Lenovo Legion Y740 17" 15"

I had hardlocks on my first Y740 which has been returned already.

For that particular laptop, I traced the issue to the Intel UHD 630 GPU being defective. Whenever the GPU was put under load long enough (like 1 second of any modern 3D games), the whole laptop would hard lock and require holding down the power button to shut down. You can troubleshoot by forcing only the Intel GPU to be used while in Hybrid mode, then starting some modern games up.

My replacement came, and under the same test conditions, no hard lock is observed, so I know for sure that was the issue.

However, I suspect my replacement was somebody else's returned laptop. The stock blue paste was haphazardly removed and replaced with a copious amount of grey paste, which INCREASED the temps by 10 to 20 degrees C. I know this because there's specks of blue paste around the edges mixed in, and many of the thermal pads had been contaminated with disgusting short dust fibers + specks of the original blue paste, which from the factory I would believe is impossible.

Based on my research, the units with serial numbers ending in "RF" may actually indicate "refurbished", but I cannot be certain of this, so check your serial number. The sloppy repaste job on mine was very suspicious though.

 

hello all,
I was wondering if this is a normal thing for this laptop. (Y740-15 inch, 8750H and 2070MQ)
The cpu idles at boost clock speed even with a very low utilization while plugged-in, but the clock speed dropped when unplugged. I use better performance settings when unplugged.

I wonder if its a bios thing, I did upgrade/downgrade bios to 1.10 to 1.08



I dont feel comfortable my CPU clock is absurdly high without the laptop needing it to do so.

 

Yea this isnt that issue. Windows will hard lock like 10-20% of the time when I try to either restart windows or shut windows down. I cant seem to figure out in event manager the cause of it unfortunately. My replacement should be brand new since iv only had it for 15 days. Im worried the new one will do the same thing though. Other then that issue, its been amazing, the best laptop IV ever owned, side from my canvas. Iv owned dozens of laptops throughout the years. This is my first G-sync enabled laptop and it sure is amazing not having tearing. Also the lighting on the KB is super sweet, it runs cool in the 70s with a slight undervolt. So many nice features. the 500 nits with HDR really makes games pop and to top it all off its 144hz. Damn hardlocks ruin the fun though even though they dont effect using software or titles i dont want to put up with a machine for years to come that I cant restart without worrying about it hardlocking for no apparent reason.

 

This is a hidden effect of the high-perfomance battery profile, probably best to leave it on balanced/better-performance (these seem to be the same setting).

Though, what I really want to know is how yours idles at 40 C and peaks at 70... heck of a lot better than mine's doing. Looking like I may need to open it up yet again to check for this terrible gray paste people are talking about.

 

@DefinityX
I have done every restore and fresh install possible. No undervolt. Disabling turbo. Different power plans. Every possible scenario I could think of- 1.6 causes the hard locks in both PRO and Home. 1809 -1903 with most recent updates. Even used driver easy to update drivers to newer driver version than what Lenovo offers. Even J95's debloated Nvidia drivers ,standard. DCH. Rolling back to anything lower 1.6 resolved It on the work laptop. But went back to 1.6 to finish testing.

1.3 with all above listed scenarios results in no lockups . The fan spins at idle 1.3 - keeping air circulating VS. waiting for heatsoak to occur before the fans ever attempted to spin in 1.6

The only thing I can put my finger on within the OS itself right before the hard lock is the Wifi NIC tossing a Netwtw06 error in the event viewer . I have used every possible driver for the wifi card - as it's made by Intel - killer branded .Even both Intel drivers and killers versions for the card.

 

My hardlocks are happening with stock settings, no overclock or undervolt, and 99% of the time when I try to restart the machine or shutdown windows, or when going into sleep mode. So no one else is having hardlocks in these situations. My machine works fine while gaming or operating windows, like I say its just when I try to turn windows off. Anyone have any ideas?

 

Hmm. I've had a couple of lockups/bluescreens on 1.6, but pretty sure that was caused by undervolting the cache too much (past -0.140), have not seen that with lesser undervolts. But maybe that's a different problem that only happens to some units for some reason? Have not updated Windows or any drivers yet.

On the other hand, I think both of the lockups (but not the bluescreen) happened while web browsing, after perfectly stable stress tests and/or games. So maybe it really is the Wifi card.

