Lenovo G50-45 stuck in continuous black-screen cycle. BIOS not showing. Help!

I have the task of fixing a Lenovo G50-45 laptop (AMD A8-6410) for a relative of mine.

The laptop suffered a fall from about knee-height after it fell off the edge of a sofa. It remained in a perfect working state for 2 days afterwards. All I know is that the laptop was pestering to do updates before it was switched off. Now it is not working at all.

When I press the power button I can hear the hard drive spooling up and the fan running. Both the power and battery LEDs are lit up. The screen backlight does not switch on, but I can see the pixels illuminate in plain black. BIOS does not appear. After precisely 25 seconds of running with a blank screen, the laptop powers off and goes quiet. It then powers back on a few seconds later, and repeats this cycle continuously.

Things I have tried, to no avail:

• Connecting the laptop to a monitor via HDMI. (No input received)
• Removing the battery and running from mains power only.
• Re-seating the RAM modules.
• Trying the Shift+F8 combo to boot recovery.
• Pressing the Novo button to boot recovery.
• Tapping F1 on boot to access BIOS.

It seems to me that this is a problem with the motherboard. I’m wondering if the BIOS got corrupted somehow.

Unfortunately the BIOS is only available in an EXE format, which is useless considering the laptop doesn’t boot. With other laptops I have been able to flash the BIOS from a powered-off state using a USB drive, without the need to fully boot the machine into Windows.

Here is the BIOS. Would it be possible to extract the files and flash it manually, somehow?

https://download.lenovo.com/consumer/mobiles/a2cn45ww.exe

I’m not sure what else could be causing the problem. Any suggestions would be appreciated.

 

I had a different issue where BIOS chip on g50-80 become Read only after updating to 1709 win 10. I had to use experimental kernel for ubuntu to sort it out. After few months, laptop gets stuck on reboot while shutdown works. I decided to flash the BIOS from a good dump knowing it fully that I will break existing licenses. I went ahead and grabbed from laboneinside.com and flashed it via ch341a programmer black edition and afterwards it booted into windows/linux w/o any issue.
https://www.laboneinside.com/lenovo-g50-45-bios-bin/

 

Thanks for the response. Yes, I found a thread on the BIOS Mods forum and it seemed to suggest that a programmer is needed to flash the BIOS on the AMD version of this laptop. But as far as I can tell they are not expensive, about £10 GBP or so for the programmer and clip.

Is there any downside to overwriting the existing licenses? I saw on another forum that the serial number and other IDs are stored in the BIOS. Would it make a difference if this was erased?

Thanks.

 

Between Lost license and paper weight PC, I chose the PC and didn't care about the license since you can get it from ebay if MS Account doesn't recognize the digital entitlement from w8.1 to w10 pro.
You could ask @t456 for guidance. I overwrote the bios image with the one provided by labone guy which gave me his OEM license and Serial No. Though I didn't use his license because it was W8 key for non-Pro.

 

Oh, thank you! I see what you mean now, it just affects the license in Windows. Well that is not a problem, I can easily deal with that.

One more question. If I am buying one of these programmers, are there any specific requirements for it to work with this laptop? I just need to make sure the clip fits a SOIC8 chip? This is what is shown on the BIOS Mods forum, seems like they are also using the CH341A that you mentioned:

https://www.bios-mods.com/forum/Thr...Pad-S405-Whitelist-Removal?pid=74575#pid74575

I found an article here that seems to discuss the same problem I am encountering, with the Lenovo G50:

http://hulifox008.github.io/2018/07/21/g50-bios-recovery.html - Picture 1, Picture 2

 

Yes, I used the same thing. SOIC 8 test clip aka alligator clip. Make sure its aligned i.e. Red lining of alligator clip must be on a notch of winbond BIOS chip. For winbond chip the model will be w25q32(4MB) or w25q64(8MB). If its not winbond, you can see it will be marked as 25Q32 or 25Q64.
Only issue is, you will see a faint red LED glowing if connection is successful and when you press on Read option the RED LED will be steady which means connection is secure and its reading from the chip.
You need to verify the chip if contents are same as flashed image. If it says, mismatch. Try again. Make sure CMOS battery is disconnected/removed.
I don't know about modding BIOS and whenever I tried I bricked my PC 100% of the time. If you need any pointers in that, ask the wizard @t456 himself.

