*** Official Clevo P65xSA/SE/SG / Sager NP8650/51/52 Owner´s Lounge ***

doesnt matter where u place then, as long as theyre right next to each other@dual channel
ive got my two sticks installed under the kb just cuz i cant be bothered to open it up again@install and forget

Sent from my Nexus 5 using Tapatalk

 

So octiceps, there is no point according to you in replacing the stock thermal pads on gpu vram by something more performant like fujipoly on these kind of mobile gpu (I am on 980m)?

How would you change the timings in prema's vbios?

 

Nah no need to change the pads really, not unless they are getting damage or you are adjusting backplates etc.

 

Yep, like Meaker said. The VRAM doesn't OC high enough to see any benefit from higher conductivity pads, and the bottleneck for heat transfer is probably somewhere else like the heatsink.

 

Hi, I have a P650SE and I love almost everything about it. My biggest gripe is the 180W Chicony power adapter which imo is less portable than the laptop itself. Are there any alternatives which are maybe slimmer or lighter? I do not overclock anything. Thanks!

 

The Dell 240W PA-9E is slimmer I think. You need this adapter with it though: http://www.bixnet.com/cntx75.html

 

Hey guys, how is your battery life on this laptop? (Mention of your 4k/1080 and 0/1/2)
Thanks!!

 

Fhd and np8652 here, normal use (balanced power plan, 75% brightness, wifi and Bluetooth on and movie loop I would say 4 hours.

Camel mode (power plan economy) plane mode on, minimum brightness and coding/word see me around 5h30

 

Hey guys...so I have one issue with the control center...i keep changing the option for the sleep button to put computer into sleep but every time I stop the computer, the option resets to hibernate.

I'm pretty annoyed as I did my best to disable hibernate everywhere in Win 8.1 and this thing keeps going to hibernate everytime I push FN+F12

Anyone have some tips on how to make the option stick?

 

Has anyone experienced overheating hard drive? I really do not like the configuration of 2.5" bays stacked right on top of each other like that...

Debating between putting a Hitachi 7K1000 or cooler but slower Samsung M9T in the 9.5mm bay, with an SSD in the 7mm. At least the 9.5mm bay is on bottom facing the access panel/away from mainboard so all heat from the HDD is rising up toward the SSD.

 

@octiceps

Word of advice/warning. I had the 7k1000 in my clevo and it indeed had the same annoying click-click head parking of the western digital drives that I was talking about in my thread.

So I just decided to stick with the samsung M9T that I pulled out of my Seagate Backup Plus , and I'm very happy with it.

 

Yeah, I know WD/HGST 7200RPM are noisier. I'm mainly concerned about heat though.

 

Yeah I saw that.

Can't really offer any input there but just wanted to inform you about it

The thing is it's not just noise... every time I'm watching a movie in VLC and that annoying "feature" comes along, the whole movie stutters and gets gray artefacts.

If it wasn't for that I wouldn't really care.

 

Odd. I don't have any such problems watching anything off my HGST drive. I can hear/feel it spin up if I run something hard drive-heavy, but I never got much "noise" out of it. Then again, my room is never really silent, but still.

 

I just c

The noise of traditional HDDs is one of the reasons I never have these in my laptops anymore...

 

Really? Im totally okay with that.
Thanks!!

 

Yeah, it's a lot better re: battery life than an MSI GS60

 

From my experience. First, make sure the power options that you're using is matching what is being defaulted for the Hotkey menu (Control Center), if not try to update Hotkey software. Second, disable/edit the Sleeping/Hibernate function within Power Options under your power plan under "Choosing what closing lid does" and then hit the "Change settings that are currently unavailable" to edit further. Hotkey should be reflecting your Power Options settings after your edits are done.

I have no overheating issues and I have a default 1TB 5400RPM harddrive on top of a Samsung EVO 840 1TB with 2x512GB M2 M550's.

 

Blimey - 3TBs total! That's quite astonishing for a laptop computer!

 

So... whenever I change the power plan in control center, they automatically change to their equivalent in windows asswell(performance to high performance, entertainment/quiet to balanced, economy to power saver)

I disabled hibernation in all windows options... and yet my control center is always defaulting the sleep key to "hibernate".

I set inside windows the options for what the "sleep" button does to "sleep" and it still goes back to hibernate in the control center EVEN THOUGH inside windows it stays saved as "sleep".

