S-XPS 1645 Throttling Info. and Updates

Dell has released their solution to the throttling issue, if you have updated to the latest A07 or newer bios, and recieved a 130w adapter from dell. Need to call the reps and get a 130w sent free of charge.

After more testing looks as if the A07 bios has indeed greatly improve the performance of the laptop. More testing will continue to see where the limit is at, as there may be close to the throttling edge perhaps.

Purpose of this thread is to educate those about the throttling issues with dell's laptops. Specifically to the xps 16 with i7 processors. Also to show community driven solutions as we wait for dells response to this issue.

This is an advancement from the power investigation thread, to summarize the thread here is a quick list.

• The 1645 is supplied and designed for a 90w adapter which is undersized and cripples the laptop under gpu + cpu stresses, such as gaming and photo editing.
• Running on battery seems to let the laptop run more towards it potential.
• The problem severity varies and depends mostly on configuration and type of program being ran.
• A 130w or bigger power supply will not solve the issue alone, as the bios actively throttles to main 90w anyways.
• Through reps Dell seems aware of the issue. As of yet there has been no official announcement and is unclear if they will actually fully resolve the problem.

All the information in this thread is not by me alone, it comes from the 2000+ posts from many different users in the power investigation thread here. Thanks to those users is how we know so much today, and a special thanks to uncleweb for the only solution thats out now to get the performance that this laptop is supposed to have.

Verify you have a problem.
**Warning, when on battery do not stress both gpu and cpu using programs like furmark and prime95 at the same time. Not recommended, extra issues/failures have been tied to this.***
All 1645 currently are underpowered with the 90w and there is no official fix from dell to this date. As mentioned the degree of being effected seems to be very dependent on setups. To see whether the applications you are currently running are effected use as described here. Do note, if you dont currently game or run demanding applications now, it is unkown what the future holds and if problems down the line will pop up for you, as things become more multi-core and gpu aware.

Symptoms usually occur during gaming or other stressful applications. Can be in the form of extremely large stuttering under certain conditions. Even if these slow downs do not occur it is very likely you could see a performance boost by allowing the laptop to run properly, given a modern game. A quick test is to run on battery, giving the system a few min's to settle, and see if performance is improved.

The problem occurs due to the cpu slowing itself down to maintain the 90w power limitation set by the adapter. The cpu speed is calculated as 133mhz * a multiplier. This multiplier should be 15x + for most games, but due to throttling can be observed around 8-10x or the minimum 7x, a huge performance hit. If this is not enough then a second type of throttling is enabled and is discussed later in this thread.

To accurately monitor your throttling it is recommended to use throttlestops logging features. Sometimes is best to play in a windowed mode, and have throttlestop off the side of the screen to watch for throttling. To get the best results its recommended to enable logging as shown below.

​

To look for throttling VERIFY throttlestop is disabled and the checkboxes are unchecked like the picture. Throttlestop is also designed to stop throttling, with them checked it will not throttle, also if your using a 90w adapter with it enabled could run the risk of damage. Throttlstop Download

Under normal circumstances clock modulation and chipset modulation should show 100% for all 8 threads shown. multipliers should be well above 12x for 720qm and 13x for 820qm. If they fall below these limits under load of gaming you are being throttled as well. The majority of games use only 1 or 2 cores and should show a multiplier well over 15x.

To watch for throttling while gaming etc, click the log function on throttlestop and run the game/application. When done come back and open throttlestop log, and verify that there is no modulation and the multipliers are above 15x etc. If its not you are throttling.

Here is a video example by user Magt1000 showing there difference between running on battery and adapter. You can see the severe throttling causing choppiness vs battery being smooth. Video link HERE
EXAMPLE: Here are my results for tf2 and crysis, clearly showing throttling. Settings are not maxed, and resolution is 1024by768 so keep in mind this is not a full stress scenario, more just a casual run. The computer is also hosting and only looking at one section.

