Clevo Overclocker's Lounge

Maybe check in the Alienware 17 Overclocker's Lounge. The one that is not in the turdbook corral sub-forum.

 

I looked, but I wanted a test near launch and there's just too many posts even if I search for 880M firestrike or something. Also, I wanted a non-overclocked run.

 

I'm glad that my old lady isn't that outdated!

OP3

Have to the tweak it further!

Gesendet von meinem ONEPLUS A3003 mit Tapatalk

 

Well I'm sure you have modded the firmware as well...

 

Maybe Prema want to go into smartphones too...
Android Premamod sounds fun!

 

some tweaking and magic later still 4.5Ghz CPU ~1975/4600 1v 1070N, what was about the Max-Q *cough*

Edit: well i ****ed even up in this firestrike run, Ram was at 2400 CL17 instead of 3100 CL15 lol

 

@Scerate glad you like it. Already looking promising!

Not yet! Only unboxed the HTC U11 yesterday, installed 3DMark and hit run.

 

This forum be like:


"Not yet" you say?

 

I'm just really impressed my little shield tablet is still getting OS/security updates great portable classic doom device.

 

How are you liking it so far?
Im thinking of getting one as well, but my HTC 10 is still going strong and i am afraid of loosing the awesome DAC thats in it. Even considering waiting things out for the iPhone 8.

 

Making consumer products more durable and easier to repair-www.europarl.europa.eu This should apply for soldered BGA filth. A small step in the right direction. But it must be worse before it gets better.

"The EU Commission, member states and producers should take measures to ensure consumers can enjoy durable, high-quality products that can be repaired and upgraded." @Mr. Fox

"77% of EU consumers would prefer to repair goods, rather than buy new ones"

Maybe post your Q in the AW17R1(5) graphics upgrade thread as well. Maybe some new or old AW 17R1 owners with 880M can see it and help yoo.

 

I usually never get a new mobile unless the old one dies. Well the old one is the M7 and it just won't die. So when the BTC surged and I saw a shop offering the U11 I decided that skipping 3 generations was enough...

Absolutely love the U11, everything is top notch especially speakers and cameras really stand out. The old school screen dimensions make sure that videos fill the entire screen.
Was a bit sceptical about the whole glass design, but fell in love with the 'amazing silver' version the moment I unboxed it.
Will test some overclocking at one point...

 

I'm on the Samsung S5 still. Personally, since so close, I'm waiting for 10nm and 7nm chips, as well as the lpddr4 advances on the way. Literally one to two more years and the phones will have a beastly upgrade with a good power reduction. Now, Samsung is waiting for 7nm because of EUV, but TSMC already has 12 companies committed to the process and has accepted tapeouts. To put this in context, TSMC has already created a demo ARM chip for HPC running 4GHz. Their process says 30% increase in performance over 14nm or 60% power reduction. GloFo's IBM based process says 40% transistor performance or 65% power reduction. So, good things coming down the chute!

Sent from my SM-G900P using Tapatalk

 

My Galaxy S5 was great, was working fine, no issues at all. Then I just had to upgrade to the S6 120gb version and there's been issues ever since. The computer part is fine, it's the phone that blows. I've tried different settings, bringing it back to the shop etc. Now my wife has the newest S7 or S8 and hers is worse than mine. It's a great hand held computer, just a terrible phone. I have been told that it's my provider "Sprint" and not the phone itself. That might be the case but my S5 from Sprint worked great.

 

I'm also with Sprint and it works great. So I can back you up on what you've seen. Samsung took a couple steps back, but many phones were stagnant the last couple years. Meanwhile, solid state batteries will be in 2019 models, so another thing to look forward to (phones not blowing up)!

Sent from my SM-G900P using Tapatalk

 

A few pages back I posted something about how my thermal paste contact looked. Could someone take a look and get back to me? Since I used the kryonaut to apply evenly to the surface before laying down the heat sink in going to assume that judging by the pictures it was spread out well.


Sent from my iPhone using Tapatalk

 

You put it on a bit too thick, but it's non-conductive so it won't be a problem, other than a bit messy.

You shouldn't remove the heatsinks after you have applied the paste as you have done, as that has now "ruined" the application / spread that you were trying to accomplish.

If you are looping back to get an opinion on your paste job because temps are too high again, it's because you pulled it apart and tried to put it back together - the paste job which might have been fine the first press (application) was ruined for the 2nd press (application).

Next time you can have faith based on this view you looked at; done once for verification that your method is ok.

Now you "should" clean it all off and re-apply it so that it spreads evenly sealed from air - you'll likely get air bubbles or coverage gaps now by trying to put it back together after it's been spread out and pulled apart - notice the thin spots - you worked hard to keep those from happening when you applied it.

I hope you have enough paste to do it again...