I managed today to fully disable hibernate and delete the "hiberfil.sys" from elevated command line by using the "powercfg" command.

Now even though the control center is defaulting to hibernate everytime, the button actually sleeps my computer because hibernate is now disabled at a very low level....the option doesn't even appear anywhere in windows anymore.

I'm still curious though why this happens and if other people have this problem.

 

Don't use the control center, its complete garbage in my opinion. Just leave it on Entertainment mode in the way you want to have it set up. Use powerplan assistant to change power plan based on if plugged in/on battery. It works flawlessly and is lighter than batterybar. In addition, it correctly detects status on startup which batterybar doesn't.

 

My laptop can have a total of (sit down for this) 8TB of storage.

3 x 2TB 5400RPM drives (one in my ODD slot) + 2 x 1TB mSATA SSDs.

If your mind is blown, please pick up the pieces before leaving this forum

 

Guys is there a way to make NVIDIA Inspector remember my OC settings at apply them at windows startup? Because I have to apply them myself everytime I reboot

 

You can "create clocks shortcut" and double click on or pin to start menu the clocks. I do that.

Also, I suggest (just in case) creating a "default clocks".

And NVI kind of pisses me off because I have to create two of each clock preset; one for each card >_>.

 

Impressive, but trying to think of a scenario where I'd actually need 8TB of storage on a laptop unless running a torrent farm

 

Is Prema mod bios required in order to lift current limits on CPU via Throttlestop or XTU? The option is either grayed out or non-existant for me on my 4710HQ with latest stock bios.

When running XTU stress tests I get by without any throttling but during gaming sessions I experience very severe current limit throttling with CPU downclocking to about 2,4 GHz frequently.

 

@soakr

I just asked the same thing on T|I forums today in PREMA's Bios thread .

Still waiting for an answer.

P.S. I found my stable undervolt @ -90mV on both Cache and Core(at -100mV it crashed after 11.5 hrs of prime95 ) and that reduced my power throttling alot.

Prime95 is stable at 3GHz on all 4 cores.

 

@Sebi97

I just modified one power plan (high performance) in the windows power options (corresponds directly to the performance one in control center) to have minimum consumption while on battery and maximum performance when plugged in.

 

I'd suggest Avast anti-virus!

It's 60Hz in P6xxx series...

Jaybeee, what's the point of faster RAM? I remember that it used to help in OCing the CPU/benchmarks/snappiness however, let's say I want to OC RAM to increase BCLK.. I have a 8GB Crucial DDR3-1600 stick and a 4GB Hynix DDR3-1600 chip.. Timings table wise, both are same in CPU-Z... Can I still OC RAM to increase BCLK and OC CPU?

 

I dunno about you, but I'd like to end up with 3TB of SSD storage at some point XD.

 

I just received my Sager NP8652 from XoticPC. Early impressions from a MacBook Air user:

-it is wider than expected. it looks more like a 17" than 15" because of it. Besides that it feels well built and looks nice. I can easily have this open at work and no one would give it a second glance. It is unfortunate that the finish is a fingerprint magnet

-quiet. I played SC2 HoS for 30 minutes at 1080p max and the fans remained at a low audible level. The machine is way quieter than my PS4. Also the machine remained cool. Top left corner was a tad warmer. Really impressed how cool and quiet the machine was over 30 minutes.

-when I was installing windows, the fans would ramp up and down every few minutes. Once I rebooted, has not happened since.

-Optimus is confusing. When I loaded up SC2, the graphics settings looked like it was recognizing the igpu over the 980m since everything was on low. When into the Nvidia control panel and everything looked fine so I went back in and raised all of the settings up manually and started the campaign. No issues.

-keyboard has flex in the middle, upper area. It bothers my OCD but doesn't effect gaming. Keyboard itself seems fine. I would not prefer typing on it over the Macbook.

-touchpad is serviceable.

-stock screen is fine. It has minor bleed. I doubt I will change it with an aftermarket alternative.

-PSU is big. I would not want to travel with it.

Overall, I am very happy with it so far.

 

BLCK and RAM clocks are two different things, one doesnt affect the other. its correct that faster ram helps with the overall snappiness of the system and also increases minimum framerates in games, for example.