 

Much appreciated, lots of helpful advice there that will be of great assistance when I am able to do this. Thank you. I have made a backup of the BIOS files from the Lab-One site just in case the Google Drive links happen to go offline.

There's just one thing that concerns me, you mentioned about disconnecting the CMOS battery prior to flashing the BIOS. But as far as I know, the CMOS battery on this laptop is soldered the the motherboard. It seems the only way to remove it would be to use a soldering iron. Is that part essential?

 

Usually Lenovo has a small plug like connector for CMOS. If its not there, then don't worry.

 

Well I bought the programmer but I am still waiting for it to arrive. I went for the green one.

In the meantime, I have watched some tutorials on YouTube allowing me to understand how to connect the SOIC8 clip and how to use the software to flash the BIOS. All of that seems OK to me.

But one thing the tutorials don't seem to cover are the spaces in the board with labels next to them. On the programmer I ordered they are in 3 groups:

• 5V - 3.3V - MOSI - CLK - MISO - CS - GND
• GND - RX - TX
• P/S

I looked up a lot of these abbreviations, but I have zero understanding of electronics, so it means nothing to me except the voltages. I'm wondering if there is a certain 'configuration' I need to use with the jumpers for flashing the laptop BIOS. For example, how do I set the programmer to use 3.3V instead of 5V?

Also, am I supposed to use the 25XX section of the ZIF socket when using the SOIC8 clip?

Sorry for these beginner questions, hope someone can help.

 

I went for Black edition since I heard they had SW link included which didn't work and had to check YT and download the black edition SW that can with ch341a green/black. You can use AsProgrammer OR Normal ch341a SW: https://www.win-raid.com/t4175f16-GUIDE-Flash-BIOS-with-CH-A-programmer.html

 

I have the same Laptop with the same issue, flashing the bios did not help me. There must be something else wrong.

 

Hmm. Well I found quite a few people online with this issue, 3 of whom were able to solve it by flashing the BIOS. So I still have hope. Right now I am waiting for the SOIC8 clip to arrive before I can try it.

What procedure did you use to flash the BIOS, did you use a programmer like the CH341A?

Is your laptop also following the same pattern of rebooting after 25 seconds?

Here are a few sites where others had success:

1. http://hulifox008.github.io/2018/07/21/g50-bios-recovery.html
2. https://www.badcaps.net/forum/showthread.php?t=79764

 

i have the ch341a and downloaded the bios from https://www.laboneinside.com/lenovo-g50-45-bios-bin/
i´m not sure if its allowed to post that, if not can a mod remove the url?

before flashing i compared my dump with the new bios, there was no difference but i flashed it anyway.

yes i have the same pattern, the laptop starts, fan spins very fast for 25 seconds then it turns off for about 4 or 5 seconds and restarts itself again for 25 seconds.

just found out that the mosfet pq801 (for vga) is shorted to ground, don´t know why right now.

 

That is concerning. I am hoping I do not still have the same issue after flashing the BIOS myself. My SOIC8 test clip should hopefully arrive tomorrow, so I will keep you updated on my progress. All I know is that these symptoms point to a problem with the motherboard...

I thought it was unusual how the BIOS available to download from the Lenovo website is larger in size than the one from the Lab-One site. When you download the BIOS package from Lenovo is comes in an EXE installer, but it can be extracted to reveal the BIOS file (Win123.bin) which is 9.63MB. The one from Lab-One is 8.00MB. And as far as I know the BIOS chip only has the capacity to store 8.00MB.

 

look at this video, it explains everything

 

I lost AMD GPU in the process since LabOne had Intel CPU/GPU model. What I did was run latest BIOS package on top of existing one and AMD GPU was back.
I did attempt thrice to get laptop booting to Windows or any OS.

 

Thanks for the heads-up about that potential problem. Good job you told me.