Crysis
fps throttled 43 fps, mult approx 9, no modulation (link to image)
fps unthrottled 64 fps, mult approx 14-15. percent increase 48% (link to image)


TF2
fps throttled 44fps, mult approx 9-10x, no modulation (link to image)
fps unthrottled 71, mult approx 14-15, percent increase 70% (link to image)

Its observed that lowering of the multipliers will reduce framerates significantly in games. If the multipliers are reduced to the lowest settings and the bios feels it needs to throttle more. It will activate modulation. Modulation is very severe and will cause extreme stutters and framerates tend to fall <10fps.



Current community devised solution to the problem. With much thanks to Uncle Webb
***DISCLAIMER USE OF A 90W ADAPTER IS NOT SAFE WITH THIS METHOD, AS THIS WILL ALLOW THE COMPUTER TO DRAW MORE THEN 90W, A 130W ADAPTER OR HIGHER IS RECOMMENDED. DELL ALSO HAS STATED THEY HAVE NOT TESTED A HIGHER RATED ADAPTER***

What is required.

• A 130w or higher ac adapter.
• Throttlestop program. by UncleWeb DOWNLOAD

Throttle stop is a program that does what it says, it disables dells throttling to allow the full performance of this laptop. By disabling the throttling the laptop then will be able to draw more then 90w, so a 130w or higher adapter is required. It has been measured through a kill-a-watt adapter these laptops need around 100-110 watts to run properly for current games. Performance increase can be double and allow hd gaming at 1080p resolutions.

The current throttlestop program v1.8 looks as so.

​

First read the disclaimer fully now to disable all of dells throttling do the following

• check disable clock modulation
• check disable chipset clock mod
• check set multiplier
• make sure set multiplier shows 13 (14 for 820qm)
• verify "ThrottleStop Is Enabled" message is displayed

Throttling is now disabled and laptop will be able to draw enough power for max performance. Extra features in the 1.8 version is listed below

• More data - increases sampling rate to give a better look at the multipliers and modulation. This allows better view of throttling
• Auto Turnoff - When switching to battery throttlestop will attempt to re-enable throttling. Some users have had problems switching to battery under high load. Possibly damaging the battery.
• Log file - Creates a log of multipliers and modulation, so programs can be run and performance later evaluated through logs.

Please donate to unclewebb if you find great success with this program. As he does this for free off his spare time, and his work is much appreciated as he continues to help us out developing more features into this program

*It cannot be stressed enough that this is at your own risk, and a 130w adapter or higher is needed to run safe at the moment.

**There is an issue when disabling dells throttling, not with throtlestop itself, but more of a side effect possibly with drivers. When keys are pressed in the game will cause servere FPS lag. To stop this, enable throttle stop and close the lid of the laptop to allow it to sleep. After a few seconds open the lid allowing the computer to wake up. Start your game and you should now be at maximum performance.

Throttling Explained
Its first important to understand how the intel i7 cpu's work. MHZ is calculated by taking the currently multiplier times 133mhz, ex 12x * 133 = 1.6ghz. The 720qm and 820qm run at a nominal 1.6ghz and 1.73ghz. To save power when the cpu is not under load, under normal conditions it will downclock to approximately 933mhz. Turboboost is the touted feature for these processors, If only one core out of the four is being used it allows the 720qm to clock up to 2.8ghz, and the 820qm up to 3.06ghz. If only 2 cores its somewhere in the middle and so on.

Throttling mostly occurs only when the gpu and cpu both are being loaded, by games and other demanding software. This is so the computer can maintain its 90w power requirement for the 90w adapter. It has been observed the cpu will first throttle the multipliers from the 15x-21x it should be during games, down to a measly 8-10x. (taking that max of 2.8-3ghz down to sub 1ghz.) The lowest multiplier it can go is 7x (930mhz) if that isnt enough to limit power consumption it will then do what is called clock modulation.

Clock modulation will kick in if the 7x multiplier is not enough to limit power alone. From what i understand, it basically is telling the cpu to ignore clock pulses from the clock, basically the cpu does nothing for some of the cycles and slows down processing immensely. This most likely is the type of throttling that coincides with user reports of slideshow type effects while gaming. Only way to monitor this is through realtemp and throttlestop as it shows the percentages.

There looks to be a third form of throttling that is used on the m15x that uncleweb is looking into. It is no known whether it effects xps 16 at all.