Both BCLK and RAM OC are fincky stuff, ull have to bring along some patience to tweak those. the former regulates the whole system including PCIe bus and RAM is crucial for the whole system stability, so stability testing is of utmost importance. with BCLK, i wouldnt expect too much, at most ull just get a few Mhz extra out of it since its pretty sensitive to overclocking. lets say ud get 105 Mhz stable, that would translate into 125 to 175 Mhz more CPU clock (105 Mhz BCLK x 25 and 35 CPU multi).

 

@Lonesyndal

@soakr

I just received confirmation from prema over at T|I that his mod allows increasing turbo power limit.

 

@-Jinx-

I checked the Prema thread that you mentioned and noticed the response that you got. The thing is that I do not experience any throttling due to either power or temperature, my only limit (with stock settings) seem to be the current limit. I notice no difference in throttling even if I undervolt ~80 mV, it still frequently downclocks to 2,4 GHz .

 

How does it go all the way down to 2.4GHz?

How does it even REACH 2.4GHz? That's below base clock.

What do you even have and why is it not RMA'd yet?

 

Downclock might not be the correct term for it then (I am new to this stuff ), it drops down to 2.44GHz for just a brief moment and then it goes back up again. I have attached a picture from how it looks while running XTU. During gaming it will go down to 2.44GHz much more frequently which causes the game to "stutter" (fps drop).

From reading this thread I actually thought that this was normal behaviour. Before I flashed the new bios the problem was even worse where the processor frequency would go much, much lower...

 

WOW that's broke to all hell. Can I see a screenshot of it while running an XTU benchmark?

And can you please set those power limit sliders both to 58W for me please?

 

Looks like I will be doing testing again soon with updated BIOS mods (if they are updated, will have to check changelogs).

EDIT: I dont see an updated BIOS mod for P650SG, so no need for me to test that...

I will be doing a new stress test soon with other settings while on 180W.

 

I have attached two pictures. One during benchmark (is it supposed to be that short? - I ran two on the picture) and one during gaming.

During gaming the processor reaches 2.44GHz just like before. During benchmark it is more stable but utilization still jumps from 99% to 40%.

So you would say it is definitely something wrong with the laptop? I still have a few days for RMA.

 

Do note that during gaming you will be using some power on GPU so you will lose some from CPU under load. 180W may do that to you. If you want consistency... try Throttlestop with Overclock forced on at 34T.

 

Hi, a question: I have a powered USB external harddisk, but it seems my Clevo does not recognise it (it works fine on all other laptops I've tried). I read in the manual that the single powered USB port (on the left) can be switched on by pressing Fn+ power button, however this just turns my laptop off. Does this work for anyone?


Sent from my iPhone using Tapatalk

 

Yeah you're power throttling like crazy @soakr I suggest RMA.

 

I did some gaming after using Throttlestop to set multipliers on 34T, it made my frequencies more stable now pending between 3.1GHz to 2.9GHz (see attached picture) but naturally also lowered my higher clocks (which is kind of annoying since I would like to fully utilize the CPU, especially considering the price tag on the laptop).

During a benchmark it looks like it did without the use of Throttlestop.

By the way, why is my power limit throttling always 0%? Should my problem not be noticable there aswell?

 

So the way that my laptop throttles is not normal then?

There are so many mentionings on this thread about throttling CPUs so I kind of accepted that all of these laptops came with it in some form. So when I RMA and get a new laptop I can (most likely) trust that my experience with the next one will be different? Different as in better .

 

Try increasing the "Turbo Boost Max Power" and "Turbo Boost Short Max Power" back to their stock values of 47W and 58.75W respectively, and see if it still throttles.

 

This is what a working CPU looks like:


Note I don't have any of throttlestop's "forcings" on (multiplier, etc). I only keep TS running to fiddle with my windows timer resolution.

In contrast, your CPU's clockspeed fluctuates like crazy, your power limit throttling is insane, and even though I got you to adjust your turbo boost power max and turbo boost short power max settings, you still are not even operating at stock turbo. It's not acceptable, bro. Not even close. You tell Sager your CPU is throttling a lot and XTU's graphs show a lot of power limit throttling and current limit throttling, resulting in clockspeed fluctuations and they'll probably fix you up right quick.

 

Hi!

I've been lurking this thread since Nov. and finally received my Sager 8652 (980m) yesterday -- so thank you everyone for all the new knowledge and insights that gave me the courage to pull the trigger!

