Page 2 of 3

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Fri Feb 12, 2010 4:02 pm
by RonS
Aleq wrote:Could you run some quick test (with SuperPI for instance) what's the performance of connected 135W and 90W adapter? I'm fine with slow charging times, but don't want to have lower performance...
Thanks a lot
I ran SuperPi with 2M iterations. System is Core i7, 820QM

Battery only: 57 secs (stuck on 1.73 MHz, according to the Core i7 overclock gadget
90W power supply: 31 secs (meter bounces around 3.06 MHz)
135W power supply: 31 secs (meter bounces around 3.06 MHz)

I also ran a 32M calculation (took 12m 56s) with the 90W power supply, and while it computed it the battery continued to charge (75%, 76%, etc). Nothing else was going on (i.e., no graphics intensive operations).

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Fri Feb 12, 2010 9:24 pm
by Aleq
I don't know why I thought my older adapter from my T60 won't fit. It does. So I started to do my own measurements and discovered a very interesting thing:

Legend:

Code: Select all

W510/BAT = W510 running on battery
W510/90W = W510 running with T60 adapter 90W
W510/135W = W510 running with original 135W adapter
T60/BAT = T60 running on battery
T60/90W = T60 running with original 90W adapter
W510 = Core i7 820QM
T60 = Core 2 Duo 1.8
Both running Maximum Performance profile set to highest speed, WiFi 802.11n connected, bluetooth connected (mouse), medium brightness (n/a for T60 which is connected to running external panel)

SuperPI, 4M

Code: Select all

W510/BAT    - 2:07, 2:07, 2:08
W510/90W    - 1:13, 1:13 <--------- Single core, no problem even with Intel Turbo Boost
W510/135W   - 1:12, 1:13

T60/BAT     - 3:05, 3:05
T60/90W     - 2:54, 3:03
HyperPI, 2M, 8 threads

Code: Select all

W510/BAT    - 1:26 avg
W510/90W    - 5:43 avg <------------- Horrible
W510/135W   - 1:20 avg
HyperPI, 2M, 4 threads

Code: Select all

W510/BAT    - 1:22, 1:22
W510/135W   - 0:55 avg, 0:55 avg
W510/90W    - 5:16 (:13-:12), 5:01, 1:38 (with disabled WLAN, BT, lowest brightness), 6:40 (with WLAN, BT, highest brightness) <-------- Horrible

HyperPI, 2M, 2 threads

Code: Select all

T60/90W     - 1:26 avg
W510/90W:
When more cores are utilised, Core Temp reported after a while Frequency 1196 MHz (133 x 9) only instead of 1861 MHz (133 x 14). It looks like during the PI calculation on 90W, it starts with 1.8 GHz, after few seconds it switches to 1.2 GHz. It keeps changing the frequencies, 10s - 15s on 1.8, 35s on 1.2. These times are variable, based on power consumption I think.

Summary:
For single core utilizing tasks, it does not matter (to be on the safe side, let's say most of the times). The charging may be slow, but that seems to be all.
For paralleled tasks, it is a completely different situation. The power of adapter matters a lot. The amount consumed by the system matters a lot. If you lower the consumption of other parts (turn off bluetooth, wifi, set lowest brightness), the speed penalty is smaller (still significant). With all bells and whistles - wifi, bluetooth, highest brightness, the speed penalty is huge.

For me it means to carry 135W allways with me, unless only office work is expected. Ok so always :)

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Sat Feb 13, 2010 2:36 am
by RonS
aleq - that was a great post. Now that I know about that limitation, I'll shy away from the 90W supply as well.

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Sat Feb 13, 2010 3:36 am
by archer6
Aleq

Thanks for the great contribution, your post was most informative.

Cheers...

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Sat Feb 13, 2010 6:32 am
by espenbago
My personal, main open question for the W510, which surprisingly no reviewer has mentioned yet, is whether the battery life consumption rate is at all suitable for some mobile work, in other words; is it possible to work with WLAN on with either 6- or 9-cell battery, for more than 1 hour? For more than 2?

Has anyone heard of any actual testing done on this area? I've tried to guess how realistic the maximum 3 hour battery life (from the Lenovo tabook.pdf) really is, from power consumption of i7 CPU, FHD (multi-touch) screen and GPU documented, but it's impossible to arrive at any solid conclusions, of course.

