Jump to content

Lenovo Y400 / Y500 - unlocked BIOS / wlan whitelist mod


svl7

Recommended Posts

svl7 can you tell me how you did the BIOS mod itself? How did you extract the BIOS from the Y500? Also what did you use to edit? e.g. Phoenix Tool to extract the modules and then a hex editor and nibitor to correct checksum?

I would be glad to know this. If you simply write the software you used that's great. If you ever happen to go away I can continue support for this laptop.

I would also like to know this. I'd love to try modifying the vbios to increase the voltage and overclocking potential.

Link to comment
Share on other sites

I would also like to know this. I'd love to try modifying the vbios to increase the voltage and overclocking potential.

svl7 tell me in PM which software you used to do the modding of the main BIOS. I really want to learn this. Great work! I am going to put everything nice together in a tutorial and one package. I am also going to put your paypal donate link in my tutorial for your work. If you want to share your knowledge and allow me to put together the tutorial and package let me know. I can be os good use to you and we can be a great team. I dont want to steal your followers or fans.

  • Thumbs Up 1
Link to comment
Share on other sites

I would also like to know this. I'd love to try modifying the vbios to increase the voltage and overclocking potential.

Be careful with the voltage. When i played with Tomb Raider (i5, SLI little OC (+135Mhz)) ,the power adapter was so hot and this is the weakest point of laptops as I know, and it's have only 1 year warranty.

We should monitor the powers during furmark+IBT to see what amount of power we have left.

BTW if we could adjust the voltage its could be good because the under voltage option.

Link to comment
Share on other sites

Great job it worked for me also now i have 2.02 moded with unlocked vbios for both cards :), thx keep up the good work !

Sorry for asking

But i searched lot couldn't find both 2.02 original and modded bios...

Can u plz navigate me to that link :)

Thank you in advance

Sent from my Nexus 4 using Tapatalk 2

Link to comment
Share on other sites

Be careful with the voltage. When i played with Tomb Raider (i5, SLI little OC (+135Mhz)) ,the power adapter was so hot and this is the weakest point of laptops as I know, and it's have only 1 year warranty.

We should monitor the powers during furmark+IBT to see what amount of power we have left.

BTW if we could adjust the voltage its could be good because the under voltage option.

Power adapters usually run hot when you're gaming or doing anything strenuous. You should test with a Kill-a-watt power meter if you're worried but I doubt you're even close to maxing out the 170W charger. Plus you have the i5 which uses less power than the i7.

FurMark + IBT is a good extreme scenario stress test but it's not at all representative of real usage. You're never gonna be maxing out the GPU's and exceeding TDP on the CPU in any game.

BTW since you have SLI you need to use the fullscreen 3D test in OCCT to utilize both graphics cards. FurMark and MSI Kombustor in fullscreen both only run on the first GPU regardless of which AFR mode or rendering API you use.

Link to comment
Share on other sites

Power adapters usually run hot when you're gaming or doing anything strenuous. You should test with a Kill-a-watt power meter if you're worried but I doubt you're even close to maxing out the 170W charger. Plus you have the i5 which uses less power than the i7.

FurMark + IBT is a good extreme scenario stress test but it's not at all representative of real usage. You're never gonna be maxing out the GPU's and exceeding TDP on the CPU in any game.

BTW since you have SLI you need to use the fullscreen 3D test in OCCT to utilize both graphics cards. FurMark and MSI Kombustor in fullscreen both only run on the first GPU regardless of which AFR mode or rendering API you use.

I Don't want to overvoltage my system. I just want to warn RedLionRisen to be careful about this for safety. :topsy_turvy:

BTW if I upgrade the BIOS to 2.02 from 1.05 and the vbios mod for 2. GPU. Can I easly rollback to the original ones for the warranity... no? And for the worst possibility is there any solution to debrick the computer?

