S-XPS 1645 Throttling Info. and Updates

I agree 100%. Im just happy we know we have someone from dell whos listening and thier corperate.. i didnt believe it would ever happen

 

Bill,

It might be something to look into with the vent facing down on this system, I have been using my 1645 ever since December with TS+130w adaptor and doing a lot of gaming, I have found slight discoloration on the bottom left of my RGBLED Screen. The discoloration is in the matte finish of the screen behind the clear protective layer, and is noticeable with the laptop being off,

I do not know what the long term affects will be, I just know it has caused discoloration that can currently only be seen with the screen off. If the vent was angled down I’m sure this would not only help keep the system cooler, but protect the screen from any possible damage over time.

 

Thanks for the reply Bill.

If this is the case, the 1645 DOES already overheat due to the screen blocking the air vent when it is open. How would we all go about lobbying for a replacement vent cover?

 

I think saying "at no point" is a bit inaccurate. I believe we acknowledged the problem here and on the DCF thread UK Graham reference earlier. It is, however and unfortunately, true that we have dropped the ball on getting the information to our front line phone and chat reps. I would not be too harsh on them, as they are just going by the information we provide them. I could say more, but I'll just step up and take the hit for not getting more information to the front line and leave it at that.

E-Series is not a gaming rig in any way, and the recent thermal table changes addressed the issue of balancing usability, performance and thermal issues as best we could. We could easily eliminate throttling altogether on that platform if we wanted to, as the AC adapter is not the issue, but the question becomes how high do we want the skin temps to go? I'm sure if those users who feel the need to push a business notebook that hard really want to, they can use TS and get all they want from the platform. But honestly, maybe they bought the wrong system, to put it bluntly.

I think there was initially a perception we could get away with a BIOS-only fix, and in talks with a "certain NBR member" I let him know you guys might see the fix coming in stages. I passed along my feeling that a BIOS-only fix was not going to make users happy, and I think they took it to heart.

I plan on getting with you in detail offline at some point tomorrow if you have time. If you do not have time, I would really like to set up a time to speak to you about a few things. We realize it's not up to the user to be the test lab, but honestly, sometimes we drop the ball and sometimes we just can't test for every scenario. I think you and I both know it was a case of the former with 1645.

This is a very insightful post. I would almost guess you work at Dell.

 

AX, we're on it. Trust me. Engineering is working both issues in parallel. I hope to have similarly good news for the 1747 crowd, but there are a few details I'm not fully comfortable with that I really need to clear up on our end.

 

You didn't believe it would happen....

When I first read Dell-Bill B posts I thought he was the creation of some dissafected 14 year old sitting in his bedroom, tired of writing viruses, wondering how he could create the maximum amount of despair and anguish.

There's a part of me that still thinks it's too good to be true.

 

Japan will be looped into the adapter swap process. If you have any problems when the time comes, hit me in PM here, and I'll make sure we square you away.

 

Got it. I'm on it.

Dialing in now, folks. Back in a bit.

 

I just got my XPS 16 on Tuesday. Will this issue and possible solution to the issue extend the 21-day return period?

I obviously don't want to have the 21-day period lapse only to find out that the solution isn't one that I'm happy with so that I'm stuck with a computer that doesn't live up to what I had hoped when I ordered it.... I posted this already, but I was already set to return the computer. I'll wait a bit longer before making that determination if the return period is extended.

 

Thanks Bill, looking forward to hearing the outcome, and thank you.

Min

 

There's a story there, Graham, but I'll just take the hit for the lack of information on DCF and leave it at that. My apologies.

 

Good luck... we'll hear from you on the 'other side'

 

Finally! Someone understands my position! And now everyone say "hi" to Bill. No one, no one saw any hint of this with my post with just a ))? No one thought that might mean something? Fine, I'll give up trying to be mysterious. Sheesh.

All right, I'll be back tomorrow; I have an Organic Chemistry quiz to study for. :/

~Ibrahim~

 

Thanks for the update.


PS: you just got yourself your first japanese groupie

 

I had no idea!

Originally Posted by bfarmer View Post
Yes, Bill, mine flickers, too.

I'm one of those folks who had been authorized a return... IKJADOON's contact with a certain Dell representative gave me hope that there would eventually be a resolution to the problem.


Wink, wink on the certain Dell employee.
Glad to help the best I can.