I'm worried that the battery unplugged might last too short too actually be usable without the power supply. 3 hours, if that's doable with the 9-cell, would be acceptable. But without knowing that, I have too keep postponing ordering until a high resolution T510 is available here in Norway (which might take months, I guess..)

So if anyone knows, or could test, I would be very happy (depending on the answer, of course :)

Espen

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Sun Feb 14, 2010 8:49 am
by Volker
Results are very much in line with my desktop i7: Single-threaded computations barely make the fan spin, but once you go to 8 threads fans are at max.

Floating-point computations are even worse than integer arithmetic (which is what the superpi test uses exclusively).

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Thu Feb 18, 2010 7:03 pm
by mspacek
espenbago wrote:My personal, main open question for the W510, which surprisingly no reviewer has mentioned yet, is whether the battery life consumption rate is at all suitable for some mobile work, in other words; is it possible to work with WLAN on with either 6- or 9-cell battery, for more than 1 hour? For more than 2?

Has anyone heard of any actual testing done on this area? I've tried to guess how realistic the maximum 3 hour battery life (from the Lenovo tabook.pdf) really is, from power consumption of i7 CPU, FHD (multi-touch) screen and GPU documented, but it's impossible to arrive at any solid conclusions, of course.

I'm worried that the battery unplugged might last too short too actually be usable without the power supply. 3 hours, if that's doable with the 9-cell, would be acceptable. But without knowing that, I have too keep postponing ordering until a high resolution T510 is available here in Norway (which might take months, I guess..)

So if anyone knows, or could test, I would be very happy (depending on the answer, of course :)

Espen
Espen, the tabook quotes 3 hours for the 6 cell, and 4.9 hours for the 9 cell. I haven't done any real objective measure, but I know with the 9 cell on my W510, I've seen Win7 report that I have over 4 hours of battery time left. With the screen turned down somewhat, but wireless left on, I think I've run for about that long. The 9 cell will certainly let you last over 3 hours with moderate use.

Martin

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Sun Feb 21, 2010 4:53 pm
by uwe_guenther
Hi all,

nice report makes me really plan to order that nice gadget. RonS you should get comissions from Lenovo.. :-)

So I am planing to buy a W510, but I would really like to know if it works with a TripleHead2Go 2 x 1920x1200 or 2 x 1920x1080 configuration.

I did some theoretical research but I could not get really find an answer, so may be RonS or some someone else could help me out before I finaly order:

1.) The Quadro FX880 from nVidia http://www.nvidia.com/object/product_qu ... _m_us.html says on its Features page:

DisplayPort Digital Display: DisplayPort supports ultra-high-resolution panels (up to 2560 x 1600) with 30-bit color, which result in amazing image quality utilizing a palette of over one billion colors.

So this doesn't sound that well for my planed setup

2.) The web based Matrox Compatibility Wizard is just falling appart when it comes to the nVidia Quadro chipset for notebooks :
http://www.matrox.com/graphics/en/suppo ... ility/gxm/

3.) So there is a native Compatibility Tool one could install in the W510, which then tells you if the W510 is compatiple with TripleHead2Go or not:
http://www.matrox.com/graphics/en/suppo ... /gxm/home/

So it would be fantastic if one of you guys could install the native Compatibility Tool (http://www.matrox.com/graphics/en/suppo ... /gxm/home/) to check if the W510 is compatible or not with the TripleHead2Go.

--
Uwe

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Sun Feb 21, 2010 5:41 pm
by RonS
uwe_guenther wrote: So I am planing to buy a W510, but I would really like to know if it works with a TripleHead2Go 2 x 1920x1200 or 2 x 1920x1080 configuration.
I checked and re-checked all of the W510's video output and it appears that none of the built-in options (including those on the dock) will support resolutions beyond 2560x1600. So your idea of using TripleHead2Go to split the DisplayPort output into two 1920x1200 outputs won't work. I bought a ViDock 2 that arrived just yesterday. I've been playing with it (with a v5200 card installed in the ViDock 2) and, with the TripleHead2Go, will output two 1920x1200 on each of the V5200's DVI ports. The ViDock appears to be a very good solution for multiple monitors on the W510, it has incredible performance, but there's a few problems I'm still chasing down. One of them is that the combination of the NVidia FX880m and ATI v5200 disables aero in Windows 7. I'll write up a long post about it soon.

