English Community

ThinkPad NotebooksThinkPad: P and W Series Mobile Workstations
All Forum Topics
Options

244 Posts

10-21-2012

Switzerland

292 Signins

1760 Page Views

  • Posts: 244
  • Registered: ‎10-21-2012
  • Location: Switzerland
  • Views: 1760
  • Message 71 of 319

Re: W530 fan noise intolerable

2012-10-23, 7:04 AM

"For me, the rotational noise of my Scorpio Black drowns out the sound of the fan up to 3000RPM."

 

While often mentioned as the primary noise source, HDD's (white or whine) noise can be easily mitigated to practically silent:

 

  1. Use SSD as primary (mostly read-only) OS partition + a silent smaller HDD (7200 or 5400 rpm) for data/programs;
  2. Use a RAMdisk/cache as HDD scratchpad, and set the main HDD to shutdown in 30s (easier in Win than Linux);
  3. Avoid external HDDs. Use only internal silent HDDs, with 'pleasant' white noise emissions - I had the luxury of selecting some of the nearly silent old 100GB Travelstars. However, many 2.5" HDDs marketed as "silent" have uncomfortable noise emissions.

 

However, all these above are finesse refinenments to a mostly passive laptop, like the T61p undocked (this one Lenovo got right, except the dock).

 

OTOH the w530 emissions and energy waste seems a rather gross TPD issue, of having raised the bottom floor by 50% vs. the w520. We're not talking here about the P0-1 and TurboBoost upper range, where the w530 actually does a better job than the w520. Here they (Intel and Lenovo) have shaved ca. 20W, while also increasing performance by ca. 15%.

 

As with cars, while it's nice to have a margin up to 6-7K RPM, the vast majority of our driving is done in the bottom RPM range, where the car makers have spent utmost resources to optimize this NORMAL (low-medium load) operation point parameters -  also valid for the Deutsche Autobahn-designed Audi, BMW, Mercedes, VW.

 

Lenovo seems to have forgotten the normal usage patterns of a premium full size mobile workstation (laptop!). Or they may think that the w530 will be running the big rendering/transcoding jobs most of the day, hence no use in optimizing the 600-1200MHz regime (disabled). Wrong design assumption or temporary lapse of reason?

 

Either way, it requires immediate corrective action, whether by BIOS patches (likely partial) or recall for fan replacement and regooping w/ better thermal paste. Anything else at this price point would be unacceptable IMO; i stand by my words and use my real name here. Mitch

Reply
Options

244 Posts

10-21-2012

Switzerland

292 Signins

1760 Page Views

  • Posts: 244
  • Registered: ‎10-21-2012
  • Location: Switzerland
  • Views: 1760
  • Message 72 of 319

Re: W530 fan noise intolerable

2012-10-25, 18:29 PM

Today i got the top end of both w520 and w530, w/ 8 and 16GB DRMA resp. Hence not fair, but next i'll remove 8GB from the 530.

 

1st observations after 6 hrs of undocked operation w/o external monitor (nvidia disabled) at 23C office, similar location/ventilation:

 

  1. indeed, by default the x30 always spins at 2750rpm, even if 34-50C.
  2. is this noise more turbulent than the x20? No. When both fans are set at 2700rpm, their spectral content is identical (by ear, since i don't have the B&K mikes handy, nor an anechoic room). Hence no major SNAFU in the air path of x30.
  3. during light load (VLC radio + Chrome easy browsing, no flash or heavy WWW), the x30 hovers around 56C, while the x20 around 48C (TPFctrl monitor). Few degrees i attribute to the extra 8GB DRAM + 1.2GHz CPU (vs. the 0.8GHz x520).. TBD
  4. x30s screen is 3-4 steps brighter than the x20 (which i prefer in dark conditions, easier on the eye, deeper blacks)... Creates the illusion of more contrast for the x30, which is contrary to the reality. Both are the hi-end 1920 panels, w/ clearly stronger backlight for the x30 (1-2W extra).
  5. Both machines can be set to "passive" cooling, which keeps them silent up to ca. 60C in TPFC Smart mode. Then it's just the HDD noise needing attention. However, the x30 heats up at a higher slope (1.5x faster) than the x20, under similarly light load. I like the steadier x20, always cooler by a wide margin...
  6. Net: A BIOS patch to bring the x30 fan in the 1500-1700rpm would render its noise equal to the x20. Ideally, both machines should go lower in energy-saving states, closer to the 600MHz and some degree of undervolting.
  7. More in the next days. If you have an x20, there's little reason to wish a change to fewer ports, arguably worse keyboard and less screen contrast (but way brighter) - in return for more nvidia cores and 18W less shaved from the top (Bravo Lenovo, this is a good thing!). Mitch
Reply
Options

22 Posts

10-19-2012

Switzerland

114 Signins

886 Page Views

  • Posts: 22
  • Registered: ‎10-19-2012
  • Location: Switzerland
  • Views: 886
  • Message 73 of 319

Re: W530 fan noise intolerable

2012-10-26, 11:10 AM

Thank you for this great analysis, Mitch. Kudos for sharing this with us.