If not, you can try to salvage the paste and redo it using the existing paste. I've had to do this a few times in the past rescuing paste applications gone wrong.

Use a non-absorptive card to scrape it off, and collect it. Hopefully with that recollected paste plus what you have left in the tube, you'll have enough to cover all the chips. Try to keep all the "fluff" out of the recollected paste, keep it clean.

Note, you can use less, the thickness on the chip should be full coverage but thin enough to hardly squeeze out when compressed with the heatsink, some will come out, but you had a lot more than I usually see.

Good luck

 

I'm not 100% sure but it looks like you're using too much paste, and I'd try a Skylake IHS. That should help with the processor fit.

 

All too much. This ??? is enough for +2 processors. Same for the graphics as well.
I have done some pressure testing today with some very thin medical gel (Basiron ) Used this instead of thermal paste for messure the pressure. You can see this small amount of cream was more than enough for cover the cpu.

Spoiler: LAPPING

Spoiler: If anybody wonder how IHS looks after being exposed to Liquid metal for 1.5 years. Before and after cleaning.

 

This is the only time I'm going to say this to another dude... but ... "Beautiful spread man"

 

So to answer questions, the pictures were from when I removed it from the last application. I didn't just pull it apart and throw it back on lol.

I cleaned everything off and reapplied with less thermal compound. I was following the method with the kryonaut where it was spread out evenly and judging by videos I had seen I went a bit thinner this time around.


Sent from my iPhone using Tapatalk

 

@clayton006 I just noticed your main computer. That's no joke. It must be a BEAST.

 

Great, so how did it work out for you? Temps upon first test, a few days later?

 

Looks like bumping up the 1070's TDP boosted its scores in 3DMark to the same range as the desktop variant - very cool.

 

Temps are about the same if not a bit lower on the CPU thanks to the shim. The temps were never horrible in the first place just looking to drop them as much as possible.

 

Thank you. It is a beast and it runs cool in my basement. I'll have to evaluate where the higher end Core-i9 CPUs land and see what my next build will be.

 

Probably a fair bit hotter but they will munch those multi threaded workloads. Not interested in thread ripper at all?

 

http://www.3dmark.com/fs/13076731

 

Decided to conduct 10 runs of Fire Strike Ultra with different TDP values (ranged from 75W to 260W) on the GTX 1070 to investigate how TDP affects performance under heavy load.
Environment temperature was set at 72F, max fan on, and each run waited the core idled below 30C.
CPU clocked at 4.6GHz, GPU core added 189MHz, and memory added 450MHz during the 10 runs:

 

Have you TDP bumped the 1080s? desktop TDP is 180-217W.

interestingly though all mobile GPUs are set to throttle at 54 Celcius while desktop is set at 60 C.

 

Good Morning.

And desktop TDP is 250W depending on the brand.

 

eh?

 

Meaning you have missed way too much.

http://www.3dmark.com/search#/?mode...0K&gpuName=NVIDIA GeForce GTX 1080 (Notebook)

http://www.3dmark.com/search#/?mode...0K&gpuName=NVIDIA GeForce GTX 1080 (Notebook)

 

i don't frequent this thread much so TL: DR yes or no?

 

Yes
Hes running the Prema Alpha vBIOS.

 

still waiting for prema to update the bios for the DM1.

 

Probably going to be waiting for a long time

 

With the @Prema test vBIOS, pulled roughly 250W per GPU on the Fire Strike run above that you asked about.

 

Awesome thanks so much for doing this!

 

No problem . Are you planning to mod your 1070s as well?

 

Yep. One day I'll get there. I doubt it will be on this P370EM platform tho

 

I'd like if you can run standard voltages/speed/TDP limits and try to undervolt as far as possible while keeping said TDP limits. For example @Scerate doesn't need to cross 1v to hit 1960MHz on his 1070N, I'm interested to see how much of a difference say a locked 1860MHz can get between 75W through 250W limits while using something like 0.94v or something (whatever's around the lowest stable value for you).

 

I will follow up my previous post above #13123 With some more info!! Sorry if this post became long, but I wanted to do it this way. It's importent for me.

This will be a big Thanks to one of the very good people here in the forum. Mr. @bloodhawk who was so generous that he would give me a gift.
He knew that it is not easy to get untraditional parts here home. He knew I wanted a Bitspower IHS for my 6700K, although I have always had good decent processor temps due my interest in modding my hardware for lowest possible processor temp.

I use what I have of resources(parts) here where I live, but this is not always easy to obtain in Norway. He sent the gift (Bitspower IHS as I showed in my tests abowe) with an expensive shipping. But he did not mention that he also laying a 7700K next to it as the main gift. Thus you all know now, why I wanted to post this <Thank you> to this nice good person. + the temp results I got from my modding on the gift.

