New M6500 Discussion Thread

@havoctex:
Ok, I found what you mean:

source: http://download.intel.com/design/processor/datashts/322812.pdf (1.3.1)
But strangely the document for the Clarksfields says exactly the same:
http://download.intel.com/design/processor/datashts/320765.pdf (1.3.1)
So obviously this information is not reliable.

And I found some other strange thing. Earlier you posted this:

In this article the M6500 is advertised to be available with 32GB of RAM in the near future. But Intels datasheet says this:

We all know this is not correct.
And to add another contradiction, this is an output of my system:

Code:

debian820QM:/home/hikaru# dmidecode | grep 'Physical Memory Array' -A 14
Physical Memory Array
	Location: System Board Or Motherboard
	Use: System Memory
	Error Correction Type: None
	Maximum Capacity: 16 GB
	Error Information Handle: Not Provided
	Number Of Devices: 4

Handle 0x1301, DMI type 19, 15 bytes
Memory Array Mapped Address
	Starting Address: 0x00000000000
	Ending Address: 0x002FFFFFFFF
	Range Size: 12 GB
	Physical Array Handle: 0x1000
	Partition Width: 0

As you can see, I have 12GB installed, and the system reports that a maximum of 16GB is supported. How can you make an advertisement for 32GB to be supported (which is apparently based on the recent availability of 8GB modules) if the system itself reports 16GB to be the maximum?
I can think of two reasons:
1st: the limitation is due to my BIOS (A00) and a later BIOS has an increased adress range. Can someone with a more recent BIOS (preferably the latest one) please run the same command like I did to verify or falsify this?
2nd: If it's not the BIOS but the chipset (I didn't check the chipset specs yet), upcomming M6500 will get other chipsets which means that current M6500 owners won't be able to upgrade beyond 16GB.

 

With a bit of spare time, I dloaded an Ubuntu image and ran the test install. dmidecode reports the same for mine as far as the maximum's concerned, so it's not your first supposition.

I'm not sure that I'd jump to conclusions just yet as far as current owners (or early adopters as the case may be, though this model's going on eight months now?) not being able to upgrade. A05 was released and revoked once already if I remember correctly. That being the case, we'll see shortly.

Maybe I'm reading more into your text than what's there, but at this point, especially after all of your apparent issues and reservations, I'm a bit surprised that you haven't taken the first opportunity to move onto another hardware vendor that would satisfy your computing needs with less resultant pain inflicted upon yourself. Sounds like daily use of this particular model is a burden.

 

Thanks for the test! Which is your BIOS version? Maybe it just needs another one or two BIOS updates.

No, I'm not unsatisfied with my M6500. Now that all my starting troubles are solved, I really like my machine. It's a great piece of hardware. What I'm complaining about are some things around this piece of hardware.

1st: Dells false advertisement on the availability with RHEL. There are different press releases and system overviews that list RHEL 5.3 besides the Windows options. This allows two interpretations:
A: The list shows the OSes which are available to purchase the M6500 with. Since RHEL is not available from Dell, A obviously doesn't apply.
B: The list shows the OSes which are officially compatible with the M6500. If a manufacturer claims an OS to be compatible with his system, I expect it to be 100% compatible unless he says otherwise (Dell doesn't). I tested the M6500 with CentOS 5.3 & 5.4 (fyi: CentOS is a 100% compatible recompile of the RHEL sources). But I found that it is not 100% compatible with the M6500 since the Ibex Peak soundchip has a bug that is not yet handled by Linux kernel 2.6.18 used by RHEL 5.x which leads to non-working speakers.
So my question is: What is Dells reason to put RHEL 5.3 on that list?

2nd: Dells almost assured bundling of the M6500 with Windows by crippling the configuration software of their phone reps. Actually my reps were very accommodating when I tried to make use of my by law granted right to unbundle Windows. They just had no chance to do it. The only way it finally worked was, that they handed my case over to their boss (who surely has better things to do than handling simple orders) who finally managed to outmaneuver his software (no idea if he has a different one) by choosing a Hungarian Windows which is not available in Germany and makes the software fall back to FreeDOS.

3rd: The RAM issue. No matter if my suspicion that Dell gets the Intel documents wrong is correct or not, they should at least train their staff to handle both configurations correctly. A case like mine, where things get mixed up can always happen, that's not my point. But that it takes two weeks and three visits of technicians to finally find out, that there is just a wrong entry in the database, must not happen - at least not if it's such a trivial thing that a system with a Clarksfield CPU should not get an Arrandale board. This means unnecessary costs for Dell and mayor inconveniences on my side, because I had to take three days of holiday that I'd rather have spent differently than sitting and waiting for someone with a screwdriver (and the wrong replacement board).

