*OFFICIAL* Alienware M17x R3/R4 Owner's Lounge Thread

Yeah, if you don't want to go to the trouble of messing about with the battery, then the best thing to do is just to forget about it & use it as normal, then buy a new one in a two or three years.

Yep, discharging to 0% is bad for the battery too. I wouldn't ever run my laptop with the battery removed. I've enabled 'write caching' and turned off 'write-cache buffer flushing' for my hard drives (improves performance), but you need an uninterrupted power supply (battery) to safely enable those because unexpected power loss scenarios can result in data/system corruption otherwise.

 

As long as your temperatures are low, and you're not passing more than 100W through your card, then I don't think it will damage your card - certainly not in the short term. It might possibly reduce the overall life of your card though. It does this through silicon degradation. The higher the temperatures, and the higher the voltage, then the faster this will happen. So, really you want low temperatures when overvolting, which helps protect the silicon. Mine stays at 65degC so I feel fairly safe with my 1.05V (I game at 1.025V regularly).

 

write-cache, i got a nice question =).
Check this picture and see it in your language, it are the options you talk about.
http://puu.sh/5KakQ.png

I heard enabling that on a SSD this will make the SSD slower, now is this true of is this a mind thing i have here?


xD

Senshin

 

Haha, that's all Dutch to me! Haha, no, I know what you mean, that's the screen where I enabled those previous things I was talking about. I ticked both of those boxes to enable both of those options. I did some quick testing with different combinations of those boxes enabled, and I saw better performance with both boxes ticked (not a big difference). I did a bit of googling at the time to find out about it, and the consensus seemed a little mixed as to whether it helped or hindered performance. I'm happy with how I've set mine though. (I have an SSD too).

 

Alot on the internet find that Intel SSD's don't like flushing and like HALF the write performance, ive seen more people talking about it and seen benchmarks. just google intel SSD and write cache flushing and see the hits.
I think I will down some HD testing program, because half in write performance is is alot.
But not today, first enjoy my 680M and some work needs to be done =).

EDIT: did some more stressing on the gpu, it works great for now, still stable on 950/2000.

It's just how much Nvidia is better for screen tearing, now I can see it with my own eyes o.0.
On my AMD cards I always have some kind of tearing/ not good v-sync or whatever, it's just plays differently, input lag is also a few big ms bigger on AMD cards.

I think I hold on to nvidia cards the next 5 years, AMD ruined there reputation for me.



EDIT2: just installed nvidia experience, what a bloatware is that lol. delete xD

 

Yep, don't worry, I'm familiar with write cache flushing, as I said I've researched it before, and also tested it myself - as I said it doesn't have a negative effect on my particular system, I saw a slightly positive effect.

Yes, your GPU is overclocking well.

Geforce Experience is kind of useless for me, I can get better performance by tweaking the game settings myself, and given that my GPU is so heavily overclocked it doesn't take this into consideration - it just sees the 670MX and recommends on that basis, not the fact that it's overclocked by 87%! I still have it installed though, because I figure it's good for driver profile updates & to alert me of new drivers.

 

I am the kinda guy that just looks on the website some now and then and pick the drivers up, i hate automatic updates .
Not really my thing.
Seen your scores with the MX670, it's really nice to see and it shows the quality is, it really looks they just downclock bigger GPU's and the GPUsamples are great in general.

 

Yeah, NVidia often seem to underclock some of their GPU's to allow more difference & differentiation between the different GPUs. So, for instance I reckon they underclocked the 670MX artificially so that it wouldn't encroach on the 770M to be released in the next generation, and also to allow for a bigger gap to the 680M. I suppose the trick to getting the most bang for your buck it to choose those GPUs that they've artificially limited. (However, it was more luck than judgement that I chose the 670MX to be honest! But, it will make me think about these things more before I buy my next laptop or GPU).

