HannStar Display. They are pretty good.
-
Larry@LPC-Digital Company Representative
-
Is there a hack of any sort to deal with a touchpad scrolling (or lack or thereof) issue? vs. up-down sensor areas
-
Can someone report the temps of the i7 at full load, especially of the 920xm?
Some screenshot will be appreciated.
Thanks. -
Soviet Sunrise Notebook Prophet
Coming right up. I just finished my CPU heatsink mod and I will be putting it through OCCT in just a few moments.
-
ss
if you could post a picture of your heatsink mod too
-nice Avatar man -
Soviet Sunrise Notebook Prophet
Pics are NDA will come once I finished everything. I am thinking of revisiting a backpanel mod as well. Only 11 minutes left on OCCT.
-
Soviet Sunrise Notebook Prophet
Here are the results of my semi-formal CPU stress test. The format for this test has been copied from my cooling overhaul. http://forum.notebookreview.com/showthread.php?t=416414
The CPU was tested on OCCT CPU Linpack. The test was executed under tight conditions. The notebook was idling at desktop for two minutes after boot prior to the test. Room temperature was at a constant 22*C. The notebook was sitting ontop of the Edova Xpad. The CPU has been mounted with a fresh application of X23 thermal paste on 14 February 2010. No other applications were running during each test.
The 920XM is clocked at 2.26GHz until the ACPI hits 60*C. From then on, the CPU is clocked at 2.13GHz. When the ACPI hits 75*C, the notorious Turbo Boost will cut back the overclock and drop the speed down to LFM for a few seconds until the temps reach around 72*C and then kick back into 2.13GHz. From that I began the test when the ACPI temperature was idling at 42*C. OCCT CPU Linpack is an adaptation of IntelBurnTest. OCCT monitors the system for one minute prior to stressing the CPU, and monitors the system for four minutes after stressing the CPU. This CPU test stresses the CPU in ~135 second sessions in proportion to it's performance with a 30 second rest period inbetween each interval. Hyperthreading and 64 bit instructions are enabled. http://forum.notebookreview.com/showpost.php?p=5883327&postcount=2100
So far I'm pleased with the results. Though the 920XM, or either of the other two Mobile Core i7's, cannot exceed 75*C due to Turbo Boost dictating the safety of the CPU, my CPU is able to remain fairly stagnant around 70-72*C for the vast majority of the test and Turbo Boost only engages to restrict the CPU only a few times during the test. With that said, compared to my QX9300 @ 2.93GHz from my M570TU, the 920XM runs cooler on this machine by 6-8*C. Many factors contribute to this, as well as my own injeted variables in the form of modding, and of course it is not an equal comparison, but overall it shows how robust the cooling is on the W870CU.
This is just a taste of what is yet to come for the W870CU cooling overhaul. I have decided as of this post to revive my plans for a backpanel mod as I believe the pinned heatsinks on my existing CPU heatsink will benefit more from passive ventilation than static pressure from the CPU fan. More details will follow once I publish my article.
Because of the inconsistency in results imposed by Turbo Boost, I will be doing this test again with Turbo Boost disabled so that we may get a truly accurate performance and temperature reading for the 920XM. Like I mentioned in my post in the link above, the 920XM will operate at 2GHz regardless of temperature with Turbo Boost disabled. -
kevindd992002 Notebook Virtuoso
-
Soviet Sunrise Notebook Prophet
Here are the results of the exact same test conducted above, but with Turbo Boost disabled. The results are much easier to reproduce on other notebooks now that Turbo Boost isn't fiddling around with the clocks. Because the CPU is operating slower at 2GHz, the time it takes for the CPU to complete each session has been increased from ~135 seconds to ~150 seconds. It seems that the maximum temperature here is ~74*C, which is very modest for a CPU of this caliber. This means that I can easily easily stop monitoring the temps on my CPU as I know that it will never exceed 75*C regardless of whether Turbo Boost in on or off.
-
Thanks a lot Soviet Sunrise.
Do you have the standard heatsink or a modded one?