Link to comment
Share on other sites

+150mghz core clock +300 memory clock both cards

https://www.dropbox.com/s/2jm7m2e5hts6ebo/3DMark_Overclock.JPG

3DMark 11 score 4022 vs 3485 without overclock.Thank you!:)

Bro you need to use ThrottleStop or enable Turbo Boost if you've disabled it or something. That Physics score is way too low. I'm getting 1000 more points than you.

Link to comment
Share on other sites

svl7 tell me in PM which software you used to do the modding of the main BIOS. I really want to learn this. Great work! I am going to put everything nice together in a tutorial and one package. I am also going to put your paypal donate link in my tutorial for your work. If you want to share your knowledge and allow me to put together the tutorial and package let me know. I can be os good use to you and we can be a great team. I dont want to steal your followers or fans.

You share my stuff freely on NBR without even asking for approval and want me to teach you anything?

You seriously piss me off.

... you're done here.

  • Thumbs Up 2
Link to comment
Share on other sites

I Don't want to overvoltage my system. I just want to warn RedLionRisen to be careful about this for safety. :topsy_turvy:

Overvolting is perfectly safe, as octiceps alluded to, so long as you are able to intelligently monitor the system temperatures and gpu stability. It's like tuning an instrument. Each individual one is different, and has its own perfect balance. From what I've seen around the internet, the GT 650m has a great deal of overclocking headroom, with fairly linear performance increases. The stock maximum voltage of 1V allows you to overclock to about 1100Mhz, but I've seen people increase the voltage to about 1.1V, and stably raise the core clock to 1300Mhz, which does correspond to about a 15% increase in performance (by 3dmark standards) over the voltage limited clock of 1100Mhz. Compare that to all stock settings, and it would represent a 50% increase in performance (by benchmark standards of course). In my opinion, that's reason enough to modify the BIOS if you want a performance boost, or just want to undervolt at lower core clocks.

Link to comment
Share on other sites

Overvolting is perfectly safe, as octiceps alluded to, so long as you are able to intelligently monitor the system temperatures and gpu stability. It's like tuning an instrument. Each individual one is different, and has its own perfect balance. From what I've seen around the internet, the GT 650m has a great deal of overclocking headroom, with fairly linear performance increases. The stock maximum voltage of 1V allows you to overclock to about 1100Mhz, but I've seen people increase the voltage to about 1.1V, and stably raise the core clock to 1300Mhz, which does correspond to about a 15% increase in performance (by 3dmark standards) over the voltage limited clock of 1100Mhz. Compare that to all stock settings, and it would represent a 50% increase in performance (by benchmark standards of course). In my opinion, that's reason enough to modify the BIOS if you want a performance boost, or just want to undervolt at lower core clocks.

I'm guessing you just want 1.1 V for benchmarking purposes and then flash it back to stock for normal use? That is an absurd voltage to be using on a regular basis and I guarantee will significantly reduce the lifespan of your card if not fry it outright. Remember, this is a 14''-15'' notebook with a single fan we're talking about, not some full tower with liquid cooling. You're gonna have to mod the cooling system in this thing while you're at it too because the tons of extra heat from the GPU will spill over to an already too-hot CPU.

Doubt it's even possible to get it that high anyway, assuming there's no hard lock in the first place. Even if it is possible, the inconvenient thing is that you have to flash the card every time bump it up. I'd recommend 0.005 V increments to be safe and you can count how many times you'll need to flash to get to 1.1 V. ;) Flashing always carries risks and the fewer times you do it the better.

Personally, I'm not going to concern myself with adjusting the voltage on this thing, and certainly not increase it. Whatever is the highest OC I can attain on stock voltage is the one I will be happy with. No sense in reducing the lifespan of this machine for some higher benchmark scores.

Link to comment
Share on other sites

As long as the cooling is fine you can easily run that card at such a voltage. Of course it will run hotter and life expectancy will be shorter, but I run my 660m at way above 1.1V and never had issues, and I reached 1350MHz easily.

