New Owner R1 multiplier stays at 6 so CPU is only 1.6

Hey all just wondering if there is a setting I am missing that is keeping the core multiplier at 6 and not 6.5 or what ever it is supposed to be to bring it to the advertised 1.73 in the basic information screen ? thanks

 

Did you enable Overclocking in the BIOS?

 

It depends on your cpu-z version as to what it shows. THis has been debated forever. It is what it is... Now... If you want more power out of your m11x, start searching the forums for m11x r1 optimization. Good Luck, StevenX

 

@Zero- yes it is overclocked in the bios and windows shows the CPU at 1.73 but every other monitor shows it to be 1.6

@Steven - thanks I'll be looking into it . I have already installed a Samsung 256GB ssd and now for the rest of tweeks..

 

As Steven mentioned, this has been debated many times over in earlier threads.

Try this utility:
Igor P. Zenkov virtual world, Freeware, TrueSpeed

You'll find it matches what Windows reports on the Sys Properties screen.

I don't have the link, but one of the AW reps followed up on this and as I recall, they stated the issued was with how CPU-Z and the other apps were reading the speed. Again, I dont have the link - sorry.

I recall from my own testing at the time using multiple versions of CPU-Z, there were different values reported by different versions.

-------------------------------------------------------------------------------------------------------------------------------------

Concerning the OC front, your results will vary. Dell has stated the following on overclocking:

Source: M11x & M11x R2 Overclocking Policy - Alienware Forum - Alienware Club - Dell Community

 

If I had known that they would take such a stance(Not supporting OC'ing, at all), I probably would have skipped over the M11X. When I see 1.73Ghz advertised on the sales page, I expect to be able to get that speed & have a functional computer. Shame on you, Dell. (You notice how they've since gone back & removed the "1.73" portion in the description. Now it just says "overclockable". What BS)


I'm one of the lucky ones that has a system that's stable while OC'd.....but still that's such BS what Dell did there.

If you're gonna advertise the overclock & use it as a selling point, then you should fully support the overclock. (be it 1.6 or 1.73)


For reference, here's what I saw when I originally ordered my system.

 

@Arak - That's what it showed when I ordered mine.

Thanks, StevenX

 

Er, um... I thought the concept of "your mileage may vary" always applies when dealing with overclocking?

The only time I have ever seen a "guaranteed" overclock in any computer component is if the product ships at the overclocked speed from the factory.

It's up to you, but I really wouldn't consider the "loss" (which really isn't a loss) of 130Mhz of CPU speed to be anything to get upset about. Small incremental changes to CPU speed have very little impact on the performance of your laptop.

 

That's also if you are in the camp that the system is not running @ 1.73. because CPU-Z says its not.

Again, old debate... lol.

With my R1, Windows sees it at 1.73, 3DMark sees it at 1.73, CPU TrueSpeed sees it at 1.73. I'm satisfied with that. Mainly the last 1.

Have fun guys.

 

And as mentioned other tools will report DIFFERENTLY.

 

TrueSpeed is a nice little utility but I wouldn't recommend using it to determine the speed your CPU is running at. On my T8100, it assumes that the CPU is always using the default 10.5 multiplier.

I used ThrottleStop to lock both cores to the 11.5 multiplier as explained here.

http://forum.notebookreview.com/win...c-acceleration-ida-both-cores-core-2-duo.html

To lock both cores of a Core 2 Duo to the highest IDA multiplier, you need to disable Enhanced Intel SpeedStep. Once EIST is disabled, the T8100 multiplier is locked at 11.5 and it's physically impossible for it to change. You can't have SpeedStep multiplier transitions when SpeedStep is disabled.

In this example, CPU-Z uses the correct 11.5 multiplier while TrueSpeed uses the T8100 default 10.5 multiplier so TrueSpeed is reporting the wrong total MHz.



In the second example, I used ThrottleStop to set the multiplier to 6.0 and then I disabled EIST so it would remain locked at 6.0. CPU-Z and ThrottleStop both report the correct total MHz based on the 6.0 multiplier while TrueSpeed is wrong because it is still assuming the default 10.5 multiplier.



