Paperclip

Silly me, somehow the max. CPU state is at 100% on battery. I must have been forgotten to switch it back to 99%

I'm starting to think that this laptop is not meant to be equipped with a quad core because the fan sucks too much juice from the battery to cool the core.

The BIOS and power options are correct. Some laptops do have this problem from what I've read....

Thanks! I did try sleeping and waking up the 2570. No issues with the turbo mode. Working exactly as it should.

Mine doesn't work after waking up or going back to AC after using the battery

At first I thought it's because the new installed 1866MHz Vengeance RAMs but turns out that even with the factory RAMs, it still behaves the same

Hey krazyazy, could you please check your turbo mode? Put your laptop into sleep mode and check if the turbo works after waking up. I'd be very grateful!

BTW congratulations for the upgrade! At first, I was thinking about buying the same CPU but at the end I end up with a 240€ 3820QM....

The Vengeance result doesn't seem to be seeing the CPU engage the turbo. Would be a good reason why you're seeing a significant decrease in the physics result : 7990 -> 6693 (-16.2%). Suggest re-do the Vengeance test.

I found out why the turbo wasn't engaged: it's because the turbo is disabled after resuming from sleep or after going back to AC after using the battery!

It's like the system can't recognize the current state it's in after a state change. Could you guys please check and report how your system behave?

The Vengeance result doesn't seem to be seeing the CPU engage the turbo. Would be a good reason why you're seeing a significant decrease in the physics result : 7990 -> 6693 (-16.2%). Suggest re-do the Vengeance test.

You're right and I didn't notice that before. It's weird because I used the same power options setting (HP Optimized) for the two tests.

This is the new score I got after changing the power setting to high performance: Intel HD Graphics 4000 Mobile video card benchmark result - Intel Core i7-3820QM Processor,Hewlett-Packard 17DF

The cores were in full force now.

Or it could be like @Aikimox mentioned, that the memory actually might down clock to 1333MHz?

Is there a sure way to check this?

I dont' think so because it shows that the RAM frequency to be 933MHz.

Okay, so here are the results from 3DMark11:

hynix: Intel HD Graphics 4000 Mobile video card benchmark result - Intel Core i7-3820QM Processor,Hewlett-Packard 17DF

Vengeance: Intel HD Graphics 4000 Mobile video card benchmark result - Intel Core i7-3820QM Processor,Hewlett-Packard 17DF

My system performs a little bit worse with Vengeance installed, weird....

I haven't tried 3DMark11 before. Should I?

Ohh I see!

Have you confirmed that they are running @1866MHz? (a screenshot of this confirmation).

Yeah, sure, it's at the end of the last page.

Also, if you could test 3DMark06/Vantage/11 with your different set of modules it would be really nice! (links to the 6 different 3dmark results).

The 3DMark versions taxes the system in, sometimes very, different ways and the RAM could show improvements in certain scenarios.

If you do this we can be certain of any improvements, also your findings and contribution would be noticed summary on the first page!

Thanks a lot buddy!

No problem!

I use the latest 3DMark because once I used 3DMark06 and it couldn't display the score because my "graphic card" (which is the iGPU) can't be registered. So, the scores:

With 2x4GB hynix 1600MHz CL11: Intel HD Graphics 4000 Mobile video card benchmark result - Intel Core i7-3820QM Processor,Hewlett-Packard 17DF

With 2x4GB Corsair Vengeance 1866MHz CL10: Intel HD Graphics 4000 Mobile video card benchmark result - Intel Core i7-3820QM Processor,Hewlett-Packard 17DF

Although 3DMark scores always fluctuate a tad bit, the results tell me that the Physics scores are a little better with Vengeance RAM and the Graphic scores are a little better with hynix RAM. That would be against my expectation but I don't know if that means something.

I read some strange thing that Corsair RAM in some cases manage to achieve 1866MHz, even with 1600MHz rated mobo (which super weird), wish I could find the source, sorry!

If someone happens to have a 1866MHz Vengeance module, please test this.

I've just got Vengeance 1866 and it works. Unfortunately 3DMark reported no performance improvement.

Ah, nando why didn't you tell me earlier! LOL, joking! At least now I have two new cool grey sticks with VENGEANCE!!! written on top of them!

Anyway, cool info..., and yeah I probably should get an eGPU.

AMD A10-5750M takes 1866 RAM

AMD APUs for Laptops

I'd imagine that their APUs would love the extra speed

Hmm I wonder why we don't get any from Intel.

Anyway, my two 1866 Vengeance have just arrived and after putting my system to the test, 3DMark said that there's no improvement. My scores keep hovering around 60000 and 6400 for the Ice Storm and Cloud Gates tests, respectively. It's a bit disappointing because my system actually recognizes the new RAMs!

Intel says that my 3820QM doesn't support 1866 although it seems that the mobo does. What's more I couldn't find any of Intel's mobile CPU that supports 1866. Now I wonder why they sell those 1866 SODIMM RAMs.... What interesting is, that these 1866 are rated at 1,5V and not the usual 1,35V. Maybe you have to somehow OC your system to enable the RAM operating at this speed?

