Silent mode

What you think guys?
If cpu in "Silent mode" have max.frequency of 798.3mhz(8x99.8mhz) with voltage 0.846 V and
and in Silent mode off it can reach 2394.8mhz with only 0.951 V that means battery should last longer with Silent mode off?
What does "ECO mode"?

 

ECO Mode is the same as Silent Mode, just with WiFi and Bluetooth disabled as well to save even more power. WiFi can be configured in Settings, whether you want it disabled.

I occasionally use Silent Mode when I am on battery, only doing light work, and want to gain a little more runtime. Under good circumstances, it adds almost an hour to a fully charged battery. Of course that's on my machine, it will be different on others.

Check this post for more info on Samsung Power management, including links to further discussion about Silent Mode. There is also the thread linked below.

http://forum.notebookreview.com/samsung/714633-thoughts-samsung-silent-mode.html

 

Power consumed also depends on the frequency so work done per clock pulse is more efficient at the slower speed.

However, if the CPU takes longer to complete its work then it's also keeping other parts of the system awake when they could be sleeping.

The only way to be sure which is more efficient overall is to create a task that will take about 1 hour to run at the full CPU speed (and longer at the minimum speed) then record the battery charge % before and after running the task under the two CPU operating conditions.

John

 

Ok guys thanks
I am interested now to see with what task cpu go higher then 798.3 mhz
,but logically with silent mode off (wifi always on)task that use higher frequency should be finished about 3 times faster with only 0.100 V more.That means longer battery life.
Also i tried that old "High Contrast Black" theme in personalization ,it seems it use about 1 V less then standard Windows 8 theme (without transparency).I only now have to see can i adjust to "everything black"

 

Hey Dannemand I was so astonished.
I didn't know that when you set cpu on 99% in advance power saving mode you will turn off turbo and cpu will stay at 1.6mhz at 0.8V.That is interesting.What you think why then if I set it to 50% it doesn't react it still goes turbo.
So it seems to me that this is the best option to save battery.To go to 1.6 MHz at 0.8V
What you think what would happened if it stay at 1.6mhz all the time.Is that possible to set up. It will use same amount of voltage as 800mhz but it will finished task faster.
Is there any software that can read SSD and wifi card voltage?

 

That information about using Max CPU=99% to avoid TurboBoost was something I had from another post, long time ago, from the Sandy Bridge days. (By now I completely forget who and where). But John has since performed tests that showed the threshold to be much lower, at least on newer CPUs. See the posts here and here. It is possible that Intel tweaks this behavior with each new generation -- or even with different CPUs of the same generation.

 

Samsung disables TurboBoost when running on battery.

Don't forget that the current (and hence the power) used by the CPU increases with frequency. Otherwise the fan wouldn't come on to remove the extra heat when the CPU is running faster.

I normally use the Samsung Optimized power plan when running on both mains and battery which gives me the maximum non-turbo speed when running on battery (but can't remember whether I had to set this manually). I keep the maximum battery in reserve when I want to squeeze the maximum time from the battery but the difference isn't much.

John

 

Yes you are 100% right John: "but the difference isn't much"
I know if I want longer battery life I need to jump to Haswell or buy 900x4c,this what I am doing now is just a play with 10 more minutes
Where I expect to see improvement at least half of hour is EVO msata ,I still don't know with what capacity will I gain more battery life is it 120 or 250gb.Maybe I take both and sell one after I make tests.But what confused me is that 2.5" basic 840 outperforms 2.5" EVO,and I cant find power test of msata PM841 nowhere.
It would be great if Samsung leaved us possibility to put battery from 940X3G in to 900x3c but they set up motherboards and connections so we don't have such crazy idea

 

May I add that the linux driver (samsung_laptop) does not suffer those restrictions:

Silent mode does not lock the CPU to 800MHz
Running on battery does not lock the CPU turbo boost.

You are free to tweak the power scheme to whatever you consider works best for you. Some more info on fan speeds, temperatures and frequencies in this post

 

Would you suggest to keep ECO mode OFF while gaming, to see some performance improvements? Mine is always on..