Link to comment
Share on other sites

As long as the cooling is fine you can easily run that card at such a voltage. Of course it will run hotter and life expectancy will be shorter, but I run my 660m at way above 1.1V and never had issues, and I reached 1350MHz easily.

I figured as much. The person who said they ran the 650m at those settings (through a BIOS mod, so no frequent switching) said the gpu stayed below 86C. I wouldn't suggest EVERYBODY go overvolting their gpu, but it's certainly doable on these late generation platforms. Still, I'd only advise it with a good cooler (which I will be using).

- - - Updated - - -

I'm guessing you just want 1.1 V for benchmarking purposes and then flash it back to stock for normal use? That is an absurd voltage to be using on a regular basis and I guarantee will significantly reduce the lifespan of your card if not fry it outright. Remember, this is a 14''-15'' notebook with a single fan we're talking about, not some full tower with liquid cooling. You're gonna have to mod the cooling system in this thing while you're at it too because the tons of extra heat from the GPU will spill over to an already too-hot CPU.

Doubt it's even possible to get it that high anyway, assuming there's no hard lock in the first place. Even if it is possible, the inconvenient thing is that you have to flash the card every time bump it up. I'd recommend 0.005 V increments to be safe and you can count how many times you'll need to flash to get to 1.1 V. ;) Flashing always carries risks and the fewer times you do it the better.

Personally, I'm not going to concern myself with adjusting the voltage on this thing, and certainly not increase it. Whatever is the highest OC I can attain on stock voltage is the one I will be happy with. No sense in reducing the lifespan of this machine for some higher benchmark scores.

I mean full performance use with the maximum state being at that voltage... But point taken. I'll probably need to be able to modify the BIOS myself to tune it to my machine and usage. I just wish I could figure out how its done, with the vbios being integrated into the system bios and everything.

Link to comment
Share on other sites

As long as the cooling is fine you can easily run that card at such a voltage. Of course it will run hotter and life expectancy will be shorter, but I run my 660m at way above 1.1V and never had issues, and I reached 1350MHz easily.
I figured as much. The person who said they ran the 650m at those settings (through a BIOS mod, so no frequent switching) said the gpu stayed below 86C. I wouldn't suggest EVERYBODY go overvolting their gpu, but it's certainly doable on these late generation platforms. Still, I'd only advise it with a good cooler (which I will be using).

- - - Updated - - -

I mean full performance use with the maximum state being at that voltage... But point taken. I'll probably need to be able to modify the BIOS myself to tune it to my machine and usage. I just wish I could figure out how its done, with the vbios being integrated into the system bios and everything.

Hey @svl7 isn't the GTX 660M already 1.1V at stock? Our GT 650M is probably a lower binned GK107 and clocks in at 790/2000 on 1.0V or 1.025V while most GTX 660M I see are at 950/2500 on 1.1 V or higher. So I definitely think the 660M has more overclocking and overvolting potential since it's already got a head start.

Don't know what system the 660M is from, but I bet your cooling system is probably better than what we have here in the Y400/Y500 as well. The CPU and internal GPU share the same copper heatpipe and fan which in this case is bad news for the CPU. It's runs 20-30 C higher than the GPU at full load. Without a cooler and everything at stock the CPU gets within 10 C of instant thermal shutdown (105 C) in 3DMark 11 which really scares me.

Based on my limited experience so far overclocking the GT 650M SLI I have found that because of the close proximity of the CPU and GPU0 the extra heat produced by the latter causes the former to overheat by about the same amount. So when @RedLionRisen says that the GPU got to 86 C that really concerns me because in our case that would mean that CPU would be past thermal shutdown. The Ultrabay GPU fares best since it's got it's own fan on the other side of the laptop and runs 10 C cooler than GPU0 and 40 C cooler than the CPU.