I've ran through a lot of the tests/benchmarks that everyone has mentioned here and just came across my first problem. When I ran the Prime95 Blend test on Priority 1, it went for 45 minutes with Max temps hitting 94C, but avg. in 88C. I had no temperature throttling, though I did get occasional power limit throttling (going between 0 - 100%). When I then tried it on Priority 10, the computer shut down after 10 or so minutes. I rebooted without issue. My setup was a -50mV at the time. This brings me to a few questions:

1) What is "power limit throttling" telling me? I.e., What should I do in response to seeing that throttling, even if it is only happening intermittently?
2) Using Prime95 at full load, my cpu temps would avg in the 85-88C, with the occasional max spike into the 90-94C range. Is this bad? What does it mean if I had 0% thermal throttling all throughout that time?
3) Is running Prime95 on "Priority 10" for 24 hours truly the "right way" to stress test your CPU ? That's what I've read, but it's obviously an extreme scenario and I don't want to read into it too much considering I've only UV'ed but did not OC the CPU. Or do those temps make the CPU or my cooling setup bad in some way?
3a) Any idea why my computer was fine running the stress test in Priority 1, but then shut down in Priority 10? Was I not taxing the cpu properly via Priority 1?
3b) When testing Undervolting, what's the minimum amount of time I should be running a stress test? I've been doing anywhere from 5-20 min, but not sure what is appropriate.

Other than that, everything about the laptop is awesome. I have the Sharp 4K screen which is beautiful. Minor light bleed on edges, but nothing that bothers me. There is a slight graininess to it that someone has mentioned before, but that doesn't bother me either. I did a 135+/250+ OC on the 980m via nVidia Inspector and avg'ed 8970-9000 on Firestrike, so I'm satisfied with that -- all on stock Bios/vBios, but with 3.4475 drivers.

4) Oh, speaking of which, I tried getting the OC'ing on the gpu to load on startup by placing shortcut of the settings (created through Inspector) but it doesn't work. Anyone know how to keep those settings between re-boots? Or should I just utilize a different program to do this?

...btw, running similar stress tests via XTU, my temps are 8-10C lower and I don't get any power limit throttling. Prime95 must really be a beast of a stress test! Does that make it the better cpu stress tester?

Thanks for any advice you can give!

 

Just a note, at stock, there is power throttling and temps are pretty high (~88 degrees)
Underclocked/undervolted, it results in no throttling AND temps that are ~25 degrees lower. There is no noticeable loss in performance under any situations I tested.
Here are my config options

 

1 - Power limit throttling is where the CPU runs into a TDP bottleneck. Aka it cannot draw enough watts to keep its clock speed, and thus throttles down.
2 - You would not thermally throttle until 95 degrees celcius on at least 1 core, so that's normal behaviour.
3 - No, that's not correct. As long as your CPU never crashes or BSODs your CPU is fine. People simply use a stress test to make sure it doesn't crash during heavy loads (like gaming), but you can have too low voltage etc and it'll crash under no load whatsoever, so just run an XTU stress test, see if it's stable, and use it. If you ever BSOD under load, you tweak settings a little bit more. That's what I've always done.
3a - I don't know. I've not used P95 because I've found it worthless to use, really. If I don't overstress a machine with streaming then I'll never overstress it.
3b - 5 minutes usually is fine, but as I said earlier, you can be fine for heavy load but crash under light/idle loads. So you'll just need to wait it out and see.
4 - You probably won't get the OC to boot at launch, but you can simply double click on it on your desktop (or click it from start menu) after booting and it should work. Other programs may apply at boot; I'm not sure personally.

Yes, XTU is a rather abuse-free test. I could easily get 10+ degrees extra rendering a video or livestreaming than XTU's bench or stress tests ever do. P95 and LinX Linpack are benches that demolish CPUs.

You have severely downclocked your CPU which likely results in the huge temp drop under load. undervolting doesn't get you -25 degrees alone. We've tried. You don't notice performance loss because your CPU is so low it doesn't matter, but under single- and dual-threaded apps, it clocks up significantly. But during such apps, it does not stress the CPU as much as when 4 threads are being used, so... explains the temps.

 

Right, that's almost a 1 GHz downclock from the max a 4720HQ can do on 4 cores (3.6 GHz), which is unacceptable to me