Jump to content

Recommended Posts

Flex VID in throttlestop set to 255 will make the CPU use 1.52V, which I would expect would make it overheat very very fast assuming the motherboard handles the power draw. Also throttlestop and cpu-z will both report the VID with additional turbo voltage taken into account, so if it doesn't say 1.52V during load, it's not working.

I think you just got lucky and it just happened to make it through the test.

What instability are you getting? Hard freezes with no warning, or do you get program shutdowns and calculation errors in stability tests like prime95? I expect the former, and all you can do about that is lower BCLK unless you want to start pencil modding your motherboard. If it's the latter though I suspect your memory is unstable instead of the CPU, so if you raised memory timings or switched to memory with more headroom you'd be fine.

You were spot on. The Throttlestop Flex VID paramter was not altering my CPU voltages at all. So I did a XTU 104.6623 BCLK overclock and didn't run Throttlestop afterwards. 3dmark11 was rock stable and I got a slightly better result too:

3dmark11 physics=8783 NVIDIA GeForce GTX 660 Ti video card benchmark result - Intel Core i7-3740QM Processor,Hewlett-Packard 17DF

aAlso I am not familiar with the CUDA-Z PCI-E bandwidth test. I have run the SANDRA pci-e bandwidth test and I know for a fact that it does show the impact of even minor pci-e clock changes and it works on AMD cards, so if you could run that instead then we have a better understanding of what's going on. Also like I said in the PM, if overclocking shows no bandwidth improvement, try underclocking. I suspect underclocking will have an impact.

@Khenglish, the min I can downclock with XTU is 99.5463 so an inconclusive test. Upclocking to 104.643 showing no improvement in pci-e bandwidth performance with the original modded 2570P ME firmware. That's not the newest one your PMed me that I'm reluctant to flash.

Performance Information
-----------------------
Memory Copy
Host Pinned to Device: 319.684 MiB/s
Host Pageable to Device: 311.12 MiB/s
Device to Host Pinned: 395.923 MiB/s
Device to Host Pageable: 382.103 MiB/s
Device to Device: 52.3205 GiB/s

Link to comment
Share on other sites

Yeah.. heatpipe bit would need to be sealed and exact width needed. Unlikely to happen. I'm using a $3 Coolermaster TIM which has not been the limiting factor here.

The 2570P heatsink is made of a large aluminium component, a small copper 'shim' and a copper heatpipe that leads to a fan-cooled radiator grill. I'd take photos but would need to repaste and I have no reserve paste latm.

Why did HP use aluminium? The beancounters would have seen aluminium is less than 1/4 of the cost. They shortchanged us of nearly double copper's thermal conductivty:

aluminium US$1717/ton conductivity=205W/(W/m K)

copper US$7313/ton conductivity=385W/(W/m K)

My thinking is this. The IVB small and hot CPU die needs to dissipate heat quickly else temps spiral out of control. HP in their wisdom decided to add a tiny copper shim at the end of the aluminium heatsink to help with rapid temps. That would be adequate for a SB 35W CPU even a 35W IVB CPU. While it's going to take up some heat it won't do it fast enough when using a 45W i7-quad.

The copper shim idea being able to remove AND store the heat until the rest of the cooling system can catch up. Store because energy flows from high to low (hot to cold). Therefore I expect the 42x42x1mm shim will improve the duration my CPU can run a TS-1024M at x35. Something I'll be able to prove once the shims arrive.

I don't know how much in the weeds you want to get on what materials to use, but have you considered using a thinner piece of copper and thermal epoxy on the heatsink? I would use a thinner piece of copper so that it still conducts the heat while storing less, and the epoxy to have a consistent connection. Ideally you want to limit the amount of materials and number of interfaces between the CPU and cooling source (air), but the shims that you've been using seem to be overcoming the aluminum limitation. Depending on how hardcore you wanted to go, I would consider doing the same thing on the heat pipe and heatsink connection as well. In doing all of this, you could probably buy yourself a few degrees C and save a heck of a lot of thermal paste between the shim and heatsink everytime you do a mod.