I asked bigun08 to do some testing for me and here are his results.

http://i722.photobucket.com/albums/ww230/Bigun08/stresstest.jpg

When fully loaded his SU7300 is only using the 6.0 multiplier. Once again, TrueSpeed is wrong. The EIST bit is locked and disabled in his processor so the multiplier is stuck at 6.0. The SU7300 supports a 6.5 multiplier but when it is set up like bigun08 has it set up, it's impossible for the CPU to ever do a SpeedStep transition and use the 6.5 multiplier.

Edit: If he could enable EIST in the bios, he could finally get access to the 6.5 multiplier but it doesn't look like this is possible when overclocking.

Can someone else do a quick test on their SU7300 by running two threads of Prime95 for a few seconds until the multiplier has stabilized and then take a screen shot of CPU-Z and ThrottleStop. Using TrueSpeed is pointless and Windows 7 is not capable of reporting the correct multiplier either. It just uses the default multiplier like TrueSpeed does regardless of what multiplier the CPU is actually using.

Edit: Here are two Windows 7 examples that show this problem.

 

Ok Unclewebb, here's a screenshot:
http://i236.photobucket.com/albums/ff266/DavyGT/e90708bc.jpg

BTW, can you enable SLFM mode without the FSB clock cutting in half?

 

Once again, at full load you are only getting the 6.0 multiplier and EIST is disabled and locked. Is there any option in the bios to enable EIST. I'm starting to wonder if Dell has this disabled and locked anytime you are using the bios bus speed overclocking feature. This would prevent you from ever using the 6.5 multiplier when overclocking.

You can also try this test at the default 200 MHz bus speed and see if EIST is enabled by default then. You need EIST for the 6.5 multiplier to work.

Unfortunately, at the moment, the only way you can access your SLFM voltages is by enabling SLFM which also cuts the bus speed in half at idle. I'll see tomorrow if I can come up with a hack to get around this limitation but I don't think it's possible.

Edit: Much easier to see that screen shot now.

When you click on Turn On and Enable ThrottleStop and click on Set Multiplier and adjust the VID voltage does CPU-Z report any change? Probably not since EIST is disabled.

 

There is no option in BIOS to enable EIST- it appears to be linked to the Overclock feature. When Overclock is enabled, the CPU stays at 1.6Ghz and is locked to the lowest VID- .9v. When Overclock is off, EIST works normally.

CPUZ doesn't report any changes when I change the VID or multiplier when the Overclock is enabled.

 

CPU TrueSpeed v1.8 says my M11xR1 is currently clocked at 1729.002MHz

That's 0.998MHz off 173MHz.

Throttlestop and CPUZ added.

1MHz lost somewhere in my system... I'll be back when I find it.

I'm just going outside I may be a while.

 

For those that bought their M11x R1 based on the promise of "Overclockable to 1.73 GHz", I guess you will have to contact Dell and ask for your money back because that claim is not true. Either that or force them to release a bios update to fix this limitation.

The way the Dell bios is presently setup, you can have either a 6.5 multiplier or you can have a 266 MHz bus speed but you can't have both at the same time so it is physically impossible to run the R1 at the claimed 1.73GHz. That must be why they removed that information from their website.

DavyGT: Just as I thought. As soon as EIST is disabled, both the FID multiplier and VID core voltage are locked. With the M11x R1, not only is EIST disabled but it is also locked by the bios. Once locked like this, it's impossible for the multiplier to ever do a SpeedStep transition up to 6.5.

DrGoodvibes: It's already been shown that TrueSpeed is not so true. Head to the top of this page if you want some examples of that. I high lighted my conclusions so it will be easier to see.

1730 MHz - 1596 MHz = 134 MHz that are missing in action.

Another way to look at this is:

1730 MHz / 1596 MHz = 8.4%

That doesn't live up to the Dell marketing department's original claim.

 

Really? 134MHz? I think you will be barking up the wrong tree with this unless its some sort of hobby.

 

Yeah, same goes for mine.
TrueSpeed shows 1729 but everything else shows 1596.
I'm on SU4100 btw.

Big sigh from me :/

 

Dell charged extra money for a feature that doesn't exist. They should either refund the money or deliver the feature they originally promised. All they have to do is change a single bit in the bios. If EIST is enabled then the SU7300 can transition to the 6.5 multiplier when needed and run at the promised 1.73GHz. Problem solved.