 

Yes, It's sure luck but if i see how much people can overclock with a 680M anyway, it's just crazy.
I geuss that's why you pay also more for Nvidia in general, you really see it back in there quality and user experiences (you always have the failing GPU's somewhere xD whatever brand).

 

OC'ing is safe as long as you don't push it to hard and monitor the temperatures.

 

I don't like you say it's safe, I think you always will take a risk when doing this.....

There is a reason why they go out of the factory as they are.

 

I usually don't recommend overclocking the video card, it is not necessary. Some people simply love to get more out of theirs. I just want to recommend monitoring the temperatures and try it carefully. We cannot support any type of damage caused due to overclocking. The warranty would not cover it because it is not there from factory, we overclock processors from factory and not GPU's for the same reason. They cause issues if pushed too hard.

I did not mean to encourage you to do the overclocking, I meant to say that you can do it as long as you don't raise the clocks too much and make sure you monitor the temperatures.

Sorry about the misunderstanding.

 

@ robbo99999
can you send a screenshot of your bios..im unable to find the option to restrict the battery level.

 

Hi, I don't know how to do that unfortunately, but I just checked in my BIOS and it's called 'Charger Behaviour', and you set it to 'Disabled' if you want to stop the battery from charging. It's on the first or second page of the BIOS. I've got the R3.

 

I think I can fully say i am successful, tonight I runned bench marking in a loop on my 680M, like 6+ hours.
It's still working and I feel confident again =).


About SSD drive, i installed Samsung magician and !@$##@$!!!
This is good, I could enable Rapid mode for what it was worth.....

Check the screenshot: http://puu.sh/5L9FW.png

My write and read speeds more then doubled!! (more then 100% improvement from 500 to 1100mb per sec.....)
Tested it in more benchmarks, my scores in some programs tripled (200-300% faster!!)
And we are just talking about a Single 840pro SSD.

Tryed some games with big loading and some programs, you can really feel it, this is a really awesome mode (RAPID).
I would suggest to everyone with a 840 EVO or PRO (I don't know what types of SSD support it) download Samsung magician 4.3 and try some benchmarks!

If it doesn't work disable Performance caching in your RST driver from intel, because they don't like each other, but it worth it every penny!!

I would say to everyone for who it possible, gogogogogogo and try xD, this is some real @#$%!


Senshin

 

That's good Senshin, did you decide to overvolt at all, what did you decide on as a final everyday gaming overclock?

I've read a bit about RAPID for Samsung, the increases in performance seen in some benchmarks like CrystalDiskMark isn't translated into real world performance - at least not by the extent shown in the benchmarks. This is because it's only a 1GB RAM cache of frequently accessed info from the SSD. (Windows caches previously accessed data in RAM anyway - shown in Windows Task Manager as 'Cached' under the 'Physical Memory' section. A lot of people say that RAPID does seem to speed up their system slightly, and some reviews support that, although some reviews also show RAPID to slow down performance in some areas too. I say that if you're noticing a benefit of RAPID, then just keep using it!

 

No decided not to overvolt the next 1 2 months just to see how everything will go , going now on the same 950/2000 what gives me great performance, maybe up the memory a 100mhz more but I don't know about it, any thoughts?

About RAPID technology of samsung, all reviews i can find now are atleast 2 months and outdated seen my findings old and they updated it last time.
I think and i am reading some comments of other people they improved the performance alot the last update, (4 days ago released).

I compared my scores with other review sites of the RAPID technology, and I am crushing all numbers (100 - 300%), even the EVO's what doesnt make any sense.
Especially when bandwidth is used it goes so much harder.

 

Well, I run my memory about 50Mhz down from where I start to see artifacts - so I consider that I run my memory at the limit; I'd recommend you doing the same if you want to get the most performance out of it.

 

Well i couldnt resist myself and I overvolted my card to 1.025V. :'( xD
But he, my score is 8044points on 3dmark 1000/2000 clocks, it's hitting 70degrees now, more like 69 but ok.