 

I would definitely not use ECO mode while gaming, Obyboby. Like Silent Mode, ECO mode caps the CPU speed (regardless of your power profile) thus preventing it from ramping up on demand. In addition to general slowness, this can also lead to the latency spikes discussed in the AB8 Owners Lounge (starting here) because interrupt processing becomes bottlenecked.

 

I wonder why Samsung decided that Silent mode should work differently under Linux than under Windows since the thermal constraints are the same no matter what OS you run. I wish we had the same tweaking options because for me Silent Mode with 800Mhz ist just a tad to slow and trying to make the Samsung Optimized profile silent via setting to passive cooling policy doesn't work either.

 

I am new to this forum and have some question about this so called silent mode. I am owner of a Samsung Notebook NP350e7C its very nice but there s nothing like silent mode.

When i turn on silent mode and eco all together again the fan starts working after 7 minutes and keeps on working at least for 2 minutes.

Even if i dont touch the Notebook, the fan will start every 7 minutes. CPu usage is about 14 % when i dont touch it.

I read nearly the whole forum and this thread is very similar to my one : http://forum.notebookreview.com/samsung/714633-thoughts-samsung-silent-mode.html

hwmonitor shows temperatures from 45-55 c, at 55c fan starts working and after at least 2 (often more) minutes temperature falls to 45, but it doesnt last 7 minutes until it reaches again 55

Its very annoying, to hear every 5 minutes the fan, although cpu is working at minimum speed.

Also the fan is very loud i can hear it from 5 meters or more, it seems fan is louder when not working at full speed, when i stress cpu and it works fastest its even less noisy or equal.

i dont know some are saying this notebook is silent, but also some wrote its loud

How is this silent mode supposed to work? Is it normal that fan starts every 7 minutes in silent mode (even eco on)?

i also installed fresh windows 8.1 on ssd, because i thought its some problem with software, but it seems even worse on fresh installed wn8.1

Fan starts with some howling and you can exactly hear when it speeds up or slows down every time with noise.

When i put my hand on the left side and do some pressure it seems quieter, the case of this notebook is thinner than 1 mm and seems to make noise when fan works.

its the i5 3210 version with 7670 amd

really i cant use it longer than 1,5 hours it makes headache.

So this is normal silent mode?

 

I don't know if Silent Mode disables the AMD GPU. If that is still running then I would expect it to create enough extra heat to cause the fan to work. HWiNFO might be able to show whether the AMD GPU is working and also its temperature.

While a CPU temperature might be enough to trigger the fan to start operating, it should only be running slowly and be difficult to hear. If pressure on the case changes the fan noise then perhaps it is not fixed properly.

Is your BIOS up-to-date? You can check using the BIOS Update program (link in first post here).

John

 

I saw hwmonitor pictures from other user and thought mine is different. I dont know why but it doesnt show the integrated hd4000 amd also not the amd graphics.

Only uncore 1.00 Watt, i dont knew what it is?

Yes i updated Bios, but i am not sure if this was when the problem began.

 

Did Samsung write the drivers for Linux? I thought they came from the community. I wonder if Samsung even published their firmware APIs for these features. It could just be Fn-keys implemented in the Linux keyboard driver (or a filter) with Fn-F11 switching power profile -- which is not the same as actually invoking Samsung's Silent Mode firmware feature. In any case do I agree that it is preferable to avoid firmware capping the CPU.

In Windows one can tweak the Power Saver profile to still get the same effect: Keeping Passive System Cooling policy and capping CPU speed at whatever desired level (say 50% or 75%). But you would have to manually switch profile, since Fn-F11 will invoke the actual Silent Mode (with the lower CPU cap).

 

Quite contradictory to:

 

No, the samsung_laptop driver provides a 'performance_level' switch which can be set to 'silent'. This has nothing to do with Fn keys or power profiles. It's not tied to anything other than fans.
Assuming the fan management is controlled by the firmware - only offering this mode switch - how could it behave any different throughout OSes?