I don't think the thermals on this system can be improved much short of a complete cooling mod. Repasting is not going to help because others have already repasted their CPU and GPU and didn't see improvement. I personally have taken apart the Ultrabay GPU and Lenovo's stock thermal paste and application are actually very decent; they don't use crap and glop it on like most OEM's I've seen. I repasted the Ultrabay with my best TIM, OCZ Freeze, and it didn't drop temps at all. My relatively powerful Cooler Master cooling pad doesn't help much either, maybe a 3-5 C drop at most.

Here's is an example of my worries regarding the thermals in this machine. When I run the OCCT Power Supply test, which runs a FurMark-like test for the GPU and Linpack for the CPU simultaneously, the CPU gets to 100 C in less than a minute.

2fUV8YU.png

BgCitDe.png

GPU0 and GPU1 were 68 C and 60 C respectively when the CPU tripped 100 C. Keep in mind I was using the cooling pad and the GPU's were at stock. I know that this is kind of an extreme scenario test, but 3DMark 11 and some CPU-intensive games like Far Cry 3 already heat the CPU to 90+ C.

So for people like @RedLionRisen who are interested in overvolting I definitely don't think the cooling system is good enough for that. No matter how cool your GPU(s) may run it's the CPU heat that will limit you in the end. With stock clocks and voltage it's already choking up so having more heat spilling over from an overvolted GPU isn't helping matters.

Link to comment
Share on other sites

Hey @svl7 isn't the GTX 660M already 1.1V at stock? Our GT 650M is probably a lower binned GK107 and clocks in at 790/2000 on 1.0V or 1.025V while most GTX 660M I see are at 950/2500 on 1.1 V or higher. So I definitely think the 660M has more overclocking and overvolting potential since it's already got a head start.

Don't know what system the 660M is from, but I bet your cooling system is probably better than what we have here in the Y400/Y500 as well. The CPU and internal GPU share the same copper heatpipe and fan which in this case is bad news for the CPU. It's runs 20-30 C higher than the GPU at full load. Without a cooler and everything at stock the CPU gets within 10 C of instant thermal shutdown (105 C) in 3DMark 11 which really scares me.

Based on my limited experience so far overclocking the GT 650M SLI I have found that because of the close proximity of the CPU and GPU0 the extra heat produced by the latter causes the former to overheat by about the same amount. So when @RedLionRisen says that the GPU got to 86 C that really concerns me because in our case that would mean that CPU would be past thermal shutdown. The Ultrabay GPU fares best since it's got it's own fan on the other side of the laptop and runs 10 C cooler than GPU0 and 40 C cooler than the CPU.

I don't think the thermals on this system can be improved much short of a complete cooling mod. Repasting is not going to help because others have already repasted their CPU and GPU and didn't see improvement. I personally have taken apart the Ultrabay GPU and Lenovo's stock thermal paste and application are actually very decent; they don't use crap and glop it on like most OEM's I've seen. I repasted the Ultrabay with my best TIM, OCZ Freeze, and it didn't drop temps at all. My relatively powerful Cooler Master cooling pad doesn't help much either, maybe a 3-5 C drop at most.

Here's is an example of my worries regarding the thermals in this machine. When I run the OCCT Power Supply test, which runs a FurMark-like test for the GPU and Linpack for the CPU simultaneously, the CPU gets to 100 C in less than a minute.

2fUV8YU.png

BgCitDe.png

GPU0 and GPU1 were 68 C and 60 C respectively when the CPU tripped 100 C. Keep in mind I was using the cooling pad and the GPU's were at stock. I know that this is kind of an extreme scenario test, but 3DMark 11 and some CPU-intensive games like Far Cry 3 already heat the CPU to 90+ C.

So for people like @RedLionRisen who are interested in overvolting I definitely don't think the cooling system is good enough for that. No matter how cool your GPU(s) may run it's the CPU heat that will limit you in the end. With stock clocks and voltage it's already choking up so having more heat spilling over from an overvolted GPU isn't helping matters.

