S-XPS 1645 Throttling Info. and Updates

what exactly is 'stepping' in colour gradients? I have a blu ray on mine, but didn't understand what you mean by that.

 

Its when the transition from a bright source to a dark source, which should other wise be smooth, shows steps from one intensity to another. This monitor is by far the best laptop monitor I've seen. But believe me a lot of desktop monitors are still better, otho this probably has as huge of a color gamut as your going to find.

Unfortunately we will probably have to wait for OLED's for perfection. This is still a TN panel, I'd take an S-IPS any day, or even a P-MVA panel over this. I could only imagine what a P-MVA or S-IPS could do with an RGB backlight.

 

I don't think they would. And probably why it won't be until Monday: Saturday and Sunday aren't standard business days. (I know, seems a bit ridiculous.)

 

If RBGLED doesn't show SEC5448 or whatever:

That happened to me, but I definitely have the RGBLED. Restart/sleep/hibernate/close the screen. That should fix it and let it report it as SEC5448 again.

~Ibrahim~

P.S. Can we keep monitor issues and EVERYTHING ELSE NOT RELATED TO THROTTLING in the Owner's Lounge, please? Where they belong??

 

Proof. If this is a TN panel then it doesn't have 100% color gamut support as per the specification.

 

"Even though the color gamut is much better than any other laptop we've tested, we do need to contain our enthusiasm somewhat. At its heart, this is still a TN panel, which means the overall quality can't match PVA, MVA, or IPS panels. Dell - or Seiko Epson if you prefer, as they're the panel manufacturer - has taken the first step by offering an improved color gamut on a laptop display. We're still a little baffled as to why no one seems to be interested in making new PVA or IPS laptop panels, as the combination of a high gamut backlight with an IPS panel would be ideal. This is still a great display, but when I want to do some serious image editing I'm still going back to my S-IPS desktop LCD." :

http://www.anandtech.com/mobile/showdoc.aspx?i=3540&p=12 Anand said this is a TN panel.

Why wouldn't a TN panel displa 100% of the color gamut, I dont see your reasoning/facts behing this assumption/truth.

 

The SXPS 16 Owners Lounge

 

Because TN panels are 6-bit and only have enough physical address space for 262144 colors; thus wouldn't quality under the Adobe 100% gamut color moniker, as Dell has advertised in the specification.

 

q-freaking-ft

aka:

MOVE YOUR PANEL/MONITOR DISCUSSION TO THE OWNER'S LOUNGE. Ahem.

 

Lol. Did it work? <_<

 

What SlyNine said, and also not just between light and dark, but on colour transitions (from pastel blue to pastel yellow via pastel green, say). To see the effect, load a colour picture into your favourite paint program (TheGimp, say) and reduce the number of colours from millions to thousands, or even just hundreds.

 

Moved to PM.

 

Sorry man, I wont say anymore on that.

 

Guys, I am writing this review having 5 calls with Dell technical hardware support. After spending 12 hours with these guys on phone, I am realizing I have been sold piece of junk.
I am having CPU throttling issues (Whole system freezes, CPU clock speed becomes 266 MHz if AC adapter is plugged in and CPU is utilized more than 80% for 2-3 minutes)
After spending $1500, I am realizing $500 laptop from Walmart will perform better. Whats the use of this so called 'premium laptop' if it performs at 266 MHz every now and then.

Dell is not accepting the return of this juck laptop. They say, they know about this issue, but do not know, when its going to be fixed.

Many users are already discussing their helplessness and anger on this issue on many forums. (google "dell xps 1640 cpu throttling" and you will see links to many forums)

I wonder, why XPS1640 and XPS 1645 were released in market, without dell doing some basic load testing ???

Dell should recall XPS 1640 and 1645 ASAP.

 

They have to accept a return. I would fill out a complaint on the BBB as the goods you sold are defective currently with no resolution planned or in sight.

E-mail [email protected]

 

I would be more curious on the 1640, and any testing someone can do using throttlestop to monitor. I assume their the same type of throttling and reason. But would be interesting yet to see how severe it is in comparison to the 1645. I can probably include a lot of that info to this main page if theres a willing person to provide the info.

 

The 1640 uses the same sort of throttling scheme. Drop the multiplier as low as it can go, enable SLFM (Super Low Frequency Mode) which quietly drops the internal bus speed in half and then when the user isn't looking, start using more and more clock modulation until the thing is down to a crawl. At this point, a Wally Mart netbook would look like a super computer compared to a well throttled XPS 1640.

maniabhinav: If you can, post a screen shot of ThrottleStop when testing. CPU-Z sometimes gets confused when clock modulation is slowing things down internally.

The BBB complaints seem to be the only thing that Dell is taking seriously. Highly recommended.