 

Bill you been great, and thanks for chatting with us.

just a note, i dont mean to be rude or anything in any of my posts. you have no idea how much i appreciate you responding, and even admitting a little about dropping the ball. No one will ever agree 100% and im glad you are wanting to talk to unclewebb more, albeit a bit late

 

Yes Bill, the 1640 throttles. I have the 2nd fastest (fastest at the time) CPU you can get in a 1640, a t9800. I had no idea about the throttling until I read the 1645 threads and on a wim decided to try out the tests on my 1640. I thought for sure my system was immune because I game on it a lot. The thing is I game only one game and its 7 years old now. I emailed Unclewebb and provided a great deal of help and information on what was going on. I loaded up 3Dmark06, something more current and sure enough on full load of cpu and gpu, my 2.93ghz throttled to 3x and 6x, producing 1500mhz and 798mhz clocks.

I think the throttling issue is a 164X problem and not a 1645 problem.

 

@Bill

My computer's currently in "In Production" status, supposedly due to deliver on 2/22. Do you think it'll ship with the 130W adapter or will it still ship with the 90W?

 

I sent the first post by Bill B (Hi Bill!) to the rep who was dealing with my XPS return. I sent him an email and asked him if he had any information. I received a call about an hour later, and he confirmed that A06 was not the BIOS to fix throttling, that a BIOS was in the works (and would be for 3-4 weeks), and that power adapters would be provided.

I have cancelled my return for the moment, with anticipation that we'll be getting closer to a solution in the coming weeks.

I'm now faced with a new problem. Before boxing up the laptop, I did a format. Now I have to reinstall Windows

(On that note, I noted that at the end of the "loading install files" section, the system sits there for almost 5 minutes before continuing. I don't recall this happening the last time I reinstalled, which was on A02.)

 

When you point the finger at someone, you leave three fingers pointing back at yourself.....

 

I can thoroughly recommend installing to a native boot vhd instead of a real partition. Makes backing up / restoring the entire system as easy as copying a single file to an external usb drive.

 

Questions for Bill:

1) Why is Dell still selling this model? Shouldn't it have been removed weeks ago?

2) Is Michael Dell aware of the problems related to the release of this model?

3) "But honestly, maybe they bought the wrong system, to put it bluntly." To put it bluntly, isn't it fraudulent to sell computers with CPUs that are prevented from operating as Intel intended? Dell charges MORE for the additional CPU power, do they not? Why isn't this fraud (since the CPU is so aggressively marketed at Dell.com)?

4) What procedures will be put in place to assure proper design and testing of future models?

 

Of course....

 

IS anyone else having this problem? I talked to the rep and agreed he'd send me a 130w adapter and i also asked for the X-Fi cd that i requested to be added to the package and that was A MONTH AND A HALF ago, i just checked the status page and it indicated that it was canceled -_-. Why would dell promise me to send those items that ive requested plus even gave me a order number but failed to have it sent for more than a month......
I emailed and called them but they couldn't do anything about it, all they said was I got to wait for it...

 

Concerning the e6400

That stuck in my craw too.

 

As someone who currently has an xps 1645 in the production phase with an EDD at the end of this month will it be possible for me to call and have a 130w power adapter included in the order so as not to have to deal with calling in after I actually receive the laptop and then waiting for another shipment?

I'm also curious about the question another member posted in regards to the fact that an 820 + RGBLED may require more than even the 130w adapter.

 

Another question for Bill. Sorry if this has already been asked. Has there been significant testing to show that 130W is sufficient to supply the power needs when the laptop is running flat out? (i.e. approaching the TDP cutouts of the CPU and GPU, and perhaps stressing the HD if needs be)

 

From Todd at the Dell Community site:

Re: DELL Studio XPS 16 1645 i7 power problem
04 Feb 2010 09:52PM

Okay, the call just finished. A lot of interesting data and I'm still working my way through it. Saw some interesting data on their tests using Crysis at different settings levels, with power draws, FPS, etc. Man, how do I get the job playing these games to test hardware?

I'm still putting together the details on what we can and can't release right now, but I will say that the BIOS that they have being validated right now makes a big difference. (and no, my saying that doesn't mean that its a bios only fix with no adapters, so don't worry). We don't have an ETA at this point on when that bios revision will pass validation. It could sail through really quickly and work perfectly or they could find a major problem with the algorithms and feel the need to start over again. When it comes to software development that just is a fact of life. What seems to work perfectly well in theory doesn't end up working the way you planned when it gets tested.