It is most likely possible geting even better max cpu temp, but as you can see, the temp results is good enough for me now. I will probably try a second round later if I can obtain a new heatsink. But we have the Clevo heatsink lottery, and you know how easy to get one from the lottery where I live, HaHa

I want to say as I said to Bloodhawk in PM (I'm still shocked that someone here would give me such a gift), so hereby a BIG Thanks to you brother @bloodhawk one-of-a-kind A very good person. In all ways!!

Temp results with 47x in Cinebench R15. I mean this is an ok cpu temp AND 2C degrees between coldest vs. hottest core and max 57C is a HELL lot better than the TRIPOD MESS from Dellienware. They are still struggling with the nice Tripod mess, now 8 months after their newest BGA Jokebook's was first released.



49x and temp maxed out 72C. With 69C on coldest core. Almost impossible with equal temp on all 4 cores @hmscott Deffective? Shouldn't all cores keep the same temp?


https://valid.x86.fr/cx5c9d

 

I'll continue on DM KBL Mod the moment you get a Killer Card for testing...

@Scerate may no longer need to undervolt.

 

I'll chuck in some cash to get it going

 

Woke up to two nice surprises, congrats to @Papusan and thanks from me to @bloodhawk for sending Papusan such a gift too ^^

And next to @Prema who is now making me dying to know what he did and how he did it

And finally, the thing I came to be happy about, is that I have worked with the heads at Notebookcheck.net and reworked the article I wrote. By "re-worked", what I really mean is, journalistic quips were cut down and I've added over 50% more volume, with references and deeper (but still easy enough to understand, I hope) explanations and definitions, and as far as I've heard a Nvidia tech PR even claimed to have enjoyed going through it

I hope this time nobody gets the vibe that I hate on the plain existence of thinner notebooks or those who want them; it was not my intention whatsoever and I've worked very hard to make sure it doesn't come across that way to the best of my current ability. Here it is for whoever wants to give it a read again!

 

As long as I am not adversely affected by their existence, I don't care. Sadly, I think all of us have been adversely affected by their existence. It has resulted in the dumbing down of everything else, so from that angle, a lot of hate vibes would be in order and appropriate for the lowest common denominators in laptop land.

 

People can choose as they like. The problems I have with notebooks like Razer's are that they do NOT put the work in. The notebooks I find an essential waste of money. But the cooling on the Zephyrus is very good and I am simply saddened that notebooks like yours or mine don't have such attention given to them.

I have said in the past that if a notebook can, without mods (even if need be purchased from a specific vendor, like Aorus from GentechPC with their free Liquid Ultra, or P7 and P8 clevos from HID/Obsidian) hold its stock performance without overheating, then it gets a pass. The Zephyrus does that. Thus my issues are not with "it" but with how the 1080MQ is marketed and priced, and that I do feel the entire "Max-Q" line's performance could have been done. That's all.

The point about the impeccable cooling however... I did need to ask why Nvidia having guidelines for their cards suddenly makes an ODM care. They should have cared the whole time, or Nvidia should have made them care. The thicker GL502VS and GL702VS notebooks cool 1070Ns worse than the Zephyrus cools a 1080MQ, while being louder. To me, that looks atrocious. Not that the Zephyrus does not deserve its cooling prowess, but that the GL models *do*.

Sent from my OnePlus 1 using a coconut

 

The only time anything gets done right is when it gets measured. But, the measuring has to be done by an entity that has the ability to hold them accountable by directly affect their bottom line. Absent that, we have the garbage that doesn't work right and OEMs and ODMs that don't care as long as they make money.

But, it still adversely affects the rest of us that don't want anything to do with those wimpy sissy-boy turdbooks. Doesn't make any difference to me whether they do a good job on that trash or not if the stuff that is good enough to be of interest to me is left half-assed and broken because they are focused only on compliance with the Green Goblin's rules for the filthbooks.

 

If this editorial article is going to fly and nobody can change it right now I think the launch of Mini-Q is a good thing. OEMs definitely care about what NBC puts out (though I may have just blacklisted myself from 98% of all OEMs) and if end users read and think what I say makes sense, I think there is a chance some good can come out of it. Imagine, Clevo, Mister Heatsink Lottery themselves, who doesn't care that OEMs are saying their heatsinks are not good enough, is putting out one of these new machines. I haven't laughed so hard in weeks. I wonder how many of their notebooks will fail the specs and if they'll have to pull the line from the market wasting R&D expenses or if they finally told Foxcomm to fix the bloody heatsinks.

Either way, what this opens up is possibly that Nvidia might consider denying companies like Razer who force-throttle and overheat their machines the ability to sell any newer notebook GPUs. I'm excited to see where the market goes from now.

Sent from my OnePlus 1 using a coconut