 

That would have been true a while ago, but not since Direct2Dell has been updated with that response a few days ago:

"In working with the engineering team, here’s what they told me: the Studio XPS 1645 and Studio 17 laptops are designed with performance in mind, and we are investigating improvements in the CPU throttling algorithms so that the systems will be performing at or close to maximum wattage more consistently than they do today. Such changes to the current state machine implementation can be made through the BIOS, and we are currently working on a test version. We will publish a release date for the test BIOS shortly – please bear with us a little longer while we qualify the mods to get the best possible performance for you."

I'm all for getting the BBB involved and putting extra pressure on Dell, but watch how you word your compliant.

@Slynine

Thanks, I appreciate it. This thread just needs to be a bit more focused, that's all.

~Ibrahim~

 

Its time to contact the BBB.

 

I would wait until Monday at least. Considering from what I'm hearing (unless incorrectly) that's when there is supposed to be a 'fix' or at least more information is going to be released?

I suppose part of the problems with throttling have only arisen until now, because (1 the people here have been making the issue known, and (2 the original specs for the 1640 probably didn't use as much wattage until they incorporated newer/more powerful parts. Therefore the problem is only being identified now. (I'm sure you guys already figured this out of course. . .) On the bright side, as/if things escalate/broaden this could ensure a better market security for this problem not occurring in future problems and quite likely security protocols so it doesn't happen/be allowed again. (Okay! Done rambling.)

 

Or just contact your credit card company and file a dispute. If Dell doesn't come up with a fix, even after the return period, you can still file a dispute. That will defintely get Dell's attention since they are charged the amount back, and will have to provide proof that they delivered what was stated on the item description.

 

I assume this question has already been asked, but i have just bought an XPS and will the situation ever be rectified by a fix ? or will DELL never sort it and I will have a laptop with these problems forever ?

I only ask as if this it true I will cancle the order.

 

Honestly, from what I've heard of past experiences involving other models, no. Although it seems in this case the 'gears are moving' and DELL might have a fix in a matter of weeks. It is rather in the air. But according to Ibrahim (see quotes above) there seems to be some concrete stuff to give you reason to believe there will be a fix on its way.

My advice is to not cancel the order if you can live with the throttling issues. (For now.) Otherwise if you do, and want to buy it after all this mess is over you'll have to of course wait again for it to be configured and shipped. (Not to mention there might be an influx of purchasers who've been waiting for this issue to be fixed themselves.) But, hey? Sending a message that 'I can't purchase this machine without the security that it will run at its optimal performance' is a good way to send a message. . . There's what we know, so I'll let you weight the odds. (And if I missed/was mistaken about anything someone can always correct me.)

 

Thanks for all the advice I think im going to keep the system however will the problem be ressolved by a simple update download somewere down the line ? or will there have to be some kind of hardware update ?

I understand you guys dont have the answers only speculations. But what the hell I had to ask

 

If you read that statement on direct2dell, its not looking great and one of the main reasons i didnt feel bad about sending mine back. It says theyre tweaking the bios to make the throttling less severe, but if they had a proper adapter you wouldnt have to tweak how the bios throttles, as it wouldnt never have to throttle in the first place.

I believe there will be a significant increase in usability, as some of their algorithms were horrible. But throttlestop will probably still be needed, in order to get the best performance.

Unless they have something else planned, but the A05 bios was a definite indication they were doing just as they say.

 

I dont know I mean everythin has problems when they come out granted they differ in severity and the XPS has been out a while. How ever im a firm believer in not jumping the gun. I have a 1 year warrenty as standard and im sure many of you have even more. That said if they do not fix the problem within in that year I am within right to send it back as tey have sold me a faulty product, Soo thats what im going to do

Perdire, do you have yourself a XPS?

 

There's a very interestng read over on laptopmag dot com.

Please see link below, and all of you who are having this problem, please add your comments at the bottom of the page as they are looking for feedback from owners. Perhaps this will start the ball rolling with generating media interest...

http://blog.laptopmag.com/dell-issu...7-with-core-i7-core-i5-now-default-cpu-choice

 

Had a good laugh with this write up you linked.......

The article reads in part ....

"When we reviewed the Dell Studio 17 with Intel’s Core i7 processor (aka Dell Studio 1747) in November, we gave the notebook a 4-star rating due to its blazing performance and relatively affordable price. However, we noted two major problems we experienced with our review unit. First, the initial system we received stopped booting (and made several beeping noises) just after testing had been completed and, second, we noticed that the exterior of the chassis got hot, with the underside of the system registering a troubling 112-degrees Fahrenheit after just 15 minutes of playing a Hulu video.

How does one give a blazing 4 star rating with two major problems ?