This is looking real nice and kinda absurd o.0.

http://puu.sh/5Lryy.png
That's my score.


Senshin

 

That's good! Not absurd though, the 680M is a good card! You don't really need a 780M if you've got a 680M - the highest overclocked scores I've seen for a 780M are in the region of 10000 points in 3dMark11, so that's only 25% faster than what you have, which is not much of a big deal.

 

Got new clocks same voltage:

1010/2200: 3dmark11 = 8185 points and 72C max.
http://puu.sh/5LKOi.png
I still have 0 crashes and 0 artifacts, what is pretty impressive.

Gonna try some more later maybe.

 

Don't be shy about it, just keep going until it crashes, and then back it down a notch (or increase the voltage some more). It's not anymore complicated than that. Just make sure your temperatures are good. In fact, just post back when you've reached your highest overclock that you're happy with, I'd be interested in seeing how far you can go with it, not much point in posting at each setting you try along the way.

 

Yeah your right, but I just like to write down my thoughts here and how i am doing with my new videocard =).

Still running 1010/2200 stable and temp peaks @ 72 degrees, and stable max temp = 69 - 70.
It's idling around 51 - 57 degrees, the fan turns on and bring it back.

I tried 1020 but 3dmark died and my computer also, needed to turn if off xD, so i guess i need more voltage now but I think i should stay with this, a extra 0.025V is gonna give me 100mhz @ the most and alot of extra heat.
Maybe I will try it but not for now, just first checking how this gonna fare.

I check my thermal padding and some where minimal/not good contact.
So I changed some thickness here and there and used some shin itsu on some thermalpads to for sure close some gaps (not vram, they where more then good).

Temps are roughly the same but I got alot better feeling now because now i am sure it is all perfect in terms of cooling.

 

That sounds good. If you run 1010 core that is actually 1006Mhz, and running 1020 is actually 1019Mhz - it goes up in 13Mhz steps. An extra 25mv won't give you anything close to an extra 100Mhz, more like 40Mhz. So, it looks like you've pretty much reached your maximum - good job!

 

Hang-on, I didn't notice this the first time I read your posts, are you saying you put thermal grease/paste on your pads?! I remember I did that once when I didn't really know what I was doing. Since then I know that there's no benefit from putting thermal paste on thermal pads! Having said that, when I did that on my M1530 it didn't cause any problems. It's just not a done thing, and probably doesn't help any, whether or not it hinders I don't know.

 

This is very good grease and very thick, not really a liquid.
I used this to ensure good contact, whether i lose a small bit of cooling dont know, but they touch very good now and sure this is best.
I also felt with my hands if the hotspots turned warm in time.

Thank you for your concern.

Senshin

 

So robo99999999999999999, you made we worry so today i took of the back cover of my laptop and check the temps on the thermal padding.

What i do is this:
I cool down my laptop completely so it will get cold.
I will start up and run a benchmark asap, then i just feel with my hands if I feel the hotspots get warm quickly, like 5 - 10 sec or so maybe.
I checked the vram and other other parts ad they turn quickly very hot, what is good.

So robo9999, thx for me letting this check again and I think i will never touch the videocard again till it dies now .

Thank you very much for helping out the last days!!


Senshin

 

Yeah, no prob, that's OK. I don't really know if it does affect the cooling performance of the thermal pads if you put grease on them - I just thought I'd let you know it's not standard practice. Sounds like it'll be OK for you, and I did that on my M1530 & it was OK too.

 

I my head I thought/think it isnt wrong, stacking thermal pads is possible but a little drop of grease on top of the elektronics is bad with it?
To ensure good and proper contact? (the 3 big squares on the videocard i used some thermal grease).
Also in numbers the grease i use is a little bit better then the actual pads (if it matters).

I am just brainstorming, nothing more nothing less, I can be wrong, I can be right, I don't know.

Any thoughts?

