S-XPS 1645 AC Power Throttle Issue Investigation

Seems the 130W adapter works wonders for us man, Along with throttlestop . However I think people with RGBLED the 820QM [the higher end 1645's] should try this with the 150W adapter and see how it goes.

Also atlstlang? how would you rate battery charge while idle/general use as opposed to charging while gaming? Does it even charge while gaming?

 

Don't you mean "won't"?

 

Might have been sarcasm?

 

Dell employees are tools, it’s sad none of them can think outside the box, and I can’t STAND their middle man support, GOD PLEASE let me just deal with real engineers so I can show them how to do their job correctly.

Hey dell, do we need to buy you a kill-a-watt and ship it to you so you can see how much you botched the Clock Modulation configuration to lock this system at 90W?

This thread needs to be closed; I’m getting sick of reading it every day.

The Studio XPS 1645's current BIOS is designed to work only with a 90W adaptor due to the fact that it is using Clock Modulation on the i7 to maintain a maximum wattage draw of no more than 90W from the AC adaptor, NO MATTER WHICH ADAPTOR YOU USE.

The only way around this is to run throttle stop to disable Clock Modulation and then you will be able to fully use the power from a 130W or greater adaptor. Under gaming this laptop does not consume more than 110W. Under stress tests it will push 147W, your system will never be run at a daily basis under stress tests loads.

IT DOES NOT MATTER what AC Adaptor you use on this system, all dell adaptors are 19.5 VOLTS, the only difference is the Amperage, the higher number allows higher Watts per hour to be pulled into the system(65,90,130,150,210, ect). If you use any other adaptor that is not 19.5 Volts you will damage your laptop.

We use 130W adaptors nonstop on a lot of our systems at my job; most do not even need it. We work on Dell laptops at my job, dell PAYS US to do repair work on all of our systems. I know from experience. Save your house from burning down and get a 130w adaptor and use it at all times on this laptop. I so cannot wait until someone’s house burns down and the cause is because of dells 90W adaptor in use on this laptop.

Hook a 130W adaptor up and go into you BIOS setup, on the 1st page down at the bottom, guess what it says? 130W!

All dell needs to do to FIX THIS ISSUE, is configure a Clock Modulation table for the 130W adaptor that will allow the system to pull 130W from a 130W adaptor when the BIOS detects the 130W adaptor plugged into the system, if it detects a 90W then it can use the Clock Modulation for the 90W adaptor and this wont screw the systems up for people that don’t care about their max performance and only people that call and raise hell about their overpriced system performing worse than a cheaper ACER or ASUS laptop can have a 130W adaptor shipped out to replace the 90W and dell can start shipping out 130W adaptors standard.

This means, WE ONLY NEED A BIOS UPDATE to fix this issue, the bios handles the Clock Modulation of the i7 on this system.

There, that settles it, PLEASE LOCK THE THREAD ;p

/Rant

I’m now off to enjoy Dragon Age with Clock Modulation DISABLED on my 130W AC Adaptor that currently pulls 110W while playing and my processor runs at 68c and video card at 81c
THANK YOU unclewebb, you did what dell failed to do, they should pay you the millions they are wasting on their (moron) support calls and testing that lead no where

 

Just testing a profile issue.

 

They seem to think the alienware as to be a performer and was still investigating, i saw that hopefully they would treat the same with the xps 16 and not write it off so fast. but more i read it, i can see how thats worrying some as well..

chookz as far as i can tell, the battery doesnt charge when loaded. Probably a good thing though, as it generates extra heat. I dunno, i dont really expect to to charge while then, but i guess some people do?

edit: siphen i feel yeah, and uncleweb is awesome for his utility. Still would rather them fix it officially though, but regardless we have a fix for ourselves just in case.

 

too bad i sent my 1645 back for refund, i would have deffinatly donated to uncleweb's awsomeness! you people should send him cash! pst it will keep him happy enough to do more awsomeness!

 

Yeah they could afford to be a lot more transparent about their engineering and investigations, it would go a long way in terms of public image too. Community liasons or whatever doesn't cut it, they still just treat engineering as a black box, "We'll get back to you when we have more information". Let the engineers post about what's going on. The Engineering Windows 7 blog is a great example of how to do this right. You could put together a textbook on OS engineering from that blog, verbose posts from engineers that don't spare us the technical details.

 

Sigh, Thanks again to UncleWebb I'll donate sumn for u on Xmas day Thanks to all the guys who did testing and gave feedback etc etc. So unless more throttlestop issues come up or Dell says something i think is issue has been sorted

/investigation

 

That is assuming, of course, that the power distribution circuitry of the laptop can handle the increased amperage. While we suspect it, we don't know it for sure.