Dell just needs to turn on bit[16] in MSR 0x1A0 to get this done but at the moment they have this bit locked. There is nothing a user can do about this until Dell releases an updated bios.

It's amazing what Dell will fix if enough people complain. The user community had to light a fire under them last winter to get them to take their throttling CPU problems seriously but now that has become a non-issue in this fall's line up.

 

Oh no! Another crap claim from Dell exposed by uncle's ThrottleStop? I'm shocked.

I really don't understand why peoples buy those Dell products. All are masochists or what?

Of course, it's nice to play ostrich policy, if three crap software tell a lie then it's obvious that fourth one (which show the truth) is ignored.

 

Kizwan tried to mod the BIOS but he couldn't rebuild the image, so I guess this is in Dell's hands.

 

Dear Dell,

Could you please update the M11xR1 BIOS to allow those using said device to be able to attain a true 1.73GHz or better.

Thank you very much in advance.

Update: Mum just rang regrading Christmas card list, if you (Dell) can get the BIOS updated, I'm sure I can get you (Dell) onto her 2010 Christmas card list. My mum knitted me a pair of slippers last year, so it may be a good thing to be in her good books.

 

That's just like asking Dell to come up with a redesign of the hinge issue and/or extend the warranty to 3 yrs for hinge problems.............not going to happen.

 

So I guess for R1 owners, we can't use ThrottleStop since our multipliers are locked

 

As soon as EIST is disabled and locked, ThrottleStop is useful as a monitoring tool but that's about it. Without SpeedStep enabled, the CPU is not able to change the FID multiplier or VID from within Windows.

Does anyone know if Asus is blocking the 6.5 multiplier on their SU7300 laptops when overclocking? I rather doubt it.

The only way to get an honest 1.73GHz out of an M11x R1 is to boot up with the FSB at 200 MHz and then use SetFSB to try and get up to 266 MHz. If you could go this way, EIST would be enabled so you could access the 6.5 multiplier.

 

Glad I stumbled upon this thread. Last night I was just reading up on your TS threads and had downloaded the program but didn't install it yet. I guess it's pointless to install since I have a R1.

 

You can always use TS to prove to Dell that their previous claim about the M11x R1 being "overclockable to 1.73GHz" was not very accurate and they knew it. As delivered, it's impossible to get the 6.5 multiplier and the 266 MHz FSB speed.

 

Realistically, what difference would that make?

Let's say that you have solid and irrefutable proof that the CPU isn't overclockable to 1.73Ghz. What are you going to do with it? Are you going to return your M11x because of it?

 

1.6ghz and 1.73ghz so little of a boost as if it will make a huge difference, if dell/alienware claimed it was 2ghz and r1 is only getting 1.6ghz then that is worth fighting for!

 

Once EIST is disabled and locked, the multiplier is also locked and in this case it's locked to 6.0. That's as irrefutable proof as you can get. You can use any software you want to read the contents of MSR 0x1A0. CPU-Z can be used to read this register and to create a CPU-Z report. The Intel documentation shows that bit[16] is the EIST bit and bit[20] is the EIST lock bit. Anyone that wants to argue about what Dell has done doesn't have a leg to stand on. The CPU-Z report also shows MSR 0x198 which shows the CPU locked at the 6.0 multiplier.

If people complain, maybe Dell will release a bios update. If no one says anything then they won't bother.

 

It's not 130MHz, it's 300 to 430MHz depending on if you're looking at 1.6GHz or 1.73GHz. That's pretty significant (~25%).

 

I'm willing to start an online petition if you guys want to help write the petition statement. It can't hurt, and once you get it going, it's just a matter of getting users to sign it, and then forward it directly to dell. An email, phone call, or text chat regarding the issue, as well as asking to escalate it or speaking with a supervisor or manager will help too. More users the better.

If it's as simple as UncleWebb attest's to, then it should be a simple implementation and require few resources from Dell.

So who's in? I'd like unclewebb's help in the technical details of the statement though as he clearly has a good understanding of it. Too bad he doesn't own an M11x.