 

I guess the only downside is that it increases the thickness slightly, so more material to transfer heat through, so temperatures might be a bit higher, probably not significant though. Hopefully the thermal paste won't degrade the thermal pad in any way. I think it's probably fine, but don't think it helps any either.

 

Finally finished up my desktop build!!! Just waiting on a graphics card now. Gonna try to play some league of legends on the igpu and see how it does though.

 

That all looks very good Bigtonyman, all very neat in there! I bet the airflow in the case will be good - seems like a lot of space in the case plus the positioning of the exhaust fans look good. Did you decide on a GPU (or multiple!)?

 

Nice laptop xD xD

But seriously, that looks awesome! what are the specs?

 

Gonna start with just a single 770 unfortunately as I've had to pay for some car parts, but I'm gonna upgrade to a 780 ti + a 1440p monitor when tax returns come in.

Processor: i-5 4670k@ 4.3ghz for now till I get a gpu to do some proper stress testing and benching
Case: Corsair 750d
Ram: 8 (2x4) gigs 2133 ram
mobo: Msi z87 mpower
Cooling: All in one cooler corsair h100i liquid cooler
Gpu: still on order: EVGA Gtx 770 Superclocked

 

Hi ppl
having problem with my 6990m , done a clean install a couple weeks back with the drivers from dell, everything running fine
install f1 2013 and started to play and had t2 gpu crashes , game goes black screen and on the desktop say amd driver crashed ans sucessull restarted , have to shut down the game and restat it , happen twice so far
any ideas or what is the latest safest driver to install??
help would be apreciated

 

I'm on the latest AMD beta driver on mine. I do have windows 8.1 and the igpu disabled though.

 

i updated to the latest one on the dell site and now the pc shuts down complety

 

what do you mean by completely?

 

it has done a few times , from what i can see and using GPUZ the gpu gets to about 106c and then the pc shuts down can my gpu be damaged?

 

I think you'll find it is shutting down to protect the GPU from damage. I don't know running temps of your GPU but I would not expect anything above mid to high 70'sC?

Take the base off (two screws under battery) and pop out the fans. You'll be able to see the heatsinks, they block real quickly and it may be as simple as a can of air to blow them out...

 

I kept a log of the temps and a screen shoot , going to call dell tomorrow because it's still under the warranty


Sent from my iPhone using Tapatalk

 

100°C sounds very high, would check the fans and gpu area for any dust before calling, a quick clean might be whats needed, although I know those cards do run relatively hot, shouldn't be over 100°C though. Won't void your warranty giving it a quick look/inspection anyways.

 

my laptop psu cables arc'd and almost fried my mobo. I got lucky. Now I'm stuck on a 90 watt power brick till I splice my old one back together.

Edit: I would also recommend cleaning out your fans. Even with the 460m fan I hit 80C max. 106C is way to damn hot.

 

What would cause the cables to arc on your psu? Do you mean something shorted out in your power adapter/brick? (Just wondering if there's anything I could do to prevent that happening with mine - thinking of protective measures!).

 

the wires on DC end just got worn down and the bare metal arced across to the ground wire. Just me being careless more than anything else.

 

I'm not that familiar with power adapter terminology, is the DC end the end that plugs into the laptop? How was it that you were careless - I just want to prevent that happening to me?

 

DC is the end that plugs into the laptop. and the wires just got stripped from being moved around and stuff to much.

 

Cheers bigtonyman, I'll make sure I'm not too rough with it!

 

I'm just lucky I happen to have someone I know who was willing to let me barrow a spare 90watt dell adapter for the time being. I can't game, but at least I can use it if I need to.

 

So i couldn't sleep cause was doing my head in the pc shutting down and getting to this temps
And guess what .... Have look at this pics



Booth were full of dust
Gave booth a good clean and also the fans now it's idling at 61c to 70c
Going to try to run a game now and see how it behaves
Thanks for all the advice , how could I forget about the dust doohhh

Sent from my iPhone using Tapatalk