Wanting to purchase this or the XPS 1645 a few weeks ago, I am so glad that I came across this forum first. I would never give Dell a break and keep a system, in hopes that there will be a fix. Nor would I allow them to send one to me after reading this very long thread. What are people on this thread thinking when some of you say, " I will wait it out before sending it back and maybe Dell will have a fix in a few days".

The only way to send a clear message to Dell is to cancel or return. Then and only then, they may get the picture.

Good luck guys. My hope in now on the Sony Z coming out

 

@SR45

Well, like they said in the review, they (incorrectly) thought it to be an isolated problem. :/

@atlstang

I've forwarded this email to my Dell guys to see if I can confirm this, but jacketberry (who I quoted) seems pretty sure. He went on to say

So...right. I can't say much, but the Dell guys seemed pretty happy at some new developments. What were they? I don't know, but could they maybe be related to what jacketberry is saying?

Sorry, I'm not trying to rub it in your face after you returned it, but just for people who still have their machines.

 

Hi guys, well I bit the bullet and thought I’d test the laptop for the sake of the community here and also to satisfy my own curiosity!
Please note: my testing was done purely from a photo-editing point of view, as far as I know – this does NOT tax the GPU in an yway – well at least the programs I used do not anyway.

Test Rig: SPXS 1645, Windows 7 64-Bit Professional, i7 620 CPU, 7200RPM HD, 4GB DDR3 RAM, RGBLED Screen at max brightness, 90w AC Adaptor
Programs Used: Adobe Photoshop Lightroom 2.6, Canon Digital Photo Professional (DPP) 3.7
Test Procedure: I took 48 random image files (CR2 Raw Files taken on a Canon 30D SLR) and saved them to a folder on my desktop. Total collection size is 385MB. I then added this collection to Lightroom and did the same in DPP.

To begin testing, I set off Lightroom to perform 7 batch conversions from RAW to JPEG. This was on the same set of images into the same folder therefore each image was being given a unique name. I chose to do 7 batches for two reasons: a] it would take longer to do and therefore give me enough time to run a Throttlestop log, and b] I wanted to ensure the CPU really was being taxed. (Even on only 2 batch conversions running, the CPU Meter Gadget was showing like 50% usage - more conversions pushed this into the 90% region)
At the same time these 7 batch conversions were in Lightroom, I opened up DPP and set it to also start batch converting a duplicate set of RAWS in another folder. This was to ensure that if for whatever reason Lightroom was not maximising the use of CPU, then DPP would use up whatever CPU resources where left.
For anyone interested – all of that only took less than 5 minutes to complete! Now bare in mind how many disk operations it is also making saving fairly large JPEG images to disk at such a high rate!

Ok anyway, with both programs running, I opened up Throttlestop (Throttlestop disabled) and set it to start logging. My results were quite interesting. What I discovered is that it STILL throttles on AC power, however on battery it is fine. I’ll leave it to you guys to interpret the results more fully, but with TS running you can see when more throttling kicks in on AC power.

See the log file I have linked to: My Throttlestop Log Here

Comments on log:
14:20:00 – Started log on AC Power
14:21:00 – Switched to battery
14:22:00 – Switched to AC power
14:23:00 – Switched to battery

Interpretations from log: note the slight dip in multiplier when I switch to battery – but then note how within 20 seconds or so on both occasions the multiplier clocks up and then stays well above 12 (hitting 13) the whole time until I switch to AC again where the highest multiplier is ~10-11 (dipping as low as 8).

I know these drops aren’t as severe as what you gamers are experiencing, but it still clearly highlights one obvious fact: there is throttling on AC power whereas running off battery gives better performance!!!

Any questions - please ask!

Manny

 

Attach it to your post (i.e. host it here on NBR).

 

hi guys i have been watching this thread for a long time. my xps1645 will be arriving tommorow. at some times i have decided against buying this system. But my bad luck in india 1645 is the only notebook with i7 and a good graphics. Hope dell fixes this issue ASAP.

nice informations regarding the throttling.. Keep it up guys and special thanks to unclewebb for his program which i will be using after my system arrives. :B (Sorry for my bad english)

 

Sorry - i didn't see that was possible. I've just created a fileden account and uploaded it. I've edited my last post with my results!

 

Your log seems ok. Your clock modulation is 100% all the time, your Multiplier are above 11. Maybe slightly throttling, but it is usable I think.
For people experience severe throttling, the multiplier will completely locked on 7 or 8 and clock modulation constantly reachs below 50%.

 

Yes i agree, only you can clearly see the average figures are higher on battery than AC. This should not be the case should it?

