Jump to content
NotebookTalk

Y740-15ichg 2070-MaxQ (ID 10DE 1F50) vBIOS flashed to 115w limit


Tofu

Recommended Posts

Did some research and grabbed some vBIOS files off of techpowerup to see if I could increase the power limit for my 2070mq since I even with both cpu and gpu at full stress, gpu was only hitting maybe 78c, and I already replaced the thermal pads on my cooler.

 

First one I tried seemed the most interesting (which is apparently a non-maxq vbios from an HP Omen) actually seems to work perfectly fine, I see it hitting a power limit of 115w now, I'm still testing to make sure this is actually stable and working well. I read about other people flashing a different vbios and losing output from their displayport connector but mine seems to be working just fine, looks like the HP has the same setup.
This is the vBIOS I used: https://www.techpowerup.com/vgabios/225730/225730
(Attaching it here as well as a backup)

 

I used the modified NVFlash from here to ignore the board ID https://winraid.level1techs.com/t/modified-nvflash-v5-667-0/36029
(Also attaching here as a backup)

 

GPU-Z:

1549874143_2022-07-0817_39_18-TechPowerUpGPU-Z2_46.0.png.c3de224036ce62dda7928f213e45fb6d.png

hp-2070mq-1F50-115w-1215mhz.rom lenovo-2070mq-1F50-90w.rom nvflash v5.670.0 patched update3.7z

Link to comment
Share on other sites

So after doing some testing it seems like there's something not right going on because even though HWInfo reports the gpu as drawing 115w I can't get more than 60w on the core itself, and it locks the core MHz to around 600.

 

But what's odd is even after I flash back to stock My GPU seems to max out at 1000MHz if I leave it on default but if I try to change it in afterburner then it pretty much always drops down to 900 or lower.

 

Only two things I can think of are an advanced BIOS setting, which I'll try resetting that and see if it helps, and the other is that I took the battery out since it was puffing and I don't really need this to be mobile any more. Would a lack of battery cause an issue with the GPU core?

Link to comment
Share on other sites

40 minutes ago, Tofu said:

So after doing some testing it seems like there's something not right going on because even though HWInfo reports the gpu as drawing 115w I can't get more than 60w on the core itself, and it locks the core MHz to around 600.

 

But what's odd is even after I flash back to stock My GPU seems to max out at 1000MHz if I leave it on default but if I try to change it in afterburner then it pretty much always drops down to 900 or lower.

 

Only two things I can think of are an advanced BIOS setting, which I'll try resetting that and see if it helps, and the other is that I took the battery out since it was puffing and I don't really need this to be mobile any more. Would a lack of battery cause an issue with the GPU core?

depends, some laptops have that "battery boost" feature *cough* where the system draws additional power from the battery during high loads when needed since the psu is only designed to just baaaaaarely handle the system wattages... ull have to check if that also applies to your model or not.

Mine: Hyperion "Titan God of Heat, Heavenly Light, Power" (2022-24)
AMD Ryzen 9 7950X (TG High Perf. IHS) / Asus ROG Crosshair X670E Extreme / MSI Geforce RTX 4090 Suprim X / Teamgroup T-Force Delta RGB DDR5-8200 2x24 GB / Seagate Firecuda 530 4 TB / 5x Samsung 860 Evo 4 TB / Arctic Liquid Freezer II 420 (Push/Pull 6x Noctua NF-A14 IndustrialPPC-3000 intake) / Seasonic TX-1600 W Titanium / Phanteks Enthoo Pro 2 TG (3x Arctic P12 A-RGB intake / 4x Arctic P14 A-RGB exhaust / 1x Arctic P14 A-RGB RAM cooling) / Samsung Odyssey Neo G8 32" 4K 240 Hz / Ducky One 3 Daybreak Fullsize Cherry MX Brown / Corsair M65 Ultra RGB / PDP Afterglow Wave Black / Beyerdynamic DT 770 Pro X Limited Edition

 

My Lady's: Clevo NH55JNNQ "Alfred" (2022-24)
Sharp LQ156M1JW03 FHD matte 15.6" IGZO 8 bit @248 Hz / Intel Core i5 12600 / Nvidia Geforce RTX 3070 Ti / Mushkin Redline DDR4-3200 2x32 GB / Samsung 970 Pro 1 TB / Samsung 870 QVO 8 TB / Intel AX201 WIFI 6+BT 5.2 / Win 11 Pro Phoenix Lite OS / 230 W PSU powered by Prema Mod!

