Jump to content

Recommended Posts

What are your current 3Dmark scores with 35W cpu?

PhysicsScore is something around 7k points ind 3dm11. But I've already sold this CPU and i've been waiting for 3635 instead of 3632.

I posted few plots with i5 and GTX670/GTX680 comparison yesterday (eGPU thread). There is a proof that I5 is really to weak for something stronger than GTX660. Core load is not stable loaded like during almost sleepy i7QM (50-60% usage).

NVIDIA GeForce GTX 660 video card benchmark result - Intel Core i7-3632QM,Hewlett-Packard 17DF

Link to comment
Share on other sites

PhysicsScore is something around 7k points ind 3dm11. But I've already sold this CPU and i've been waiting for 3635 instead of 3632.

I posted few plots with i5 and GTX670/GTX680 comparison yesterday (eGPU thread). There is a proof that I5 is really to weak for something stronger than GTX660. Core load is not stable loaded like during almost sleepy i7QM (50-60% usage).

NVIDIA GeForce GTX 660 video card benchmark result - Intel Core i7-3632QM,Hewlett-Packard 17DF

Ok, my physics score is 8520 with the 3820QM so I seems to be a point using a 45W CPU.

I haven't tried BF4 myself yet but when I do I'll apply the 'unparking core fix' to hopefully be able to wake the i7 from 50% usage.

Link to comment
Share on other sites

Ok, my physics score is 8520 with the 3820QM so I seems to be a point using a 45W CPU.

I haven't tried BF4 myself yet but when I do I'll apply the 'unparking core fix' to hopefully be able to wake the i7 from 50% usage.

o yeah I forgot about parking - I'll try it tomorrow If i get my CPU hopefully. if find enough time, i'll do test with GTX670/GTX680, with parking and without :)

Link to comment
Share on other sites

so is there any sense to putting 45W TDP CPU in 2570p?

U can run it at the 35W performance level and have far better temps than with a 35W quad. You could also run it at the performance level of a 35W dual core and have even better temps. That alone is big pro in my book.

Other than that it's about bragging rights mostly and potentially making your system a bit more future proof. In everyday life and even gaming, I don't notice any difference between a 3632QM and 3840QM. But I do a ton of virtualizaiton and there those extra clocks shine. I was hitting 70% utilization on a 3632QM when running my virtual environment. Now it's like 35-40% tops.

  • Thumbs Up 2
Link to comment
Share on other sites

Hi Tech Inferno Fan, just wondering how you managed to view the bios whitelist like this? I've got an EliteBook 8470p which I guess has a similar bios to the 2570p. I'm planning on buying a wwan card for it and want to know what is compatible. Have tried using PhoenixTool to view the bios info but can't figure out how you did this? Any help would be much appreciated! Cheers

Link to comment
Share on other sites

Hi Tech Inferno Fan, just wondering how you managed to view the bios whitelist like this? I've got an EliteBook 8470p which I guess has a similar bios to the 2570p. I'm planning on buying a wwan card for it and want to know what is compatible. Have tried using PhoenixTool to view the bios info but can't figure out how you did this? Any help would be much appreciated! Cheers

I disassembled the 2570P bios using phoenixtool then did a binary search for known whitelist wifi PCI IDs to identify the whitelist file(s). Then just perused the whitelist file(s) to see the whole whitelist.

See the ultraedit screenshot at http://forum.techinferno.com/hp-business-class-notebooks/2537-12-5-hp-elitebook-2570p-owners-lounge-25.html#post67796 to have an at-a-glance understanding.

Though you can probably save yourself the effort. I believe the 8470P and 2570P will have the same whitelisted WWAN cards. In which case just refer to the list at http://forum.techinferno.com/hp-business-class-notebooks/2537-12-5-hp-elitebook-2570p-owners-lounge-23.html#post67475 .