 

There doesn't seem to be enough R1 users that are angry about this issue. I think both the SU4100 and SU7300 have EIST locked and disabled so the originally promised, "overclockable to 1.73GHz", is impossible. Count me in if you ever need someone to explain exactly what Dell has done.

 

Count me in for the petition.

 

I'm not sure I can help technically, but I can do a pretty good angry.
Inside that gas mask, I'm really quite angry. Grrrrhhh!!

Add one more on the petition.

 

Thanks guys. So what should I put in the petition statement? Make it general like have Dell update the BIOS for overclocking to 1.73GHz? Or put specifics like unlocking the EIST in order to achieve the promised 1.73GHz (or more)?

 

so i just wasted my money ? :< my order will be at my house in a day or two it's in Kuwait just waiting for customs :< so i wasted my money or i will not notice the difference ?

 

I'd do it in three parts.

1. Introduction. Addressing the issue and outlining it in layman's terms. Attempt to reference Dells deliverables were they have mentioned them.
2. Technical details without labouring the point.
3. Summing up. i.e. the outcome we would like.

I wouldn't go over one page. Most managers above senior technical manager suffer from ADD.

Gosh, remember the good old days, when Brain of Alienware would be in here work shopping the whole thing with us.

 

Ok I'm in.

 

Well, it's usually just a statement less than 1000 words, as this is what is used on many online petitions which is what I was intending to use.

 

Dell no longer markets the M11x-R1 as "overclockable to 1.73GHz" so if you recently ordered one of these, you will get exactly what you paid for. I have no idea when this statement was removed from the Dell website.

The only people that can complain are the ones that purchased this laptop based on the above statement which has now proven to be false.

There's not a huge difference in performance. It's just a matter that some people don't like being lied to by the marketing department when the engineering department was well aware that the M11x-R1 can not possibly run at that speed. Kind of like a car company advertising that an engine upgrade will put out 450 HP when in fact engineering knows that it was deliberately limited to only 400 HP.

 

Unclewebb, is the below statement true?

 

That statement is definitely FALSE. CPU-Z 1.56 is reading these CPUs correctly. Go back to page 2 in this thread and read about these "other programs" and Windows. They are all wrong. Dell has locked the multiplier at 6.0 when overclocking so it is impossible for it to increase to 6.5. 1.73 GHz is a marketing myth.

 

I'm in for a petition. Im sure if we all just call and chat with them and keep bugging them itll have some effect.

 

Well, off topic but in Engineering like Automotive, the BHP they stated are usually true, without taking parasitic loads into account that's it. But in this CPU case...lol.

 

Well, as someone who also got an M11x with 1066 memory.... that never runs at 533 but at 400... I am not too surprised about the lock on the multiplier at 6. I think they fell in love with a cooked test mule and that became the specs. Bet IT didn't have hinge problems....

 

Funny story...I called alienware today just cause i wanted to know what they'd say. Guy ends up running prime95 and messing with power settings for almost an hour before telling me that they'll send a technician to replace something in the computer (mobo im guessing). I told him everything id seen here about how the multiplier is locked at 6 through bios when oc'd but he insisted that the multiplier should be on 6.5x and frequency should be running on 1730 through cpu-z. So im not sure if I should let them touch it, if it'll be no benefit and ive heard they use refurbished parts for repair? The guy wasnt from the us either. Im pretty sure i hear him yelling at someone in another language one time during the phone call.

What do you guys think?

 

Did you call Alienware support (866-287-6727) or Dell support?

As to your question, I'd tell them not to bother sending anyone out to replace the motherboard unless there's something actually wrong with it. The chances that something gets screwed up are not worth it.

My guess is that the support rep simply wasn't aware of the facts about the clock and the limits that Dell has in place on it.

 

I called the alienware support number. Thats what i thought about the rep and i dont think ill be having anyone coming over to fix something thats not broken...just could have been made to work better lol

 

How about after the tech does the motherboard swap, if it is still stuck at 6.0 then make sure he swaps back to your original board. I think this is a good first step. If Dell promised you 1.73GHz and they can't deliver then it's up to them to find a solution. Would a M11x-R2 with a Core i7-640UM help you forget about this problem?