M1330: Temp Probes

Hey,

I noted that Sandra's reporting 4 seperate temps on the M1330... ! For example, the following temps were reported with system @Idle/CPU@80%/3DMark03 system, ambient @22C:

Board Temp. (41-47-52C)
CPU Temp. (33-?-61C)
Board Temp.2 (41-49-52C)
Board Temp.3 (34-67- 83C)
System FAN (0*-2960-3075- 4352** RPM)
CPU Core Power (29W)

*Fan RPM can fall below measurable RPM @Idle, but this does not necessairly mean that is completely stopped.
**Damn, at that RPM, the fan sounds like a hair dryer, lol

Maybe it's correct, but I find it hard to believe that a Notebook mobo would have 3 separate mobo probes...?!? Board temp & temp2 look to similar, for one thing. On the other hand, the ultra-high temps measured by Temp3 seem to be the onboard 8400M GS GPU. On my PCs, GPU temps always surpass the CPU's during gaming / 3D-benches.

Can anyone confirm this with their own preferred monitoring tool(s) please?

Thx

 

could you run TAT( http://www.techpowerup.com/downloads/392/Intel_Thermal_Analysis_Tool.html) for a while and then tell us the max temps for both CPU cores?

 

I dled it from your link, installed it but it won't run:

"Administrator login is required to run the tool. Please look at the known issues section in the readme and restart the tool from the administrator login."

I've had this error once before w/VISTA and one other SW app; of course my user has Administrator rights (& is currently the only profile defined on this system).

Is there something I need to set in order to bypass this issue, or do I really need to use the default "Administrator" profile?

[edit] OK, I noticed the run as administror opt. in the sortcut's context-sensitve drop-down & now it runs

 

Thanks, that cleared up a few things! Max CPU temps for both cores was 73C and Sandra showed similar temps for both "CPU" and "Board temp 3".

Sandra's other two values measured 54 & 57C respectively, and were very close in value throughout the tests (60%-100% loads in 5% increments every 10mins).

Thus, no real connection to the GPU; a shame because a probe for the GPU wouldn't be a bad idea!

Btw, fan RPM was at 2980 (70% load), 3754 (80&90% loads) & 4352 @ 100% load.

[edit] The above values are from TAT's workload test with a gradual climb from 60% to 100% load. The entire time VISTA's "Sidebar" cpu guage always showed 50%.

Then I tried using the Monitor and manually set workload levels to 80, 90 & 95% respectively and left "on demand trottle" at 0%. Wow, what a difference! First off, Sidebar was pegged at 100% during all settings and I hit some pretty scary temps:

%Load...............Core0...........Core1........MHz
80....................75................79............2200
95....................62................66............2200-1600-2200-1600-... this throttle pattern started instantly.
90....................78................83............2200 for 3mins, then throttled down to 1600 for 12 secs...

It took about 4mins @idle for the temps to get down to a constant 40/43C respectively.

 

I was wondering about this auto-throttle business and installed & ran TAT on both my Lenovo (see sig.) as well as my gaming PC (C2D E6400). Both of these dual-core systems managed to go to 100% without throttling. The Lenovo hit 81C at 100% and the PC hit 63C @100%.

So this makes me wonder if I should be worried that my T7500-equipped M1330 is throttling! Can/ has anyone else run TAT on their M1330s at 95% or better and did their cpu(s) throttle too? What max temps were reached?

Thx in advance for any info/ experience you may have on the subject

 

I recently got a M1330 with a T7500. My idle temps are in the low 40 range. I've found that it is impossible for me to have my CPU run at the full 2.2Ghz.

Intel's thermal monitor always kicks in and throttles my processor under a full load. This happens even if my cores are in the low 50 range. If I disable thermal monitor 1/2 via RMClock then I can get it to run at the full 2.2Ghz and my temps get up into the 70's. I didn't test for a max temperature though as I was getting a bit uneasy.

Not sure what's going on. I thought it might have been a problem with the cpu heatsink but I'm not sure now hearing that your lenovo hit 81C. I'll try TAT later when I have time. I'm still within my initial 30 days so I may return my M1330 if I can't get this fixed.

 

That's definately not cool. My cores jump to 2200MHz when I open a web page. If the system's idle, it'll drop to 800MHz (lowest value recorded). Btw, what did you use to monitor the core clocks? I am ALSO assuming that your performing all tests using AC (mains).

But as I mentioned in my post, if I use TAT to set both cores with a %95 workload, then the system throttles the cores immediately, regardless of temps... .

what we need is for a few more M1330 owners to give us their values to figure out in the end what "normal" is!

 

my T7500 runs cooler than T7300 in full load, with a max 66C(idle 33C amb:25C).

 

Well, that's very cool compared to our results, but you do have a different NB and maybe the cooling is better...

 

I was on AC am using CPU-Z, Everest and I8KfanGUI (only to monitor temps) to monitor my FSB and temps. Everest does induce an occasional 10% using or so I ended up using I8KfanGUI to monitor temps on the cooldown.