  • Thumbs Up 1
Link to comment
Share on other sites

I have a new CPU (3630QM) for a couple of hours, I did some tests but I have a feeling that even in idle mode, the average temperature is greater than the 3632QM. And it's not 2-3 degrees but about 7-9C. After a few hours the average in idle is around 45C on the cores, while some tests with 3dm11 max recorded on the cores was 81C.

In my opinion there is no chance to achieve this temps I got with 3632QM (btw it has only 500 point less in 3dm11 physics - 7500 vx 7000).

http://www.nhl.pl/index.php?s=&showtopic=69774&view=findpost&p=724796

Link to comment
Share on other sites

Im considering buying this over a dell m4800 for budget reasons and the fun of upgrading a laptop cpu and making a egpu (670) system.

I have a laptop with a broken screen with a socketed i7 2630qm processor. Would it be possible for the ivy bridge chipset to use this cpu?

Sent from my LT26i using Tapatalk

Link to comment
Share on other sites

Im considering buying this over a dell m4800 for budget reasons and the fun of upgrading a laptop cpu and making a egpu (670) system.

I have a laptop with a broken screen with a socketed i7 2630qm processor. Would it be possible for the ivy bridge chipset to use this cpu?

Sent from my LT26i using Tapatalk

I can't recommend swapping CPU from Sandy to Ivy, I've heard stories of faulty motherboards as results.

Link to comment
Share on other sites

Yea, that was probably main reason why we fried 2570p MB with i7-2630QM

I need to resign for 45W TDP. there is no visible improvement in playing/GPU loading but in temps bump is too very visible.

Few plots from different setting BF4, it doesn't matter. I suppose You figure out which one is for i5, i7 35W TDP and 45W TDP

post-10292-14494996764687_thumb.png

post-10292-14494996765426_thumb.png

Link to comment
Share on other sites

I can't recommend swapping CPU from Sandy to Ivy, I've heard stories of faulty motherboards as results.

Thanks for the reply jacobsson. However you are talking about upgrading a sandy mobo to a ivy cpu. Im talking about upgrading a dual core ivy to a quad core sandy on a 2750p. I gues this is a different outcome. Might even be no1 tried it?

Sent from my LT26i using Tapatalk

Link to comment
Share on other sites

Thanks for the reply jacobsson. However you are talking about upgrading a sandy mobo to a ivy cpu. Im talking about upgrading a dual core ivy to a quad core sandy on a 2750p. I gues this is a different outcome. Might even be no1 tried it?

Sent from my LT26i using Tapatalk

I assume by "2750p" you ment 2570p?

Anyway, using a quad core Sandy on the 2570p Ivy Mobo is not a good idea since there seems to be a compatibility issue between the two 'components'.

Again, I'm referring to the issue when @bjorm (I think?) put a Sandy quad in his 2570p and resulted in a fried mobo.

It would make me sad to see this happen to you as well.

Link to comment
Share on other sites

Thanks for the reply jacobsson. However you are talking about upgrading a sandy mobo to a ivy cpu. Im talking about upgrading a dual core ivy to a quad core sandy on a 2750p. I gues this is a different outcome. Might even be no1 tried it?

Sent from my LT26i using Tapatalk

Yes, with @jot23 we tried use i7-2630QM in 2570p and it was working. After thi, when we were tryingo to swap i5 back, motherboard suddenly fried. Thats over.

Link to comment
Share on other sites

Yes, with @jot23 we tried use i7-2630QM in 2570p and it was working. After thi, when we were tryingo to swap i5 back, motherboard suddenly fried. Thats over.

Yeah read the posts about it now, just seems extremely strange it was working and then failed upon changing back to ivy. Im a stubborn tweaker (i gues you all are as otherwise you would never do this stuff :P) so im just gonna go on a limb and say it has been another reason it was fried. Just seems very weird to me that a motherboard fries upon installing the original cpu because it changes back from an cpu it should not work with but did! Also there is absolutely no mention of how broken the motherboard was, no post?, just random errorness?, burned and eaten by aliens?