Couldn't it rather be the other way around, with Samsung Settings stepping in and locking the CPU when activating silent mode / running on battery?

 

I see, thank you. Did Samsung provide that driver? Is it possible that it simply switches to the equivalent of Passive System Cooling in Windows power profiles (prioritizing lower CPU speed to stay cool instead of increasing fan speed, but not actually capping CPU speed unless temps demand it)?

I don't think Settings could throttle the CPU without firmware support (though I couldn't say for sure). And indeed, I remember back when I tested this that the Windows System log showed events to the effect that "CPU speed restricted by firmware" (I forget the exact wording). Admittedly, that was back in the days of Easy Settings (Win7) and I wasn't able to reproduce those events just now with Settings (Win8).

Either way would I prefer if Silent Mode relied entirely on the power profile to control fans and temps, so that one could adjust how much to cap the CPU (if anything). As I posted earlier, manually switching to the Power Saver profile will accomplish that -- but not with that convenient Fn-F11 button.

 

In my experience setting the cooling policy to passive does absolutely nothing.
But i now found an acceptable workaround: I use silent mode with Throttlestop 6 and set the clock to 12 x 100 Mhz. This is fast enough to browse the web and scroll documents and still doesn't go over the 55C threshold where the fans turn on in silent mode.

 

Certainly for browsing the web, regular Silent Mode should do. My old NP700Z3A (i5-2450 2.45Mz) caps at 800MHz in Silent Mode, and when using Chrome, I hardly feel much difference -- even with my charting platform running in the background. Loading large programs or performing computationally intensive tasks is where the cap becomes evident. I have another platform written in Java, and that one definitely becomes slower in Silent Mode.

In my experience Passive System Cooling policy does what it claims to do: it slows the CPU down before increasing fan speed. But if temps are still creeping up, it has to run the fan regardless -- it cannot just let the CPU overheat. I assume that's why Samsung chose to cap the CPU as well, to make sure that Silent Mode really is silent.

Are you saying that Throttlestop overrides Silent Mode's CPU cap to allow a higher CPU speeds during Silent Mode? That's an interesting idea?

 

Yes, that's what Throttlestop does, i now have a completely silent system which doesn't feel laggy. I use Firefox and my machine is the 900X3E with i5-3337U, maybe things are a little different with an ULV processor.

 

I just hadn't heard about Throttlestop used in the context of overriding Silent Mode before. That practically proves that Silent Mode is implemented on the firmware/CPU level (at least in Windows), not on a software-triggered level, otherwise I don't see how Throttlestop could work. Thank you for that.

 

Why?
On linux you have the intel_pstate driver to control the frequencies. I guess Windows is also able to control the CPU w/o Settings installed!?

But even if the CPU state in this case was controlled by the Samsung firmware, couldn't it also just be a non-depended firmware function with Settings switching it according to the silent mode state?

No, silent mode does neither relate to CPU frequencies nor to temperatures!

I can fully stress the CPU in turbo mode w/o any throttling occurrence while in silent mode - even on battery . With lighter usage / dropping temperatures the fans then switch off automatically one after the other. So there is no reason for me to ever leave silent mode.
Of course fan speeds are set in the firmware themselves, but they are lower for silent mode. Some observations about silent mode's temperature thresholds.

samsung_laptop further provides access to radio switches, BLE, USB charging & keyboard backlight. Those things all need to be accessed via firmware imho, so I don't see why silent mode isn't.

 

This threshold is the same for silent mode under linux. That makes me want to believe silent mode is the same regardless of the OS. Only difference is that Settings under Windows additionally locks the CPU (which Throttlestop is able to override - by accessing the CPU, not Samsung firmware)

Can you also unlock the turbo when on battery with Throttlestop?

 

If CPU speed capping were to be implemented solely in the Settings software -- independent of the power profile, as IS the case in Windows -- that means Settings would have to monitor and trap any events that could trigger a change in CPU state. If the user changes to a different power profile, Settings would have to trap that and override the Max processor state setting in the new profile. That's not a very likely implementation, if you ask me.

