's a law of electrophysics that heat dissipated by a resistor is I^2 * R. If you'd like to argue over it, go talk to any degreed physicist or consult a pirated college physics text.

 

Your conclusion is wholly incorrect because your understanding of physics itself is quite lacking. You are not creating that heat energy out of nowhere. The extra energy comes from converting mass to energy by way of quantum mechanical processes which result from the motion of particles, motion accelerated when heat is introduced.

 

The law of conservation of energy is an absolute, but think about collisions or friction. If you're running and then plant your feet, skidding along on your shoes, you are transferring some of your momentum to the ground, an exchange of energy, but some of that energy is also lost to heat, and you also wear down your shoes. Some of the mass in your shoes was not only torn off by the rough ground surface, but lost entirely. As I keep saying, if you don't buy what I'm selling, you can consult any physicist you want. I'm only the messenger. Existence sucks that way.

[Hatsune Miku 「 」]

 

You need energy to move those particles in the first place inorder to wear out the material that you claim is generating the energy into a form of heat.

 

Tell me what law of quantum mechanics explains your point of creating energy(heat) due to quantum processes?

[Lewellyn]

You need energy to move those particles in the first place inorder to wear out the material that you claim is generating the energy into a form of heat.

 

Tell me what law of quantum mechanics explains your point of creating energy(heat) due to quantum processes?

 

He doesn't have a point, it's just nonsense. Basic thermodynamics doesn't have anything to do with quantum mechanics and "mass being shaven off". As a physics student this misrepresentation of the subject is making my head hurt.

[Hatsune Miku 「 」]

He doesn't have a point, it's just nonsense. Basic thermodynamics doesn't have anything to do with quantum mechanics and "mass being shaven off". As a physics student this misrepresentation of the subject is making my head hurt.

I was curious about that too. Just read the laws of quantum mechanics and have found nothing on the removal of matter from conductive materials due to heat or the conversion of non potent materials to energy.

Unless there is an unexplained loophole. I really want to get a massive computer to compute the possibilities of zero degree celcius super conductors and their compositions and finally just remake nvidia's kepler with maxwell elements.

Also any physics student or academic will realise that maxwell in nvidia refers to maxwell's equations of light. Which is why nvidia has it "inspired by light"

 

The closest thing heat can do to a conductor is melt it since most conductors are metals. And shaving off matter by heat is mostly applicable in insulators wearing out. You can't shave gold using heat can you?