I've used the epoxy on a 25W ceramic resistor in an automotive application. I applied it back in 2005 and it's still holding, and is still effective at moving heat.

Link to comment
Share on other sites

I don't know how much in the weeds you want to get on what materials to use, but have you considered using a thinner piece of copper and thermal epoxy on the heatsink? I would use a thinner piece of copper so that it still conducts the heat while storing less, and the epoxy to have a consistent connection. Ideally you want to limit the amount of materials and number of interfaces between the CPU and cooling source (air), but the shims that you've been using seem to be overcoming the aluminum limitation. Depending on how hardcore you wanted to go, I would consider doing the same thing on the heat pipe and heatsink connection as well. In doing all of this, you could probably buy yourself a few degrees C and save a heck of a lot of thermal paste between the shim and heatsink everytime you do a mod.

I've used the epoxy on a 25W ceramic resistor in an automotive application. I applied it back in 2005 and it's still holding, and is still effective at moving heat.

The two 42x42x1mm copper shim sandwich mod may be the simplest cooling system mod for the 2570P

Thank you for the info. To reiterate, my efficient i7-3740QM has no thermal throttling after installing the 25x25x1mm shim. Only issue now is the Intel Turbo boost algorithm drops my multi from x35 to x34 at 86 degrees during TS-1024M test, the point at where I appear to cross 45W TDP. So I'm seeing TDP throttling, something I see in 3dmark11 eGPU results of other larger HP/Dell/Lenovo systems with factory 45W i7-quad CPUs and beefier cooling system configurations.

Unlocking power limits to say 50-55W, something I'd have thermal headroom for, via say a bios mod isn't feasible due to HP's RSA protection. @Khenglish says that TDP will increase with temps so the idea is to drop temps if possible. Furthermore, my aim is to keep the system in stock form for warranty/resale purposes so no glueing/drilling it.

To achieve those aims the next plan is to install two 42x42x1mm shims (once they arrive). Jacobsson has these and advises they sandwich perfectly between the heatsink and CPU without needing to drill or cut them and I can have two stacked ontop of each other. The heatsink screws supposedly still have sufficient length to reach the screw mounts.

In theory, adding copper shims in this way may turn out to be the simplest, most inexpensive way of maximising 45W i7-quad performance by dropping temps while under load. Even owners of stock dual-core i5/i7 CPUs would benefit with lower load temps.

Only other minor mod would be surrounding the elevated copper plate on the heatsink with some thinner and smaller (0.1mm?) shims. That would make the whole heatsink flat. That would give a larger surface area for thermal conductivity between the heatsink and the shims. Though having those two stacked shims would (I believe) already be doing a great job of soaking heat and doing the critical task of passing it to the heatsink copper plate + heatpipe for dispersion out via the radiator fans.

Link to comment
Share on other sites

From http://forum.techinferno.com/diy-e-gpu-projects/2109-diy-egpu-experiences-%5Bversion-2-0%5D-256.html#post79847 that may be of interest to other 2570P owners:

Well my SB 2560P 4.6W vs IVB 2570P 5.4W idle test was with SSD vs HDD. A SSD can consume virtually no power in HIPM+DIPM slumber mode as discussed at AnandTech | SanDisk Extreme II Review (480GB, 240GB, 120GB) , whereas my HDD consumes 0.8W MKxx61GSYN Series in low power idle. That's exactly the 0.8W difference I noted.

I had already been using an extensive set of 2570P tweaks as documented at http://forum.techinferno.com/hp-business-class-notebooks/2537-12-5-hp-elitebook-2570p-owners-lounge.html#post38399 . Without the ASPMs tweak, the 2570P's SD/MMC upgraded controller was consuming an addition ~1.8W idle power making it significantly worse than a 2560P.