They showed us some of their test data on CPU usage and spiking. I'm paraphrasing but they basically said "the current algorithm is not optimized to maximize processor usage. There are issues in the BIOS that, after the system spikes from heavy use it was taking too long to reset to normal power states". (Like I've said before, I'm no electrical engineer). They were taling about different CPU performance states (P-State and T-State), and a lot of that was over my head.

They are pushing to get that validation completed as quickly as possible but there are hurdles that need to be overcome. Since the systems affected all have different configurations possible they're working to validate the different needs. That means you may see a BIOS come out for one model before the others. That won't mean the others won't get updates.

It's nearly 10pm here. I'm going to get together with Brad and Bill tomorrow so we can go over all this information we got tonight and make sure we all understand everything so there aren't any communication errors moving forward.

 

I think I'm the first with a question for Unclewebb lol

1. I know you mentioned that CPU cycles are wasted as an inevitable side effect of Throttlestop doing battle with the dark side (A.K.A the chokehold BIOS ) Since there a pending official fix, does this mean we'll see even better performance than with TS? This is just a curiosity of mine.

On a side note. Kudos to Bill for being straightforward and reopening lines of communication.

 

OK. Meeting's over. Let me take this one first. We have to look at the 1640 with a different engineering team. I asked the 1647 lead to take this issue to the core team. Once they start looking at it, we will pass along updates as they become available.

 

nevermind I understood wrong sorry

 

Coolio. Great to hear. If I'm lucky I'll be able to (hopefully) see this all resolved when I get back. (And my model should be here by then.) I hadn't actually expected there to be any information so soon considering it is after normal 'work-hours' so my thanks to you guys for taking the time to get the news to us as soon as possible.

 

So, what are you doing with your e6400 when it throttles? I have had one for a month or two now, and it has not throttled yet.

 

1) Neither of those questions are within the scope of my role to answer.
2) Not 100% sure. Guessing what Mr. Dell knows or does not know is not a wise practice on my part.
3) If the system gets too hot in your usage model, ie pushing the CPU at or near its peak Intel-rated capabilities for a period of time, then I think a business class notebook is not what that user should be using. Is your E-Series throttling on you? If you feel defrauded, get with me in PM, give me your service tag for your E-Series, and I'll be glad to go over your options.
4) I can't really tell you exactly what procedures will be put in place, but I can say anecdotally that no one likes to deal with issues that cause so much customer angst and damage to our brand, so I'm pretty sure whatever procedures are put in place will be the best foreseeable improvement we can come up with. I'm simply not privvy to that kind of information.

 

Fen, throttlestop the way it works as unclewebb has explained it to me is very efficient. All it is is fighting to set a value in the processor. Which is very quick to do, even if you do it every so milliseconds. But at 2ghz a cpu core is clocking 2 000 000 000 clocks per second, your not going to notice some of those missing. You will and can notice like i did kind of, is the dead time between the bios setting a throttling state, and throttlestop reversing it, which appeared as a slight "stutter". Those little "blips" i mentioned earlier and can see in more data option checked. Only game i could ever notice that though was dragon age and that was barely. So there shouldnt be much difference.

 

Bill,

I have been debating over ordering the XPS 16 for a few days and was wondering if you would recommend waiting on the issues to be corrected before ordering. I was also wondering why many of the options for it have been removed from the order build pages. Blu-Ray was the biggest item that I noticed was removed the past few days.

Thanks for your input.

 

RGBLED doesn't seem to be available anymore either.

 

Wow! RGBLED was available earlier today.

 

I checked. The RGBLED is still available for the 1640 and 1647, just not the 1645 it seems.

Heh. And did anyone notice they took down the Dell SXPS 13? I'm guessing it got discontinued. . .

 

Bill,
Thanks for your feedback today.. It's about time. Dell needs more open communication like this, as it helps relieve the pressure. They need to focus on getting to it sooner. If you have forum posts growing by multiple pages per day, you need to get in front of it sooner. Most people understand that it takes time to develop a solution, but if dell is not out there in front of the issue and talking to the customer, the pressure and disgust only gets worse.

I do not understand why you seem to go through these issues so often. This is the second time in a row for me, the first time was resolved several months later (XPS 700 desktop), and it was that reason I decided to give dell another chance. I don't know where I'll be the next time, this helps, but twice burned is one time too many. I wish you luck and again thanks for sharing some info. Keep it up.