1.3 kinda sounds like too much in the opposite direction for fan speed, would prefer the fans not spin at idle; but on 1.6 they seem to wait around until like 10 seconds after hitting thermal throttling temps before the fans gradually begin to spin up. A middle ground between those would be nice. But maybe this CPU just can't stay under 50 without running the fans, even on idle?

 

I could see the middle ground part aswell. But at least with 1.3 it's moving air before it's too late lol.

 

I thought that was the case too, but I am using better performance instead of best performance battery setting while charging, which is why it still bothers me.

I was using the computer straight above hvac system that blows cold air into the laptop, real-desk usage I was seeing something like 94s for the CPU. Feels like I also need to repaste to get a better thermals.

 

Don't run your laptop on top of an HVAC system. It can cause condensation that will kill your PC.

 

Anyone else having wrist issues when gaming on the 17 inch or 15 inch? My arms get "cuts" from the edges and after a while it starts hurting ..?

 

My advice would be to re-examine the way you are using your laptop and go see a doc if you've got wrist pain.

 

The issue is that unlike my old laptop this one has very large distances between its keyboard and the end of it (lets call it "large keyboard edges") and no matter how i use it those edges press on my wrists or when gaming just don't feel "right" and stab my wrist while playing..

The only fix i found for this was using my adjustable desk that lowers the table and eases the pressure..but 99% of places i go to with the laptop don't have that and again 99% of places use standard tables/desks that are too high..thinking of returning it and getting something with much smaller keyboard edges. Y540 15 inch comes to mind.

Hope i can adjust to it since it's a great laptop overall, leaps and bounds over the Asus ROG heatfest/noisefest.

 

If you are having problems with your wrist there, you can always get a simple solution like a wrist rest (GRIFITI FBA_24021 Large Slim Palm Pads Notebook Wrist Rests and Laptop Wrist Pads Made with Silicone to Easily Reposition and Remove while Travelling (2 Large 4 x 3.12 inches) https://www.amazon.ca/dp/B00CMJ1S8A/ref=cm_sw_r_cp_apa_i_gasADbXRK7NRF) or a wrist support for below the laptop such as ( VicTsing Mouse Pad Keyboard Wrist Rest, Memory Foam Non Slip Gaming Mouse Pad, Durable, Lightweight &Comfortable for Pain Relief & Easy Typing Office Mouse Pad, Support for PC,Computer & Mac https://www.amazon.ca/dp/B0795CD8Q8/ref=cm_sw_r_cp_apa_i_odsADbQTDHBMM).

Sent from my LM-G710 using Tapatalk

 

Honestly never ran into this issue , I leave my machine on 24/7 usually only restart for updates and apps that require it , I don't undervolt.

I did have a weird issue where it rebooted on me out of nowhere a while ago i think the bios fixed that .

I do have an issue though When i have mpc-ht open and i have obs open and streaming gameplay , streaming and playing any game then exiting the pc game and exiting out of rust i get a error i think it's software related or nvidia driver related but other than that i run the laptop stock no overclocking or anything like that /

 

I do too, literally, but its nice before running some benchmarks or getting ready for a gaming session to restart the machine. I really hope my replacement does not do the same thing. I was told it was going to be a brand new machine since I just got my first unit only a few weeks ago. Does anyone know how to tell if lenovo is trying to slide a refurb off without telling us on the machine. The sad thing is the machine runs everything perfectly while its on. The cores do seem to get off center though temp wise, I think their is a bad paste job on this one as some cores on the CPU would hit in the 90s and others would be in the 70s.

 

I was also told my replacement was going to be brand new, and my replacement was done 4 days into owning the original.

Things you could use to tell:
1 - Do the screws on the bottom cover look excessively worn, like improper tools or technique were used on them or they were played around with more than once?
2 - In HWiNFO64, is the SMART uptime count for the hard drive or SSD excessively high? I'm thinking more than necessary for a company to do QA tests, like 12+ hours.
3 - In HWiNFO64, is the SMART total lifetime written GB for the SSD excessively high, like multiples of the stated capacity? (indicates somebody did a secure wipe after a return was made)
4 - If you do a repaste, is the paste is not the stock blue paste, or has dirt/dust that you won't expect from a clean-room assembly area?
5 - Does the Serial Number end in RF? (I suspect but cannot be certain about this one)

Interesting thing to note, I was speaking to a close friend today, who ordered a brand new Lenovo laptop 6 months ago and after a lengthy delay he received what he claims appeared to be a used laptop. Please take this with a grain of salt because you are reading this from an anonymous person posting a friend's story on the Internet.