I still have some questions, and perhaps you can help.

  

1. How would you characterize the difference between the fan noise of the W520/W530 at 2750 rpm and the W520 at the near or below 2000 rpm level? Does the fan noise at 2750 rpm contain a high pitched whining sound, and, if yes, does the near 2000 rpm level significantly reduce the high frequencies or perhaps even eliminate them completely? As I tried to describe earlier, it's mostly the high frequencies in my W530's fan noise that bother me, and that - even though they are not very loud - add to my overall impression of listening to a jet engine, rather than a ventilation fan.

 

2. You write that "during light load (VLC radio + Chrome easy browsing, no flash or heavy WWW)" the W520 hovers "around 48C". But does this also mean that the W520 stays near or below the 2000 rpm level during these tasks?

 

I'm wondering because harrisb (message 53) wrote that "the W520 fan speed using the BIOS setting is 2730 rpm according to TPFcontrol", and that he only achieved a lower rpm speed by either setting TPFancontrol to Fan 1 (yielding 2140 rpm), or by using smart mode (yielding 1730 rpm). It's not clear what programs harrisb was running when testing the BIOS setting. But if it was just easy browsing, or typing in word for, example, his results would suggest that, even if Lenovo were to introduce lower fan speed options for the W530 one would still have to use TPfancontrol to enforce them on such easy tasks.

Reply
Options

2366 Posts

04-20-2008

United States of America

19439 Signins

110581 Page Views

  • Posts: 2366
  • Registered: ‎04-20-2008
  • Location: United States of America
  • Views: 110581
  • Message 74 of 319

Re: W530 fan noise intolerable

2012-10-26, 12:37 PM

floppy wrote:

 

I'm wondering because harrisb (message 53) wrote that "the W520 fan speed using the BIOS setting is 2730 rpm according to TPFcontrol", and that he only achieved a lower rpm speed by either setting TPFancontrol to Fan 1 (yielding 2140 rpm), or by using smart mode (yielding 1730 rpm). It's not clear what programs harrisb was running when testing the BIOS setting. But if it was just easy browsing, or typing in word for, example, his results would suggest that, even if Lenovo were to introduce lower fan speed options for the W530 one would still have to use TPfancontrol to enforce them on such easy tasks.


The programs that I was running were browsing, e-mail, etc. I stated that the CPU temps were around 41-42 degrees C and that the fan was running at a reported 2730 according to TPF. The fan is not set to run passive. I could barely hear the difference between 2730 and 2140 rpm when sitting at my desk using the systems. The difference was about 2db which is almost imperceptible (3db is the threshold to normally detect sound levels). I suspect that there is a difference in the noise spectrum created by the fan in the W530 that is yielding the difference in perceived sound levels. I do not possess a spectrum analyzer to determine the precise noise pattern. In any event, the noise of the standard HDD in the primary drive bay more than drowns out the fan noise if it is running.


P53 XEON 2276 BIOS 1.25 OLED UHD 4K Multi-touch display, 96GB RAM, RTX5000, RAID 1, 1TB x 2 Samsung PM981, 1TB PM981, WWAN
P70 XEON 1505, BIOS 2.35, 4k Display, 64GB, M3000M NVIDIA GPU, RAID1 1TB Samsung PM981 PCIe-NVMe SSD x 2, 2x Samsung 850 Pro 1TB SSD. EM7455 WWAN
P1. BIOS 1.25,Xeon Processor, 4k UHD IPS multi-touch display, 32GB, PM981 1TB x 2, RAID1

I am not an employee nor an agent of Lenovo.
Reply
Options

22 Posts

10-19-2012

Switzerland

114 Signins

886 Page Views

  • Posts: 22
  • Registered: ‎10-19-2012
  • Location: Switzerland
  • Views: 886
  • Message 75 of 319

Re: W530 fan noise intolerable

2012-10-26, 13:11 PM

Thanks for elaborating harrisb. I also suspect that it has far more to do with the spectrum
than the db. In my case I would say that the fan noise clearly stands out from
the sound of the hdd, and that this is mainly due to the high frequencies.

Reply
Options

63 Posts

10-01-2012

Germany

269 Signins

990 Page Views

  • Posts: 63
  • Registered: ‎10-01-2012
  • Location: Germany
  • Views: 990
  • Message 76 of 319

Re: W530 fan noise intolerable

2012-10-26, 14:58 PM

Hello,

 

my T61p always stays at the first, nearly unhearable fan level (the HDD's short "clacks" are much louder) - only after some minutes of a continuous CPU load of nearly 100%, the fan gets a little bit louder, but still much more quiet than the first fan level at the T/W530.