Hmm, your temperatures seem a little bit odd to me :/. I've run multiple full-load 3dmark, furmark, WEI, etc. tests, and done hours of gaming on my system with the gpu overclocked AND cpu on constant turbo mode, and the absolute highest I ever got out of my cpu was 92C, and 78C on the gpu during furmark (which is 8 degrees higher than I have ever seen it go in any other usage). No external fans used right now either, plus I'm running the smaller 14" Y400 model. All of that amounts to a rather large difference in cooling between our systems, and I'm not sure why yours seems to heat up that much. Maybe because you're running two gpus?

Out of the bag, the cpu does run kinda hot for sure though. I regularly peak at 90C during gaming, but not a degree more. I'd prefer to keep the cpu under 95 at all possible times.

Link to comment
Share on other sites

Hmm, your temperatures seem a little bit odd to me :/. I've run multiple full-load 3dmark, furmark, WEI, etc. tests, and done hours of gaming on my system with the gpu overclocked AND cpu on constant turbo mode, and the absolute highest I ever got out of my cpu was 92C, and 78C on the gpu during furmark (which is 8 degrees higher than I have ever seen it go in any other usage). No external fans used right now either, plus I'm running the smaller 14" Y400 model. All of that amounts to a rather large difference in cooling between our systems, and I'm not sure why yours seems to heat up that much. Maybe because you're running two gpus?

Out of the bag, the cpu does run kinda hot for sure though. I regularly peak at 90C during gaming, but not a degree more. I'd prefer to keep the cpu under 95 at all possible times.

Der because you're not running a GPU burn test and CPU burn test at the same time. ;)

Go ahead and try it out yourself: Prime95 torture test on all 8 threads with FurMark or MSI Kombustor burn test in fullscreen. Or OCCT Power Supply test like what I did above which is easier to use because it incorporates both tests in one program. Guarantee you your CPU will exceed 100 C in short order if it doesn't throttle before then. I have to you warn you though that running this test repeatedly isn't good for your system. Even though I've got the 170W adapter the OCCT test makes Windows Explorer crash and restart after I exit it. :( And I'm on stock clocks too ATM. This program will test every ounce of stability in your rig.

Like I said having the CPU in the 90's when gaming is way too close for comfort IMO. One core spike to 105 C and BAM say goodbye to your game. Which is why I'm so wary of overvolting the GPU, especially the main one, because it essentially shares the cooling system with the CPU and any increase in its heat will transfer over.

Running two GPU's has nothing to do with it. I get the same temperatures on the CPU and GPU0 regardless of whether GPU1 is plugged in and running or not. Like I said the Ultrabay GPU is actually the coolest of the bunch since it's all the way on the other side of the laptop and has it's own heatsink and fan.

- - - Updated - - -

@octicep - That's why I'm thinking of putting some ramsinks on top of the heatpipes, do you think it will lower the temps on the cpu? my cpu temps reaches 100c in less the 30s running prime95 then throttles down to 1.2ghz.

btw I have the Y400 model.

Don't think it's gonna help much if at all and I don't think there's enough thickness in the case to fit any RAMsinks. Plus I don't think airflow is good enough in such a small case to seen any benefit. The main problem is the proximity of the two components and the sharing of the single copper heatpipe and fan.

Link to comment
Share on other sites

Der because you're not running a GPU burn test and CPU burn test at the same time. ;)

Because you will never conceivably run anything that will put the computer under that kind of forced load? Running one burn test, much less two, only indicates the extreme maximum your system can handle. It's well beyond what you'd ever see in normal operation. Using that as a basis for claiming the sky will fall under normal usage isn't... correct. If you run a more reasonable assessment of the temperature response in the system and find it to be 95 max (for example), you should be able to expect the temperatures not to spike beyond that, certainly not to the boiling point. And even in the unlikely event that it does spike that high, the system will shut itself down before any damage occurs, and you will know to dial back the demands on your system.

But in any case, if you're worried about it, definitely don't mess with it.