Link to comment
Share on other sites

That is interesting.

I have the 2080maxq also with plenty of head room for higher thermals.

 

I was going to go with a shunt mod but never got around to it as I pretty much stopped playing games.

 

 

Thunderchild // Lenovo Legion Y740 17" i7-9750H rtx2080maxQ win10 

RainBird // Alienware 17 (Ranger) i7-4910mq gtx860m win8.1

 

 

Link to comment
Share on other sites

On 7/9/2022 at 12:09 AM, jaybee83 said:

depends, some laptops have that "battery boost" feature *cough* where the system draws additional power from the battery during high loads when needed since the psu is only designed to just baaaaaarely handle the system wattages... ull have to check if that also applies to your model or not.

 

I'll have to find my battery somewhere to test this, and if that is the case then I'll have to order another battery.
I believe the laptop shouldn't be limited without the battery since the power supply is rated to 230w, but I've seen more confusing design choices.

 

On 7/9/2022 at 5:03 AM, Eban said:

That is interesting.

I have the 2080maxq also with plenty of head room for higher thermals.

 

I was going to go with a shunt mod but never got around to it as I pretty much stopped playing games.

 

 

 

If you're interested you could try finding a vBIOS that matches your device ID and has higher clocks on here: https://www.techpowerup.com/vgabios/?architecture=Uploads&manufacturer=&model=RTX+2080+Mobile&interface=&memType=&memSize=&since=
So long as you're in hybrid mode you can flash multiple times without having to reboot since the intel gpu is driving the display (though I usually unplug any external displays before flashing), so flash, test if it works better or worse, then flash again if it has any issues.

What's odd for me is the core clock seems to change proportionally to the memclock in that driving the memclock higher makes my core clock go lower.

I did replace the thermal pads at one point and now I'm wondering if maybe the VRMs are overheating? Or maybe it's something with the board itself as I did send my laptop in for repair some time ago and they replaced my 2060 board with a 2070mq board...

Along with forgetting to flash the windows license and Lenovo product name into my BIOS, and they also left out the keyboard backlight ribbon entirely, though I had a spare from a replacement top plate they sent me when I complained about my touchpad clicking poorly.

Link to comment
Share on other sites

13 hours ago, Tofu said:

What's odd for me is the core clock seems to change proportionally to the memclock in that driving the memclock higher makes my core clock go lower.

I did replace the thermal pads at one point and now I'm wondering if maybe the VRMs are overheating? Or maybe it's something with the board itself as I did send my laptop in for repair some time ago and they replaced my 2060 board with a 2070mq board...

thats a sign that there is some kind of power limit applied, but not from vbios but rather the mobo side. seems like there is some TBP envelope which is being completely used up now by the vRAM, GPU core and other components.

as for VRMs, u can double check the thermal pad contact with the heatsink (indentations) so make sure u have proper contact.

Mine: Hyperion "Titan God of Heat, Heavenly Light, Power" (2022-24)
AMD Ryzen 9 7950X (TG High Perf. IHS) / Asus ROG Crosshair X670E Extreme / MSI Geforce RTX 4090 Suprim X / Teamgroup T-Force Delta RGB DDR5-8200 2x24 GB / Seagate Firecuda 530 4 TB / 5x Samsung 860 Evo 4 TB / Arctic Liquid Freezer II 420 (Push/Pull 6x Noctua NF-A14 IndustrialPPC-3000 intake) / Seasonic TX-1600 W Titanium / Phanteks Enthoo Pro 2 TG (3x Arctic P12 A-RGB intake / 4x Arctic P14 A-RGB exhaust / 1x Arctic P14 A-RGB RAM cooling) / Samsung Odyssey Neo G8 32" 4K 240 Hz / Ducky One 3 Daybreak Fullsize Cherry MX Brown / Corsair M65 Ultra RGB / PDP Afterglow Wave Black / Beyerdynamic DT 770 Pro X Limited Edition