 

Thanks for reply, interesting. If anyone with the Y740, specifically those who have the 15" with the 9750H and 2070, if you could do a test for me. If you could restart windows for awhile and in between a few of those use the shutdown, im talking the windows restart and shutdown from the startmenu. IF you could do that for 30 or 40 times and see if your machine hangs on any of those. My machine will randomly hang 10-20% of the time while restarting, i mean hardlock to where you cant do anything but hold the power button down. Oddly enough the keyboard lighting still functions, but pressing any key will do nothing. Every 10 or so times of restarting my machine it will hardlock like im trying to describe. Something is not right here, i need a reliable machine first and foremost. If I Cant even shut down windows jeez, their is problems. Im not trying to make a big deal out of nothing, its unacceptable from my stand point. I have half a dozen laptops sitting in my room that I have never had an issue with restarting windows thousands of times.

So yea, just do the standard windows restart and let it come back into windows. Give windows 30 to 50 seconds to get everything loaded after you have put in your PW if u use one and then restart again. If any one has any issues please reply back here so we can compare.

PS. on a side note I was going to deal with it as is, as like I mentioned ealier I rarely shut down my laptops but it did end up hardlocking once on me while the screen saver was in use sitting on the desktop which was the last straw for me.

 

Try one of these: Keyboard wrist rest.
It could help and it’s fairly cheap, don’t need one myself but it could help pushed up against the front of the laptop.
They have different brands and sizes available

https://www.amazon.com/dp/B07RRW4QV...jbGlja1JlZGlyZWN0JmRvTm90TG9nQ2xpY2s9dHJ1ZQ==

As a bonus it can be used as a pillow in emergencies if you get sleepy at your desk

 

Idle temps 36-38 on CPU and GPU. And 3 hour Warframe run with Max temp of 68c on the GPU and 71c on the CPU. Ahhh yeauuhhh

 

How the hell did you hit the silicon lottery there?!

Sent from my LM-G710 using Tapatalk

 

You have no clue the hell I have been through getting the heatsink flat. Lol LM played a big part in that too LOL

 

Those temps are insane! How do you go about flattening out the heatsink?

 

You get the pads off of the heatsink. And back on the appropriate places on the Mobo.
I highly highly highly advise using .5mm pads where there are 1mm thick pads. Clean the possible large amounts of thermal paste that is staring you in the face all like "woohoo ! I'm a whole tube of paste !!" From the heatsink and GPU and CPU on the Mobo.
Then the next part is cumbersome and a pain. ** Read all of post before attempting !!*DISCLAIMER ****
I am NOT responsible for what you are doing, going to do, possibly do and or thinking about doing in the below steps. By agreeing that I ( ME) am not responsible for any implications-faults-fires-explosions-deaths of mobo, it's surround components ; That you are an adult or someone who posses the ability to understand flatness , the basic understanding of scew removal, cleaning, electrical components etc and liable for your own actions here forth ! Continue below
Ok - Find a razor blade . Or something very thin, but with an edge that you can gently lay across the Copper portions of the heat sink. While doing this peep out between the blade and copper. Lifting it up towards light to see if there are any large gaps between the blade and copper. Gently move the the flat portion of the blade around checking different areas of the copper portions of the Heatsink.
Once you have verified deformations and or gaps. Look for your vise, press or c-clamp- having spare CPU IHS's from desktop CPUs work great for this next part. Or if you don't have access to the spare IHS's find something metal but Flat, also thick enough to not bend under pressure- (you will need 2 pieces) - for the love of all things holy- make sure it's flat !!! and sized small enough to be able to safely lay across the CPU cooler portion of the heatsink and the other piece of flat metal can lay on top of the opposite side of the heatsink. Palllease use common sense while doing this. If your flat pieces of metal is big enough to cover the top side of the heatsink.
Make sure you are clear from CRUSHING the heatpipes. You will need to position each piece of metal and heat sink sandwich in the vise, press or c-clamp-. Very slowly compress this sandwich.
DO NOT and I repeat DO NOT get all heavy on the pressure YOU have been warned!
Apply pressure while watching all corners, heatpipes , and baseplate of the heatsink.
What you are doing with this is pushing the copper plate back into the base plate in a Flattened manner. And will possibly correct the warpage of the base plate as well. Out of the 3 I have done . The CPU is the only portion that has been warped. Dropping the thermal pad sizes to the sizes listed above had been all I did to get the GPU temps down .
The CPU portion has been warped on all three I have came across so far.
After you've gotten what you can hope is flat- reapply your favorite thermalpaste.
***** ALSO Note !! There should be some black tape on the Cooper plates. Remove that as well. **** Before attempting to flatten!!!! Removing the tape will more than likely show you the imperfections that they have hidden under it.