And Can you please report the temps of 1core at maximum turboboost (3,2Ghz)?
P.s.: I see that you have changed your avatar, nice. -
Soviet Sunrise Notebook Prophet
My heatsink is heavily modded so the results on untouched heatsinks will cause the CPU to fluctuate more frequently at 75*C, and will definately begin to near 80*C if not surpass it with Turbo Boost disabled.
The temps for only one core at 3.2GHz is going to be less than four cores on load. Also, even in a worst case scenario stress test where OCCT CPU Linpack is locked to using one specific core at 3.2GHz for the entire duration of the test, the temps are most likely going to stay under 70*C if not ever pass it. However, because of how Turbo Boost and the thread ordering instructions on the CPU works, the CPU will juggle the load from one core to the next for legacy apps that only utilize one or two cores. This means that the CPU will run the thread on one core for a few seconds, then pass the load onto the next idle core and so on so that all cores are used equally and doesn't wear out a single core by always running on it. This is good for evening out the overall temperature across the entire CPU. The only way you can have an application have it's thread locked to a single core is if you force it to do so. The same concept applies for dual and triple threaded apps. -
I see that many people here already have the gtx 285m.
Can one of you please send me (upload and give me a link in a pm) your 285m bios (just extract it with gpu-z)? Id like to flash it over my 280m..
Someone pretty please -
I don't know if you're stress test can be singlethreaded, but if you can do something to simulate a single threaded heavy application (an application that require 3,2Ghz) and post the temps, i will really appreciate. Sorry before i didn't explain myself correctly when i was asking for 1core on load temps.
You know you're test is a worst scenario one, but in the best condition, because you have a modded heatsink (average man like me, won't have one) and the room temperature is 22°. So if the 3,2Ghz turboboost can't last long, let say for the 29 minutes of your test, maybe for a singlethreaded user like me, without the modded heatsink and with a summer room temp of 32°+, it is better to opt for a 620m.
Sorry to bother you, but i really need to have some kind of evidence, before buying my notebook.
Thanks in advance and excuse me if sometimes i make some grammar errors, i'm not english motherlanguage -
Soviet Sunrise Notebook Prophet
Correct. Synthetic stress tests are unrealistic stability tests developed for hardware diagnosis and error checking. I'm not going to post my results as it would be a bit redundant. I did the test again using Prime95, which is very similar to Intel's Linpack. Core zero did indeed stay at 3.2GHz for the test, though with a few dips here and there due to the dataset reloading each time. The best I can do is get the active core up to 65*C, but this was done by using a tshirt to block the CPU exhaust vent as it was impossible for me to get the temps up above 60*C. I'd say you're good to go if you're going to be running single threaded apps. The clock speed for that single core will not pull back as it gets hotter, or if it ever gets hot I should say. Surprisingly, when I ran a dual threaded test locked to two cores, the CPU still did not drop the clocks from 3.06GHz even when the hottest core got up to 76*C, again from me blocking the vent as it was difficult to get the cores over 70*C.
I guess that covers everything in regards to Turbo Boost on this notebook. The Mobile i7's will do their job and will be kept cool in real applications. You need not worry of overheating with this notebook.
The only reason that I can think of as to why Clevo decided to set the restriction to 75*C is to protect the MCH, which is conveniently located on the CPU die now as opposed to previous systems. The MCH is more delicate than the CPU and cannot withstand the high thermal stress. Since the MCH is on CPU die, it needs to be protected from the heat being produced from the cores, as well as protected from the own substantial amount of heat that it produces itself. Instead of designing a larger heatsink that would pull this notebook out of it's targeted class, Clevo in their typical conservative style, setup the temperature restriction via ACPI and the Turbo Boost algorithm. This goes back to what I stated last year; the i7's are "hot" compared to the previous generation because it's not just the cores that are producing heat on a single substrate anymore. http://forum.notebookreview.com/showpost.php?p=5441851&postcount=17. The temperature that the user is seeing is the heat produced by both components. The MCH temperature cannot be digitally monitored on the Mobile i7 and on previous generation notebooks, so users generally disregard that the MCH is producing heat at all and just focus all their efforts on the CPU just because they can "see" the temperatures. I know some of you are thinking that it's a power issue. But I can assure you it's not.
