Why do Desktop Processors need/have Heat Spreaders?

[Riley-NZL]

Why do desktop CPU's have heat spreaders? I work in the portable computer market and none of the processors in  laptops etc have a heat spreaders, the thermal compound and cooling solution are mounted directly onto the CPU die. 

 

I'm sure there is a good reason, or they wouldn't bother.

 

 

[tp95112]

Im guessing the die is alot smaller and the heat spreader helps add surface area to help cool the chip

[WhiteWolf117]

1 minute ago, Riley-NZL said:

Why do desktop CPU's have heat spreaders? I work in the portable computer market and none of the processors in  laptops etc have a heat spreaders, the thermal compound and cooling solution are mounted directly onto the CPU die. 

 

I'm sure there is a good reason, or they wouldn't bother.

 

 

i could be wrong but maybe for aftermark coolers and such

[Riley-NZL]

1 minute ago, tp95112 said:

Im guessing the die is alot smaller and the heat spreader helps add surface area to help cool the chip

Mobile CPU dies are generally somewhat smaller then Desktop dies.

[Lays]

Probably cuts down on cracked dies from people that don't know what they're doing.  

[Hackentosher]

The die is much smaller than the heat spreader, look up what de-lidded CPUs look like. It also protects the CPU when you clamp a cooler onto it.

[intelCore]

2 minutes ago, Riley-NZL said:

Why do desktop CPU's have heat spreaders? I work in the portable computer market and none of the processors in  laptops etc have a heat spreaders, the thermal compound and cooling solution are mounted directly onto the CPU die. 

 

I'm sure there is a good reason, or they wouldn't bother.

 

 

Commons sense would tell me that since the CPU is laying on pins and is not soldered would need more spread force (applied by the cooler) to not lose contact. Also the CPU socket and the ledge needs something to press down on and if it was just the substrate I bet that would bend or brake.

[Godlygamer23]

Surface area.

[Godlygamer23]

5 minutes ago, intelCore said:

Commons sense would tell me that since the CPU is laying on pins and is not soldered would need more spread force (applied by the cooler) to not lose contact. Also the CPU socket and the ledge needs something to press down on and if it was just the substrate I bet that would bend or brake.

What about AMD CPUs? In both cases, the pins on the CPU/CPU socket bend from the pressure.

You can use a CPU without a heatspreader just fine. But heatspreaders also add rigidity - if too much pressure is applied by the CPU cooler, it can actually crush a CPU die that's "exposed."

[intelCore]

2 minutes ago, Godlygamer23 said:

Surface area.

Why no headspreader on graphics cards then?

[intelCore]

2 minutes ago, Godlygamer23 said:

What about AMD CPUs?

You can use a CPU without a heatspreader just fine. But heatspreaders also add rigidity - if too much pressure is applied by the CPU cooler, it can actually crush a CPU die that's "exposed."

AMD CPUs have yet to lay on pins as far as I know since intel CPUs used to not have a heatspreader when they still had their pins on the substrate aswell.

[Godlygamer23]

Just now, intelCore said:

Why no headspreader on graphics cards then?

It turns out GPUs are actually easier to cool.

[intelCore]

Just now, Godlygamer23 said:

It turns out GPUs are actually easier to cool.

Huh how's that?

[kirashi]

2 minutes ago, intelCore said:

Why no headspreader on graphics cards then?

Because adding aftermarket heatsinks to Graphics Cards is a lot less common than your regular CPU heatsink.

[Godlygamer23]

4 minutes ago, intelCore said:

AMD CPUs have yet to lay on pins as far as I know since intel CPUs used to not have a heatspreader when they still had their pins on the substrate aswell.

In both cases, the pins are bending from the retention pressure. There is no need to think about the loss of contact - the amount of pressure present in the mount is sufficient to hold down the CPUs.

Edited August 6, 2016 by Godlygamer23

[Godlygamer23]

Just now, intelCore said:

Huh how's that?

Some oldie posts:

https://linustechtips.com/main/topic/27337-gpu-vs-cpu-heat-output/?do=findComment&comment=347530

https://linustechtips.com/main/topic/27337-gpu-vs-cpu-heat-output/?do=findComment&comment=348279

 

[xentropa]

Probably because CPU coolers have to be installed manually by clumsy human hands and too many people were dropping screwdrivers onto the die and breaking them or something.  Laptops and GPUs come with cooling solutions installed.

[intelCore]

14 minutes ago, kirashi said:

Because adding aftermarket heatsinks to Graphics Cards is a lot less common than your regular CPU heatsink.

Waterblocks on graphics card have become a thing I think.

 

13 minutes ago, Godlygamer23 said:

In both cases, the pins are bending from the retention pressure. There is no need to think about the loss of contact - the amount of pressure present in the mount is sufficient to hold down the CPUs.