How did you check it, Aikimox? Crucial says that my (our) mobo supports up to PC3-14900 (1866).

Good question! I recently read about this, can't remember where atm. The guy made comparisons between 1333MHz vs 1866MHz and 1600MHz vs 1866MHz.

The latter resulted in 2-3 FPS increase in gaming. So I'd say it's not worth it.

Edit: this might be interesting for you:

DDR3-1866 Benchmark Results - Seven 8 GB DDR3 Memory Kits For Your AMD A75 Motherboard

Hey thanks for the link! But all of those RAMs are faster (maybe just a tad bit) than our standard CL/CAS11 1600MHz hynix.

Linus has also this investigated this issue:

But this review mentions that they got 45% FPS increase in Aliens vs Predator D3D11 Benchmark v1.03: Corsair Vengeance 8GB 1866MHz CL10 Laptop SODIMM Memory Kit Review - Page 6 of 7 - Legit ReviewsAvP & Hyper PI

The two RAMs compared on the last review are exactly what I have on my machine and what I'm interested in.

I'm thinking about upgrading the hynix 1600MHz CL11 RAM to Vengeance 1866MHz CL10 but I'm not sure if the increase in performance (gaming) would be noticeable. If I only use the iGPU, how much performance increase in your opinion, let's say FPS wise, can I expect to get?

Yeah I tried to cut the NC8430-fan down with a dremel in order make it fit our system, unfortunately I wasn't much left after I was done.

How about the NC8430 heatsink-pipe-radiator part? I guess it also cannot be fitted correctly?

@jacobsson

Have you tried mounting the copper NC8430 fan? How is it going? I'm sorry if you have already done it and I missed the post!

BTW I've kind of "successfully" managed the CPU temp just by using HPFanControl, which I stupidly didn't know to exist before.... Now I don't need to limit the CPU using Power Options anymore and thus disabling the turbo mode, but I still want to lower the working temp especially when summer is coming. One thing for sure though, It's definitely not silent..., but I'm ok with it.

BTW again, here's my VID and TDP data from ThrottleStop (1024M, background processes as minimal as possible):

12x / 0,8556V OR* 0,8456V / 10,7W

23x / 0,8556V / 17,3W

24x / 0,8706V / 18,5W

25x / 0,8906V / 19,7W

26x / 0,9056V / 20,9W

27x / 0,9257V / 22,4W

28x / 0,9457V / 24,0W

29x / 0,9657V / 25,6W

30x / 0,9857V / 27,4W

31x / 1,0157V / 29,5W

32x / 1,0408V / 32,0W

33x / 1,0658V / 34,2W

34x / 1,0958V / 37,5W

didnt continue because I'm afraid that my CPU would explode because it had reached >90 degrees!

*) I got different values when I went up and then down again after reaching 34x multiplier.

Am I doing this correctly?

EDIT: forgot to mention that my CPU is i7-3820QM

EDIT: (perhaps mods can merge this post with the one before, thanks!)

I decided just to limit the CPU to 80% under power options (it makes my 3820qm functions more like 3632qm I think). I get no noticeable performance drop while playing games except when there's heavy physics processing involved (confirmed with 3DMark). What important is, now my CPU temp never gets above 70 degrees (it was 90+ degrees before)! I know that it kind of defeats the purpose of installing a 3820qm....

And one more thing, the fan spins only up to around 3300rpm under "BIOS" no matter how hot it is which then permits the CPU's (3820qm) temp to get dangerously high (90-95 degrees) under load (games or 3DMark)!

Is it a good idea to set your fan at full blow when playing games (I use TPFan)? Using this method I can slash the temperature down close to 20 degrees (from 85 to 65 degress).

SEC4149

I bought the machine two months ago.

I just upgraded my i5-3320m CPU to a i7-3820qm.

BTW when I was browsing ebay looking for some CPUs, I stumbled upon a x230 motherboard. It shows that the cooling system has 2 pipes. Now I'm wondering whether the temp would get down if 2570p was also equipped with 2 cooling pipes.

Welcome to T|I and congratulations on your 2570P acquisition. Imho a far superior piece of gear to a X230 in terms of build quality and performance potential. It now rightfully out-ranks the X230 at Best Laptops - The Top 50 Most Popular Notebook Computers, as it should have always been.

I have no problems with my keyboard.. certainly not mushy and feels far nicer to type on than any of the older Thinkpads I've used previously (T61, X201). Perhaps see if you can test on another 2560P/2570P to ensure yours isn't a dud, even asking HP to send you another under warranty if need be?

Hey, thanks for the welcome!

Yeah the CPU (quad) upgrade capability of 2570p is what attracted me the most. Regarding the keyboard, I don't think I can get my hands on another 2570/60p for me to test. I'm pretty sure though that mine is original because when I check under warranty on the HP Support Assistant program, it acknowledges my machine. I can't pinpoint what's lacking on the keyboard, but I swear it was better on the two Thinkpads that I've tested (actually you've got me doubt myself - I've to compare it next time with my friends TP).