Upon viewing your comments I wondered if I could possible squeeze out some more battery savings? Yes, a further up to 0.3W gain was possible by (1) Disable USB 3.0 - function disable bit 27 (2) Use Throttlestop to decrease multiplier from x12 to x8. Function disable is at 0xFED1F418 on my HP. That may be different on yours. Find it by doing a PCI read of 0:1f.0 f0 (32-bit) but set the rightmost digit to zero, then add 0x3418 it. Use the RWEverything utility to do the pci read then the memory read/write for disabling USB 3.0. Doing a device manager rescan afterwards to see it disappear.

Now if I had a SSD I'd love to show a ~4.6W idling 2570P. That would be at least 2W better than as supplied by HP. The Haswell "power efficient" ULT (ULV) platform idles just under 4W. There might not be much in it at idle.

Anybody want to test a US$34.95 Macbook Air style 802.11AC WLAN card (BCM4352) to gauge battery life improvements?

13" mid-2013 Apple Macbook Air with Broadcomm BCM4360 802.11ac wifi, SSD and 54Wh battery reviewed here:

Idle (without WLAN, min brightness) = 13:28 (4.01W average)

WiFi Surfing (legible brightness) = 10:23 (5.2W average, +1.19W over idle)

12.5" 2013 Lenovo X240 wth Intel 7260 802.11ac wifi, SSD and 48Wh battery reviewed here:

Idle (without WLAN, min brightness) = 13:41 (3.51W average)

WiFi Surfing (legible brightness) = 7:56 (6.05W average, +2.54W over idle)

14" 2013 Lenovo T440s with Intel 7260 802.11ac wifi, HDD, 48W battery reviewed here:

Idle (without WLAN, min brightness) = 11:08 (4.31W average)

WiFi Surfing (legible brightness) = 5:57 (8.07W average, +3.76W over idle)

12.5" 2012 HP 2570P wth Intel 6205 802.11n wifi, 256GB Plextor SSD and 55Wh battery reviewed here, presumably without applying 2570P battery life tweaks:

Idle (without WLAN, min brightness) = 11:14 (4.90W average)

WiFi Surfing (legible brightness) = 6:03 (9.09W average, +4.19W over idle)

The MBA uses only an extra 1.2W when driven with greater LCD brightness AND wifi websurfing. Incredible. The Haswell Lenovo X240, using the same ULT platform, sees a 2.54W increase when wifi websurfing using it's Intel 7260 card, which is probably already better than a 2570P's 6205 card. The Haswell ULT platform, the newer WIFI cards and newer more efficient SSDs being the reasons why they both outclass the 2570P in battery life. The latter two can be retrofitted to a 2570P.

That begs the question: is the X240's 1.34W increase over the MBA due to the Intel WLAN card being used instead of the Broadcomm one ? It's either that or the MBA's pci-e SSD that bypasses the SATA controller that is giving the MBA miserly wifi websurfing power consumption compared to the X240 even.

By induction then does that mean the 2570P's Intel 6205 WLAN card is inefficient compared to a Broadcomm card? Is there a pioneer in the US that wants to investigate this? That can be done by testing a Macbook-Air like Broadcomm 802.11ac BCM4352 WLAN card in place of our stock Intel one. Those can be had for a low US$34.95 price at Azurewave AW-CE123H 802.11ac/nbg WiFi+BT PCI Express Module<br> Broadcom BCM4352 / Bluetooth 4.0+3.0 HS Class II, a good price as they are > US$64 on ebay + delivery from China. Their advertising claims "Low power consumption and high performance" . Broadcomm's press release also saying:

Broadcom's new 5G WiFi chips deliver better coverage and longer battery life in a small form factor that is interoperable and compatible with existing technologies.

Given our BIOS has the WLAN whitelisting disabled this card should work fine. There are Win7 drivers for it too Broadcom BCM4352 WiFi Driver for Windows 7 (32-bit and 64-bit) - ThinkCentre Systems . Don't expect the BT 4.0 to work though. I've looked at photos of these cards on ebay and it looks like USB is wired (pin 36, 38) for the BT 4.0 component but our wifi slot doesn't have USB connected.