Dell's Update

Update 2/4/10
Bill_B from dell has finally come and answered some of the lingering questions. After 3 months of near no communications and conflicting reports, it would seem dell is finally coming out. Here are some of the more relevant posts to the issue.

Seems like dell now agree with us, and the adapter will be the proper fix. Looks as well like 90w adapter may be replaced with a higher wattage as well. ETA is still unclear at this point in time, but hopefully bill will stay in contact and know more after a few meetings time.





Extra related information.
1645 power related discussions. [0] , [1]
M15x response to throttling issues can be read here. link
Tinkerdude's report on dells throttling of latitudes (similiar mechanisms). link
See this post to revert back to an older bios (if it wont let you downgrade) Link


Any input/clarification is welcome. Its not my thread just a summary of the past thread, and all the info that may be buried in it. along with new stuff as well.
Updated 1-30-10

 

User submitted results (non-synthetic)

User ------ application ---fps --- fps with tstop --- powerdraw --- powerdraw w/ tstop --- screen ---processor
romills ------ Demigod --- 1-20 ---- 30-50 ----------- 65w --------- 115w peak -------------- rgbled -------- 720qm
A.H.A ------ burnout P - <10-15 --- 35-40 ------------ n/a --------- n/a --------------------- rgbled -------- 720qm
Siphen ----- D.A. max -- 10-20 ---- 15-30 ------------91w --------- 100 --------------------- rgbled -------- 820qm
Siphen ----- D.A. min -- 17-28 ---- 34-58 ------------91w --------- 100 --------------------- rgbled -------- 820qm
Siphen -borderlands min -- 8-22 ---- 34-62 ------------77w --------- 107 --------------------- rgbled -------- 820qm
Siphen -borderlands max -- 4-23 ---- 12-24 ------------95w --------- 105 --------------------- rgbled -------- 820qm
*post your results

Power consumption
100-110w for normal gamin on my wled + 720qm screen, peak 130 with furmark + prime 95, others with rgbled screens have reported up to 146watts with different configurations.

Siphens power numbers for max wattage per component, also agrees with testing done with my system as well. As you can see the laptop is very power hungry if it where to run at its max potential.

A test by mitchell2.4 that is more proof that a 90w is not enough, as it is pulling 93w from the adapter and the computer i severely throttled. LINK

This is power from the wall, not necessarily power to the computer, but as the 90w adapter just about catches a house on fire when it draws 90w from the wall. It wouldnt be safe in trying to get the actual 90w it should have.

Frequently Asked Questions

Q: Has this issue been resolved? When will it be?
A: The issue is not resolved. Despite what a rep may tell you if you order a 1645 today it will have throttling issues as described. Dell has not made any official announcements and any response to the problem may not be for a while.

Q: What other adapters will work?
A: The adapter should be rated 130w or above. While supplying 19.5v. Buying from dell is expensive and is suggested buying from other retailers.

Known adapters that work.
130w

• PA-4E slim 130w adapter
• PA-13 brick 130w adapter

150w

• PA-15 150w adapter

Fore more info on the 130w vs 150w, dimensions etc. Look HERE
*please send me model numbers that you are using with success

Q: The laptop seems to play my games just fine on the 90w.
A: As mentioned before, it is very dependant on the application and system configuration. Those with rgbled screens seem to have a lot more problems then those with wled. If the game is a recent game it is still very likely to see up to double the performance and smoothness with throttling disabled. May allow for full 1080p gaming for a lot of games.

Q: Mentioned other laptops have the same throttling?
A: It seems the i7's are a very power hungry processor and is likely the systems with i7's and 90w adapters are indeed throttling, although some core 2 duo laptops also have throttling issues but may not be power related. Though it is unclear how severe their issues may be with actuall performance, as this is mainly for the 1645. These are the systems that users have posted here recently stating similiar problems.

• Studio XPS 16 (1645)
• Studio XPS 17 with i7 processor
• Alienware M15x Investigation thread. LINK
• Latitude series (specific models anyone?)
• Studio xps 1640 (some signs needs additional testing for confirmation)
• Studio xps 1647 (recently confirmed via fear2)

updated 1/12/2010

 

Ahh, this was very much needed. I will add a link to this thread at the beginning of the investigation thread since the investigation is pretty much concluded at this time.