 

I know this is whats been told to jackertberry, but it was from a rep again even if a lil higher up. And how have weve been told so many different stories by these reps before. I even think someone contacted the same department shortly after and got a whole different story.. Unless they keep repeating what they told jackert, i have big doubts still.

a multiplier of 9-10 is only 75-80% of what the cpu should be running at 100% load. Multiplier throttling is far more common then modulation, as it does this first. a few multipliers is a pretty substantial performance loss, about 8% per multiplier with all 4 cores maxed. And gets even more insane on single/dual core programs throttling. Such as games, when they should be at 16-19x, theyre at 9x over a 50% decrease but still no modulation. anything below 12 isnt good, to me that log is no where near okay.

 

I second that atlstang. When you see the new i7-620 Dual Core able to run Prime95 with an average multiplier of 21.7, suddenly multipliers of 10 or less is looking really pathetic. Throttling is throttling. No one should be happy giving up 25% to 40% of their computers performance because Dell didn't use a 130 watt adapter. Other manufacturers use 120 watt adapters and up. Dell's next solution needs to include something better than the current 90 watt adapter. 90 watts is just not enough.

Can someone run 4 threads of Prime95 to see what multiplier you get? It's not much over 12 is it? Even in a multi threaded application, it looks like the new dual cores are going to be able to almost equal the performance of these Quads and in single threaded applications it's no contest; the Dual cores will kill a Quad, especially in gaming when the GPUs are the same.

 

Does anyone know what adapter is best ? I looked at the start of the thread and saw 2 130W and a 150W, which is better ? and am I risking breaking my computer by using them ?

 

I'd like to say that that is a more than just a little higher than your regular representative you get on the phone. We'll find out if it's true in a couple of days, though, I imagine...

~Ibrahim~

 

hopefully it will be soon. But i wasnt talking about the normal phone reps. Users have reported early on from the supposed higher ups through the same methods. And they gave some wacky stories at the time. I hope that is the case though and 130w adapters all around, but im still highly doubtful until further confirmation that more then 1 person in the company makes the same claim.

 

Thanks for adding the adapter answer I was beginning to think no one had nticed lol, is the 150W orverkill ? the 130W wont harm my system will it ?

Im guessing we will have to play the waiting game, sort of put up or shut up situation. And off topic whats the Envy 15 like ? are they as good looking as the Dell

 

I'm planning on taking my laptop to a client site to give a 2 week training class. I've been having random lock ups on the system which I end up having to reboot for. From what I understand, the lock ups are related to the throttling issue. I'm currently running bios version Z29 provided by Dell.

Have others noticed if version A05 corrected the random lock up issue? Thats the only concern I have with using this laptop to give a training class. I'm not doing anythign that involves heavy GPU. Mainly just presenting over ppt and word. I'm hoping someone can verify that A05 solves this so I can install it and avoid the random hangs in a training.

 

Yes, that was me, I had emailed Michael Dell also and I received a call back from the Executive Escalations department like 2 days ago and was told that at this time the 130W Adapter is still not approved and to give them some more time for the Bios Update.

However, I also want to state that on 1/2/2010 I was told by the Executive Escalations department that there would be a Bios Update by the 3 week in January and that did not happen so it also looks like the higher-ups don't always get the correct information as well.

You know stuff happens and in no way am I bashing Dell, I really hope that the Bios Update/130W Adapter is available to us.

For now I am going to wait and see if they can fix the issue, as all of us have different needs with this computer I am able to wait as I do not game and use it for general computing.

Best of luck to all in whatever decision you make.

 

I see the 1645 is good for gaming again.....

http://www.dell.com/home/laptops

Expand 'How fast does it need to be?' on the left. Tick 'Gaming'. 4 results, two of them 1645s.

(And I guess I'd better add "True at 16:50 Eastern Time on 24 January 2010, on my laptop, in my location, on Firefox 3.5.7, but YMMV )

 

Yeah, that was ruled out a longgg time ago. The funny part is if you click, 'power - ultimate speed for complex applications' you get every model except the underpowered netbooks, the Inspirion 11z, and Inspiron 13. Not much criteria for that is there.

 

M15x isn't included in any category.

 

Just my own update: I just ran the Fritz Chess benchmark with all 8 cores checked, and CPU-Z reported 1.73 GHz consistently through the benchmark. Task Manager reported of course 100% CPU usage. This is with the RGBLED screen on lowest setting (which is still pretty bright).

 

I sure hope we have some good news soon, because that Envy 15 looks better every day. How does the 5830 perform compared to the 4670, anyone know?

atlstang, when you get your new laptop you'll have to let us know what you think of it.

 

Have they fixed the throttling on the envy 15 while on battery? Seems like some of you are going from one issue to the other.

Just curious.

 

I don't return sxps 16 because of RGBLED, beside this , envy15 is much better than this . (At least , it won't lag when I play COD6 but sxps 16 does )