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Mon Feb 22, 2010 4:32 am
by uwe_guenther
Hi Ron,

thanks a lot for your insight I will have a lock in the spec of the ViDock2 and will lock forward for your upcomming post. :-)

--
Cheers Uwe

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Mon Mar 01, 2010 4:15 am
by zenit

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Mon Mar 01, 2010 1:34 pm
by RonS
I just checked -- I didn't see any of the red ghosting effect. To be sure, I took my W510 into a dark room and tested at full brightness and several dimmer settings. Still, I didn't see any ghosting at all. I did the test by opening a (black) command shell window and right-clicked to bring up white menus at different places as in the video.

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Mon Mar 08, 2010 2:30 am
by Bring_back_IPS
RonS wrote:

Heat, etc.


The W510’s fan is just slightly audible. It seems to be always on. It gets slightly louder under load. It slightly louder than the fan on my T60p, which was very quiet.

The system is very well cooled. The only warm area is the memory access cover underneath the system. Sitting here typing this post, the exhaust vents are just barely warm and I can hear the fan only if I put my ear up to the vents. But, when put under load, the W510 heats up quickly and the fan kicks in.
1. Does it get back to quite mode when the CPU gets back to idle?

2. Some people complain about too saturated colors of icons, etc., with wide gamut monitors. Is that a problem for you?

Very helpful review, thank you.

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Thu Mar 11, 2010 2:00 pm
by RonS
Bring_back_IPS wrote:Does it get back to quite mode when the CPU gets back to idle?
Yes, it returns to normal speed within a few seconds or so. Cooling on the W510 is the finest I've ever seen.
Bring_back_IPS wrote:. Some people complain about too saturated colors of icons, etc., with wide gamut monitors. Is that a problem for you?
Nope, no problem at all. My only real complaint, having used the system's screen for a while now, is that the picture quality deteriorates a little when taken off full brightness. But full brightness is to bright for me. Oh how I wish they'd bring back IPS, or push forward with OLED. Still, I think everyone who ends up with the FHD screen will be very happy with it, as I am.

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Thu Mar 11, 2010 11:48 pm
by Bring_back_IPS
RonS wrote: Nope, no problem at all. My only real complaint, having used the system's screen for a while now, is that the picture quality deteriorates a little when taken off full brightness.
Thank you for responding. When you say the picture deteriorates - what do you mean? I hope that lowering brightness is analog (dimmer backlight), not done digitally by clipping levels, etc.

I use my laptop for photography often. :(

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Sat Mar 13, 2010 12:20 pm
by Anduril
RonS wrote: Nope, no problem at all. My only real complaint, having used the system's screen for a while now, is that the picture quality deteriorates a little when taken off full brightness. But full brightness is to bright for me. Oh how I wish they'd bring back IPS, or push forward with OLED. Still, I think everyone who ends up with the FHD screen will be very happy with it, as I am.
Hi RonS,

Would you mind dimming the screen (e.g. to 1/2 or 1/3 of the max brightness) and try looking at the screen through a camera? My T400s (LED backlight) would occasionally flicker except when it is set to max brightness and I am wondering if it is the case with W510/T510 as well.

Thanks!

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Sun Mar 14, 2010 3:48 pm
by zenit
Anduril wrote: Hi RonS,

Would you mind dimming the screen (e.g. to 1/2 or 1/3 of the max brightness) and try looking at the screen through a camera? My T400s (LED backlight) would occasionally flicker except when it is set to max brightness and I am wondering if it is the case with W510/T510 as well.

Thanks!
I believe all LED screens flicker with low brightness. The brightness of LEDs is usually adjusted by two methods - lowring the forward current or lowering their "flickering" rate (pulse width modulation). When you have multiple LEDs trying to keep area evenly lit, the forward current method does not work well due to manufacturing variabilities with LEDs. LEDs are not all created equal (usually roughly separated by manufacturers into "bins" based on their luminance) and almost never emit the same amount of light, and the variability gets worse with lower forward current. Pulse width modulation is a way to get them all to dim evenly. Human eye usually can't detect it as it is generally too slow to detect changes in flickering. With dimmed pulse modulation dims LEDs it lengthens the "off" intervals in between flickers and our eye detects it as less light. Basically the system takes advantage of crappy temporal resolution of human eye. Camera's resolution is a little better, so thats why the flickers shows up on there. I am almost 100% positive you will see flicker with camera in T510 and W510 screens simply because they are LED backlit.