Thanks again for the great job you did at reconciling all the information.

EDIT: I think its important to emphasize that despite what Dell reps are telling people in various sales departments, THIS ISSUE HAS NOT BEEN FIXED. There is NOT a new motherboard. There is NOT a new power supply. There is NOT a new BIOS (Yet). If Dell tells you that this is not going to be a problem with your machine THEY ARE EITHER LIEING TO YOU, OR NOT INFORMED ON WHAT THE ISSUE IS (probably the former).

As of 12/27/2009, if you order a i7 1645 XPS, you WILL experience throttling on some/all of your applications without taking some additional steps to get your laptop performance to where it should be.

 

they totally know that they are doing this fellas lesson learned dont buy dell

 

will most definitely. Ill borrow your line and stick it on there for now. Give me a few days to get this up and going. I ran out of steam pretty quick tonight due to illness and my mind is just on cloud nine at the moment. I want to dedicate a section on the dell experiences/updates and will definitely mention what you say. As their methods/answers are a bit ridiculous.

I want to try to have anything worth knowing on this main page. mainly to allow everyone to discuss/post their experiences even if a bit offtopic. As nothing should get lost in a sea of posts.

 

This needs to be posted in the Dell Community Forums. Much larger audience there..

 

I actually just posted a link to this thread there. I wont hold my breath as it will be deleted in no time.

http://en.community.dell.com/forums/t/19312160.aspx

 

Wow, I hope they won't be so low and delete it

 

I'm almost positive they will. Thats why I even sorta called them out on it in my post.

 

thanks! great work!

 

if they do, guess that might tell us a bit about their stance on the issue then wouldnt it. lol

 

I think it Definitely would. I'm about to jet out of work, but if some one would submit this thread and the other ones to Engadget and TheInquirer and maybe Anandtech or some other sites maybe we could get some more attention.

 

Done!...

BTW, thatks. O.P. for posting this, and everyone who contributed the software - I was actually just made aware of this cause of this forum and my recent sXPS 16 purchase which has shipped already, and have already ordered a PSU off of ebay.

Can we make this a sticky?

Cheers and happy holidays!

L.

 

awesome!! rep

 

GREAT work .....

 

I played it up a little too.. might as well twist the knife if someone from dell will eventually read it.

 

Funny thing: for like 1 week ago when i googled "xps 1645". The throttle investigation thread came number 3 on the list. But now its not even on the list? Its only a small "underthread" for one of the review thread. Odd isnt it?

 

It still shows number 3 for me. In the US here.

zlog's post in dell forums is still there heh. hopefully others will see it and become aware. As i think the bottom line is, if a lot of us knew about this issue probably wouldnt of bought this laptop.

I'd like to make a little database comparing real world results through games etc, of the difference throttlestop makes. I'll be throwing in my results for tf2 l4d2 and maybe even dragon age as i now have that game. if ya could post your fps before/after results along with processor and screen configuration. And any observed multipliers, wattage or modulation.

 

Dell is aware of this issue and are working on it. A similar problem with some other models has been resolved via a BIOS revision.

No one has been banned on the Dell forum nor have any threads been deleted (unless they violate forum TOS, just as is done here).

The thread to follow is this one, if you care to keep track.

 

This cant be resolved by a bios revision alone. Givin the history with the latitudes, and the response it seems the m15x has gotten. Its not looking good for us with the 1645. If i were to guess there will be a bios update that dell will say "resolves" the issue, that might help slightly. But will not be a solution to the problem, although we do remain hopeful.

It is seen in our eyes, that the only proper fix would be to issue a bios to allow full use of the 130w. Then replace the 90w's with a 130w or higher, but will that happen?

Its important to note, this isnt really a bios problem, its doing exactly as its supposed to. That is to limit the power consumption to 90w and not explode our 90w adapters the computer come with. Our computers are pegging the 90w adapter causing these huge crippling issues.

 

Just added a results section above the faq. Post your results/logs and i will add.

 

Where do we search/monitor for Dell's official response/solution?

 

Link please.