 

Whereas the T530, which seems to have exactly the same fan as the W530 (I compared both noises in a Lenovo shop) with the only difference that at the T530, the fan is not always running, but only with light (or more) CPU load. The fan noise at the T530 had much higher frequencies than the T61p, that's why it was much more bothering (as you can hear on my uploaded MP3 files further above in this thread).

_______________________________________________________________________________
T460s 20F9CTO1WW (i7-6600U, 20 GB, 512 GB SSD NVMe, WQHD [LG], Win 10 Pro, Ultradock)
W530 N1K59GE (i7-3820QM, 8 GB, 240 GB SSD, 1 TB HDD, FHD, Win 10 Pro, Minidock Plus 3)
T61p NH0BQGE/UK (T9500, 4 GB, 200 GB FDE, WUXGA, UMTS, Vista Ultimate 32-Bit, Adv. Minidock)
Reply
Options

22 Posts

10-19-2012

Switzerland

114 Signins

886 Page Views

  • Posts: 22
  • Registered: ‎10-19-2012
  • Location: Switzerland
  • Views: 886
  • Message 77 of 319

Re: W530 fan noise intolerable

2012-10-26, 16:45 PM

Hm, doesn't  sound very promising that you got the same impression when listening to the W530 and T530 in a shop, dark_rider.

 

However, I am still hoping that there are only some machines affected by the high frequency problem. The reason for this is that one can find several statements from users that say that the T530 is very quiet. Post number 245 in this thread is an example

 

I also find this statement from post number 374 interesting: The user has two T530s, one with NVIDIA and one without, and while the one with NVIDIA is said to be quiet the one without is said to be noisy with the fan always on. One theory in that forum seems to be that the noise might be connected to having an i7 quad-core CPU instead of an i7 dual-core. However, the user from the first link who claims that her machine is dead quiet seems to have a configuration with quad core and NVIDIA.

 

Regarding the W530 I found this review that claims that, on what seems to be very light load, the W530 is very quiet.

http://blogs.technet.com/b/keithcombs/archive/2012/07/05/nearly-12-hours-on-battery-with-the-lenovo-thinkpad-w530.aspx

 

This matches with my experience that when you do not anything on the W530 or just launch Paint, for example, the fan stays off, at least when the room temperature is average or low. Even light browsing, however, can cause my fan to kick in.

 

(No way for me, by the way, to return my W530 to the vendor and see if a solution for the noise problem can be found. Unlike in Germany there is no legal obligation in Switzerland for online vendors to take back items for unspecified reasons).

Reply
Options

12 Posts

10-11-2012

Germany

30 Signins

220 Page Views

  • Posts: 12
  • Registered: ‎10-11-2012
  • Location: Germany
  • Views: 220
  • Message 78 of 319

Re: W530 fan noise intolerable

2012-10-26, 19:32 PM

(No way for me, by the way, to return my W530 to the vendor and see if a solution for the noise problem can be found. Unlike in Germany there is no legal obligation in Switzerland for online vendors to take back items for unspecified reasons).


If you bug the machine as individual in germany online you return it within 2 weeks, but not when you buy it as a business or company. This is why I am really interested at a solution here, or figure out which configurations make the noise. I really need new machines for my business, but when I buy them I cannot return them. We want to stay with Thinkpads, but they must be silent.

Reply
Options

244 Posts

10-21-2012

Switzerland

292 Signins

1760 Page Views

  • Posts: 244
  • Registered: ‎10-21-2012
  • Location: Switzerland
  • Views: 1760
  • Message 79 of 319

Re: W530 fan noise intolerable

2012-10-26, 20:59 PM

Part 2. More details

w520, aka x20 here, Model 4282-23A (from memory, not sure)

w530, aka x30 here, Model 2449-A22

 

  1. Removing 8 of the 16GB DRAM from the x30 has reduced the operating temps to precisely the same as x20's: Now under light load, both hover 39-46C (undocked, placed on metal trays for passive cooling, ambient 23.5C). Removing the CD spindle and the battery, cuts another 1-2C (Sorry for the metric, it's been years since i don't have to convert C/F anymore). This mode is fanless iff TPFC is installed and active in Smart mode (worth a donation!).
  2. x30: On max power savings and passive cooling in Lenovo's PM, the fan 'always' spins @ 2750rpm. Noisy enough to drown any other nearby noise source, incl. HDD.
  3. 'always': Down to 31C, then it stops.
  4. Restarts at 39C, w/ an 8C hysteresis.
  5. Practically, in offices and home w/ ambient around 22C, the fan on BIOS will keep spinning @ 2750.
  6. This is indeed not a pleasant white noise. It contains whine-like harmonics, annoying in quiet offices.
  7. x20: Is spectrally identical @ 2750. However, this is rarely heard, because it  rarely hits 2750: Most of the time it spins @ 1800, quieter than most cheap HDDs.
  8. Both systems under medium load and BIOS control will quickly rampup the fan above 3-4K rpm, when i don't bother measuring anymore: A heavily used machine of this type is NOT quiet. Better get a tower PC w/ silent cooling if you spend hours daily at high load. Or launch your sims, renders, big jobs after hours.
  9. x30, undocked, w/ TPFC in Smart mode remains fanless up to 60C, a nearly 20C of silent range... that's for all light jobs, 98% of the time. Then you'll want SSD and really silent HDDs (another topic).