And that is exactly my point. While the whole "pins on the CPU going into the socket" mounting variant does not rely on pressure from above LGA sockets do. Whether it is from the retention bracket of the socket itself or the cooler doesn't matter. It needs force to push it down and if that force was just pushing down on the substrate it would damage it before getting sufficient pressure.

[Godlygamer23]

10 minutes ago, intelCore said:

It needs force to push it down and if that force was just pushing down on the substrate it would damage it before getting sufficient pressure.

Then explain why people can use Intel CPUs without a heatspreader.

[intelCore]

Just now, Godlygamer23 said:

Then explain why people can use Intel CPUs without a heatspreader.

I don't know exactly how that works but I assume the pressure you need is either applied by a custom made retention bracket or just the cooler itself with maybe some kind of support bracket on the substrate.

 

And if it was just for cooling why is there a heatspreader then? You would get more surface with the cooler anyways and obviously better cooling without another layer.

[Godlygamer23]

Just now, intelCore said:

I don't know exactly how that works but I assume the pressure you need is either applied by a custom made retention bracket or just the cooler itself with maybe some kind of support bracket on the substrate.

 

And if it was just for cooling why is there a heatspreader then? You would get more surface with the cooler anyways and obviously better cooling without another layer.

Heatspreaders also protect the die from damage from the cooler, but people use the stock retention when using CPUs with no heatspreaders.

 

Some manufacturers of water cooling parts have made special parts that elevate the cooler from the CPU to keep the mounting pressure from crushing the die, which is also much higher than the pressure to keep the CPU inside the socket properly.

[Mira Yurizaki]

Heat spreaders were originally designed purely for preventing users from damaging the CPU die when installing aftermarket parts. You get better cooling performance if the heat sink directly touches the die, so there's no real point to having a heat spreader besides offering a mechanical safety. Though apparently delidding doesn't do a whole lot anyway so...

[Riley-NZL]

49 minutes ago, Godlygamer23 said:

It turns out GPUs are actually easier to cool.

Er what? High end GPU's put out 2x as much heat as high end CPU's.

 

56 minutes ago, Hackentosher said:

The die is much smaller than the heat spreader, look up what de-lidded CPUs look like. It also protects the CPU when you clamp a cooler onto it.

And Mobile CPU dies are even smaller again and do not require heat spreader?

 

49 minutes ago, Godlygamer23 said:

 

56 minutes ago, intelCore said:

Commons sense would tell me that since the CPU is laying on pins and is not soldered would need more spread force (applied by the cooler) to not lose contact. Also the CPU socket and the ledge needs something to press down on and if it was just the substrate I bet that would bend or brake.

Many  (particularly older) mobile CPU's also use a pin/socket system. It has only become common practice to solder them in in the last few years. The old socketed design mobile CPU's also had no heat spreader.

[intelCore]

6 minutes ago, Godlygamer23 said:

Heatspreaders also protect the die from damage from the cooler, but people use the stock retention when using CPUs with no heatspreaders.

 

Some manufacturers of water cooling parts have made special parts that elevate the cooler from the CPU to keep the mounting pressure from crushing the die, which is also much higher than the pressure to keep the CPU inside the socket properly.

Okay so what point were you trying to make again exactly?

 

I wanted to say that this is for the way of how the CPU is mounted into the socket.

 

5 minutes ago, Riley-NZL said:

Many  (particularly older) mobile CPU's also use a pin/socket system. It has only become common practice to solder them in in the last few years. The old socketed design mobile CPU's also had no heat spreader.

Then I have to admit I do not know.

[Mira Yurizaki]

18 minutes ago, Riley-NZL said:

Er what? High end GPU's put out 2x as much heat as high end CPU's.

It's questionable that they do. People are still confusing TDP as power consumption and I'm afraid it's even seeping into NVIDIA's product literature. Also is it the TDP of the GPU itself or all of the parts of the video card? The RAM and VRMs generate a non-trivial amount of heat too. So I'd wager the GTX 1060 is no hotter than say a Core i7-6700K once you remove those things.

 

EDIT:

As some measure of proof about this, let's take a look at Tom's Hardware's power consumption measurements on the i7-6800K. Their test setup is measuring the 12V CPU power cables using a current probe. Here's their power consumption measurements of the GTX 1060 using the same method (current probing the power lines, including the PCI Express slot).  And according to http://www.anandtech.com/show/9266/amd-hbm-deep-dive/2 , GDDR5 at least on the R9 290X took up 15%-20% of its power.

 

So if we take the GTX 1080's power consumption (110W for the entire card), subtract what VRAM is possibly taking up at the worst case, we end up with about 88W. Not sure what the VRMs, fan, and other house keeping circuitry do so let's just say 80W is what the GPU takes up. This is within 30W of a stock i7-6800K or 15W with an overclock to 4.0GHz (and matching it in the worst case).

 

EDIT 2:

Oh, before I get called out on misunderstanding, I'm saying GPUs aren't as hot as you think they are.