Welcome New Surface Pro


actually the 1600 euro is for the lot…
pen and a keyboard… hence >…see the link: https://www.wehkamp.nl/Zoeken/ArtikelDetail.aspx?ArtikelNummer=877002&Ntt=surface%20pro%20&PI=0&PrI=2&Nrpp=96&Blocks=0
I geuss I can wait till monday…or till mid next week…
Unless I can buy a Eve V by then… I am going for the SP… …


No you can’t buy it next week. The web shop will only open in July… But at least you can know the price :slight_smile:


Does the new SP destroy the reason of EVE to go with a Y CPU? :confused:


V also support microsoft’s dial. Konstantinos showed that at video.


The only two things I like better with the SP5 is the (truly great!) kickstand angle and the new pen. And we might be able to use the tilt sometime in the near future, so really it’s just the one issue.
I am so looking forward to my V, I need it yesterday - uni has become quite a hassle with my 8yr old 7 pound Laptop with its broken battery…


Yeah that’s what I thought… I had already thrown around thisidea, but I wasn’t sure that it would be possible. But now we see other devices doing it and we know that V cooling solution would be almost suitable for this. But Y processor still has advantages: it consumes less energy, resulting in longer battery life, and it produces less heat, which means the body doesn’t have to become as hot.

I actually envision the future of computing as scalable processors that can scale between desktop and phone depending on cooling. Currently we have the ability to change clock speed (turbo boost) and Core M is doing that pretty well. But what if we could also disable physical cores when the processor runs hot?

Of course we would need some new technology that redistributes workload every time a core is enabled/disabled without slowing down multithreaded applications, but I believe that is doable given the clever hyper threading design we already have.

Then, the cooling part - I was thinking about that, blowing air onto the back of a phone isn’t enough to make a quad core x86 processor run in there. If it’s air cooling, it needs to have a heatsink. Then there is a possibility to blow air into the device and leave the vents open otherwise. But air is a very good thermal insulator, so if there’s an air gap between the heat pipe/plate and the body, it would automatically make passive cooling less effective.

But then I remembered that new passive liquid cooling, like in Lumia 950XL. Would it be a good idea to use this solution but also leave taps for the coolant to flow? That way you could connect a pump with big fans to a tiny computer and vastly improve its cooling. I mean liquid cooling solutions handle over 100W overclocked systems easily…

But I’m getting offtopic. Perhaps I should’ve created a separate thread for this…


“Almost” doesnt count. If you say that your computer has 15W CPU, it better has 15W cooling solution, which we dont have.

Otherwise youre basically lying, as for example, the i7 will run at i3 base clock speed after the cooling reaches its limit. Youre selling an i7 laptop that runs as fast as an i3 laptop.

Acer did that exact same thing in their products. The best example is the TimelineX, where the i7-2620M CPU in it, with a base clock of 2.7 GHz on the specs sheet, throttles all the way down to 1.2 GHz after just a couple of minutes, making it slower than even a Core i3 at the time.


Are the U cpus in the new surface rly running at 15w standard or are they under clocked to fit the passive cooling ?


I’m not impressed with the new Surface Pro for many reasons, including no Thunderbolt 3. I like that they stuck with the U series processors and kept Iris, but the lack of moving to USB-C or Thunderbolt 3 is a huge disappointment. Honestly, I believe the Porsche One (Porsche’s take on the Surface Book) will outperform every Microsoft Surface device subjectively and via the benchmarks.

As far as fanless devices, the quality of the device depends on the manufacturer and cooling process. The Skylake Core M version of the HP Elite X2 was impressive as it hit maximum turbo and stayed there. That was the only HP device that was forward thinking with cooling because all of the other devices had cooling issues and were bigger. Acer in its Switch Alpha 12 used liquid cooling and they’re repeating that trick again.

As far as the Eve, I need to see some more benchmarks for the i7-7y75 because benchmarks for this processor are mixed and depend on how the device is cooled. The Eve is subjectively ahead because it offers full TB3 support that other tablets, convertibles and even quad-core devices do not offer. Many devices only have 2 PCIe lanes available for their TB3. Looks like the Eve has access to the 4 PCIe lanes. And honestly, with all the ports it has…an external fan can be added for those who are worried that the internal cooling isn’t enough. USB fans have been shown to drop degrees and help these PCs stay cooler for improved performance.


Yes “almost” counts, it could always be slightly improved :slight_smile: And I doubt SP5 has 15W cooling either - U-series chips never really reach 15W.