Im searching for ether the 2560 or 2570 but whatever is cheaper i will buy, just try to make sure the warranty is good when i buy a 2570 :)

Edit: Okay thanks to you all i just bought a 2560p on Ebay for a grand total of 304 euro including shipping and the Dutch 21% import tax :) :) :) I have checked for ages on the web to find something comparable new but other then the 1700+ euro M4800 from Dell i couldn't find anything like the good old Elitebook series. Have a i7-2360qm and a 840 evo 256gb ready for some upgrade action.

Btw this snow effect from the website makes me panic and think i have screen tearing all the time :(

Link to comment
Share on other sites

Edit: Okay thanks to you all i just bought a 2560p on Ebay for a grand total of 304 euro including shipping and the Dutch 21% import tax :) :) :) I have checked for ages on the web to find something comparable new but other then the 1700+ euro M4800 from Dell i couldn't find anything like the good old Elitebook series. Have a i7-2360qm and a 840 evo 256gb ready for some upgrade action.

Theory: installing a Ivy Bridge CPU on a Series-6 HP 2560P

Congrats on the purchase and joining the exclusive Elitebook users club :)

Given you are a tweaker you might be interested in doing a proof of concept. There is speculation that the Series-6 systems could work with Ivy Bridge CPUs (if you have one to test) *if* ME 8.x firmware was flashed onto them. Series-6 systems have Intel provided ME 7.x firmware. REF: http://forum.techinferno.com/general-notebook-discussions/2091-lets-enable-overclocking-all-6-7-series-laptops.html

2560P has a 8MB BIOS versus 2570P's 16MB so can't use the 2570P larger ME 8.x VPRO firmware. However, Series-7 6570b has a non-VPRO ME 8.x firmware that fits into their 8MB BIOS of which I just happen to have a dump of it here . Use the instructions at http://forum.techinferno.com/hp-business-class-notebooks/2537-12-5-hp-elitebook-2570p-owners-lounge-12.html#post65904 for flashing of this file. You could even try it with your SB CPU. If the system boots and turbo clocks work then the firmware was applied successfully. Then the final proof would be testing a IVB CPU. My only concern being whether the HD4000 iGPU on the IVB could be driven by the HD3000 vbios integrated int the 2560P bios.

I'd be interested in the results of doing this. That's because one negative going from a 2560P->2570P is higher idle power consumption, requiring the battery life tweaks on the first page to partially correct. Even after doing those it still was not possible to get idle as low as I've had on a 2560P. I believe the reason for that is a gimped Intel USB 3.0 implementations in the Series-7, something that the Series-6 2560P doesn't have.

So if this mod was successful, then it a user doesn't need RAID-0, USB 3.0 or the slightly nicer aesthetics of the 2570P, a 2560P modded with ME 8.x firmware could provide users a cheaper IVB upgradeable alternative with better battery life. An inexpensive way of getting one of the best notebooks ever made.

Link to comment
Share on other sites

I have a new CPU (3630QM) for a couple of hours, I did some tests but I have a feeling that even in idle mode, the average temperature is greater than the 3632QM. And it's not 2-3 degrees but about 7-9C. After a few hours the average in idle is around 45C on the cores, while some tests with 3dm11 max recorded on the cores was 81C.

In my opinion there is no chance to achieve this temps I got with 3632QM (btw it has only 500 point less in 3dm11 physics - 7500 vx 7000).

http://www.nhl.pl/index.php?s=&showtopic=69774&view=findpost&p=724796

INFO: How to choose the most efficient 45W i7-quad

Summary: recommend a i7-3720QM or i7-3740QM/i7-3820QM as best value CPU upgrades with at most 9% performance variation between the three. The i7-37xxQM or i7-38xxQM being more power efficient, if battery life is important than the i7-36xxQM CPUs.