I am just working on the laptop at the moment, unfortunately I did not forsee that it would require stripping the entire laptop down, I had hoped that the BIOS chip would be accessible via the maintenance cover, or at worst it would require removing the base from the machine. But it seems this laptop dismantles from the keyboard-side which is total inconvenience.

I do have another question - I have seen a few guides for different machines that mention using a 1.8V adapter when connecting the programmer to the BIOS chip. Is this necessary? I do not have one myself and purchasing one will add further delays and expense... The CH341A programmer I have is the green one which supposedly provides 3.3V. I don't know anything about the electronics side of computing and I am concerned about 'overloading' the chip... Did you use one?

 

I got a small jumper along with ch341a black and I used a video which showed me to short 1 and 2 pin which has 5v, GND printed on right side of the ch341a when you flip it over.
Never short GND and 5V while its plugged in. You'll see USB port failed message shown by windows 10. I thought my PC was a goner. But after reboot it was alright. I think the USB port was grounded.

 

That sounds very confusing. It is kind of embarrassing that I am so illiterate when it comes to electronics. I can already foresee a disaster approaching.

I think the procedure you mentioned was only applicable to the black version of the CH341A - I remember seeing comments from people who warned that the programmer could provide too much voltage, and it was advised to do a certain 'modification' to remedy it... And that's the reason I bought the green CH341A, because it seemed like the more hassle-free option for a novice like myself, with less opportunity to make a mistake.

This is exactly what my CH341A looks like underneath. I had just been planning to use it exactly how it is showed here, with the same 'configuration' as it came out of the packet. I also attached a picture of the SOIC8 clip I purchased:​

​

 

I believe black edition supplies more voltage 3.3V+ so that's why its shorted. I didn't see anything being needed on green edition.

 

Got mine working again, it was a bad cap on the vga rail and not the bios.
now i just have to figure out how i can reduce fan speed, seems to spin full speed all the time. the "lenovo fan speed control driver" doesn´t work on win10.

i have the black version of the ch341a, i would check youtube for your version

 

Glad to hear you got it working again. Though I am not really sure on what you mean, can you expand on what was causing the problem? I am really interested in case mine also has the same issue.

Yes I have looked on YouTube already for tutorials for the green CH341A. Sadly it seems that 95% of the videos on there are covering the black model, not the green one, and the ones that do cover the correct unit are in different languages other than English... But the latter videos are still helpful to a certain extent because I can gain information just from watching the pictures.

 

before flashing i saved my bios dump and compared it with the downloaded bios, they were the same. i flashed anyway hoping that would solve the problem.
it didn´t.

i downloaded the schematics and startet measuring the voltages.

every Laptop has a main power rail which goes to each component, each component has its own powersupply because of different needed voltages. each powersupply has protection through mosfets, if something is shorted the mosfet won´t open. the component would not start.
i had a shortet capacitor on the vga rail, thats why pq801 (vga mosfet) didn´t open. the bios trys to start vga and if it can´t it reboots (in loop), that´s why i first thought it was a bios problem.

after removing the capacitor the short was gone and pq801 opened, i had picture

 

@regi2020 Thanks for the explanation, makes a lot more sense now!

Well here's a little progress report.

It seems my CH341A has a little fault in that the ZIF socket will not stay locked... When I insert the pins from the SOIC8 test clip lead into the programmer, and I push the handle on the ZIF socket to lock everything in place, the handle immediately springs back upwards as soon as I let go. What I have done as a temporary solution is mount the CH341A onto a piece of foam-covered plastic secured with a zip-tie. Then I use a plastic peg to act as a clamp, to hold the handle in place and restrain it from springing upwards.

Anyway, I managed to successfully make a dump of the BIOS chip (GigaDevice 25B64BSIG) using the CH341A software in Windows.

As mentioned previously, the latest BIOS image I downloaded from the Lenovo site was 9.63MB, so I used my own dump as a guide to show me where exactly to trim the header-space from the BIOS file in order to reduce it down to 8.00MB. I flashed the resulting file to the BIOS chip using the CH341A programmer.

I have not yet tested the outcome. But if this does not work then I will resort to the BIOS file from the Lab-One site instead.