When compared to Haswell ULT, using my 2570P battery life tweaks I can get the 2570P idle down to 5.4W. That would be ~4.6W with a power efficient SSD (~0.0W idle) instead of my HDD that uses 0.8W at idle. So at idle there isn't much in it. However, when I do wifi websurfing I get somewhere between 5:30-6hrs of battery life from my (now) 55Wh battery. If I assumed a straight 0.8W gain using a SSD then that would give an addition 35mins. If I use HP's 15% extra battery life when using a SSD over a HDD I'd be seeing an extra 1hr.

So here I speculate that by using a Broadcomm BCM4352 WLAN card instead of an Intel one we may see some additional tasty battery life gains. Wonder how close it would see a IVB 2570P come to Haswell ULT power consumption levels?

  • Thumbs Up 2
Link to comment
Share on other sites

@Nando,

You got a lucky chip, bro. In my case it's not about temps, but rather a bios locked power envelope. Even when I run my machine without the back panel, elevated with a modded HS at very low room temp + maxed out fan, while I never cross 80c during load, it still won't hold 3.6 on all cores.

  • Thumbs Up 1
Link to comment
Share on other sites

I have a 7260 coming in on a slow boat from Newegg. Should be here in a week, so I'll report back when I get some time.

I'm not sure if I'll be getting any power reductions, maybe Intel is counting on the increased bandwidth to lower the time with the chip in an active state to save power? I found two fact sheets that show exactly the same power usage estimates:

http://h18000.www1.hp.com/products/quickspecs/14650_div/14650_div.PDF

http://h18004.www1.hp.com/products/quickspecs/14801_na/14801_na.pdf

Either they copied and pasted the numbers or they're banking on the bandwidth to pick up the slack. I can't find anything from Intel on the power usage, or anything on the Broadcom chip. Also, the review for the Lenovo laptop said that the "radios" were on, did this include the WWAN + GPS module that they had installed as well? GPS receivers can use a decent amount of power when active. The Samsung 840 (in the Lenovo) is one of the lowest power consuming SSDs out there, so I doubt the increase comes from that. I do like the dual battery setup and screen on the Lenovo, but no TB/ExpressCard + BGA CPU = fail for me (and most of us here I'm sure).

I've been trying to use a ramdisk as a browser cache to reduce power consumption and increase performance. What else am I going to do with 16GB of RAM? I also disabled my pagefile to try and reduce read/writes to my SSD. My current setup is using SoftPerfect RAM Disk, but it keeps thrashing my HDD with a 1MB/s write and occasionally loading up one of my cores to 100%. Does anyone know of any good RAMdisk software that will let me save an image file in a specified time interval and on system shutdown? I'll try Dataram's free version to see if I can find anything useful. The performance improvement I do have right now is very much worth it though!

Link to comment
Share on other sites

Hello everyone,

After reading the hole thread (took me 2 days) I'm sure that this is the right place for me as I'm as techy as you are ;-)

since a few days I'm a proud owner of a 2570p, Atm it's a redundancy for my 9470m, which has more advantages for me (higher resolution display, msata SSD, core i7 (even if it's an U mobile), WWAN, WLAN AC), but I already did some testing of what is possible with my new machine.

the 2570p is set up with an Intel 520 SSD, 16 GB of RAM and an Intel Centrino 6235 WiFi card. The WiFi card comes from my 9470m, which has an Intel 7260 AC right now. I put it into the 2570p to test if there are any differences in power consumption and I was right.

When you swap the stock 6205 against the 6235 you'll save 0,5W in idle and under "load" (surfing and downloading some files) the difference goes up to a maximum of 1W.

I also tried the 7260 AC from my 9470m and it works too, but there was no better result in battery time.

Also the HP un2430 EV-DO/HSPA card is working, but I was expecting this.

Have to investigate which BIOS version I have, that was something which wasn't neccecary for me at first.

