*OFFICIAL* Alienware 17 Ranger (2013) Owner's Lounge Thread

Test also your maximum overclock with 3DMark Sky Diver. This test draws more watts than Fire Strike (about 20-25w more). Test also with 3Dmark11. If ypu have the payd version try all benchmark in loop.

 

OK , I will get 3DMark downloading. I just tested Furmark and it runs for about 5 seconds before the GPU automatically throttles, looks like that was the protection Prema was talking about.

 

Just finished with Sky Diver. 1228/[email protected] pulls ~215W throughout with 190W low and 230W peak. 1339/[email protected] pulls 240W right at the start and the laptop immediately shuts the charger off.

 

What was the wattage during Furmark(gpu shall throttle in Furmark)? Is psu registered as 330W in bios? It is long time since I tested now but I know that I have been on over 255w in 3Dmark11 and Sky Diver. My oc gtx780m certainly draws less watts than your gtx980m from mmx slot when gtx980m is overclocked properly... It's strange that you can not pass 240W without laptop turns off in benchtests. How high do you think you can overclock GPU without touching the CPU speed during these tests?

 

I can't get an accurate number in furmark as it throttles instantly.

Yes the laptop sees the charger as a 330W in the BIOS.

The laptop doesn't turn off, only the charger does. Unplugging and then plugging it back in turns it back on briefly until it pulls 240W again and turns off.

I can overclock the GPU to 1339MHz if I disable my CPU turbo boost. It keeps my wattage right around 230W. At 1228MHZ and the CPU turbo enabled I see around the same 230W.

I am going to try my 240W charger tomorrow night when I bring it home from work. I swear I've seen 250W+ as well but I can't get it to work now...

 

I'm quite sure that I've seen a post or several where one in this forum run + 285 or even over 290w on its AW17R1. Is quite sure of this.

 

That's possible; inefficiency is at least 10% and the closer to max., the higher the loss (adapter heats up). Measuring from the socket you have to take into account 240W + loss, so 270W is not unexpected. The issue with Dell is the 'id adapter' and downclock bit in their bios. It's been there forever ...

Since it is validated as 330W, that means the trigger isn't properly implemented in the bios (latest version?). Either that, or the adapter is defective, self-terminating under load.

 

I just confirmed my 240W does the exact same thing. 240W and off...

So is there anything I can do? Both the 330W and 240W are shut off at exactly 240W which leads me to think it must be something to do with the laptop. I was hoping I would be able to take the ID chip out the 240W and put it in an adapter but since the 240W doesn't work I don't think the ID chip will help with the 330W.

 

How long can you run with 240W power before psu turn off?

 

Literally as soon as it goes past 240W it trips it off. So in game like 2 minutes maybe, depends on what my OC/voltage is at. At stock volts it usually won't shut off but sometimes it goes over 240W and does.

 

How high can you overclock gpu before reaching problems? Have you connected something else on your laptop? Such as a cooler or similar ?

 

1338/[email protected], vRAM speed might have been a bit higher/lower but it didn't seem to effect power draw very much. Only a mouse is connected, I have a single SSD and a single hard drive, no optical drive.

 

Indeed; an artificial trigger.

Swapping the id chip won't help here and guess there's no bios update that fixes this? Probably not an official one, at least; this ' safety feature' will keep their RMAs down, after all ... and sales up, too.

Since the adapters are switched off by the bios (or ec), it should be possible to mod the adapter to ignore the 'off' signal, yet still keep it id'ed as a Dell adapter. One way to make any adapter id as 'Dell' is to desolder its id chip and relocate it to the motherboard. That way all adapters are approved. And because there's only the single pin for the id+off signal that means there's no more 'off' signal either. Well ... it's still there, only it will not reach the adapter anymore.

Either that or splice-in a switch to the id wire; close on boot and open after boot. Might use a relay instead, that way it's automatic. Of course, that's not an option if bios does post-POST id checks as well.

 

Thank you for the info, I'll have to look into that. I'm on the latest BIOS, A14, and there won't be another from Dell, at least not one that changes anything. I'd love to flash a custom one but they decided that wasn't allowed either, gotta love these "features" they keep adding to make sure us poor stupid customers can't do something that isn't approved.

 

i wants to upgrade my CPU and be done with it but i haz no monies

i was thinking of the 4810MQ or a 4910MQ if i can find either for a reasonable price. looking to buy in mid to late september.

 

There is almost no point in upgrading from i7-4700 to i7-4810mq. Minimum i7-4900mq.

 

so a 4910MQ

 

If you select a 4900 or 4910 does not really matter. Choose the one you can obtain cheapest. It has proved that the i7-4930mx is a better overclocker than the latest revision. The same could be also on Mq series.

 

getting flickering when playing mad max

 

Hi Guys,

Can anyone help me? I've recently purchased a 980m and installed it in my R1. The problem is I am unable to get Intel HD + Geforce 352.48 to work. It hangs everytime. When I install it in non-optimus mode, running 3dmark11 causes the card to only use 2d clocks. By the way, I am using pure UEFI win 8.1 pro.