Hmn.. rmclock must do something when it it gets loaded even after i close it. I rebooted and didn't load RMClock. Using Orthos to load both cores, this time I was able to keep the cores at 2200 until I hit around 70C or so. At that point, throttling kicked in and I jumped between the 8 and 11 multiplier so 1600 to 2200.

Sorry about the earlier eroneous information. It looks as though it's pretty much impossible to run CPU at full capacity. After shutting down Orthos, and letting things get to a stable state, my temps go down to 47/48 at idle. At idle my mutiplier jumps between 6 and 8.

To be honest this is a bit annoying. I can't actually run my CPU at full speed for more than a minute or two before throttling kicks in.

On my 640M, the T5300 never enduces throttling. At max performance my T5300 runs at 1730 (133 x 13) with the voltage reduced to 1.00 and the temp maxes at 66C. The fan never kicks into high speed unless I've done something dumb like put the 640m on my bed and blocked the bottom vent.

unsure of what to do given I'm still in my first 30 days. I don't think the M1330 is very efficient in dispersing heat. The cpu exhaust is usually blocked partially by the screen. TBH i didn't even want to get a T7500. I wanted the T7300 but the rep I was dealing with was sick for a week and in between that time the T7300 was discontinued.

Edit - BTW Orthos is just a modified version of prime95 that runs 2 instances.

 

M1330's CPU and GPU share heat sink while 1400 with X3100 has independent heat sink for CPU and GPU respectively, that might be a reason.

 

BTW, this test was run using the 6 cell battery. I wonder if I should just return my M1330 and wait till the new Penryn laptops are released. I believe Penryn runs at 26W vs the Merom's 34-35W.

 

I haven't exactly understood your main concern; is it the max temps (altough you mentioned 70C, which is no where near as bad as I managed with mine), or is it the fact that you noted that the CPU throttles?

I'm going to have to try running PCMark05 /3DMark0x while running TAT in monitoring + logging mode to see how soon / how often it throttles. In any event, I know that my particular unit scores on par with all the reviews I read (scores in my sig.).

Last but not least, Vostro1400user made a very good point that the M1330 uses the same cooling pipe for both GPU & CPU (& North Bridge). I doubt my Lenovo's on-board ATI X1300 causes as much heat...

 

technically, the penryn chips should work with santa rosa

you should be able to just drop the new chip in, and maybe with a BIOS flash before that, be good to go, if you don't want to return your lappy

***uh....maybe not? theres news that intel is changing the Voltage requirements....so the Mobo needs new voltage regulators.

(slaps self)

 

considering GPU yields significant amount of heat, i would not buy a laptop which has its CPU cooling pipe shared with GPU.

 

Idle temps run high 47-51 range for me. My max temps get up to the high 70's until the throttling starts to kick it. From idle to max temp before throttling is maybe a two minutes.

My issue is that the M1330 does not appear to be able to handle the temperature of the T7500. If the T7500 has to throttle itself then really I'm not getting 2.2Ghz of performance from the chip. Some people may say it's designed that way but throttling is a safety mechanism so you don't burn out your chip. If I can get it to kick in within two minutes, obviously the M1330 isn't equiped to handle this amount of heat and I'm not getting what I paid for.

 

Hey,

I still don't have a clear picture of your testing environment and I don't trust RMClock because it doesn't seem to report my actual clock speed(s) and/or cpu utilization (always shows a min. of 70% which is incorrect & documented in RMClock's own forums). I recommend that you download TAT - it's from Intel, it doesn't change any settings, only monitors CPU and can produce an artifical workload.

Speaking of which, I used TAT to monitor my cpu(s) while I ran 3DMark06. I noted the times when 3DMark06's CPU tests started & completed, etc. I also noted when the fan shifted up & down a gear.

The good news is, from the beginning of the CPU tests until pretty much the end of all tests (6mins elapsed), both CPUs were pegged at 2200MHz with no throttling whatsoever (except shortly between tests)! CPU temps recorded during these 6mins ranged from 61/68C to 69/70C for CPU 0 & 1, respectively. Admittingly, the fan was at 100% ("hairdryer" speeds) for the last two mins of the elapsed 6mins.

I started 3DMark after noting the idle temps of 37(0) & 39C(1) [ambient temp was 21C] and stopped "logging" after the temps were back down to normal (38/39C) which was about five mins after the testing was completed.

So, considering the above results, I do not see anything to be concerned about (anymore) and I assume that artifical cpu-loading apps like TAT, Prime95 etc. do not represent real working conditions. What other explaination can there be for the various results we've experienced?

As for vostro1400user's recommendation of seperate cooling solutions, I really doubt there exists a 13.3" or sub/ultra-compact Notebook that can offer this, so the final question (which is/ should be also the initial question) is what's more important, mobility or power...

 

I will try TAT today.