 

No offense, but you have a very canned response. I'd like to think that you have enough intuition to know that when a big company says "we're aware and working on it" it really means "We'll get around to it if we have to."

I'm gonna be blunt here. Dell doesn't have to. They wont. They aren't. They'll likely never. They have our money, they have their return policy to protect them from the majority of us sending our units back.

The only thing we have left in our favor is repercussion, i.e. we all have a PROBLEM with our notebooks, and they are all under warranty at this time. If this starts costing Dell A LOT OF MONEY, they will respond.

What do you as an owner need to do? Call in, and stay on the phone with tech support until they decide to replace parts. Let them replace those parts, then tell them the problem is still there. Keep doing this until the problem goes away, which will likely be never. Eventually Dell will realize "Oh shoot, we thought these chumps would like the smell of our farts but we were wrong."

That being said, I plan on making my first call on my next day off.

 

Just want to inform everybody that if the laptop is defective you can still legally return it and get a refund despite of any returns policy.

Legally, I think the laptop can can defined as defective at this stage.

Send back those laptops!

 

Can you post some further information and actual testing to back that up?

The Latitude E6500 that motivated tinkerdude to write a 59 page pdf still throttled like crazy during his last test.

Does anyone know why he stopped complaining? I'll give you a hint. NDA.

Muzzling people that complain is not equivalent to fixing the problem.

 

you really think he got hit with a NDA? if thats true, then that is why we must try and spread the word to many people. Hopefully reviewers will catch on, if they ever review this laptop. Making it to widespread, limiting dell's options and hopefully they will just fix it.

I know some are unsure. But they had to know this was a problem some point down the line. A good engineer would of known from the start that this was possible, nevermind what they should have caught during testing. Most likely made a calculated risk at one point, thinking not many of us would notice theyre selling a premium laptop that doesnt perform. Really the 4670 etc means nothing being so crippled to many of us.

That why i made this thread, and hopefully we can spread the word. To those that already have the laptop, and to those thinking of buying it. Show them our findings and let them come to their own decisions. Do this enough and i would think reviewers etc will catch on in time, to spread the word even more. Spread it to enough people, and the option to "contain" the issue wont be on dells list.

quiet is right, ive emailed the engineers that i am supposedly in contact that are "capturing" the laptop for any updates they can tell me. If nothing comes from it or the answer is another run around, im probably going to try for a full refund for this laptop. Even though we have uncle webbs solution, still feel cheated and will rather go without a laptop for a while.

I really dont think many of us would of jumped on this laptop and bought it if it had a warning about this issue.

Help me build this thread guys as its our thread not mine, let me know what i should add/change/clarify. Im doing my best to remember all the details from the 2000 post investigation.

 

Can 150w adapter harm your components more then a 130 adapter? do we have any info on that ?

 

90W, 130W, 150W doesn't matter. The adapter is not forcing the notebook to run at 130 or 150W.. it only makes it available if it needs it. the device DRAWS power FROM the power supply, it's not pushed from the power supply.

In my opinion, putting a 150W power supply has less risk than the 90W since we have already seen that the 90W is being run pretty close to it's capacity. It is a function of the bios and onboard circuitry on the laptop to limit the amount of current consumed.

You can't make a laptop that only needs 80W consume 120W. Likewise, it is a bad thing if you can't provide the power required, if it requires 95W and you can only provide 85W, bad things will happen.

The only thing that can be really bad is if you use an adapter with wrong voltage, unclean power delivery (spikes, sags etc... IE knockoff adapters instead of genuine dell or at least high quality replacements)..

Because this issue has a known remedy, I am actually thinking that Dell may step up and fix the issue. With Throttlestop the issue is resolved. All they would need to do is move that functionality into the bios. They can see the wattage of the PSU in the BIOS.. so dependent on what PSU is installed, different throttling routines or power profiles are applied.. Can be done pretty easily but will require them to develop it, test it and then release it. Since it is the holiday season, I don't think they will get to this till mid January..

 

Just had an interesting conversation with a Dell rep. It seems that you can request a 150W PSU with the Alienware 15X. However they still won't say there is an issue.