 

Anyone try slamming 64GB of RAM into y740?

I know the psref states 32gb, just curious to see if anyone has done it.

 

Smart idea with the razer blade, and thank you for taking the time to write all this down.

I'll save this for when I finally scrounge up enough to get one

 

Not a bad question to ask... especially since the 9750h variant supports XMP Profiles, which I haven’t heard anyone mention. (Mine supports the 2666mhz profile on Ballistix Sport that everyone was originally telling people to avoid).

I would test it, but I don’t have 64gb of laptop ram sitting around...

 

I have the crucial kit as well, but hmm don't see any 64gb kits 2 x 32 at 2666. Unless you buy the two sticks at 32gb each but then you lose dual channel, cas is like 19 etc.

 

Hello everyone, new guy here.

I guess this topic has been gone through many times on this forum already. BUT... Ive been having very strange behavior with my unit (i7 8th gen. 2070 maxq).

So I was sticking with the 1.06 bios that has the manual controls over the fans. Temps in BF V was about GPU 76C (around 1600Mhz@ 58% usage) and CPU 89C with 3,7Ghz and 45W (70% usage) on Ultra settings.
I read somewhere, that bios 1.10 actually had no effect on cooling and actually had better performance, so I took a change on it. And temps went up to 94C. And on top of that, OC'ing the GPU crashed the system everytime in games (BFV and battlefront 2 namely).
So naturally I went back on the bios, this time to 1.08 as this seems to be very popular.
Yesterday I tried some BFV, (without OC) and got following results:
GPU 82C, 93 % usage @ 1440Mhz
CPU 94C, 76% usage @ 3,5 Ghz and 44,5W.
And those were all settings LOW?!?

Getting freaked out a bit what went wrong, I have UV -144,5mv on core and -135,7 on cache and have had around the same UV since June.
Is there something i could have messed up? Should i go back to 1.06 bios or is there actually that big of a difference. That GPU is the strange bit here, as its on more load on low as it is on ultra settings.. In battlefront 2 its always around 90-96% also, no matter Im a playing on low or ultra, or even low with resolution scale in 133% (looks so good thou..)

BTW has someone done liquid metal? Im getting real frustrated, Lenovo support just told me to run some diagnostic tool in Vantage software and the basic "Those clocks are within the usual limits of the cpu" (but in Finish).

Thanks in an advance and sorry about the mistyping!

 

Paste and thermal pad degradation

I've tried 2 sets of the hyper-x 3200mhz . Absolutely trashed the system when the xmp profiles were set. Those sticks have two xmp profiles a 3200mhz and a 2986 or something. It was rock stable at the tighter 2900mhz xmp profile. The 3200hz was a crash fest.

 

With all the reported hot CPU temps does a notebook cooler help at all?
Also if someone who hade the older screen vs the new 500 nit screen can compare , I know ones brighter but does the new screen have better color reproduction
Thanks

 

...after only a couple of months? Sounds more like they're having the same problem as many/most of us, temps being too high from the start and getting even worse after bios updates (82C on the GPU, yikes) because the fans are slowed down for whatever reason. Bad paste/pads/etc could be the entire problem but I can't imagine they'd wear out that quickly.

Greatly reducing the boost wattage limits for the CPU is the only thing that worked on mine, the high GPU temps may just be because the CPU is sitting at 94 while both are under load (shared heatpipe and all). Though if I don't end up returning mine this week, I guess I also need to repaste at some point.

 

If there is so much paste used that its insulating and baking around the CPU die, and the pads aren't big enough (half the size of the chip they are trying to cover) and if the pads are the wrong thickness to begin with. Lol id fancy to guess if they are all not being used correctly to begin with, degradation would be an understatement. Lol

 

Almost feel like this is a disagreement over the definition of "degradation"... if the pads and such aren't doing their job then, if anything, they should wear out more slowly, because less heat is passing through them. More to the point, they said the temps increased after a bios update, as many other people have reported. Though none of this rules out the possibility that the heat-dissipation system is badly constructed in one or more ways, making the fan speed a much bigger problem than it needs to be.