-romills

P.S. I do not like what Dell has done to their forums, they were better a couple years ago, but still not great, now they are slow and hard to navigate. Notebook review is a much better example of a forum done well. In general, all your community sites suffer for slow response and poor organization. I know Dell keeps redesigning them, they should just find a site with a good design and emulate it, and stop trying to be fancy and unique for something they don't need to sell.

 

Want to optimize your HD experience?

Choose RGBLED for 100% color gamut, 16.9 aspect ratio and full 1080p.
Full HD Widescreen 15.6 inch WLED LCD (1920x1080) W/2.0 MP [Included in Price]

thats funny.

 

Hi Bill B. I just got a call (5 min ago) from a Dell rep stating that there was a new bios that was available for my xps 1645 that wasn't on the website and that he could help me install it. Is this A07 already? It would seem awfully soon...

 

The site just changed again. Monochrome is the only option for the 1645 now. Does it draw less power?

 

Now for my promised update and answers to the questions I had gathered before the call. Studio 1747 owners, we talked about your throttling issues as well and will share details on your threads. We expect to see the final fix for you sooner than the 1645/1647 owners.

Firstly, let me talk about what I am not prepared or will not talk about first.

Killawatt findings- We use different tools and methods to measure total power draw the system takes from the AC adapter, so I will not be addressing any questions or counterpoints based on findings derived from Killawatt meters.

Japanese replacement AC adapters- there are some factors involving 3-prong to 2-prong adapters that we currently do not stock, so you folks may be waiting for the additional time it takes to design and get line of sight to parts supply. Sorry about that.

Now for what I can talk about.

We will get into our benchmark findings in a little more detail on Direct2Dell at a later date. We plan to pull some specific posts from this thread and address them there in detail.
The A-rev of BIOS A07 went to test today. Once it passes testing, we will web post it. We will begin the process of sending out 130W adapters at that time. We will take NBR and DCF members to PM and get yours to you, so you do not need to call in. We will let other users know when the Direct2Dell post goes live how we will be getting them their adapters. It may or may not be a proactive replacement. The adapters will be the slim models and not the bricks. General sentiment is it should be proactive, if at all logistically possible. We do not have an ETA for factory cut in for A07 and the new adapter (in other words, for the orders that have not shipped yet). We will post details on the ETA for factory cut in when we get them.

The questions I had from the NBR community that I was able to get answers to:
1. Is this for all regions? yes. Depending on regulatory requirements per country, the timelines may vary.
2. What about 1647? We will need time to look at closer. Component and chipset factors are different and need a seperate investigation. I am sure the timeline will be a little quicker than it was for 1645.
3. What about customers who bought their own adapters? This really was not a question for the engineers to answer. Among the non-engineers, there were 2 different opinions. We will discuss the question tomorrow and get back to you on this one.
4. Will the bios officially allow the use of 150W adapters too, since laptops with RGBLED and i7-820 (like mine) are more power demanding of a WLED+i7-720? Or 130W will be the only choice? 130W is the only adapter needed. You can use a 150W if you like, but it should make no difference.
5. Can the customers have better release notes on bios and driver updates? The engineer could only answer for this platform, but he assured us he will take this question to the team and do his best to make sure the notes are more informative for the next BIOS release.

One more thing: I brought up the exhaust vent question. This will have to be opened as a seperate service request. I will get that in the works. We had the same problem with Studio XPS 1340. We wound up designing a new exhaust vent cover that fixed the problem by pointing the exhaust down. On my test 1340, I just popped the cover off and ran around with a gaping hole there until I got the new cover. It worked as a temporary workaround, but it was a little goofy looking.

 

Generally speaking, we are facing some challenges with some specific parts and extended lead times. I'd be guessing if I had to address them part by part.

 

Thank you sir!

 

Thanks Atl, I was being overly optimistic I suppose lol. Have you already shipped your 1645 back? If not are you planning on keeping it?

 

Whoa!! Super interested in getting that rep's name. If he's got a copy of A07 and he's offering it to you without putting you under NDA, I am more than super interested. Please get with me in PM if it's actually A07 instead of A06.

 

Thanks a ton for the updates. now i can go to bed and sleep better!
Was waiting for the updates. Much appreciated...

 

Sweet. Now, I can have some peace on my vacation. Meaning I'M OUT. Thank you very much again Bill for your assistance and availability and responsiveness to those of us here. (Although I can only really speak for myself.) Shalom!