Settings (and before that Easy Settings) implements most of its hardware features through SABI (Samsung Advanced BIOS Interface) which passes calls to BIOS. (This is generally not good form, and probably one reason why Samsung has had problems keeping up with drivers in Win8 and 8.1.) Settings implements SABI as a DLL, whereas in Easy Settings it was a kernel driver -- which has much more liberal hardware access. So I am not ruling out that some features which were implemented through firmware (BIOS calls) in Easy Settings could be implemented as software i Settings. As I posted earlier, I was indeed unable to reproduce the "CPU speed limited by firmware" system events in Win8 that occur in Win7.

If someone had the time and patience to find and study the SABI interface (you can Google it) I would be curious if it has a Silent Mode call or state. If it doesn't, that means Silent Mode is whatever a given programmer decides it should be. Whoever wrote Settings (and before that Easy Settings) had their idea of Silent Mode (different implementations, but same net effect); whoever wrote the samsung_laptop driver for Linux had a different idea; heck, I could write a "hello world" program and call it Silent Mode.

Again, I like this implementation better than the one in Easy Settings and Settings. But the mere fact that its effect is different from the two Windows implementations (again Settings and Easy Settings are implemented differently, but arrive at the same effect) made me wonder if someone didn't just add a (really nice) feature in samsung_laptop and called it Silent Mode. I don't see any problem with that, particularly since it sounds more useful than Samsung's implementation.

Let's leave it there, I was just curious about that Linux implementation, since I don't use it myself.

 

Yup, with the multiplier high enough i can turbo up to 2.7 Ghz when on battery. It really seems that there is a Samsung software doing all the CPU limitations in Windows and silent mode itself only turns the fan off and sets the fan limit to 55C. Making an 'unlocked' silent mode my standard setting is really a tempting idea.

 

I would conclude the exact opposite: If the CPU speed cap in Windows is implemented purely by Settings software trapping and overriding CPU speed changes (including changes to the power profile, which have no effect on the cap) how would Throttlestop be able to undo that? It would have to "eliminate" Settings or steal its hooks and undo every time Settings overrides. (And again, I don't believe Settings works like that). But if Throttlestop simply pulls some of the same firmware levers as Settings (or the CPU directly) that would explain how it can override the cap -- which makes much more sense to me.

OK, once again, I'll let this go. Feel free to conclude otherwise. We know the resultant behavior in all three OSs (Win7, Win8 and Linux) which is what's important.

 

Well, we both agree silent mode is implemented in the firmware. But if this silent mode firmware implementation was also controlling the CPU then neither samsung_laptop nor throttlestop could circumvent this w/o some dirty firmware hacks - which I doubt any of this programs have that knowledge.

Whether or not the firmware has a CPU frequency implementation (other than a plain thermal one) is not the point here. I just don't see silent mode itself controlling it, especially after hearing Matthias description matching my own observations..

 

Fan ON thresholds should be: 55°C (1st fan) / 64°C (2nd fan)
Fan OFF thresholds should be: 55°C (2nd fan) / 47°C (1st fan)

Having a non-locked (including GPU freq) silent mode there is no reason for any other mode, as those just let the fans spin (with even higher rpm) for no reason.
Furmark will find out about the GPU.

 

Settings might be successful locking CPU access from Windows power profiles. But then again throttlestop might just take over the CPU and doesn't let loose.

So does samsung_laptop:

Code:

samsung_laptop: detected SABI interface: SwSmi@

Whoever wrote all these programs, they might not have any idea about silent mode except using the interface provided by the firmware..
Even if you have a program with the magic "hello world" name you can't just easily adjust the firmware's fan control to your liking

 

@oled:

Silent Mode's CPU cap in Windows (Settings and Easy Settings) is NOT implemented through power profiles. That is what I have been saying. On the contrary, it is completely independent of power profiles. You can can change and tweak the profiles all you want, it won't affect the CPU cap. The fan policy, OTOH, IS implemented through power profiles.