 

Just use 353.06.... They work perfectly... other drivers are crap... Also use DDU (Display Driver Uninstaller) to remove all NVIDIA and Intel drivers, install Intel and restart and then install NVIDIA...

 

353.06 has the stupid chrome TDR bug...353.38 or 353.49 should have that bug fixed.

 

Thank you @TomJGX and @thegh0sts ,

It is now working. However, my core clock stays at 286Mhz max when running 3dmark

 

No such proiblems for me...

Is this even happening in games or just 3D mark? Did you use DDU like I told you to? Check in games.. If this happening all time, your GPU is on the way out... After Windows 10 install, this is what happened to my 970M... The damm clocks stayed on 135/450 which was lowest 2D state even in games... I hope this isn't the case for you

 

@TomJGX

Yeah even in games like battlefield. I did use DDU. The core is as 286 while memory is at 1262. I was thinking a vbios unlock could solve this.

Update:
Good: Was able to flash the vbios, it is already at 1200/2500.
Bad: Switched to dedicated in UEFI + legacy enabled, now getting 8 beeps haha

 

legacy should be disabled.

 

Yup, learned that the hard way. Was able to recover it now by plugging the 780m in, however, fn+f5 (dedicated) results in a green screen before post and when it loads into Windows the screen is flickering. Reinstalling and flashing the bios did not work. LCD EDID is okay. Screen refresh rate is 64hz. I think this should go to the service center.

 

yay or nay for external audio cards for laptops?

 

I'd say yay if you've got the extra cash and a decent set of headphones & you're an audiophile - I have one. On board sound for laptops is pretty good anyway - there's no real need for an external soundcard really. I feel that the sound is crisper & contains more detail with my external sound card.

 

i was wondering if an external device would solve my audio popping issue.

 

Maybe, DPC latency you'll want to have a look at before you invest in an external sound card.

 

I had this issue before, it got resolved by downloading the latest rst from the intel website. It had something to do with caching drive. Can you try this one out?

 

gimme a few mins.

EDIT: new version of IRST doesn't help. so i may just uninstall it to see what happens cos i don't run my drives in RAID configuration anyway and one less thing to load.

EDIT2: the following seem to be using a lot ...

dxgkrnl.sys
sdbus.sys
USBPORT.SYS

EDIT3: Narrowed it down to dxgkrnl.sys as the top.

 

still clueless as to how to remove the DPC latency.

EDIT: hmmm.....as soon as i turn off HWInfo64 Latency Mon stabilizes out and gives the all clear. as soon as i load up HWInfo it becomes unstable. to make sure it wasn't a fluke i closed HWInfo and again the latency Mon stabilizes.

Is there a way to stop this while I still have HWinfo active? the issue is that it might have to do with the EC sensor and the EC sensor data is needed to be able to control the CPU and GPU fans individually.

 

Try the latest version Hwinfo64 if you have not already done so.
http://www.hwinfo.com/news.php

 

i did and the DPC latencies still exist. I am not sure if there's a setting to fix this.

 

Write about the problem in Hwinfo forum. Martin might help.
http://www.hwinfo.com/forum/

 

i've also posted on techinferno to see if there's an answer. This wouldn't be so much of an annoyance if they allowed us full fan control.

 

Sounds like it's using SMM/SMI:

Try disabling 'SW SMI' and see if EC fan control still functions without that:

 

Nope, as soon as it's disabled I can't access the fan control tables (that little fan button is no longer visible).

Enabling EC support removes my ability to control the fans individually.

 

Has run some 3DMark 11 tests. Several of tests with wattage from 248 and up to 265W with ease. I can not understand that your 330W psu turns off immediately after you pass 240W in bench tests. My old 240W psu was very unstable. Could turn off at 220-225w while other times it worked well over 240W in bench tests. Can you test again with 3Dmark11? Test 1 in 3DMark 11 is the one that uses the most watts.

 

maybe it's the PSU at fault??

also, looks like there's nothing i can do about the DPC latency caused by HWinfo

 

It will turn off right at 240W, I'm pretty sure the laptop is sending the shutdown command or whatever. Maybe I have a different motherboard revision on mine that has a hard restriction, it would make sense if the power delivery was only designed to handle 240W that it would disable the charger rather than damaging anything. I've actually got a Clevo arriving tomorrow, I'm selling the AW but I'll still be around.

You could install physical switches like I did, then you wouldn't have to run HWiNFO to control your fans.

 

You known the board's part number? It should be visible in cpu-z

Sent from my HTC One using Tapatalk

 

I don't, I'll check when before I pull the SSD tonight.

 

I am just wondering if it is a fault specific to a part number or of something else has happened.

Sent from my HTC One using Tapatalk

 

Yeah it's really odd, I have no idea why it does it but I guess for 99% of the time it's fine. I'm just hoping my new 980M clocks as well as the one in the AW.

 

If you can get it between 1200-1300 then you should be ok

Sent from my HTC One using Tapatalk

EDIT: so if i disable reading from the EC sensor the DPC latencies will be less....hopefully.

EDIT 2: if you load up CPU-Z in the mainboard tab it should show your part number. mine is 068R5X rev. A03

 

Same motherboard as mine.

 

so if i cut the PWM wire the fans go full speed?