Is it possible I can just test the motherboard using an external monitor? It would save time in reassembling the laptop. I guess I can just plug the motherboard into mains power, ensuring that a monitor is connected via HDMI and that the power-button is coupled to the motherboard? Presumably I don't need to bother connecting the speakers, disc drive, HDD, keyboard, fan, internal display, etc just to make sure it works.

 

yes, you can do that.
no need to connect anything else, just power and a monitor

 

how did it go?

 

I was going to post another update tomorrow. Good news... I tried connecting the motherboard to my television via HDMI and now it seems to be working, I am getting display output and no more reboot-loop. Haven't tried connecting the HDD yet but I assume it will boot into Windows 10 no problem.

The only thing that worries me is that the BIOS from the Lenovo website will not include device-specific identification such as serial number, MAC address, UUID, etc.

I did find some BIOS dumps that users provided on another forum, which include their own IDs, and I figured how to transfer my serial number to the new BIOS by searching for the serial in my own corrupt BIOS file and copying the hex numbers across. However I have not been able to find the other IDs like MAC address and UUID, since I don't know what these numbers were to begin with, hence I cannot search for them as easily. I was warned by @Vasudev that it may cause licensing issues in Windows 10 without them.

 

Well I just booted the F1-key BIOS info screen (for the BIOS provided on the Lenovo site) and it shows this:

OA3 Key ID: 0000000000000
OA2: N​

I'm guessing there should be some values in here. I will try flashing one of the BIOS dumps that I found on the other forum. Perhaps those will have some numbers that I can use as a reference, to help me locate the correct figures in my old corrupted BIOS. Then I can transfer the values across using a hex-editor.

 

Hello, new guy here. I have the exact same AMD motherboard and original issue as you with my G50-45 (only lower grade APU installed) so thanks to this thread I decided to place an order for a CH341A programmer. It is arriving tomorrow.

I’ve read through the entire thread and as much as I could find in other places, but I’m still left with some question marks:

1. How did you successfully extract the Win123.bin file (latest official BIOS from Lenovo) from the .exe?
2. I read you bought the green programmer, I only had the choice to buy the black one - in any case, was it OK to use 3.3V when reading/flashing or does this specific chip require the 1.8V adapter in order not to get fried?
3. Which Hex Editor software did you use (sorry for asking but I’ve never really dabbed in editing hex code before)
4. What Windows program did you use for the CH341A and where did you find the download?
5. Where you able to read out and transfer your Windows license info to a new, modified BIOS in the end? If so, how? If not, did it work buying a new Windows license from eBay or similarly?
6. What other bios dumps from another forum are you looking at, I’d love to have more information to increase my chances of success.

Apologies for the heap of questions. Thanks in advance! And thanks for creating this thread!

 

@Kenta1337 Sorry to hear you are having the same problem, but yes, I think the CH341A is your best bet at fixing it.

1. The BIOS from the Lenovo site can be extracted from the EXE using 7-Zip and is called Win213.bin. When you download the file you realise it is too big. The chip is only 8MB but the extracted BIOS is about 9.5MB. You need to trim the file in HxD. Compare the file against the dump from your own chip, to see precisely where at the beginning and end to trim the file. You'll know if you have it right because it'll end up being exactly 8.00MB in size once saved.

2. I had the same concerns as yourself. I also bought the black one at first, but I returned it to Amazon after reading comments from people saying this reader sometimes inadvertently provided 5V instead of 3.3V. There was some kind of modification people did to correct this problem, but I decided it sounded too complicated for me, so I returned it and bought the green one instead which everyone said was safe. However I think a lot of these concerns are overblown as I have never heard of anyone frying their BIOS chip with one of these, and there is even someone in this thread who used the black programmer, so I think it is fine. - I was also concerned about whether I needed a 1.8V adapter, but in the end I concluded it was not necessary, which turned out to be correct as I attached the programmer several times without consequence.