Condition wise, my condo is 23C. I clean booted my system. Connected to the AC adapter I set my system to high performance. I then loade Everest and CPU-Z to monitor my FSB, multiplier and temperatures. Once I've let my system idle for a while to heat up to a stable temp, I load Orthos and start it with settings to minimize ram usage. Both cores are immediately loaded to 100% and things heat up till they reach the 70's and throttling kicks in.

While a CPU load application doesn't necessary represent most day to day activities, I'm sure if you were to do something such as media encoding then you would max your processor for an extended period of time.

If I had known the m1330 would not be able to handle the heat of the T7500 I possibly would have gotten a T7250.

 

Did you get the chance to test with TAT yet? Same conclusion(s) reached...?

 

Sorry took a while to get around to testing with TAT.

I was unable to test my system off the AC yesterday as my condo's power was out but I didn't figure out at least one part of the problem.

For some reason my 6 cell makes my system throttle early. The issue is not heat, the system will throttle regardless. I checked to make sure I was snapping the battery in firmly and it was. I may have a defective battery.

With my 9 cell, I'm at 50-55C at rest and 85C maxed out. I ran my system at 85C for about 8 minutes yesterday and did not have the system throttle at all. Don't know if 85C is too hot but system seemed ok. Definately not temp related because when I used the 6 cell, I made sure I was not covering the vent.

Dell's XPS support has been piss poor. I spoke to a tech a week ago, he said he would research the problem and get back to me. He was suppose to email me after our first call, didn't email so I couldn't get in contact with him via email. I called the tech back this week after giving him 7 days.

The tech tried passing it off as being a feature of Intel's to save power. I told him that he had to be kidding. When I have the battery profile set to high performance and I'm maxing both cores... Intel decides that I need to save power then? Huh?

He then tried to get me to disable dynamic acceleration and speedstep.... no dice... my processor defaults to 1200Mhz (this was on the 6 cell). What's more, I need the battery saving features so I would not be happy with this solution.

Lastly he suggested that I call up Intel and ask them if this was normal for the processor. At that point I started to get quite irate. I bought the system from Dell not Intel so why are they referring me to them.

We scheduled a remote desktop appointment for this Saturday. After my call I discovered it was the battery by chance. Yes it's not a very easy thing to diagnose but I don't appreciate the tech trying to tell me that the instant throttling was a feature and not a problem. Telling me to contact Intel after I've already wasted too much time diagnosing this problem really irritated me. This is bad enough that I think I will write Dell Corporate.

If the tech didn't know what the problem is then they should say they're unsure.... don't give the customer BS information especially, when said customer has demonstrated that he's technically proficient. My first call with the tech was very long and we ended up talking about technical things not related to my system, so he should know better.

 

More weirdness. It's not just a defective 6 cell.

While using my 9 cell with the AC adapter, my processor will throttle to 1600Mhz. It's somewhat funny, running off battery on high performance it will run at 2200Mhz. When I plug the adapter in, it throttles to 1600Mhz and yes it still is on high performance profile. I give up. Not sure where the problem lies.

Edit : Now it is running at full speed off the 6 cell battery. I give up figuring it out. Still throttles on AC with both my adapters. I don't know what's going on... flaky laptop/batteries/adapters?

 

Sorry to hear it...! Can I assume that your still quoting results noted while executing Orthos? If you used TAT, at what load % did it (immediately) throttle at?

I just got a 6-cell today and I haven't noted any difference between batteries, so in all honesty, I think you're barking up the wrong tree with your flaky battery theory.

My money's still on internal processor settings...

 

No I'm not using Orthos, just TAT. Just tried TAT again w/o the battery and it throttles immediately. Goes from 2200 to 617-619 and then up to 800. Right now it occasionally goes up to 2200 but mostly bounces between 800 and 617-619.

Can't figure things out, my results are too inconsistent aside from my processor always throttling to 1600 under the AC adapter.

 

Under what kind of load??? If you don't put a load then the cpu will throttle by default. If you tried a load 95% or higher, then you have the same odd behavior as mine...

 

This was under 100% load. A bit annoyed with Dell's service. I had a phone appointment booked for 10:30 Saturday for a tech rep to do a remote connection. Waited around till 12:00 for him to give me a call. Finally called and he was sick, thanks for Dell not giving me a call and wasting my time. At least the rep I spoke with was more capable than the one I had dealt with before. He didn't tell me to call Intel which was a plus. Anyway had me do various diagnostics and then had me do a factory recovery. Problem still isn't solved so this system is going back. On the fence on whether to reorder my system or go with a Lenovo and not have to deal with Dell anymore. Ordered 4 other computers from Dell over the years but this may be my last. I will speak with customer care on Monday and voice my disatisfaction over my current experience.

 

My GPU got up to 88C when running 3DMark06 & using the 6-cell battery and 86C when using the 9-cell. The 2-degree difference being due to the 9-cell's better air-intake allowance.

As scary as 88C is, I just logged 99C from my 7900GT running 3DMark06
...as if that wasn't scary enough, I logged 104C running Gears of War...