@bjorm, I recently tested three i7-3740QM CPUs, all SR0UV spec. Notebookcheck says these have up to x35 4-core mode here. Results below for them. I skipped from x13 to x22, starting at x23 to see the ramp up to the the highest non-turbo freq (x27) of the lowest end 45W i7-quad (i7-3610QM).

Comparing the best (#1) and worst (#3) we find there is quite a difference between them. The more efficient one runs x12 at 0.1051V lower, can run x34 even engaging x35 for some time (2mins with bottom cover off) until presumably thermal limits start seeing it pull back the multiplier. It runs x34 at 0.04V and 3.6W TDP lower than the worst one's x33 multiplier. It's x34 mode using the same power as if the the worst was running at x31.5!! The worse one will not run at x35 let alone x34.

Q30PBhx.png

NOTE: Haswell results from a Dell E6440. Ivy Bridge results from a 2570P. Sandy Bridge results from a 2560P.

// = missing data in sample set

XX = specced multiplier unavailable due to TDP throttling or because is an engineering sample (ES)

BOLD = unlocked multiplier

YELLOW = x12 TDP

CYAN = x27 TDP used to gauge comparative efficiency

MAUVE= highest 4-core turbo multiplier used during benchmark

Note: (i) ordered by voltage/TDP at x27 multiplier, lowest to highest

(ii) values reported are the consistent ones observed after initial benchmark start fluctuation

(iii) TDP reported reflective of CPU temperature. Temperature value not being used due to significant individual machine variance, eg: ambient temps, cooling mods and other internal engaged hardware such as HDD, wifi, WWAN, bluetooth.

(iv) Throttlestop can be used to limit the turbo frequency. Eg: Can limit a 45W i7-3630QM to x29 so it runs like a 35W i7-3632QM.

<span style='background: yellow'>If scouting for the coolest running i7-quad off say ebay, suggest ask the vendor to report the x27 TDP using Throttlestop as I did above. That quick check being a good indicator of overall CPU efficiency.</span>

Maybe @Aikimox can his i7-3840QM CPU multiplier/voltage/TDP table too? I can then add it to the table so we can build up knowledge about how these CPUs differ.

Use of copper shim increased max turbo

What's interesting is now I have my 25x25x0.8mm copper plate sandwiched between the heatsink and CPU, my temps are noticably lower and the i7-3740QM can hold x35 4-cores for at least 2 minutes. Previously it would downclock from x35 to a locked x34 within seconds. This makes me wonder if the CPU temps are the limiting factor here? Now after a number of minutes of a locked x35 4-core mode where the temps have risen I see the turbo then start oscillating between x35 and x34. Presumably to either keep TDP or temps in check.

I'm now wondering if I now installed a US$7.59-shipped 42mmx42mmx1.2mm copper shim instead whether I'd be able to hold x35 indefinitely? That's about the max size of a shim that would fit within the constraints of the heatsink screw mounts, though the 1.2mm thickness might be too much. 0.8mm or 1.0mm would be better.

Details of how to configure Throttlestop to do such a table as above

Click "turn on" , set the multiplier you want then run a a TS bench. I used 1024M as the 32M finishes in like 8s. The CPU will be running at 100% load. Below is the Throttlestop window that I set and I observed while my other 1024M TS bench window was still running to determine the x12 table entry:

eqap.jpg

Regarding the TDP: It jumps up and down like crazy so I don't think there be any meaning to it, meaning no accuracy? I'm not sure if you guys have stable TDP when measuring?

The way I get a pretty stable TDP is I start a x12 TS 1024M 4-core/8-thread benchmark, let it settle into a rhythm and not do not stop it. Then I just change the multi along to the next one .. x23, wait a few secs for it to stabilize and take the volt & TDP reading, then change multi to x24, wait a few secs for it to stabilize and take the volt & TDP reading and so on until the I gather the results for all available multipliers.