 

Thanks for all your replies! I feel like this is an odd one here. With bios 1.06 I didnt have that much of a problem with over heating, went to 1.10 where the problem started and went back to the bios 1.08 that SHOULD have same thermals as 1.06 with the max fan setting.... But i dont. So thats the odd thing in my mind. And yes the unit was bought in june so only couple of months of use...

I did limit the cpu boost tdp from 65w to 50w, and that helped some. Still hitting over 90C thou. On a closer look, there seems to be like 1 core that is hitting over 90C and the rest stay around 85c. GPU still at 80C and over too. And this in games like battlefield V and battlefront 2, quite demanding games...

Perhaps i need to take a look at the thermal design and do a repaste with the smaller thermal pads like kfxsti suggested. What do you kfxsti mean with that the pads arent big enough? That the copper plate isnt covering the whole die?? Is there something that can be done to that?

Do you guys know if this voids the warranty? I should suspect that in principle yes but in practice not so much.

EDIT:

So... Was about to repaste the system. I could actually see that some of that blue stuff all over and some thermal pads were peeking under there. Got 5/6 screws out, but one is stuck. And because those are soft (like Bob warned...), the screw just got round.

I think im stuck at this point unless I find some replacement screws so that I can force out the one that's round and stuck. For the time being, Ill have to do with the 3,4 ghz so the temps dont go high in more demanding games...

Thanks for Your time anyways!

 

When they are less than half the size they should be in surface area, and too thick allowing the pad to bake. That would be degradation. When the smaller surface area can't dissipate the heat due to its extra thickness , the side laying on the surface creating the heat bakes that surface of the pad.
Check and see how slimy or "wet" it is under your thermalpads on the Mobo side of the pads not the heatsink side.

 

If you were to repaste and redo the thermal pads what size pads and how many of each would you need and any brand recommendations? Using LM did you creat a “barrier” around the CPU so it doesn’t seep out?
Besides LM what thermal paste would you recommend? Little scared using LM since I never worked with it before so I would like to use a traditional TM.
And about pads any other recommendations of where to add them to help with additional cooling?
Thanks

 

Wait, back up, aren't the thermal pads on various extra motherboard chips, not the CPU/GPU? Things that don't get nearly as hot? Wouldn't think that would be all that important unless that extra 0.5 mm you were talking about lifts the heatsink off the main components. Surely the pads are soft enough for that not to happen?

I don't know if you tried just repasting without the whole arduous heatsink-flattening process, but I'd think that would be enough for most people...? If not, I should probably return mine, that's a little too DIY for me with something this expensive.

 

I ended up returning my 1660ti model due to the G-Sync thing and ordered a 2060 model which arrived today. Unfortunately it has a keyboard rattle when typing on the top row. Anyone experience something similar and is there a fix?

 

The pads are on vram chips- vrms, voltage regs etc surrounding and or above the CPU and GPU. And yes, they do get that hot. Also, the .5 is enough for to much gap. They may be soft yes... But your talking about a very very low mounting pressure with the way the heatsink is mounted and designed. It's only able to compress but so much.
And yes, repasting alone was tried. Results were just as hideous as the factory job. Very possible cause that the vids out there with people who've repasted and ended up with worse results or the same results were due to being clumsy with an already not so very rigid heatsink-a warped heatsink- and or wonky sized thermal pads. The heatsink as a whole is flimsy, and 3 out 3 y740 Heatsinks were warped where the heatpipes leave the cold plate. A whole corner of the cpu was nearly untouched by the compression of the thermal paste. Warped Heatsinks are not an uncommon thing with laptops - yes the laptops may be expensive. But they are made about like everything else nowdays. Use it until ~ toss it, buy another.
I'm not saying that this is a common problem on the legion y740. But it has been the problem for the three I've received.
I have only provided info on the temp problems- if you want to attempt any of it or not is on you. But I have enough experience with laptop cooling , heatsink designs and modding to know what works and what doesn't. I have an entire work bench dedicated to heatsink shimming, correcting and soldering lol.

 

A question to the owners of the Y740 "17"
Can you recommend it?
I have seen some Videos of YT, i think its a good Notebook. I read in the reviews that the fan noise and the temperature lower than another Gaming Notebooks.
Lower Temperature is i hope better for the lifetime.
Is the Quality from the Notebook good?