Once again, I have no experience with this in Linux; but that is how it works in Windows -- based on fairly extensive tests (admittedly most of them back in the days of Win7/Easy Settings).

 

I also have some question about silent mode.
my Notebook is series 3 350e7c with i3210m

If i understood right you have to install Samsung Settings Software to get Samsung power profiles and silent mode.

There are 3 modes in Settings for silent mode , Auto, low and out.

Who knows the difference between Auto and low? I didnt find any place where this is described.

For me Setting silent mode to Auto or low from Samsung Settings Software or fn11 only changes actual power Profile to energy saver Profile.

So i dont believe theres anything in the Background or Firmware, ist only powersetting Profile.

If this is different on your Samsung please let me know.

now my question, how does your Notebook act in silent mode? how Long is the fan off?

it would be interesting to hear other owners with i3210 processor.

i mean 10% cpu usage is not much, but if i have 10%cpu usage my Notebook fan starts in silent mode at least every 10 minutes for 2,5 minutes.

even when i dont touch the Notebook and cpu usage is at 2-3% the fan will start every 15 minutes...

is your i3210 Notebook really silent in silent mode?

 

My understanding is that the two options behave thus: "Auto" turns the fan on and off as needed while "low" keeps the fan running continuously at low speed which can be less intrusive than the fan starting and stopping. "Off" means Silent Mode is disabled.

The main way that Silent Mode enables a quieter computer is to slow down the CPU so it uses less power and produces less heat. Depending on the CPU power rating, the minimum CPU speed may not be enough to prevent the fan from running. If you mean you have the i3-3210 CPU then it is a 55W part and is likely to use more power on idle than one of the lower rated CPUs.

John

 

Yes, it would make sence, if it worked like that.

I was asking myself why samsung didnt make the fan work at a low speed all the time, so it could work qieter.

Instead cpu temperature goes up to 55 c and fan starts working for 2,5 minutes until cooling down to 45c and then in max 10 minutes again at 55 c fan starts working, (and this in silent mode!)

my cpu is i5 3210m

Bios is newest, also settings downloaded via swupdate and installed....

but it isnt working as it supposed to be, auto and low changes only profile to powersafe......

I bought this notebook with i5 3210m because i thought when i set it to lowest frequency possible it would be still more powerful than my 5 years old athlon notebook, and the fan wouldnt turn on ever..... but seems my thoghts were wrong, i even cannot surf internet or listen internet radio without every 10 minutes loud fan.... only 10% cpu usage and this fan goes mad, i cant believe its normal?

Noone here can tell me how often the fan of his notebook works in silent mode? Especially with i5 3210

 

There is no doubt that Silent Mode and other aspects of Samsung's proprietary approach to power management could be more intuitive and/or work better. But knowing what it does and how it works makes it a lot easier to get it working for you instead of against you.

On my model (an older Series 7) I've found it to work quite well, including the infamous Silent Mode -- but only after I stopped fighting it and began using it the way it was intended, with my own tweaks of Samsung's power profiles.

The post here and the links within it try to sum it up. Again, that's on my computer and for my uses. Other models and other users may need other tweaks.

 

This is exactly what i am trying to do here, i want to know how this system works and what is ok and what is maybe a disfunction.

Unfortunetely i dont have anyone around here with samsung notebook, so i have to ask users in this forum.

I have read many topics and will read also this one, but i am not sure if i should send this notebook for repair because of noisy fan.

 

I understand. What I meant was Samsung's way of doing power management (at least in Windows) works differently from what we're all used to on other computers. So the first thing I recommend is always to make sure members understand how it was meant to work on Samsung's part (which is described in that post and the links within it).

If it doesn't behave as designed, certainly there is reason to look into whether bad, worn or defective fans are an issue: Bad or worn fans are notoriously noisy. And if one fan is not running on a two fan system, the other will run much faster.

 

Hey guys,

i finally decided to register, after reading quietely for several months. Thank you for all the great support offered practically in every thread.