  • Thumbs Up 1
Link to comment
Share on other sites

Does the GT650M actually throttle at 70C? I successfully applied the mod but noticed it doesn't throttle.

I also overclocked using evga precision x and when returning to default clocks they simply stay at 790mhz if this is normal behavior. Overclocking seems to help a lot.

I noticed there was no temperature setting for the HIGH FAN SPEED TRIPPOINT. So it gets triggered around 95-105C probably? has anyone tested this ?

Link to comment
Share on other sites

Because you will never conceivably run anything that will put the computer under that kind of forced load? Running one burn test, much less two, only indicates the extreme maximum your system can handle. It's well beyond what you'd ever see in normal operation. Using that as a basis for claiming the sky will fall under normal usage isn't... correct. If you run a more reasonable assessment of the temperature response in the system and find it to be 95 max (for example), you should be able to expect the temperatures not to spike beyond that, certainly not to the boiling point. And even in the unlikely event that it does spike that high, the system will shut itself down before any damage occurs, and you will know to dial back the demands on your system.

But in any case, if you're worried about it, definitely don't mess with it.

True, it's an extreme stress test. But just running FurMark by itself isn't very representative either since you're basically just maxing out the GPU and putting little to no load on the CPU. You need to have something running at the same time that will also put a significant load on most or all threads of your CPU to be representative of a very demanding game like Crysis 3. As you know Crysis 3 is very CPU hungry and will utilize all available threads on the i7-3630QM. Running a simultaneous CPU and GPU burn test is a little beyond that on the CPU side to be sure. I think an even more accurate representation of thermal conditions during gaming would be to run FurMark while transcoding a video on all CPU threads and using 50-70% of the CPU overall.

This kind of stress testing is causing me to reminisce about my ASUS G73Jh I sold prior to getting this Lenovo. Man that thing had a beastly cooling system. Not many gaming laptops out there can do FurMark+Prime95 24/7 without crashing or overheating and it was one of them. It was built like a tank and weighed 2-3 lbs more than the Lenovo but that dual-vent design and large case airflow was so much more efficient. I ran the extreme test for 1.5 days non-stop when I first got the machine for stability testing and the CPU and GPU both stayed in the 80's the entire time. Absolutely amazing considering that it had to manage hotter components like the 45nm i7-740QM and 40nm Mobility Radeon 5870. The Lenovo has the benefit of two shrinks on the CPU and one on the GPU but it still chickens out after mere minutes. :rolleyes:

Link to comment
Share on other sites

@octicep - That's why I'm thinking of putting some ramsinks on top of the heatpipes, do you think it will lower the temps on the cpu? my cpu temps reaches 100c in less the 30s running prime95 then throttles down to 1.2ghz.

btw I have the Y400 model.

Putting on some RAM heatsinks won't help a lot. If you guys want lower temps then repaste and see whether you can get some more pressure on the GPU die (and CPU), this helps a lot. I modified the backplate of my GPU and temps really went down. Just be careful not to crush the die.

Does the GT650M actually throttle at 70C? I successfully applied the mod but noticed it doesn't throttle.

I also overclocked using evga precision x and when returning to default clocks they simply stay at 790mhz if this is normal behavior. Overclocking seems to help a lot.

I noticed there was no temperature setting for the HIGH FAN SPEED TRIPPOINT. So it gets triggered around 95-105C probably? has anyone tested this ?

That's probably caused by your overclocking software. Try Nvidia Inspector. I had a lot of issues with precision.

What is the max temp for GPUs? I can't found any information about this. I know the CPU limit is at 105 C

You don't want your CPU to run at the limit, same for GPU. GPU is probably about 105°C as well, but it should start to throttle at about 95°C or more.

Also @all... don't use furmark as long as you want to have your card functional an running.

I'll prepare a volt modded BIOS when I get the chance, but there really doesn't seem to be a lot of thermal headroom at the moment.

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.