3. I used HxD 2.4.0.0 - https://mh-nexus.de/en/hxd/

4. I cannot remember where I downloaded the program, I believe someone was hosting a 'collection' of all different versions of the software compiled into one archive. I deleted everything aside from the one I used, but I reuploaded it underneath. Note the drivers in the other folder, CH341PAR.EXE and CH341SER.EXE - You have to install these first. I sourced the drivers direct from the WCH website to be safe:

https://drive.google.com/file/d/10ue7apL0ikbiAbYYNWoTpY6X6Yjo-DVS/view?usp=sharing

5. I was not able to do it. Transferring the details turned out to be a disaster and caused the laptop to stick on the BIOS screen when booting. In the end I had to revert the changes. I wouldn't bother with any additional hex-editing, since I did not experience any problems by skipping it. Windows has not complained of any problems with the license as of yet.

6. Not sure, I will try to find it, but it might not be required anyway.

 

@CivicJDM Thank you so much for providing the files and such detailed and polite responses! Regarding the 5V problem, I think @Vasudev mentioned something about how to do the modification earlier in the thread, but I wasn't entirely sure on what steps to take or what solder points to bridge. I'll look further into it before making the rescue attempt tomorrow when the programmer has arrived.

Besides that, I think I'm all set thanks to you! I'm also relieved to hear that Windows didn't complain about the license either. Perhaps it has verified it through some other unique ID that isn't tied to the BIOS. Here's hoping.

I'll report back as soon as I have a chance to try out the procedure! Thanks again!

 

@Kenta1337 No worries, glad you found my responses helpful, I will try my best to assist if you have any further questions.

Yes, as I mentioned some of the values in the BIOS were erased after flashing, but I checked the license status in Windows and it still showed that it was registered. Hope it will be the same for you.

My frustration with this laptop is the fact you even have to do any of this... Most laptops have the ability to flash the BIOS from a USB stick. Can be done within 15 minutes. But with this one you have to purchase the programming tool and the test clip, which is OK, but the BIOS chip is not accessible via the maintenance hatch since there is a thin piece of plastic covering it, and because the laptop disassembles from the keyboard side it means you have to disassemble the ENTIRE LAPTOP and remove the motherboard just to access the BIOS chip... So something that would take 15 minutes in most laptops ends up taking countless hours (even days) in research and labour with this one.

 

@CivicJDM Yes, the disassembly part is annoying indeed. I disassemble laptops all the time so for me it isn't as big of a pain but I did consider cutting out that plastic part in the case covering the chip in order to make the testing process much simpler. It is one thing to disassemble and reassemble a laptop once, but every time between a repair attempt? No thanks we'll see what I end up doing. The programmer should arrive within a couple of hours at my door!

 

Success! I almost can’t believe it worked, haha!


The Programmer
As suspected, testing the black board CH341A with the multimeter it would only spit out 5V despite advertising also being able to support 3.3V out of the box.

So what I ended up doing was modifying the board after this excellent guide:
https://www.chucknemeth.com/usb-devices/ch341a/3v-ch341a-mod



Garbage soldering job, I know, but it worked and doing another test with the multimeter - sure enough, 3.26-something volts which I deemed acceptable.

If you don’t have the necessary tools or experience soldering, I highly recommend going for the green board instead.

I learned how to assemble and connect the parts thanks to this video:


The BIOS
Using the manufacturers programming software (the one @CivicJDM ) I got an error upon installing the driver, but it seemed to work anyway. The programmer was detected by the software and could detect and read the chip.

The hardest part was actually to successfully extract the Win123.bin BIOS file from the .exe provided on Lenovo’s website. Using only 7-zip threw errors upon trying to extract the second .exe. So looking closely at the steps outlined in this video:


I tried the tool the guy in the video used, “Uniextract”. I was a bit worried about the security regarding the software since it only seemed to be available on shady download sites, but the download I found in this thread ended up doing the trick as well as not having Windows 10 throwing an unsafe alert:
https://www.portablefreeware.com/forums/viewtopic.php?t=22394

As for trimming the unnecessary hex code in the new BIOS file, I followed your advice, read out and saved a copy of the old, corrupted BIOS, compared it with the new and trimmed the beginning and end parts of the file, ending up with exactly 8MB as you said.