Next steps I want to change on the machine:

- Order an Intel 7260 AC card, just because it's AC and will give me some advances in the future

- I have some heavy copper coolers which originally belong to some HP server blades, I will try to modify and apply them on the 2570P heatsink

- Modify the Heatsink, I have ordered a second one from ebay. I will try to grind as much aluminum as possible from the cooler and want to apply cooper instead. This would deliver a better cooling I think

- apply a cooper shim below the heatsink, I think about a 40mmX40mmX1,0mm piece.

That's in my head for now, any suggestions what I could do next?

  • Thumbs Up 1
Link to comment
Share on other sites

@Nando,

You got a lucky chip, bro. In my case it's not about temps, but rather a bios locked power envelope. Even when I run my machine without the back panel, elevated with a modded HS at very low room temp + maxed out fan, while I never cross 80c during load, it still won't hold 3.6 on all cores.

Agreed. I have one very nice i7-3740QM. I see throttling between x34-x35 under TS-1024 mode. Though running the XTU bench seems have far higher load. There it drops down to x32.

Wondering if you would be willing to chart the VID and TDP for your i7-3840QM as shown at http://forum.techinferno.com/hp-business-class-notebooks/2537-12-5-hp-elitebook-2570p-owners-lounge-37.html#post77432 . I'd summarize it in that table making a very good buying guide for i7-quad CPUs.

Your comment supporting my observations as well. The TDP limit will be silicon specific and while higher-end CPUs should be more efficient, it's not always the case. More luck of the drawer. So when limited to 45W TDP as our 2570P is the best CPUs to get are a i7-3630QM (x32 4-core), i7-3720QM (x34 4-core), i7-3740QM/i7-3820QM (x35 4-core). If have deep pockets then could get a i7-3920XM which will raise TDP to 55W. I'd guestimate that it could hold a x36 4-core multipler at that TDP. 55W would definitely need cooling mods.

I have a 7260 coming in on a slow boat from Newegg. Should be here in a week, so I'll report back when I get some time.

I'm not sure if I'll be getting any power reductions, maybe Intel is counting on the increased bandwidth to lower the time with the chip in an active state to save power? I found two fact sheets that show exactly the same power usage estimates:

http://h18000.www1.hp.com/products/quickspecs/14650_div/14650_div.PDF

http://h18004.www1.hp.com/products/quickspecs/14801_na/14801_na.pdf

Either they copied and pasted the numbers or they're banking on the bandwidth to pick up the slack. I can't find anything from Intel on the power usage, or anything on the Broadcom chip. Also, the review for the Lenovo laptop said that the "radios" were on, did this include the WWAN + GPS module that they had installed as well? GPS receivers can use a decent amount of power when active. The Samsung 840 (in the Lenovo) is one of the lowest power consuming SSDs out there, so I doubt the increase comes from that. I do like the dual battery setup and screen on the Lenovo, but no TB/ExpressCard + BGA CPU = fail for me (and most of us here I'm sure).

I've been trying to use a ramdisk as a browser cache to reduce power consumption and increase performance. What else am I going to do with 16GB of RAM? I also disabled my pagefile to try and reduce read/writes to my SSD. My current setup is using SoftPerfect RAM Disk, but it keeps thrashing my HDD with a 1MB/s write and occasionally loading up one of my cores to 100%. Does anyone know of any good RAMdisk software that will let me save an image file in a specified time interval and on system shutdown? I'll try Dataram's free version to see if I can find anything useful. The performance improvement I do have right now is very much worth it though!

Agreed, Samsung SSDs are very low powered. Though what Apple have done is use a x2 2.0 Samsung pci-e SSD which bypasses the Intel SATA controller and instead uses it's own onboard one. So the 13" Macbook Air is seeing better wifi websurfing battery life compared to the X240 either because of it's different LCD, Samsung pci-e SSD or the BCM4360 (3-wire) WLAN controller.

After you have reviewed your ramdisk options would you mind creating a guide? This could be another way of dropping power consumption by a noticable amount.