I previously owned a 900X3E, and i liked it really good (well?). I loved the keyboard (short travel, keys tigfhtly fit into the case, great feedback), and the screen was beautiful, if you disregard the light bleed every LED-type model suffers from.

The one thing that kept me from being perfectly content, and returning it, was the noise.

As far as I understand, the cooling system is in fact one of the best currently to be found in ultrabooks. However, due to poor driver management on Samsungs side, the left of the two fans ran continuously. In fact, due to its small form factor, it only sat a few inch above the table. The air vented out by the fan would somehow not disspate, but stick under the Notebook, causing even more heat etc. Consequently, the left fan would kick in even in idle, and never turn off again.

That was my deal breaker: I spend lots of time in a library, and in my room there is often absolute silence. I could make out the fan running softly, it was never truly silent. Even my old Macbook 2,1 from 2007 had been completely silent with no fan, even though it roasted the table it sat on with its heat-emmission.

When i last checked the 940X3G (touchscreen one), Samsung updated its drivers: In fact, the fan would stay off most of the time (Office, simple browsing etc.). This is probably owed to the lower power consumption of the Haswell-Series (same conclusion: Notebookcheck, german article). Unfortunately, glossy touchscreen .. ugh.

Seeing as the Series 9 was in fact the best Notebook i shortly owned, I decided to go with the 900X3G. I then stumbled upon this thread.

Finally, ages later, my question:
I want my 900X3G to run fanless most of the time (ie silent mode), but the 800Mhz cap is too low to work comfortably. The linux driver doesnt cap at all.
It has been suggested to use Throttlestop 6 to circumvent the throttling.

so: What do i need to set the multiplier to in order to stop all capping? Will the Notebook run silently until reaching the 55 degrees C as stated previously by a user, while being able to explot all CPO-power - ie a permanent silent mode, as with the linux driver? Is there any chance of thermaly damaging the components?

Sorry for the ridiculously long post. Keep up the good work!

 

Perhaps the easiest way to achieve your objective would be to customise a power plan to set the maximum CPU state % (under processor power management) at a value that keeps the temperature below the fan turn on setting.

I was experimenting with this a few days ago and it appeared (on my NP900X4C) that anything above 78% resulted in the CPU running at up to full speed. 78% and below reduced the maximum speed and the lower the % then the lower the maximum speed. You can quickly test this out by running HWiNFO which shows the CPU speed (and the Sensors window shows the temperatures) and then run wPrime 1024M (perhaps with only one thread to represent a less severe CPU loading condition). Then open the power plan and change the max CPU state setting. It gets applied the moment you press apply so you can immediately see the effect on the CPU speed. Once you have found what you think is an acceptable speed (eg 1600MHz) then leave wPrime running until the temperature stabilises and see what the fan activity is like.

Another possible heat mitigation measure would be to put a pair of bigger rubber pads at the back of the notebook to improve the airflow. The hot air should be blown out of the back but it helps if the cooler air can easily flow into the vents on the bottom.

John

 

For the Ivy Bridge i5 the highest multiplier is 27 T ( T stands for turbo) and Throttlestop doesn't let you set anything higher than what your processor can handle anyways. Note that if you run under full load with turbo activated the CPU will hit the 55C treshold very quickly even with the Haswell generation. But no, you will not damage anything since 55° is ridiculously low compared to the 100° that is usually the point when the CPU automatically shuts down to prevent damage.

If your workload consists just of browsing the web and office programms then you will be able to get a completely silent machine. But if you do something CPU intensive like video streaming or virtual machines then the fans will come on after a few minutes. Personally i use a multiplier of 14 for the "enhanced silent mode".

 

Sure, that goes without saying. I just wanted to verify that Throttestop only removes the cap, whereas keeping the higher fan threshold of silent mode,

However, why do you use a multiplier of 14 instead of the max one? If Throttlestop works as described, ie remove the cap, then there should not be a benefit of running a lower multiplier than max. You would never have to leave silent mode, after you enabled it once.