I found this gave accurate numbers that can be repeated with subsequent runs hours or days apart.

  • Thumbs Up 6
Link to comment
Share on other sites

Huge thanks for this guide.

Do You think x12 would be enough for good assesment for CPU or should we push to the limit and check the highest settings as well?

I'll try to do it with 3230, 3630 and 3632 (ordered back).

What order should I choose to ensure good comparison?

1. Set the lowest possible mutliplier (don't know why 8x100MHz is possible for Ivy when they supposed to work on 1100MHz as the lowest),

2. Save setting and turn on button, set proper threads (how many if I'd like to compare 3230 vs 3630 vs 3632. 8? 4?).

2. start 32M bench and write down stable values VID and TDP.

Is this correct?

Link to comment
Share on other sites

Huge thanks for this guide.

Do You think x12 would be enough for good assesment for CPU or should we push to the limit and check the highest settings as well?

I'll try to do it with i5, 3630 and 3632 (ordered back).

If you do the same test with your current i7-quad (i7-3635QM?) then we'll have a further sample to decide if this is true to simply request the x12 multiplier voltage as a reflection of overall efficient. I'm curious what voltages your i7-36xxQM sets across the board compared to the more premium i7-3740QM CPU.

I didn't include any dual-core CPUs in my result because a dual-core at any of those multis is only doing half the work (4 threads).

I've done the same.. managed to reorder back my efficient i7-3740QM. With it I was the first eGPU user to get a 3dmark06 result greater than 26k. The 'efficient' i7-3740QM also got me a 8200pt 3dmark11 physics score whereas the less efficient one was 8011pts.

1st i7-3740QM "less efficient one" - runs x33 4C mode

3dmark06: 25802

3dmark11: physics=8011

2nd i7-3740QM "the more efficient one" - runs x34 4C mode

3dmark06: 26121

3dmark11: physics=8200 (don't have the link, result noted during discussion)

Link to comment
Share on other sites

so is there any point to make comaprison between i5M and i7QM's at 4 thread or should reject i5 and focus on comparison between i7QM's with 8 threads setting.

please, take a look on my question about order, from above.

I have i7-3630QM, I bought 3635QM but it appeared to be 3630QM ;)

Link to comment
Share on other sites

What order should I choose to ensure good comparison?

1. Set the lowest possible mutliplier (don't know why 8x100MHz is possible for Ivy when they supposed to work on 1100MHz as the lowest),

2. Save setting and turn on button, set proper threads (how many if I'd like to compare 3230 vs 3630 vs 3632. 8? 4?).

2. start 32M bench and write down stable values VID and TDP.

Is this correct?

While IVB can do x8, Windows sets the default lowest multi to x12. That's why I used x12 instead. I didn't do i5 testing as it's a dual-core and so only does half the work at each multiplier.

Yes, turn on Throttlestop, set the multiplier you want then run a a TS bench. Actually 32M completes in like 8s so I was using the 1024M more often. The CPU will be running at 100% load. I've explained this under the Details of how to configure Throttlestop to do such a table as above section at http://forum.techinferno.com/hp-business-class-notebooks/2537-12-5-hp-elitebook-2570p-owners-lounge-37.html#post77432 .

Link to comment
Share on other sites

Were You checking maximal tempereatures under load, for example 5-7 minutes prime or maximal temp during 3dm11 performance settings?

How about idle? what are the average temps for few hours of internet etc :)

I used the CPU reported TDP rather than temps. Thats because temps will vary depending on the ambient temperature, any other heatsink/fan/enclosure mods the user has and whether they have additional devices active generating internal heat (HDDs, WWAN, Bluetooth). Too many variables prevent doing a meaningful temperature comparison.

FYI: right now my ambient temperature where I am in Australia is 30-35 degrees. Yes, it's hot. In Europe you'd be what hitting 0 degrees around this time??

  • Thumbs Up 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use. We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.