 

Heh i emailed them for an update and mentioned returning the laptop last night. Promptly received a call from dell about that. first told me that the laptop i received fixed the issue. I think what she meant by it tho, she though it was a replacement to resolve some issue, as that seems how the capture stuff is processed (just like replacements). After saying it wasnt and was the same as the old problem. She pulled up some things and said it looks like the 11th of janurary will be a bios update that may resolve the issue. I asked her, if i could wait to see if that resolves the issue and still if i could return it for a full refund if it doesnt, she said sure just give the case number and that should be no problem.

Dunno if i was getting the run around, but looks like i can get a refund now or later. so heres hope dell does a little bit more then a bios update on January 11th

 

Finally got around to testing this myself. My results are typical.

Intel Core i7 720QM
ATI Mobility Radeon HD 4670
1920x1080 WLED Display

Testing Apps: ThrottleStop, i7 Turbo, GPU-Z

FurMark: Stability Test, 1024x768, 8x MSAA
Prime95: Torture Test, Small FFTs

Idle:
CPU Multiplier: 7x
Clock Modulation: 100%
GPU Core Clock: 220Mhz
GPU Memory Clock: 299Mhz

Furmark:
CPU Multiplier: 10-15x
Clock Modulation: 100%
GPU Core Clock: 675Mhz
GPU Memory Clock: 797Mhz

Prime95:
CPU Multiplier: 12x
Clock Modulation: 100%
GPU Core Clock: 220Mhz
GPU Memory Clock: 299Mhz

Furmark + Prime95:
CPU Multiplier: 7-8x <= THROTTLING
Clock Modulation: 75-100% <= THROTTLING
GPU Core Clock: 675Mhz
GPU Memory Clock: 797Mhz

 

That is the problem. When running a CPU + GPU intensive application at the same time, these computers can slow down to a crawl. This isn't a major issue for many users that do mostly 2D work with their computers. If you like to play a game beyond Solitaire, then your computer can slow down significantly and kill your frame rates in many modern games.

Can you post a screen shot of your final test so that it is easy for other users to do their own testing and to compare their results?

 

Screenshot here: http://dl.dropbox.com/u/123465/xps1645-throttling.jpg

This time I also ran a HDD benchmark to let the HDD join in on the power draw fun.

 

zengei: Most of the throttling in your screen shot is from the drop in the multiplier. With this load a Core i7-720QM should be using the default 12X multiplier so the total should be 133.3 X 12 = 1600 MHz. A 7X multiplier shows you are only running at 933 MHz. That means if it wasn't for the Dell bios throttling your computer, it would be running over 71% faster.



The only good news is that it could be worse. Clock modulation is just starting. Core 1, 2 and 3 are still running OK with a C0% of 100.00 like you should be seeing. Both threads of Core 0 are clearly starting to use clock modulation to slow down your CPU even further. As clock modulation kicks in, your C0 State Percent will drop lower and lower.

If you run Prime95 and manually set Clock Modulation in ThrottleStop to 87.5% then you will see what C0% that corresponds to. At the moment it looks like clock modulation is cycling on and off on core 0. When it is fully engaged, the C0% is usually around 88%.

Thanks for the test.

 

I don't know if this has been posted, but if anyone is looking to just purchase the 130 Watt adapter without the joy of talking to a Dell rep, here is a link to a site that sells it for substantially less than Dell charges:

http://search.store.yahoo.net/cgi-bin/nsearch?query=PA-4E&Go.x=9&Go.y=15&catalog=getpartsonline http://search.store.yahoo.net/cgi-bin/nsearch?query=PA-4E&Go.x=9&Go.y=15&catalog=getpartsonline

 

who did you email?
does Dell have a 21 or 30-day return policy?

 

well the email i sent it to the people that supposedly were capturing the 1645 for their own investigation. I dont think im allowed to post their email here, a rule of the forums or something?

anyways, The guy didnt really respond anyway and had someone else call about my thoughts about returning it, the person didnt know what was happening to be honest. did say a bios update was coming the 11th, dont hold your breath it will fix the issue, hopefully there is more to it. Im still looking at my options etc for if i return the laptop is why i asked, and they said since i been having problems and have a case number shouldnt be a problem come the 11th and things arnt fixed.


MrAceXPS that is a good price for a brand new adapter. have you had any problems etc with your laptop/programs?