If anyone doesn’t believe me and has a 15W CPU, they can install Throttlestop and see for themselves :wink:


You guys quote Lumia 950XL & Acers Switch Alpha 12 as liquid cooled, but that is just a marketing gimmick these companies used. Heat pipes are liquid cooling the same, the difference is only the mechanics of how the fluid flows back to the heat source to get evaporated again / where the fluid condenses. Heat pipes, the mentioned types of “liquid cooling” systems cant be used to connect to an external pump/heat-sink as they include a fluid in vacuum, these are not your regular liquid cooled systems where the fluid is used to transfer the heat to a place where it is cooled down.
If i remember correctly V has this large thick copper plate to absorb heat from cpu etc and a small heat pipe that moves the heat away to the other copper plate that radiates the heat into the body near the battery.
You can see from the pictures of SP4, SP5 - MS uses heavy heat pipes, in SP5 it collects heat from 2 sources & has 2 spreaders + heat sink with fan, the no fan version would probably look different with a different setup of pipes / spreaders.


Here’s a photo of V’s heat pipe:

I didn’t take the heatpipe off as the ends are firmly stuck to those shiny plates and the shiny plates are firmly stuck to the shell.

On closer examination I noticed that the mid section is not attached to the shell and that there is probably some kind of thin insulation underneath the mid section.

This makes sense, as you want to move as much heat away from straight under the processor and distribute the heat on a larger area.

Heat pipe with wick:

Please remember that the photo is of DVT3 prototype. The production heatpipe solution is probably similar, but execution is neater.


Correct me if I’m wrong, but not all heatpipes have fluid inside… Anyway, if there’s a pipe with fluid inside touching the processor, why couldn’t you pump the fluid? While not pumping it would work as that fake liquid cooling gimmick thing or whatever…


Heat pipe is using thermodynamics, the fluid evaporates and absorbs heat, the heat pipe includes a special structure inside that provides a large surface area that then causes the evaporated fluid to go into its fluid state again radiating the thermal energy (similar to if your car front window is dirty inside, water condenses on it to a much larger degree than when clean), the surface area then is using the capilar effect for the fluid to travel back and so on. It is a closed environment with the fluid in vacuum. I especially love how it is a wonderfully simple thing and works in a self sustained mode…
Regular water cooling is heating up the water a couple of degrees and a pump lets the water circulate to somewhere where the water is cooled down.
A heat pipe and what acer implemented is only different in that acer uses some closed loop, so the fluid is not traveling back to the heat source but forward. Hard to even say if it is better when compared to a regular heat pipe, but what is for sure better for them is that they could name it differently (as it is not a pipe anymore) and they refer to it as liquid cooled just for marketing purposes. I remember when heat pipes started to be used in laptops, someone too advertised it in a similar fashion for the wow effect. Heat pipes work similar to the regular fridge - just that there the state change of the fluid is artificial by injecting fluid into a low pressure environment causing it to flash evaporate and so absorb heat (cooling down), when the gas is pressurized and changes its state to fluid it emits the absorbed heat outside of the fridge…


I know how it works… but there’s nothing preventing you from connecting a pump abd making that fluid flow in one direction.


There is actually a big difference. One requires moving parts and one does not. There is a battery drain and a potentially failing part in both circumstances (the heat pipe can potentially leak rendering it virtually useless, I am not sure if anyone yet knows the failure rate due to cracks as the heat across the carbon fiber differs substantially across it’s surfaces - it will expand and shrink as the heat is dissipated, which is effectively how it works, if it is using carbon fiber tubes (aka, graphene), which I doubt, it would be quite stable, however if it is using carbon fiber sheets, the substrate used with the carbon fiber can be subject to leakage over time).


That’s the whole point. You connect an EXTERNAL pump to pump the liquid and turn it into a liquid cooling loop just like in desktops. When you don’t need it, you disconnect the pump and it acts as a simple heat pipe.


One basic difference between a heatpipe and liquid cooling :arrows_clockwise: loop is, that the amount of fluid in a heat pipe is minuscule :droplet: and the liquid cooling loop is filled full :ocean:of fluid.

You probably cannot hear the fluid making any noise, if you have a heat pipe in your hand and you shake it. There really is very little fluid inside a heatpipe. Most of the fluid is contained within the wick and there is hardly any free flowing liquid at all.

So, no you cannot attach a traditional electric driven radial-flow pump, as there is no fluid to pump.

Please remember also that the heatpipes are usually vacuum/near vacuum devices. So very little to pump or move around also in that sense.

It is possible to move fluid with a capillary pump, but that’s another story :rainbow:


Well, that doesn’t depend on how much fluid there is, if there is no air inside… But anyway, I was just shooting out a random idea :slight_smile: what I meant was that while using a pipe designed for liquid cooling, maybe someone could figure out a way to use the same pipe as a heat pipe. I’m not saying it’s doable right now, but IMO it would make sense in the future, which is what I was discussing here.

It’s very simple, just connect a tank of fluid together with the pump lol… A tank is needed either case, because where else would you put the fans?

That’s why the connection wouldn’t be very simple. When disconnecting the pump, it should depressurize the heat pipe to make it work without a fan.


23 posts were split to a new topic: On the virtues of a cloud-computing future

On the virtues of a cloud-connected future