 

I was under the impression the i7 clock (FSB) speed was 133MHz....It should be more?

 

Some people are reporting random crashes. Is the throttle problem releated to other people's 1645 crashing?

 

BTW, Core i7 actually doesn't have a FSB anymore, but I understand what you're saying. The multiplier actually doesn't go below 7x, but they also turn on "clock modulation" that periodically turns the clock on and off and slows the processor down even further. So the "effective clock rate" just happens to be 133 mhz in the worst case with the lowest multiplier and most aggressive clock modulation.

So the processor ends up behaving like the multipler is 1x, due to the clock modulation.

 

ok, i just seen bus speed in cpuz64 and assumed fsb.
in the i7 white paper it says input frequency 133MHz under internal clocking... so i got it now.

 

Not to be rude, but who exactly are you asking and how exactly are you testing this?

Because I can verify without a shadow of a doubt that using a 130W adapter with ThrottleStop does work quite well and that the issue is resolved.

 

Yeah, I think Brad meant just using the 130W alone doesn't fix the issue.

 

Just because it works now, doesn't mean you will not have problems 6 months later.

As Deathwalking pointed out, this does not mean that the power distribution circuitry can handle the increased amperage. Dell is getting a reputation for cutting corners and I'm starting to think they cutting them way too tight for my future business. I will wait for dell to have an official solution, as I don't know their circuit ratings. I'm rather disappointed in Dell's handling of this and other issues, they seem much to patient in letting customers stew over this.

I was checking SW compile times today using it, just to see if I could get throttling, when first powered up, it was fine, but later runs it was limiting it to 10 and 11x overall. The main difference was how much time the machine had been sitting at idle vs just powered up. It's still much faster than by older dual Core laptop, but I'm disappointed that the longer it was used, the more it throttled the system down.


_____________
Dell Studio XPS 16|Core i7 720QM|4GB DDR3-1333MHz|ATI 4670 1GB
16" RGBLED|500GB HD|DVD+/-RW|Intel 5300 (A/B/G/N)
Bluetooth 370|2-year (Warranty+Accidental)|Windows 7 64-bit

 

Perhaps. I also could have worded my post slightly differently, instead of saying that the problem is resolved.

But there's a reason I posted like that and that I dislike PR posts like Brad's. Initial tests show that it doesn't work? Who has he been talking to? What kind of tests?

It just doesn't inspire much confidence when a company comes and says stuff like that when you can throw a stone here and literally find 10 people who have actual, verifiable proof that the 130W adapter DOES work, at least in the short term. That's what I have a beef with the most in his post.

There's 1 of 2 things going on here. Either Dell didn't provide the proper circuitry in this laptop that can reliably handle more than 90W, but the laptop itself needs more, in which case absolute shame on them. What are we paying our money for?

Or, the laptop actually can handle 130W just fine (which I still suspect is the case) and Dell is either too reluctant to have to shell out and give everyone new adapters, or they are really that clueless.

 

im willing to try throttle stop. im sober today. so umm if you are willing to send a 150w adapter to me then i will try it thats on you. i live in near T. so that areas close. message me if you want me to cuz i got the 820qm and oh yeah its in my sig. k.. by the way, why on earth... please tell me would dell make a 4gb + 1gb configuration? the 2+2 and 4+4 are the best in performance, but the 4+2 and the 4+1 dont make any sense besides.. no it still doesnt make any sense unless someone wants to find an identical 4gb and replace the 2 later on. i put my external hard drive in the freezer and it still doesnt work. i think the pcb is damaged and soldiering is needed. my laptop is still gone. i need to stop doing things or else im gon end up in the hospital on morphine. come to think of it morphine sounds good right about now.

 

Wow.

That was one of the most coherent posts I have read yet from you, razr

 

mine crashes... only under throttling conditions or about to be throttling conditions. Throttlestop seems to be the fix to this problem. i havent talked to dell yet and if the people from dell are reading this thread they are probly biting thier nails.

Kay so we have what seems to be a fix...

the best i see dell doing at this point is cave in to people asking for higher watt per hour bricks and not directly adress this issue any further.

on a side note, a very important side note... very very

the battery says 85Watt per hour
the ac is 90 watt per hour.

how are people getting better performance from the battery than the ac?

i think i have thottling my brain right now.

i found a perminent fix to the throttling issue... simply get a piece of metal, jam 1 end into your laptop ac power hole and 1 end into an outlet that has a current flow. your screen might explode but trust me, there will be no more throttling on the cpu, gpu, ram, keyboard, flickering screens, light, normal air smell, or your house.