The Flashing
Programming the newly trimmed BIOS to the chip, the programmer software threw a mismatch error upon finishing up the writing process.

What I ended up doing was erasing the chip, doing a blank check, then writing the new BIOS again with the Auto command (Erase and blank unchecked). This ended up working on the first try.



Connecting the board to power, HDMI and the stupid power button I was able to check the results and low and behold - the computer booted straight into the EFI netbooting tool! Which means it worked!

Reassembling the computer (changing out the thermal paste in the process), everything worked.



One thing that surprised me even more was that the license info seems intact even after the fact! I used the BIOS file from Lenovo’s site as a base for the BIOS I wrote to the chip, so perhaps the license info isn’t in the BIOS after all? Who knows. The W10 home license check was OK even after a format and OS reinstall.



In the end, great success. Thanks to you guys and other helpful people online I was able to turn this brick into a usable computer again. Now if only the AMD E1 CPU wasn’t so slow…

 

And the fan is constantly spinning on 100% when in Windows. I don't mind it that much since even at 100% the noise isn't that bad, just a bit irritating that it doesn't auto adjust itself since the CPU isn't getting hot at all. If someone has a nice solution to this, let me know - I haven't found anything yet.

 

Try ' reset to defaults', if you haven't already. If that doesn't work then flash a different bios with the regular procedure, though perhaps you first need to use the programmer again prior to that if you happen to have used the latest bios version.

The issue seems to be that the ec you have flashed isn't communicating properly with the keyboard controller chip and hence isn't receiving the rpm signal from the fans. Safety-wise it'll then do the 100% speed thingy (same as what you see during a normal flash and for the same reason). Could run HWiNFO and check the values there.

 

@t456 Thank you for the input! Here's a followup:

I currently have two issues with the computer:
A. Fan spinning at 100% (besides in the BIOS, for some reason)
B. Display backlight turning off, a couple of seconds after waking from sleep (the backlight was also off during the entire Windows 10 install). Only goes back to normal after a reboot.

These issues could potentially both be related to what you were saying.

1. I tried the "Load Default Settings", unfortunately it had no effect.



2. To be honest, I had no idea what "EC" was or what it did. After some quick research, it seems similar to what an SMC on a Mac is responsible for (I just got back into PC's after being mostly a Mac user for 15 years), so what you're saying makes complete sense. Having the BIOS and EC versions mismatched could be a potential source of trouble indeed. It is probably untouched while the BIOS was updated to the latest version with the programmer.

Does updating the BIOS the regular way (InsydeFlash's Windows app) normally update the EC version as well?

3. HWinfo does not register any RPM values at all. It does get a CPU temperature reading though. AIDA64 however neither gets a temperature nor RPM value.



4. I found these options in the BIOS as well:




Should I give the "Reset to Setup Mode" a go as well? What exactly does this option do?
I am very hesitant to trying the "Restore Factory Keys" options since I'm guessing that it will wipe the computer's embedded Windows license info (quick research online ended up with conflicting info on what this option actually does).

5. I would like to have the EC to match the BIOS version to remedy the problems. Hoping for an easy fix, I ran the regular BIOS update from Lenovo in Windows, only to be greeted by a message that says the update version and current BIOS version being the same, thus exiting the program without moving further.



I would rather not disassemble the entire computer and fiddle with the programmer again if possible. Could it be possible to take advantage of the "BIOS Flash Back" function?




Problems with this approach:
- Lenovo does not seem to provide older BIOS versions for this model on their website, only the latest one.
- I've read somewhere that the procedure will entail putting the BIOS file on a FAT32 formatted USB stick and pressing the fn+R key combination at boot in order to force flash. However, the BIOS file must be renamed to something else than .bin in order for it to work?

The easiest approach at the moment seems to be if you could bypass the version check that InsydeFlash does. Is it possible?

Apologies for the long post. Thanks again!

 

Update:

Turns out, it was possible!

By going inside the "app" folder that I extracted previously to get the Win213.bin file, editing the flags Platform_Check and Bios_Version_Check in the platform.ini file to 0 and then running the InsydeFlash.exe inside the folder, it worked!