Hello everyone,

After reading the hole thread (took me 2 days) I'm sure that this is the right place for me as I'm as techy as you are ;-)

since a few days I'm a proud owner of a 2570p, Atm it's a redundancy for my 9470m, which has more advantages for me (higher resolution display, msata SSD, core i7 (even if it's an U mobile), WWAN, WLAN AC), but I already did some testing of what is possible with my new machine.

the 2570p is set up with an Intel 520 SSD, 16 GB of RAM and an Intel Centrino 6235 WiFi card. The WiFi card comes from my 9470m, which has an Intel 7260 AC right now. I put it into the 2570p to test if there are any differences in power consumption and I was right.

When you swap the stock 6205 against the 6235 you'll save 0,5W in idle and under "load" (surfing and downloading some files) the difference goes up to a maximum of 1W.

I also tried the 7260 AC from my 9470m and it works too, but there was no better result in battery time.

Also the HP un2430 EV-DO/HSPA card is working, but I was expecting this.

Have to investigate which BIOS version I have, that was something which wasn't neccecary for me at first.

Next steps I want to change on the machine:

- Order an Intel 7260 AC card, just because it's AC and will give me some advances in the future

- I have some heavy copper coolers which originally belong to some HP server blades, I will try to modify and apply them on the 2570P heatsink

- Modify the Heatsink, I have ordered a second one from ebay. I will try to grind as much aluminum as possible from the cooler and want to apply cooper instead. This would deliver a better cooling I think

- apply a cooper shim below the heatsink, I think about a 40mmX40mmX1,0mm piece.

That's in my head for now, any suggestions what I could do next?

Welcome aboard. You are in fine company if like to tinker a system to max performance :)

Did you test power consumption of the 6205 with latest drivers from Intel's website? I found the HP supplied older (Intel) driver had noticably higher idle and load power consumption.

Still, it may very well be the 6235 is a more efficient WLAN than the 6205 with the 7260 is at least as good if not better than the 6235. Though if you are in the US I'd suggest grabbing a BCM4352 instead as discussed. It may be the the reason why a 13" Macbook Air has noticably better battery life than a 7260-equipped Lenovo X240.

The shims to get are the 42x42x1mm ones from ebay: 5pcs Laptop GPU CPU Heatsink Copper Shim 42mmx42mmx1 0mm | eBay .Jacobsson advises that two can be sandwiched between the heatsink and CPU and still have the heatsink screws be able to reach and the screwmounts. This will lower load temps on both dual and quad core CPUs. Quad cores benefitting by being able to stay under thermals limits as well as being able to hold TDP-limited turbo boost, which is also affected by temps, for a longer duration.

Next would be a quad-core CPU, NVidia eGPU, RAID-0 SSD storage, battery life savings tweak. All documented on the first page, when applied they morph a 2570P into an ultraportable beast of a desktop-replacement machine.

Does anyone made a test with shims and without them?

Still waiting for mine to arrive. I have a 25x25x1mm in there which allows my i7-3740QM to provide a x35 multiplier under TS-1024M load for what is at least 20s. Prior to adding the shim it would run x35 for a split second then throttle down to x34. So the shim decrease the temp spike under full load and provide additional heat soaking ability. I don't as yet have a chart showing by how much.

  • Thumbs Up 1
Link to comment
Share on other sites

Welcome aboard. You are in fine company if like to tinker a system to max performance :)

Thanks for the welcome :)

Not ever, I want a rockstable and long running system, that is capable to work on VM's or some emulation. But I think it aims in the same direction ;)

Did you test power consumption of the 6205 with latest drivers from Intel's website? I found the HP supplied older (Intel) driver had noticably higher idle and load power consumption.

I used the latest drivers available at the Intel page for my test but this could vary if someone uses another OS than Windows 7 X64 or has more services running than me.

Still, it may very well be the 6235 is a more efficient WLAN than the 6205 with the 7260 is at least as good if not better than the 6235. Though if you are in the US I'd suggest grabbing a BCM4352 instead as discussed. It may be the the reason why a 13" Macbook Air has noticably better battery life than a 7260-equipped Lenovo X240.