4th: The information politics which is effective in advertising, but poor in real valuable technical information. Let's take some of havoctexs statements in this thread. This is not meant to be offensive against him, actually I'm very happy to have someone here from "the other side", it's just the closest example for the general situation.
There is this article about the memory increasing. I guess the ones who have seen the M6500 from the inside know, that there is neither space nor a connector on the board to add additional memory slots to the four that are already available and the ones who slightly know the Intel specs know, that further slots wouldn't be supported by the chipset. If we now take into account, that just at the time of this article 8GB RAM modules were about to enter the market it's clear that Dells only "achievement" will be to put those new bigger modules in their laptops. Hey thats still great for the customers, but Dell shouldn't treat this as their own exclusive achievement.
Then, there is this 2-slot vs. 4-slot issue. Before the yesterdays conversation with havoctex all I heard from Dell about it was "It's simply this way." Nobody was able or wanted to give me some reasonable explanation. All I knew was "Dell says so." Then yesterday I got a bit further when havoctex correspondingly changed this statement from "Dell says so." to "Intel says so." So I know now that it's not an artificial restriction of Dells BIOS and I should find the reason in Intels specs. I've looked them up, and after all I understand that according to Intel the memory-relevant specs for Clarksfield and Arrandale are the same. Both say, only two slots are supported. Apparently the Clarksfield specs are faulty. So why should one assume that the Arrandale specs are correct when both are derived from Lynnfield? If Dell has documents that show the difference, it should be possible to provide a link to tell me: "There you get it wrong." According to the documents I have, I can say: "I think, there you get it wrong and point at a specific chapter."

So to conclude this rant: The M6500 is a great laptop, thumbs up for the designer and manufacturing teams. But Dell needs to sort some things out behind the scenes when it comes to logistics.

 

Hi All

Just got my M6500. Ordered it with a 128GB SSD (Samsung PM800) drive and a 250GB standard drive to use as a backup.

They are configured for non raid. OS is on the SSD drive.

I've noticed in the bios that under System Config>SATA operation, that out of 'ATA', 'AHCI' and 'RAID' that I have 'RAID' selected. This is how the laptop was preconfigured when i received it.

My gut feel is that the SSD isn't giving max performance. Under AS SSD benchmark I get a seq. Read write benchmark of 198/148 MB/sec which I thought was OK...not great.

Any advice appreciated

many thanks

 

Debguy, I feel your pain. Nothing against Dell engineers, but they seem too fast in giving meaningless statements, instead of checking deeper.

Dell says so, they say. You ask, "Why". Then it is, "Intel says so." You point out the apparent discrepancy in the literature. To me, that says that the engineers or support personnel, didn't check the literature. Dell isn't a small-scale operation. You'd expect that a large business, with such a large pool of engineers would notice this little thing?

As pertaining to my own situation, I was indeed called back on Monday by a gentleman from Second Line Support. I sent him both the screenshots as well as some posts from this thread with regard to the situation.

Today, I got another call from him. They've sent it up the line, including to the engineers in Ireland. Good, they're taking it seriously now. What did sour this was that he insisted it had to be a software problem, since the USB 3.0 spec is still in draft formation. He also seemed to imply (sounding a bit frustrated) that there's a deflecting of responsibility between the various parties ie. NEC/Dell with regards to the problem.

Assuming it is a software problem, why does it not affect everyone who has USB 3.0 on this machine? I've tried the official Dell as well as the subsequent official NEC drivers. If it were a software problem, shouldn't USB 3.0 be kept back until it is fixed? And why sent the problem up the chain to the engineers? Or is it a software problem related to some other drivers in the system?

I want facts, not suppositions, guesses or tea-leave reading. If you don't know, either say so, or don't say anything until you've found the information. Don't get me wrong, I'm happy they're finally moving on this, however, the way they try and think-out-loud before they have all the facts grates on me. In the meantime, I'm still here with the blue USB ports which don't work.

 

@Mr. Chapel:
If you have a possibility to actually check if USB 3.0 works or not (i.e. a USB 3.0 device for testing, not just some system panels that claim it works) you could boot the laptop from a recent Linux Live CD (at least kernel 2.6.31) and try if the USB 3.0 device works as such.
If it does, you know it's not a hardware issue, but if it doesn't a hardware issue is very likely.

 

I don't have a 3.0 device at the moment. I have tried what you suggested, using several different USB 2 devices (three different USB keys, a USB dvd burner, two external HDs, Sharkoon Quickport). Tried it through W7 and with Ubuntu 10.04. The two USB ports don't even make a sound when plugging in a device nor show it being mounted.

 

Ok, I didn't retrace your full story. I just assumed that the USB 3.0 protocol doesn't work, not that the ports aren't working at all. In this case testing them with USB 2.0 under Linux is fully sufficient to tell if there is a hardware issue.
I guess the Dell support is as inattentive as I was in the first place.

 

all you have to do is make sure your power options are set to performance on windows

open throttlestop and set your clock-chipset modulation to 100% and set multiplier to 16 i guess that is the max on the 920XM and click save , after that right click on throotlestop and you will see two options turbo power limit " and turbo ratio limit"

set your ratio limit to 23 on all cores and now your power limit has two options TDP and TDC increase that slowly and keep eyes on your temps you can increase by 10 at the time each if you have wprime running you will see how the speed is increasing and better scores also but don't go to crazy on the TDP-TDC because you can fry your power supply or motherboard

 

Don't beat yourself up over that Often enough, we go looking for complicated problems while it's a simple thing, most of the time. Been there, done that.