Become a fan of Slashdot on Facebook

 



Forgot your password?
typodupeerror
×

Comment Re:So they want to get less viewers? (Score 1) 307

Is there a way to detect the slowness, to at least tell the users this is happening to them?

Try watching a 4K video. It will stop shortly after you hit play, as though it were buffering (remember that?), but won't actually buffer anything if you hit pause. If you go down to 1080/1440 the video will play. I've also seen instances of the page itself not loading completely: missing images for likes, channel logos, that sort of thing.

Comment Re:So they want to get less viewers? (Score 3, Informative) 307

To be honest I wouldn't mind paying for Youtube, but I refuse to pay the extra for Youtube music. Until they bundled the two together I think it was only €7 per month, which seemed quite reasonable. Now they're going back to cable-style bundling and forcing people to pay extra for something they simply don't want.

Comment Re:Repeat after me: (Score 2) 31

> the jet impingement thing. I assume that it reduces the thickness of the boundary layer

Probably, though the illustration reminded me of a Ranque-Hilsch vortex tube - not sure if that's the heat transfer principle but it sounds that way. After removing the heat from the chip surface, that is.

You could be right, but I was thinking more along the lines of thinning the insulating layer of air between the heat source and the air in the device. I don't think this is the same as a vortex tube; remember that gases act more and more like viscous fluids the smaller you go.

Maybe they can replace their vibrating membranes with MEMS devices witth sub-micron gaps to make it more solid state?

Being "solid-state" is not really a question of scale; it's not even a clearly defined term now that such things as MEMS exist. A transformer in an electrical substation has no moving parts. Is it solid-state? Obviously not. Personally I tend towards thinking about how many parts there are physically interacting with other parts: a quartz oscillator is solid-state, as is a PTC heating element. In any event, I don't foresee anyone etching millions/billions of MEMS mechanisms on a chip to move some air any time soon.

Work wirh BIOS vendors to occasionally kick the MEMS into reverse and eject a poof of nasty dust out the inlet from whence it came at 200 KPH.

Somebody somewhere will start claiming laptops are farting, you just know it.

Now this I can believe. I remember installing Lightsaber.app on a portable Mac with new-fangled accelerometers* and using the say command in Terminal.app via SSH to scare the shit out of people in the next room.

*IIRC the idea was to park the heads on the HDD - this was before SSDs were common - if the device was dropped.

Comment Re:Repeat after me: (Score 1) 31

Peltier cooling has been tried and found to be a non-starter. The problem is that they only move the thing that needs to be cooled a few millimetres; one is still left with the task of actually getting rid of the heat, which falls back to traditional heatsinks + fans. I saw people experimenting with such things - along with actual, honest-to-goodness evaporative coolers, i.e. miniature refrigerators - back in the early days of YouTube. The reason we don't see them now is that heatpipes were so much better.

Peltier devices are frankly terrible in terms of how much heat they can move for the amount of power they need. You can compare this with using mechanical devices (i.e. steam turbines and other heat engines) to extract power from a heat differential vs. using Seebeck devices. With our current technology they just work so much better than solid-state devices. The difference is significant enough for people (NASA I think) to seriously consider using Stirling engines, which have moving parts, over thermoelectric devices on space missions, despite the fact that one can't fix the former if it breaks down and the latter probably never would.

Just an aside... back in the really early days no-one needed to care about hot chips in laptops because they were basically tower cases with a handle attached. A little later on we were still using chips that gave off very little heat (386's didn't need any cooling at all, didn't they? I forget.)
When processing power started to really take off... again, no-one really cared because AT/ATX cases were roomy enough for air cooling and no-one expected desktop performance from a laptop** anyway.

Now we find ourselves in a situation where we're trying to cram a relatively modest thermal load into an increasingly small space. We're still using forced-air cooling because there's really nothing else better that can fit in the space we have.

**It's been said that the word "laptop" fell out of favour among marketing types at Apple because the metal cases often made them too hot to comfortably use as such. My PowerBook G4 could be very hot in the middle where the RAM slots were so I think there's a grain of truth there.

Comment Repeat after me: (Score 4, Informative) 31

Solid-state does not mean "no moving parts". Bad, bad summary writer apparently didn't even watch the video...

This thing basically works in a similar way to piezoelectric aquarium air pumps: there's a membrane that vibrates and some kind of check valve to force air to move in one direction only. What is quite clever is the jet impingement thing. I assume that it reduces the thickness of the boundary layer to allow better heat transfer; I think Sandia labs used the same concept but in a different way and for a different type of cooler.

On the other hand, I'm not convinced by the dust mitigation. This thing seems like it would be very sensitive to dust, which means it needs much better filtration that traditional coolers. The company says there's a filter on the device itself, which is all well and good, but if needs to filter more and smaller dust particles it will clog more quickly (on top of being very small, looking at the video) and I don't see that working too well when you have to take the computer apart to clean it.

Also to go out on a limb here, I have the feeling that dogs and teenagers will absolutely hate these.

Very interesting concept (like those genuinely solid-state dehumidifiers for electrical cabinets) but I suspect it's not going to find many applications in cooling computers. I'm curious if it only works in a small range of atmospheric pressure; though if you take a MacBook to the top of Everest you'd probably be grateful for a little extra heat.

Slashdot Top Deals

If a train station is a place where a train stops, what's a workstation?

Working...