 

hi guys, sorry but i'm just getting slightly confused. Am I supposed to put a check in the 'Clock Modulation' and 'Set Mulitiplier' box coz the screenshots I'm seeing doesn't have a check in it?
But as mentioned above, "make sure the Disable Clock modulation and Set Multiplier check boxes are checked".
Hope you guys could kindly advise.

 

Nope, haven't had any problems. But, for that price, I can rationalize it as a second adapter...

I figure Dell will fix the BIOs but then require an adapter exchange. Not sure how they do will do it logistically though. If they fix the BIOs alone, won't that pose risk to the laptop?

Anyway, I can't handle calling Dell to talk to a rep. It's too painful. I mean really painful. Money well spent if I decide to use the adapter.

 

aiee, yes it is. and fixed for clarity

mracexp, not a bad price i agree. I spent a little over 20 dollars for a refurb. Dell sells them for over 160 i think so yeah good deal. Dont run any game etc to have problems?

 

Hi guys.....
A doubt crept in.... Dont knw whether it was discussed in any of the other forums, but here it is. If the Throttlestop allows more power to go in then wont that mean that more heat will be generated by the CPU n GPU. Now the BIOS restricts the power to only 90W and hence we are not facing any heat issues.
Got my laptop yesterday and have not tested it with ThrottleStop.

Looking forward to your replies.

 

interesting well apparently dell has is discontinuing the 1645 due to power issues and had allready removed the i7s from its build.. i just called dell canada and they confirmed this, although some dell sites still offer the i7 apparently the orders wont be honored...

 

Myself, the laptop seems to run on average 75C during normal gaming and barely peaks above 80C. Which is normal operation for most laptops

ironically, if its just the first type of throttling thats affecting the game, the temperatures are very similar to throttle or not throttled. It seems so because the fan is kicking on higher speed to keep the same temperature when throttling is disabled.

IMO the heat dissapation is very good, and seems to be a non-issue, the system is simply being throttled to maintain the 90w power restriction. At least this is my case, and is similar to a lot of others i believe

A bit early to tell whats going on, but yeah i await to see what this is all about.

 

EDIT:

@XPS 1645 EOL?

This has nothing to do with fixing the throttling issue. Let's not make this thread into a repeat of the last one. This type of discussion belongs in THAT thread.

 

im guessing replacements for all? lol not likely.. maybe dell certificates for everyone who whines?

 

it has everything to do with throttling.. xps1645 is bieng discontinued due to "inadaquite power supply issues"

 

Ok, I have found out conclusively (at least the one that was tested) that the PA-3E power supply is 71.8% efficient. I had one of the guys at my lab who has a dell with this power supply to concurently test the current on the AC side and DC side of this power supply since things were slow this morning and this is what he found out.

This was with the PA-3E running at about 55% load on AC side with 118VAC and .82 Amps (68.4 watts) and the DC at 19.5 and 2.52 Amps (49.12 Watts).

So keep this in mind when you are measuring your power on the AC side that the 90 watt rating is for the DC side. To find out what the side power consumption is just multiply by .72 or there abouts.

 

ikjadoon its okay, anything thats worth knowing will be added to the front page as it comes. Nothing hopefully will be lost like the last thread.

and yeah it is related if it is due to the power issue like it seems? this might be dells solution, by killing the 1645 project.

its to early to tell. If true though, can probably safe to say this is the worst laptop decision ive ever made.

edit: justmark, the efficiency can vary a lot based on load. a efficiency reading around 90w probably be better if ya could. just a suggestion for better numbers

 

Of course it has to do with the throttling. But not fixing the throttle problem. Speculation/talk/rumors about what Dell is doing belong in the original thread. This thread is for users who want to fix the problem. Unless it's official, it does not go in this thread. At least that's the feel I was getting for this thread.

I wouldn't put it past Dell to do something like that, though, but, again, this thread is for users who want to FIX the problem, not hear us discuss whether Dell is getting the 1645 ready for an EOL timeline. But, when it's official, bam, let's put it here.

Fine. lol...

I don't want a stupid Dell certificate, I want money so I can buy a real system.

~Ibrahim~