@John: Thank you, but reducing the potential CPU power is what i am trying not to do. Sure, you can simply manually switch between plans, but i am trying to automatically get the needed processing power (up to max) and having at the same time the lowest noise level possible

 

Actually, a given Max Processor State setting combined with Passive System cooling Policy in the power plan should accomplish just that. But it is quite possible that Silent Mode invokes a different set of fan rules in the DSDT, so that using it in combination with un-throttling the CPU is a more ideal solution for those who want full CPU potential with minimum fan.

Note: For any casual passersby who sees this, it is important to point out that there is no secret trick that can safely keep fans completely off while continuing to run the CPU at full speed, regardless of temperature. Computer vendors don't put fans in laptops just to annoy us, they are there to prevent damage to components (not just the CPU). But tweaking the fixed rules implemented by vendors can be fine as long as you know what you are doing.

I am sorry guys, but seeing some of the questions we receive from time to time, I just felt that bit of disclaimer was in order

 

I owned the 900X3E briefly. I guess the 900X3C wasnt, and the 900X3G will not be different. Changing the cooling policy to passive did absolutely nothing, which i tried for several days before returning it.

I understand the disclaimer . However, this is more about the ridiculously low "fan-on" threshold when NOT being in silent mode. Notebooks far worse or potent (ie more heat intensive) manage to stay silent during idle. It never got to me why not the Series 9, with its 4 GB RAM, average CPU, on-board graphics and good heatpipe system.

What I am trying to do is raise the fan threshold to the limit employed by the Silent mode, not completely disable the fans As you said, there have to be fans

 

I understand completely. I think combining Silent Mode with Throttlestop (as suggested by Matthias_K) may be the way forward. But I am afraid I cannot help with details of that.

Puzzlingly, my NP700Z3A (which uses far more juice than yours) remains silent during almost all normal use. For me Silent Mode is nothing more than a quick shortcut to enable a power frugal mode for extra battery time, since fan noise was never an issue except during very heavy operation.

 

With such a thin chassis Samsung's designers were stuck between a rock and a hard place: The thin chassis increases the risk of heat transfer through the metal base and unacceptably hot surface temperatures while the thin fan(s) needed to fit in the thin chassis have to work harder to shift much air.

I've got an old Fujitsu notebook on which part of the bottom was covered in felt. This was their workaround for the problem of the heat transfer - the felt was over the CPU. However, the fan noise still makes the Series 9 sound quiet.

No manufacturer seems to have yet come up with what I think would be an effective solution: build a flexible heat pipe that shifts heat up into the display which can act as a big passive radiator.

John

 

Clearly the slim size of the Series 9 compared to my bulkier Series 7 is a big difference. I hadn't thought about the fact that the fans themselves must the thinner too, thus having to work harder to move the same amount of air.

OTOH the NP700Z3A and NP700Z3C are fairly packed inside with both HDD and ODD in a 13-inch chassis, and regular (non-ULV) CPUs, yet I don't remember complaints about fan noise -- while I DO remember complaints from owners of the roomier 15.6 inch models, including the 2013 models that don't have ODD.

Regarding the heatpipe going into the display lid: Wouldn't that risk damaging the LCD? I was worried about that already with all the hot air flowing up along the display, particularly on my older HP HDX18.

 

I don't think those settings can achive the lower fan speeds that silent mode provides.

This is is what silent mode is all about.

As long as the temperature stays below 55°C the fans stay off regardless of the CPU frequency.
If temps go beyond that threshold the fan(s) still spin slower than in normal mode, but with sufficient force to cope with even the turbo on full load (tested with 3GHz Core-i7).

It is very unfortunate that Samsung caps the CPU on Windows. But I guess that's what they had to do to guarantee a "true silent" system. Anyway throttlestop can overwrite that cap.

 

I agree, they cannot. If you read the post I was responding to and the full paragraph you quoted, you will see that was indeed the essence of my response.

 

Sorry, I didn't read the 'should', I guess I was blinded by the bold characters