Lowe and behold! The BIOS and EC version are now the same:



What changed:

- The backlight turning off after sleep issue seems fixed! After putting the computer to sleep and waking it several times, as well as booting the Windows installer again to check if the same issue occurred, everything now works as expected!

EDIT:
What didn't change

- HWinfo still does not see any RPM values.
- AIDA64 still can not see any CPU temps.
- The fan speed seems the same as before. However, while the computer was flashing the EC, the fan speed ramped up to even higher than what I assumed was 100%. So maybe it is working as intended after all? Since the fans are quiet in BIOS, I thought that was supposed to be the default for when the computer is idle.

Current fan curve:

BIOS: Quiet
Windows: Slight "woosh" (same both idle and during load)
During EC update: Full on big "WUUUUSH"

I'm still curious about if the "Reset to Setup Mode" in the BIOS is worth exploring and what "Clear PK" does, but maybe it is not worth bothering with?

 

Once Windows 10 is activated online then bios settings and content should no longer matter. Both options relate to Secure Boot btw and ought not to have any influence on fan sensing and control.

Very nice .

You're quite lucky; many new systems would refuse to flash regardless of what is set in the ini. The 'Back flash' option in your bios is even visible and editable, so you might even revert to earlier versions without needing a programmer whereas many system have this option hidden. Lenovo has been quite lenient with this particular system so far, but do take care; one day there'll be a bios update with 'enhanced security' that might set and hide this option. Dell and HP tend do this with the very last bios version they provide.

Could also try the '[ForceFlash]' and '[Region]' settings; these provide a few more options to write to otherwise protected (skipped) sections. Mind that the programmer ignores all of that though, so it has written every byte to the eeprom.

That does not necessarily have to mean something; ECs vary wildly in laptops and not all are supported by HWiNFO. Best check whether rpms are visible on another user's G50-45. Also tinker with the options under ' Safety'; one of these might do the trick.

Whether the cpu or gpu is doing any work or nothing at all has no influence on fan speed, only temperature has. The EC contains a temp-rpm table that it is using to send either a specific pwm signal to the fan (0-100%) or, more rarely, a variable voltage (0-5V) and with the fan spinning up the tachometer sends back the effective rpms to the EC. If you were to hold back the little fan with a finger then the rpm value returned would be '0' and the system should automatically shutdown.

Run a benchmark while keeping tabs on the temperatures in HWiNFO; if all is working well then fan speed should increase (unless you're using LN or happen to be outdoors in Antarctica ). If this happens with noticeable steps then the fan table is a bit crude, but if there's a nice gradient then it has many entries and might sound somewhat more natural to the ears.

Clear Platform Key.

The system is already in Setup Mode though, according to the bios screenshot. And oddly enough you have SB enabled, but the status claims ' Disabled' ... perhaps because it hasn't yet written the Platform Key.

None of the options have anything to do with the workings of the EC itself. They're only there for ' driver validation purposes', but think it more likely entails ' cash cow' for the big names in the UEFI Forum since device manufacturers and third party driver programmers now have to pay certification fees in order to get their product running on SB-enabled systems. In theory they ought to be performing tests on varying platforms in return (what else is signing for?), but considering the easily detected and avoided Synaptics troubles some time ago it is more of a pipe dream than anything else. Also, the purported motivation behind Secure Boot is not a little ridiculous since it assumes that new drivers would otherwise be coded, compiled and thrown to the consumers without doing any testing at all.

HWiNFO also uses its own driver, so you might simply disable SB and see whether rpm reporting magically re-appears.

 

@t456 Thank you for all your comments. I did disable secure boot and tried HWinfo again but no change. Also, the fan speed didn't seem to change much at all during a stress test. It may very well be the result of that the E1 CPU in the machine is so underpowered (10W TDP) that it just doesn't get hot but at the same time runs att pretty much 100% all the time since it has a hard time keeping up with just running Windows as is.

Anyhow, I'm happy with the result. The fan noise isn't unbearable, sleep now works again and most of all - the computer works again! And it's all thanks to the efforts of the community. So thanks!

And yeah, Secure Boot seems like a big mess.