That's probably all that I'm going to do for the rest of today other than find out why my GTX 285M is playing tricks with me. English isn't my native language either, so you're not alone. -
I'm not an expert, so take this words as an opinion, but i think that the 75° limit is put by Intel, not Clevo.
However thanks for the info, I was really afraid of 920xm temps, because the 55W TDP really scared me.
There is one thing that i don't understand, what you mean with "Instead of designing a larger heatsink that would pull this notebook out of it's targeted class"? This is a 17", i think the only larger notebook i have seen has been a 18", not a big difference. Does bigger notebook exist (19"+)?
P.s.: i have seen some screenshots of the bios of a W870CU on this forum, if i remember correctly Babyhemi posted them, and i can't see the option to turn off or disable the hyperthreading. Can it be disabled or not? -
Soviet Sunrise Notebook Prophet
No. The limit was put by Clevo. Other notebooks that are equipped with the 920XM are capable of exceeding 75*C without being held back. The reason why Intel lists the 920XM as a 55W TDP chip is because it is a 55W TDP chip, plain and simple. Clevo has a reputation to keep as being a manufacturer of well cooled notebooks, and they wouldn't give you the option to stick one in there if it would overheat.
Notebooks larger than 17 inches do exist. Clevo makes one called the M980NU, now transformed into the X8100, an 18 inch P55 SLi notebook.
Unfortunately, there is no option to disable Hyperthreading in the BIOS. I wanted to disable it as well when I first got this notebook. Even on my 1201N, there is no option to disable Hyperthreading. -
After having the unit less than a week, I sent it back because of power issues. When the cpu/gpu was maxed out it would shut down. I've written about this before. Soviet told me about getting a bigger PSU. Which was an option I was considering.
Well, as I was discussing it with sager, after a few weeks they stated there was nothing they would do about the issue. Not advising me either way about getting a larger PSU, and that my only option was to deal with it or get a refund. Whether or not using a larger PSU would void the warranty.
Well with them being so rude, I opted for a refund. About a week and half ago I sent back the manual, cd's and other things I didn't send with the unit since it was going in for a repair. Fast forward to today. They are saying I am not entitled to the 3 Day Select return shipping label provided to me when I sent the unit because I didn't send in for repair, but actually sent it in for a refund. $68 for that.
They going to deduct $250 for scratches on the top case. I was extra careful with it. There were absolutely no scratches on the unit at all. They have had it for more than a month now and trying to blame me for scratches. It remained on my desk the full time I had it as I was trying to figure out what was wrong with the unit. Only going into the laptop bag when I sent it to sager.
The laptop just like the 8690 would be amazing if not for the power issue. I'm just angry at them trying to charge me for these things. Venting here a bit. -
Soviet Sunrise Notebook Prophet
This is why I don't buy from Sager.
-
This scares me because, as much as I loved the machine, I decided to return it for a refund. I hope this isn't the case with mine.
-
I bought it from PC Torque.
-
They are now saying that the unit wasn't defective because it was only one software causing the power issue. Where as I ran other software/games that caused the same issue if the dvd drive was active, and or the battery was recharging. Plus they were saying I didn't return the unit before the 30 day refund time limit and that they, PC Torque spoke with sager so that I would get any refund at all.
Where as I sent it back within a week of receiving the unit and had to wait several weeks for them to tell me they weren't going to do anything about the defective unit. -
Oh wow. I am sorry to hear that. I brought my laptop from PC Torque and they have been nothing but helpful to me. I hope you are able to get a quick and tidy resolution to the matter.
-
Soviet Sunrise Notebook Prophet
-
Lesson. Learned.
-
Are you going to at least get the beast back?
-
-
Soviet Sunrise Notebook Prophet
-
tnx 10char
-
I didn't know this power issue. Is it only a problem with Sager branded Clevo or not?