 

The 85Whr rating on the battery has nothing to do with the wattage it can put out - it's just a rating.

The AC adapter isn't rated in Whr, just Watts.

You are confusing these two things.

 

haha i agree. At least my head doesnt hurt this time after reading it

I will say though, regardless of if dell fix's it or not. Or regardless if its not "spec'ed" for anything greater then a 90w (which im confident like ash it can). I find it hard to settle for lack luster performance found in 900 dollar laptops.

Like sleey says. Watt/hours means the battery can sustain 85 watts for one hour. They measure that for battery as its an indication of how much time you can spend on battery.

 

Yes and it seems there isn't a cap on the power for battery OR its higher than 90W, but there is a Cap from power from AC. So the laptop can draw more power from the battery than the adapter.

 

Never heard this before. Is ash particularly confident?

When I get home on Friday, I'm going to open up my laptop, while it's running on both battery and AC and use my multimeter to poke around and find out what components are only involved in AC power and not battery power. I then intend to look up the parts and see what they're rated for.

On that note, would someone with a high-res camera get me a shot of the area I marked on the photo? Greater than 2000 pixels wide would be great, but I'll take what I can get. Thanks!

 

Has anyone confirmed, that using thermal paste on the cpu to improve heat issues will not void the warranty?

cause arctic silver 5 would do a good job.

If you've done this, was opening the 1645 and applying it difficult or fairly straight forward?

Thanks

 

Hello I hate to rehash something that has already been asked, but I am an avid gamer and I'm experiencing these problems with the laptop I recently received. I'm very anxious to test ThrottleStop but I didn't quite understand what was meant by emailing the RealTemp About.. box and I appear to be unable to send PM's. Could anyone clarify for me how I can contact unclewebb?

 

oh and also, can anyone confirm if a universal ac adapter works similarly for this issue or is it important to get one from dell?

 

uncleweb is the author of realtemp so email him and you can get the beta of throttlestop. Make sure you have a 130watt or greater powersupply and not the stock 90watt. As the program will cause more power to be allowed in than the 90w can provide.

throttlestop stops the modulation done to lower the power. Unfortunatly clock modulation is horrible as the benchmarks i just ran at the request of uncleweb really opens the eyes. It effects on performance horrible. At full modulation its a 10 fold decrease in performance, and thats using a 12x-17x multiplier, where when throttling its usually 7x.

edit: i would get a dell powersupply, but not from dell. I got my 130w from traingle laptops for 20bucks. its been said they are sold out at the moment. but look around!

 

excellent, do you think general electronics retailers would carry dell supplies? and i hate to be a bother but could you or someone link to the appropriate realtemp website? i cant find a clear address and id hate to email the wrong person.
also, based on what you said, does that mean that throttlestop prevents throttling which helps the situation, but the laptop is still not reaching its full potential because clock modulation itself reduces performance?

 

modulation is a way dell limits the power of the laptop to 90watts. But its been proven this laptop needs at least 100watts-110watts to operate like its supposed too. So a 130watt adapter would still be modulated/throttled to 90watts. But combined with throttlestop which disables dells modulation/throttling it allows the full power to be drawn through the 130watt adapter.

heres realtemp page. His email is in the readme filie of the program. at http://www.techpowerup.com/realtemp/

as for the adapter i dunno would have to search. As long as its the 19.5 volts and can handle the extra power it should work.

 

Real_Temp ATyahoo.ca

replace the word AT with the appropriate symbol @ and send me an email and I will send you ThrottleStop

 

ahhh i think im finally close to understanding! hah. but thank you for the info, friend. much appreciated

and thank you unclewebb! you folk are lifesavers

 

Could you confirm this using the RealTemp program?

 

I'm going to do some C2D testing today, and will send my results to unclewebb

 

Don't use anything that has not been designed to power a computer. Because what I think a "universal adapter" is, isn't able to modulate the output power and will deliver a constant amperage, which could really dammage the computer.
As a general advise, don't use anything not coming from Dell.

 

Just sent you an email. Thanks for all your hard work on this.

Jim

 

Great work from everybody here.
Now we need to get this news out to the wider media for dell to fix this issue and prevent it happening again for another dell/other manufacturer of laptop.
There is nothing better than bad publicity and revenue loss to change a companies view on the matter.
This is huge.

What boggles my mind is that they are still selling this laptop.

I guess money is the bottom line....NOTHING else matters.

Could a legal action be brought against them.

PS No response from dell yet!

 

Yeah, unfortunately I'm not really expecting dell to fix this issue, well not properly at least.