I don't think this is becaus of the used WiFi card, but because of the used OS. Maybe I will have time this month to get OS X 10.9 up and running on the machine to test the battery time under OS X. This would be my first step before buying a card which could not work or consume more power than the card before ;)

The shims to get are the 42x42x1mm ones from ebay: 5pcs Laptop GPU CPU Heatsink Copper Shim 42mmx42mmx1 0mm | eBay .Jacobsson advises that two can be sandwiched between the heatsink and CPU and still have the heatsink screws be able to reach and the screwmounts. This will lower load temps on both dual and quad core CPUs. Quad cores benefitting by being able to stay under thermals limits as well as being able to hold TDP-limited turbo boost, which is also affected by temps.

I did understand this and that's why I wanted to do this as one of the next tasks. But I wanted to know if there is anything else you would suggest me to do ;)

Next would be a quad-core CPU, NVidia eGPU, RAID-0 SSD storage, battery life savings tweak. All documented on the first page, when applied they morph a 2570P into an ultraportable beast of a desktop-replacement machine.

I think the quadcore and eGPU has to wait because I use it only as a mobile replacement for my native 8-core Desktop PC ;)

I will sacrifice the RAID capability for more storage instead, this goes handy with the allready applied (and extended by myself) battery tweaks.

As I mentioned, I read the whole thread before asking something which was asked 200 times before ;)

If I made any grammar or spelling mistakes I apologise for that but I'm not a native english ;)

Link to comment
Share on other sites

Heh, few minutes ago I realized that fan is not spining. I took a look on HPFC and there was no reaction on manual settings. I tried to restart but after restart I got info abou system fan error 90B, very common error. I thought that is a reason of my mods, I changed a fan and after this I got the same message so reason is on settings/main board. Despite that, I tried to chalange it during BF4 without working fan :P

But there was no chance, during waiting for map I got 95C on cores so start a game would give fast restart ;)

Im going to try hard reset and upload BIOS, if doesn't help I will be forced to call fo onsitebulance ;)

Link to comment
Share on other sites

phillofoc

I use the program Primo.Ramdisk.Srv.Mui.Setup.5.6.0. It works perfect.

Thanks, but I'm looking for a free solution. Softperfect just updated their offering that addresses many of the bugs that I've been experiencing. Once I get everything sorted out I'll report back and post a guide on how to put everything together. Right now I'm not seeing much battery savings as Firefox is running a constant 2% on my CPU, but maybe we can see some savings while actively browsing. HDD users might see more savings than me as it can potentially allow the drive to spin down while browsing, depending on the rest of your setup.

Link to comment
Share on other sites

Just got my 2570p. I like it, but it feels kind of big! Is the 2170p comparable but just smaller?

Comparing 2570P to 2170p

- 12.5" versus 11.6"

- 35W i5/i7 socketted CPU (upgrable to i7-quad) versus 17W i5/i7 ULV CPU.

- SSD/HDD+ODD (two drive bays) versus SSD/HDD (1 drive bay)

- expresscard slot for eGPU use vs no expresscard slot

- 55/62/100Whr battery versus 48Whr battery

- no backlit keyboard (but has nightlight) versus backlit keyboard

- 4.2lbs versus 3.2lbs

Overall I'm not a fan of the 2170P due to the small battery, low grade LCD, no expresscard slot and ULV CPU. A Lenovo X230 straddles middle ground between the 2170P and 2570P (35W CPU, expresscard slot, 6/9-cell battery options) and even includes a better IPS LCD option. Or consider HP's newer 12.5" "820 G1" Haswell system.

Link to comment
Share on other sites

Comparing 2570P to 2170p

- 12.5" versus 11.6"

- 35W i5/i7 socketted CPU (upgrable to i7-quad) versus 17W i5/i7 ULV CPU.