"How would you characterize the difference between the fan noise of the W520/W530 at 2750 rpm and the W520 at the near or below 2000 rpm level?"

 

Without TPFC in a typical 22C ambient, but idle or lightly loaded, the x30 spins by default @ 2750 (bad), while the x20 @ 1800 (acceptable, if the HDD is not SSD). Huge difference in perception, and not measurable by dB w/o spectral analyser.

 

With TPFC i manually controlled both systems at various speeds above 2700rpm: They're identical to fine instruments (my ears :-)).

 

"Does the fan noise at 2750 rpm contain a high pitched whining sound,": YES. Not acceptable for longer than a few minutes (in quiet places and libraries).

 

"and, if yes, does the near 2000 rpm level significantly reduce the high frequencies": YES.

 

 

"or perhaps even eliminate them completely?": NO. Must reduce under 1700, likely somewhere in the 1200-1500rpm. However, i can't measure these modes on the x30 unless somebody from Lenovo sends me controller s/w or a BIOS hack for testing (I'm part of IBM Research, Zurich Lab, i worked w/ our TP Division before divesting it to Lenovo). I understand their worries, since it's about warranty.

 

"2. You write that "during light load (VLC radio + Chrome easy browsing, no flash or heavy WWW)" the W520 hovers "around 48C". But does this also mean that the W520 stays near or below the 2000 rpm level during these tasks?": YES, @1800 on BIOS. Fanless/passive in TPFC Smart, up to 60C (both the x20 and x30).

 

"even if Lenovo were to introduce lower fan speed options for the W530 one would still have to use TPfancontrol to enforce them on such easy tasks.": Remains TBD. I speculate that 3-5 steps under 2K rpm would enable even BIOS/Win7 to enter in a quieter (if not fanless) mode. These control steps are a must IMHO. Mitch

 

P.S. I'm having problems w/ the x30 new chiclet KB; Home/End/PgUp/PgDn are essential, yet spread all over. Also hybernation's gone from F12. But my most severe x30 issue is the display.

a) its 5-6 steps higher LCD backlight; not only it killed the black level and contrast (worse than x20!), but it's too high for the evening. I simply can't use it, and all my external screens are brighter. I'm left w/o a night screen... :-)

b) low resolution for this class. A 3K$ machine should be in the Retina class, w/ 4-5Mpix LCD. I miss the real estate on the main screen, even if my monitors are 4 to 9Mpix (zr30w and T221). Since the A23p, all our workhorse TPs had 1200 lines, not 1080. Not to mention Apple's retina MBP and iPad, or the upcoming Nexus 10 (speculated over 300dpi).

 

However, none of these can be fixed in the x30, which i count as a lost generation :-). But the fan control can and should be.

 

 

Reply
Options

22 Posts

10-19-2012

Switzerland

114 Signins

886 Page Views

  • Posts: 22
  • Registered: ‎10-19-2012
  • Location: Switzerland
  • Views: 886
  • Message 80 of 319

Re: W530 fan noise intolerable

2012-10-27, 0:34 AM

Thank you very much again for doing all this research, answering my questions, and sharing your insights, Mitch.

 

Also: "I'm part of IBM Research, Zurich Lab" Funny coincidence - who knows, maybe we are neighbors :-)

 

Despite your rather somber final verdict. Let's hope Lenovo can fix at least the noise issue. I have not focused much on the screen of the W530 yet, but, coming from a T61p, I find the brightness to be a vast improvement. Hopefully this will provide much better readability outdoors in sunlight.

 

One more positive note. I still have faith in the Thinkpad tradition. My T61p has served me very well for many years without problems and has survived, not just one, but even two beverage spills. So, I guess, we all make mistakes. Hopefully we also learn from them, and try to treat those who suffer from them fairly.

Reply
Forum Home

Community Guidelines

Please review our Guidelines before posting.

Learn More

Check out current deals!

Go Shop
X

Save

X

Delete

X

No, I don’t want to share ideas Yes, I agree to these terms