 

Well, Dell only bothered to release a new BIOS for the Latitude customers a few weeks back after several sites announced what was being called ThrottleGate. I have my doubts that Dell would react the same for an issue affecting a smaller portion of their consumer line, but regardless it seems that the BIOS didn't even resolve the Latitude problem.

In short, Dell is really dropping the ball all around.

 

Now if only we knew when...

 

If only we knew this wasn't another LIE from DELL

 

The dark side I sense in you, young padawan. Much anger, you have.

 

Here is the rest of the quote from the Dell guy. I am happy to hear that Dell acknowledges that the Alienware M15x issue is different than the SXPS 16 issue in that our problem occurs under normal usage scenarios when using the adapter.

"Thanks for dropping by. I just want to assure you that a fix for the 1645 is still coming. I also wanted to assure you that I am a Dell employee, not an agency or middleman. I'm also the Dell employee authorized to speak here regarding this issue.

I do want to make this very clear to everyone: the issues that XPS 1645 owners are experiencing are in no way related to the behavior you're seeing with the M15x. I can not stress this enough.

How are they different? Well, I'm glad you asked. They have different causes. They also occur in different scenarios, the 1645 issue occuring at normal CPU loads when the system is plugged in, the M15x throttling occurring only when all 4 cores experience a sustained 100% utilization, not a real world usage scenario."​

Now, in deference to Razr, I realize that the above post may make too much sense. Accordingly, I offer the following:

Man i hope that D fix this prob i like da screen but fuzz bunnies get under there i can just clean it why do they use these ram configs when u know they shld be different.. I am returning my machine but i like it i will say that it is a good machine but where did that pink monkey go he was right here a sec ago i think i will find him.

Stream of baked consciousness mimic operation concluded.

JimmyB

 

Thanks! I was gonna say tl;dr, but then that last paragraph made complete sense.

 

It seems to me my 720 cpu will only run at 1.6G when all 4 core at 100% on adapter or battery. For single core, it will run at 2.5g on battery, but 1.6g on adapter. The LCD seems also brighter on battery.

 

The i7-720 is limited to a maximum multiplier of 13 when all 4 cores or 8 threads are fully loaded. This is the default 12 multiplier plus 1 for turbo boost. If this exceeds the built in power limits of this CPU then turbo boost is automatically disabled by Intel (not Dell) so the CPU can continue to work within it's designed limits.

12 x 133 MHz = 1.6 GHz

When fully loaded, if you are getting 1.6 GHz then it is working like Intel intended it too. If it is dropping below this value at full load then that is a sign that the Dell bios is in charge of how fast and hot your CPU should run. To get rid of that Dell limitation, buy a Dell 130 watt adapter and use ThrottleStop. Problem solved.

 

I updated the DELL forum post as follows, please PM me if you want anything more to be added or deleted.
I also added a poll, so please cast your voice.

"Update

Here is our group findings so far

1. The power conservative BIOS forces the system to throttle to 7x when the power consumption is around 90 watts due to one or more of the following reasons:

a. Limitation of the power supply capability

b. Limitation of the motherboard to accept more than 90 watts (This reason is from DELL, but we suspect it)

c. To increase the battery life

d. To limit heat generation

obviously DELL should be blamed for poor design if it is a, b, or d.

2. Thanks to unclewebb there is a software now to unlock the throttle, this one will almost certainly put the 90 watt charger to rest or fire.

3. Due to the power limitation, it looks like the people with the RGBLED are most affected by the problem due to the added power consumption.

4. DELL is aware of the problem, but according to them it only affects a small percentage of people that put the system on full performance.

5. DELL keeps saying that the 130 watts adapter is not the solution of the problem, which is quite true as it needs both an adapter and BIOS update.

6. After a long wait, DELL acknowledged that there is a problem and that they will address it soon without announcing a time frame.

7. We as a group lacks the trust in DELL fixing the problem to our satisfaction.

I think most of our group will agree on the above.

With all the above said, i am quite happy that mine was returned back to DELL after a frustrating experience with DELL level 1, 2, and 3 technical support. This doesn't represent the other members as most of them are happy with their notebook but expect action from DELL."

 

It seems that the same "throttling" problem affects also the New Dell Studio 17; a Dell 130 (or more) watt adapter + ThrottleStop solve the problem anyway?
Wich AC adapter is compatible for this NoteBook? The 150 Watt adapter from the Alienware M15x it's compatible? The 210 Watt adapter from the Precision M6500 it's compatible? They charge also the battery?

 

Well I can't wait for Dell to fix this. I'm so glad that someone actually acknowledged the problem and that a fix will come. This lappy has a lot of power under the hood and it would be a shame that it never reached its' potential.