- SSD/HDD+ODD (two drive bays) versus SSD/HDD (1 drive bay)

- expresscard slot for eGPU use vs no expresscard slot

- 55/62/100Whr battery versus 48Whr battery

- no backlit keyboard (but has nightlight) versus backlit keyboard

- 4.2lbs versus 3.2lbs

Overall I'm not a fan of the 2170P due to the small battery, low grade LCD, no expresscard slot and ULV CPU. A Lenovo X230 straddles middle ground between the 2170P and 2570P (35W CPU, expresscard slot, 6/9-cell battery options) and even includes a better IPS LCD option. Or consider HP's newer 12.5" "820 G1" Haswell system.

Thanks! Looks like there's no reason to get the 2170p over the 2750p then... Anyway, does anyone know if it's possible to change the function of the wireless button and internet button on the computer to be volume up and down buttons? there's a mute button right next to it. i have no use for wireless on/off and an internet button...

Link to comment
Share on other sites

Is anyone able to explain me where are the differences in BIOS, hardware which allows me to do this kind of things in 2530p but not in the 2560p/2570p?

2570P and 2530P are manufactured by completely different companies with different architecture, including the thermal sensor:

2530P = Compal LA-4021 using EMC2103-2 Thermal Sensor. You can use hwinfo to control the fan on these.

2570P = Inventec Simba using GMT G781 Thermal Sensor. You can use TPFancontrol to control the fan on these.

Link to comment
Share on other sites

Thanks. :)

But as mentioned many times before, on 2530p we can choose every single value between 0 and max rpms.

In 2560p/2570p only 0 or another one is 3200rpm. Shame... ;)

The two diode solution was proposed earlier on to reduce the fan cut-in rpms. Do you no longer have the diodes in place?

http://forum.techinferno.com/hp-business-class-notebooks/2537-12-5-hp-elitebook-2570p-owners-lounge-10.html#post64705

http://forum.techinferno.com/hp-business-class-notebooks/2544-hp-probook-4330s-working-silent-fan-mod.html

Link to comment
Share on other sites

Yes, I have them all the time but I really regret 2530p (2540p also?) fan controll ability=simplicity ;)

I understand completely. Yes, hwinfo has a nicer interface for fan control. Maybe @Mumak, the author of hwinfo, could work with the author of TPFancontrol to add newer SB/IVB fan control to his software?

  • Thumbs Up 1
Link to comment
Share on other sites

are You able to do test with 30x and 25x?

I'm not sure what you want tested at x25 and x30 mode? As mentioned, the shim improves heat dissipation from the CPU so I get a longer run at my full x35 4-core mode in say TS-1024M. That heat doesn't get removed fast enough and is backlogged unless some of the cooling mods are implemented. Basic one is to just run the fan at 100% and remove the bottom cover.

Eg: Ambient temps are now way cooler at 22 degrees, sort of what you'd expect in an office environment. With my cover off, TPFancontrol on full (100%), my CPU never goes beyond 86 degrees. That's the point where I see TDP throttling between x34 and x35. Means I nearly get a full TS-1024M run at x35 4-core mode. Result being:

261.1s (@100.00Mhz BCLK - 86 degrees max - nearly a full x35 run)

253.6s (@104.63Mhz BCLK - 86 degrees max - nearly a full x35 run)

Some 3dmark11 scores

Since you like 3dmark11 physics scores, here's what I get with a i7-3740QM@BCLK=104.63 with a GTX660Ti and HD7970 Ghz edition eGPU:

http://www.3dmark.com/3dm11/7871291 : physics=8798 - GTX660Ti

http://www.3dmark.com/3dm11/7845155 : physics=9102 - HD7970 Ghz edition

Note the higher physics score with the AMD card. That illustrates the slight CPU overhead the NVidia x1 pci-e compression uses.

Here's a comparison to a 130W i7-4820K Haswell desktop CPU + HD7970. Its not even 10% faster:

http://www.3dmark.com/3dm11/7825601 : physics=9764

  • 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.