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Sun Mar 14, 2010 3:55 pm
by QFoam
RonS wrote: My only real complaint, having used the system's screen for a while now, is that the picture quality deteriorates a little when taken off full brightness. But full brightness is to bright for me.
RonS:
I'm wondering if the contrast ratio is what you're finding to be too high at max brightness rather than the brightness itself? One of the things that RGB-LED backlighting does is substantially increase contrast ratio.

I ask this because your display is ~280 NITs, and I've seen lots of White-LED backlit MacBook Pros that have 300 NIT displays, and they don't seem too bright to me in normal room lighting. And that's when they're cranked up to full brightness. In fact, they seem a bit dim/dull compared to my W700's 400 NIT display, which I keep at 90% brightness in room lighting (and at that setting is probably a lot brighter than what you're looking at on the W510).

So what I'm wondering is, will you be able to make more effective use of your display's brightness if you turn it up all the way, run the color calibrator, then while still in the calibrator software, adjust the display's Gamma setting (that adjustment is on the same screen of the Huey software that lets you adjust color temperature). By adjusting Gamma, you can effectively adjust the screen's contrast to be more pleasing. That way, you can make better use of the screen's full brightness, while at the same time maintaining color-quality, which you say decreases when you turn down the brightness.

High contrast-ratio is great when you're watching movies on a display. But it's not neccessarily so great when you're using the display as a computing display. (In fact, if you're doing something like photography or image processing, as I do, high-contrast prevents you from seeing shadow detail, which can be a real bummer.) So perhaps adjusting Gamma to effectively modify that contrast will let you make more effective use of your display.


Anduril:
Regarding display flicker, see my post in the following thread:

http://forum.thinkpads.com/viewtopic.php?f=48&t=85800

I suggest a way that you may be able to eliminate the flicker in your T400s's display.


Good luck!

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Mon Mar 15, 2010 4:50 pm
by Anduril
zenit wrote: I believe all LED screens flicker with low brightness. The brightness of LEDs is usually adjusted by two methods - lowring the forward current or lowering their "flickering" rate (pulse width modulation). When you have multiple LEDs trying to keep area evenly lit, the forward current method does not work well due to manufacturing variabilities with LEDs. LEDs are not all created equal (usually roughly separated by manufacturers into "bins" based on their luminance) and almost never emit the same amount of light, and the variability gets worse with lower forward current. Pulse width modulation is a way to get them all to dim evenly. Human eye usually can't detect it as it is generally too slow to detect changes in flickering. With dimmed pulse modulation dims LEDs it lengthens the "off" intervals in between flickers and our eye detects it as less light. Basically the system takes advantage of crappy temporal resolution of human eye. Camera's resolution is a little better, so thats why the flickers shows up on there. I am almost 100% positive you will see flicker with camera in T510 and W510 screens simply because they are LED backlit.
Thank you for the explanation. I checked some other laptops with LED screens and this is indeed the case. However, I tend to use the laptop with low brightness level and unlike most people, my eyes can't stand the backlight flickering... I've decided to return the T400s and purchase a T500 w/ WSXGA+ screen instead.
QFoam wrote: Regarding display flicker, see my post in the following thread:

http://forum.thinkpads.com/viewtopic.php?f=48&t=85800

I suggest a way that you may be able to eliminate the flicker in your T400s's display.

Good luck!
Thanks, that link is really helpful. The flicker I mentioned is a different kind of flicker, though.