 

My Lady's: Clevo NH55JNNQ "Alfred" (2022-24)
Sharp LQ156M1JW03 FHD matte 15.6" IGZO 8 bit @248 Hz / Intel Core i5 12600 / Nvidia Geforce RTX 3070 Ti / Mushkin Redline DDR4-3200 2x32 GB / Samsung 970 Pro 1 TB / Samsung 870 QVO 8 TB / Intel AX201 WIFI 6+BT 5.2 / Win 11 Pro Phoenix Lite OS / 230 W PSU powered by Prema Mod!

Link to comment
Share on other sites

  • 2 weeks later...

Did some more testing and I'm just confused. On stock vBIOS my gpu core hits .750v with furmark at 1440p 8x AA, but without AA it doesn't go above .615.

On 115w vBIOS I don't see the GPU core going above .685v no matter what furmark settings.

Attaching an image of hwinfo with 115w vBIOS running furmark at 1440p with 8x AA

 

Anyone have an idea of what might be going on?

2022-07-21 00_06_53-HWiNFO64 v7.26-4800 - Sensors Status.png

Link to comment
Share on other sites

3 hours ago, Tofu said:

Did some more testing and I'm just confused. On stock vBIOS my gpu core hits .750v with furmark at 1440p 8x AA, but without AA it doesn't go above .615.

On 115w vBIOS I don't see the GPU core going above .685v no matter what furmark settings.

Attaching an image of hwinfo with 115w vBIOS running furmark at 1440p with 8x AA

 

Anyone have an idea of what might be going on?

2022-07-21 00_06_53-HWiNFO64 v7.26-4800 - Sensors Status.png

i wouldnt rely on furmark for such testing, there are way too many algorithms in place that changes the default gpu behaviour when it detects furmark running. rather check with somth like 3DMark stress tests or superposition.

Mine: Hyperion "Titan God of Heat, Heavenly Light, Power" (2022-24)
AMD Ryzen 9 7950X (TG High Perf. IHS) / Asus ROG Crosshair X670E Extreme / MSI Geforce RTX 4090 Suprim X / Teamgroup T-Force Delta RGB DDR5-8200 2x24 GB / Seagate Firecuda 530 4 TB / 5x Samsung 860 Evo 4 TB / Arctic Liquid Freezer II 420 (Push/Pull 6x Noctua NF-A14 IndustrialPPC-3000 intake) / Seasonic TX-1600 W Titanium / Phanteks Enthoo Pro 2 TG (3x Arctic P12 A-RGB intake / 4x Arctic P14 A-RGB exhaust / 1x Arctic P14 A-RGB RAM cooling) / Samsung Odyssey Neo G8 32" 4K 240 Hz / Ducky One 3 Daybreak Fullsize Cherry MX Brown / Corsair M65 Ultra RGB / PDP Afterglow Wave Black / Beyerdynamic DT 770 Pro X Limited Edition

 

My Lady's: Clevo NH55JNNQ "Alfred" (2022-24)
Sharp LQ156M1JW03 FHD matte 15.6" IGZO 8 bit @248 Hz / Intel Core i5 12600 / Nvidia Geforce RTX 3070 Ti / Mushkin Redline DDR4-3200 2x32 GB / Samsung 970 Pro 1 TB / Samsung 870 QVO 8 TB / Intel AX201 WIFI 6+BT 5.2 / Win 11 Pro Phoenix Lite OS / 230 W PSU powered by Prema Mod!

Link to comment
Share on other sites

20 hours ago, jaybee83 said:

i wouldnt rely on furmark for such testing, there are way too many algorithms in place that changes the default gpu behaviour when it detects furmark running. rather check with somth like 3DMark stress tests or superposition.

Fair enough, did some more testing with superposition and still seeing similar results, 115w gpu vcore won't go above ~.685 while 95w hits .800 fairly often.
Maybe it's just some kind of hard limit of my motherboard, in which case I guess I'll just have to stick to traditional OC/UV

Link to comment
Share on other sites

This is still very interesting! Thank you for sharing your discovery.