And does the possibility to disable the hyperthreading exist in other notebook brand or not?
Damn, one problem goes out from the window (the heat issue) and another one come through the door. -
I've noticed another laptop from a different company with similar specs that ships with a 150 watt psu, where as Sager ships with 120 watt. I am curious of their reasoning.
-
They just sent me pictures of the unit showing scratches covering pretty much every surface, including the screen. Obviously the technical support at Sager weren't as careful with the unit as I was. Also the fact they had it three times longer than I had and I presume removing parts and testing the unit trying to figure out why it was shutting down, obviously contributed to all the damage pictured.
I am glad I didn't get the unit back in that condition. Also they keep denying that the unit was defective. Saying that it was only defective for my "particular and specific uses and not a generally defective unit". It's like buying a car and then being told that you can't drive it 65 on the expressway or it'll shut off. But it's not defective, since it only pertains to my "specific use".
I fear the only way to settle this will be take it to court. This has long since not been about the money but they way I have been treated. With PC Torque stating I'd have to "convince Sager" for a more full refund since they have no choice in transferring all deductions from sager to me. I did not buy the unit directly from Sager, so why would I have to go convince them of anything pertaining to a refund. I would understand if I bought the unit with them and they were issuing the refund directly to me. But I bought from PC Torque and they are the ones I am doing business with and they will be the ones ultimately issuing any refunds. -
Soviet Sunrise Notebook Prophet
RJTech now has the Intel 6300 wireless card up for individual sale for $48. If you can somehow shell out the cash and perhaps make bank on your old 5300, $48 won't be too much of a blow to your wallet.
-
PC Torque offered to split the difference in the amount not refunded.
Is the 6300 that much better than the 5300? -
Soviet Sunrise Notebook Prophet
I don't know. I haven't tried it out in the field yet ever since I got it last week. So far, all I've done is transfer the heatsinks from my 5300 onto the 6300. It's been turned off most of the time as I use a wired connection at home.
-
-
Soviet Sunrise Notebook Prophet
That's not news. Clevo has been using the same 120W power supply on all of it's top models, other than the D900x and M980NU/X8100, for the past six years.
-
And I agree with gedekran, what is the meaning of buying a monster notebook if you can't use all its power?
It is ridiculous, because the only thing needed is a more powerful psu. I know that we can buy a better psu, but why do they choose to not upgrade the psu? Are they not aware of the problem? -
Soviet Sunrise Notebook Prophet
Part of the issue is with the contract that every notebook manufacturer has with the PSU manufacturer they purchase from. As to the details between Clevo and Li Shin, I don't know. But it's not just technical compromises alone that stand in the way; there are political and business reasons as well. But you are right. The bottom line is eventually we will be seeing realistic situations where the PSU can't churn out enough power from mid tier and lower tier configured W870CU's.
-
Soviet Sunrise Notebook Prophet
-
nice job
10char -
howd you punch the holes
-
Soviet Sunrise Notebook Prophet
With sunlight and a magnifying lens.
-
-
Please stop modding your notebook, it is bad for its health.And mine, because I envy you
-
guys I had a question for all owners, How big is the noise of this laptop?
-
When you buy a machine like this, there are downsizes as well, just like when you buy a Ferrari over some other average care. -
Larry@LPC-Digital Company Representative
Last edited by a moderator: Jan 29, 2015 -
-
Soviet Sunrise Notebook Prophet
You mean these forks?
The CPU heatsink is 100% complete. The Intel 6300 also got some nice forks. I will be doing a preliminary stress test of the finished design within the next hour on OCCT. The GPU is incomplete and is undergoing a redesign. Expect it's completion within the next two or three weeks depending on my schedule. RAM heatsinks will come soon after the GPU heatsink is finished. -
forks indeed!
*** OFFICIAL Sager NP8760 / Clevo W870CU Owners Lounge ***
Discussion in 'Sager/Clevo Reviews & Owners' Lounges' started by Larry@LPC-Digital, Oct 8, 2009.