 

So I had a weird gaming session yesterday of the new Ghost Recon Breakpoint and The Division 2. I forgot to change the performance profile and was playing. Went and checked Throttlestop after cuz I had noticed that it wasn't throttling and fans were at maximum so that's good; temperatures peaked at 73C. Went and checked Lenovo Vantage and it was on Quiet mode. Sure, I know it runs the GPU at 80W instead of 90W but that fan profile and those temperatures? I think I would rather have that.

Sent from my SM-G970W using Tapatalk

 

What is the difference in frame rate between Quiet mode and Performance mode?
Was that CPU temp?

Can you try something for me?
Can you run that game in all profiles ( Quiet, Balanced and Performance ) and report back the difference in CPU and GPU temps also the difference in frame rates?
Thanks

 

I had previously played on performance mode solely, and temperatures were maxing at 94 and tripping PROCHOT. Framerates were great at about 75-80 on Division 2 but obviously throttling. I didn't notice much of a fps difference so it was certainly above 60 but I didn't have RivaTuner running. I will check tonight but temperatures in the Division 2 were at a maximum of 73C on the CPU after gaming on both those games. Again, I'll check Framerates tonight and report back but so far, I like those temperatures for sure lol

Sent from my SM-G970W using Tapatalk

 

That's definitely interesting. I haven't really messed with the other modes either, I keep mine on performance also to get the GPU pulling 90W. I've been tweaking with TS and MSI afterburner for a long while now, trying to get the best performance possible with the lowest temps possible.

Yesterday I tried keeping my 150mv Undervolt but also downclocking all CPU cores to 3.5Ghz and staying on Performance mode. I also used MSI Afterburner to do an OC + UV, +200 core OC and Undervolted 1,785Mhz @ 793mv. This allows a +200 core OC at the lower voltages when GPU load is very high, and then maxes the clock speed at 1,785Mhz at 793mv when GPU load is low to medium. I tested the Division 2 benchmark on High, and with the CPU at 4.0Ghz and a normal +150/+400 OC, I get around 100 fps. But max temperatures with the CPU at 4.0Ghz get into the mid to high 80s. With the CPU Underclocked to 3.5Ghz and the OC + UV applied to the GPU, my temperatures on Performance mode hit a MAX of 80c, I haven't seen anything higher. And on the Division 2 benchmark on high, I got 101 fps yesterday. So performance is exactly the same while temperatures are much lower. Definitely something to try if you want lower temps and the same performance. I tested Apex, Destiny 2, and Overwatch with these same changes and I got the same results: Lower overall CPU temps with almost the same performance. I can upload pics of my voltage/frequency curve if anyone would like to see how I did the OC + UV. The GPU also has lower temps by about 5c, leading to lower CPU temps and allowing the GPU clock speed to stay much more consistent.

I'm on the 15in 9750H/2070MQ model btw.

 

Would love to see pics and maybe some sort of guide on how to do it.

 

Sure thing, I'll post a guide on how I did it with some pics of my voltage/frequency profiles later on today when I get home. I have noticed even though average temps are much better, I don't think there's a way to avoid the CPU hitting around 80c due to how the fan profiles work on the current Bios. The fans often don't ramp up until the CPU hits a certain point it seems. But with the CPU UV + UC and GPU OC +UV, average temps are great and performance remains about the same, so it's a win-win. I still hope they update the Bios at some point to provide manual fan profiles or control of some kind.

 

Getting mine any day now and I’m excited to see what the new 500 nit HDR screen is all about with the same specs as yours!
What is a stable UV you use on the CPU? ThrottleStop of XTU?

 

Alright, so I played around with the Division and realized I was only getting about 80 fps because I forgot to enable DX12... Oops! So I now on to the comparison. Simply put, I can OC and play with no problems on Auto and Quiet, but on Performance, there was continuous freezing on startup or close when Riva Tuner was running (I had to start it after I started the game).
Playing on Ultra but some things turned down that don't matter much:
Suffice it to say, auto with no OC would be ideal it seems. Quiet it still above 80 fps and very nice temperatures. Performance is nearly unusable right now. Funny enough, Performance mode acts more like a quiet mode in that the fans don't even ramp up until it's too late. So I will be sticking to Auto or Quiet from now on.

 

How loud are the fan noises under maximum power?

If the keyboard hot under maximum power?