 

Could try the shunt mod if your into extreme :classic_ohmy:

Thunderchild // Lenovo Legion Y740 17" i7-9750H rtx2080maxQ win10 

RainBird // Alienware 17 (Ranger) i7-4910mq gtx860m win8.1

 

 

Link to comment
Share on other sites

7 hours ago, Tofu said:

Fair enough, did some more testing with superposition and still seeing similar results, 115w gpu vcore won't go above ~.685 while 95w hits .800 fairly often.
Maybe it's just some kind of hard limit of my motherboard, in which case I guess I'll just have to stick to traditional OC/UV

sounds like your power delivery is holding you back. even when flashing a higher wattage gpu bios, theres still a total system wattage limit in place. and, of course, u never know what weirdness can happen when flashing unofficial stuff haha 😋

Mine: Hyperion "Titan God of Heat, Heavenly Light, Power" (2022-24)
AMD Ryzen 9 7950X (TG High Perf. IHS) / Asus ROG Crosshair X670E Extreme / MSI Geforce RTX 4090 Suprim X / Teamgroup T-Force Delta RGB DDR5-8200 2x24 GB / Seagate Firecuda 530 4 TB / 5x Samsung 860 Evo 4 TB / Arctic Liquid Freezer II 420 (Push/Pull 6x Noctua NF-A14 IndustrialPPC-3000 intake) / Seasonic TX-1600 W Titanium / Phanteks Enthoo Pro 2 TG (3x Arctic P12 A-RGB intake / 4x Arctic P14 A-RGB exhaust / 1x Arctic P14 A-RGB RAM cooling) / Samsung Odyssey Neo G8 32" 4K 240 Hz / Ducky One 3 Daybreak Fullsize Cherry MX Brown / Corsair M65 Ultra RGB / PDP Afterglow Wave Black / Beyerdynamic DT 770 Pro X Limited Edition

 

My Lady's: Clevo NH55JNNQ "Alfred" (2022-24)
Sharp LQ156M1JW03 FHD matte 15.6" IGZO 8 bit @248 Hz / Intel Core i5 12600 / Nvidia Geforce RTX 3070 Ti / Mushkin Redline DDR4-3200 2x32 GB / Samsung 970 Pro 1 TB / Samsung 870 QVO 8 TB / Intel AX201 WIFI 6+BT 5.2 / Win 11 Pro Phoenix Lite OS / 230 W PSU powered by Prema Mod!

Link to comment
Share on other sites

8 hours ago, Eban said:

This is still very interesting! Thank you for sharing your discovery.

 

Could try the shunt mod if your into extreme :classic_ohmy:

I've definitely thought about it, but it doesn't seem worth it when this is my only personal computer currently.

I might try flashing some other vBIOS files that are clocked higher but not higher wattage to see if it makes any difference.

 

What's interesting is from everything I can find the only difference between 2070mq and 2070m is the mq is clocked lower with a lower TDP.

Part of me now wonders if maybe the whole Max-Q thing was a plan to sell more lower binned chips that don't pass requirements otherwise, as theoretically I should be able to clock pretty to similar levels as a stock 2070m.

 

3 hours ago, jaybee83 said:

sounds like your power delivery is holding you back. even when flashing a higher wattage gpu bios, theres still a total system wattage limit in place. and, of course, u never know what weirdness can happen when flashing unofficial stuff haha 😋

What still gets me is that the sensors are actually reporting 115w going into the GPU but most of it just gets listed under Misc0 Input Power, makes me wonder what the sensors are actually reading from, or if that's just what the chip itself is allowing but the motherboard itself isn't providing more than 95w, which then makes me curious if a shunt mod would even work, maybe Lenovo put a limiter before the GPU somewhere.


There's also an allow overclock option in the debug bios settings for this laptop but whenever I enable it, it just defaults back to disabled when I save and reboot, otherwise that would be a nice option to tweak the core voltage.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...

Important Information

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. Terms of Use