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Mon Apr 26, 2010 3:47 am
by hondalee
Hi all ,
I find some 120W ac adapter. maybe can replace the big 135W in W510 ..
Kensington 120 Watt see http://us.kensington.com/html/6330.html
Acbel Power King Universal 120W see http://www.trademe.co.nz/Computers/Comp ... 424343.htm

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Mon Apr 26, 2010 6:00 am
by Greg Gebhardt
Aleq wrote:Could you run some quick test (with SuperPI for instance) what's the performance of connected 135W and 90W adapter? I'm fine with slow charging times, but don't want to have lower performance...
Thanks a lot
I have not run tests on my 510 yet but with the 90 watt adapter the CPU "boost" is limited, so there would be some difference in performance under heavy CPU intensive use. The 90 watt did run my 510 and even charged the battery from 87% to 100% is a reasonable amount of time. The 90 watt charger did get much warmer in comparison to the 135 watt supply. I got two more 135 watt adapters ordered but Lenovo says they will not ship until 4-30-10. Nice to know the other adapters work. I love my slim 90 watt powers supply and hope they come out with a 135 watt version.

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Mon Apr 26, 2010 9:50 am
by hondalee

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Tue Apr 27, 2010 2:50 pm
by pae77
My i7 quad core based HP DV8 came with a fairly slim 120 watt power supply. Perhaps these could be adapted for use with the Thinkpads and perhaps they would supply sufficient power for full speed operation.

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Thu Apr 29, 2010 1:21 am
by hondalee
I'm fine with slow charging times, but don't want to have lower performance...
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
me too, i test by Nuclearus Multi Core 2.0
My W510:I7 720QM,HD+

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Thu Apr 29, 2010 5:23 am
by Greg Gebhardt
hondalee wrote:I'm fine with slow charging times, but don't want to have lower performance...
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
me too, i test by Nuclearus Multi Core 2.0
My W510:I7 720QM,HD+
Under normal use I do not even think you would notice lower performance. But then it depends on what you are doing. I would bet that you might notice with intensive CPU use but after 5 days of running on both the 90 and 135 watt power supplies, I have to say I do not notice any difference except the 90 watt brick get a little warmer. I am still waiting on Lenovo to send me my two extra 135 watt chargers.

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Tue May 18, 2010 12:24 pm
by StevenD
Hi, Ron,
Thanks for the post. There's a question that I don't see was quite answered: Do you feel the screen is as bright, or maybe brighter, than the T60p, or is it just the best so far now that the T60p is no longer available?
--Steve

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Wed May 19, 2010 10:41 am
by RonS
StevenD wrote:Hi, Ron,
Thanks for the post. There's a question that I don't see was quite answered: Do you feel the screen is as bright, or maybe brighter, than the T60p, or is it just the best so far now that the T60p is no longer available?
--Steve
The W510 screen is FAR brighter than the T60p's Flexview screen. I usually run with the screen slightly dimmed because it's too bright.

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Sat Jun 12, 2010 11:45 am
by dfgalletta
All I have open is IE and Outlook, and scrolling in IE is very jerky when plugged into my Kensington 120 watt adapter. When I unplug it, IE is smooth as silk. That is ridiculous. There is plenty of headroom. I have about 8 of those adapters and was really confident that they would be close enough to 135 watts.

I hope someone is able to hack this deliberate and needless (when within 15 watts) crippling.

DG

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Sat Jun 12, 2010 12:00 pm
by dfgalletta
Please allow me to amend my previous post. If I only have IE running, the same thing happens--very, very sluggish, jerky web page scrolling when plugged in to 120 watts, and very smooth, fast scrolling on battery power alone (or on 135w power, of course). I verified Outlook was closed by going to task manager and making sure it didn't appear in the list of processes.

On the 120w power supply, this machine performs pathetically--besides the scrolling problem, The characters cannot appear as quickly as I type in this very window, so I'm just hoping that the words catch up with me accurately.

I'm only using a tiny amount of processing power! My 1 ghz $200 Droid Incredible is faster than this $2,100 machine. Of course, I do have a 15.6 FHD touchscreen, which is clearly more memory and power intensive than my 3.7" Droid screen.

I must confess, though, that I have a second hard drive in the ultrabay. A new ultrabay caddy was another unwelcome expenditure--due to the redesign of the ultrabay for the w510.

Perhaps Lenovo will "fix" this to make it at least a little more tolerant--of 120 watts, that is. But there is no incentive for them, of course.

Re: Thinkpad W510 First observations and testing (Embedded Pics)

Posted: Sun Jun 13, 2010 7:57 am
by infinus
Maybe the BIOS isn't correctly recognizing that power supply and scaling back performance. If it's just fine with a 135 then I'd assume that to be the case.