My month old laptop is performing a lot worse than the one I got rid of

Page 3 - Seeking answers? Join the Tom's Guide community: where nearly two million members share solutions and discuss the latest tech.

Tom Dyal

Estimable
Jan 25, 2015
16
0
4,560
I switched from the Lenovo Y410p to the Lenovo Y50 after an unfortunate accident. I had extra money so I decided to get an upgrade. But when I try to play games on the new Y50, it performs a lot worse than the old one. Such as game freezing, entire system freezing, low fps at times in game, and weak wifi connection (I've had to restart at least 6 times today because the entire system froze). Some people say the fps problem was from the resolution change (the Y50 has a bigger screen) but when I change it, it really does nothing. As for the constant freezing, it is a new laptop so overheating from dust blockage shouldn't be that big of an issue. This never used to happen with my old one. Can anyone help, or did I just make a bad choice for my new laptop?
 
Solution
Screen size has nothing to do with resolution. Any size screen can run higher or lower resolutions, up to whatever the supported maximum is. If the older unit had a 720p resolution and the newer one has a 1080p resolution, then yes, you might take a hit on FPS. That doesn't account for any of the other issues though.

The Y50 is known to have overheating issues during intense extended gaming sessions, so it could be related to that, but doesn't necessarily have to be.

I'd start with doing a clean install of the GPU drivers, by which I DON'T simply mean installing the latest ones over older drivers or just uninstalling and re-installing the drivers. Following the steps in the CLEAN driver installation section at the following link...
Oh and now in game my cursor is randomly stopping and turning from the game cursor to the desktop cursor and I have to open task manager then re open the game to get it back to the game cursor and allow me to continue playing 🙁
 
Since my very first instinct was a heat issue, considering the Y50 has poor cooling and is known for overheating, and you have a warped case, and you have a 60°C core temp with 0% core activity and 0% GPU load, I'd say it's getting or has previously been, seriously hot. At 60°C with no load, and no activity, I'm quite sure this unit is going over 100°C when gaming. What is the ambient temp in the room normally?

Maybe run a game for ten minutes or so with HWinfo open and then take a screenshot of the temps again. I have a bad feeling there will be a pretty high thermal response.

Do you still hear the fans kick on high when you start gaming or running intensive processes?
 
Sorry, I'm going to be at school for the next week. I was just able to respond through the day because I was off on break. ( Don't want you to think i'm ignoring you ) But anyways yeah that was with the game on and my fans kick on when I run games.
 
Hmm. Very strange that there was 0% CPU or GPU usage with a game running. Something is definitely suspicious here. If those temps were under load I'd say it's probably ok but since I don't see any loads, dunno. Try screenshotting the HWinfo sensors while running Furmark after about five minutes. Leave Furmark running when you take the shot.

You might also run one of the cpu stress tests and take a second set of screenshots. Have you at ANY point in the past installed or used one of the CPU or GPU overclocking utilities?
 
Well I ran Furmark and here are the results :

fcqxw9.png


10if4eb.png


23hwpjo.png

 
Considering only one core is at 100% and it's already showing 80°C on the CPU temp, plus the GPU is at 79°C, I'd say it's not unlikely that the unit is getting too hot. Furmark only uses a single thread to run but many games will use two, four or more cores, although most are single threaded still, so temps when the system is using two cores at above 80% are likely to even be higher than the threshold you're already at. It's not exceptional heat right now, but it might have been or could be. I don't know. I wouldn't condemn it entirely but I'd say it's getting near the badlands.


What is the average ambient temperature of the room the unit is in?
 
Would this be more common on an older computer? Or is this normal for the Y50? If it is normal have I just made a bad choice for my new laptop ? Or if its not do you think I should send it back ?
 
So, for your CPU configuration it seems that 100°C is the Tjunction or TJmax, at which the system begins throttling the CPU to prevent overheating and when the the system reaches ThermalTrip, which is about 20-25° hotter than that, it will shut down. Clearly you do not want to reach even TJmax, as that means the system will decrease the CPU multiplier and voltage to reduce temperatures, thereby also reducing heat. At 80°C with 100% load on a single core, it might not be too bad. I'd like to see what it's doing with all cores on 100% load, just so I can positively tell you that it looks ok to me.


Please, if you would bear with me once more, download Prime95 version 26.6 here:

http://windows-downloads-center.blogspot.com/2011/04/prime95-266.html

Open the HWinfo sensors window as before. Run Prime95 and select the Small FFT option. Run the program for 10 minutes with HWinfo open then take a screenshot of the CPU temps section of HWinfo. Exit Prime95 and post the results here. Your GPU temps were within specs.
 
I'd save that image in case you decide to RMA the unit. There is a 10°C core temp difference between Core 0 and Core 2 on the max temps. That is abnormal. There is either an issue with incorrect mounting pressure of the CPU heatsink assembly, the thermal past was incorrectly applied or there is an internal issue with the CPU. It's not TREMENDOUSLY bad, but core variance should never be more than 5°C difference at worst. At best they should be the same, or nearly the same.

One last time, just to be sure, can you make the HWinfo window bigger by dragging the bottom border to stretch the sensors window and dragging it as high up the page as possible. I'd like to see the core load usage along with the temps to verify that the core temp variance isn't due to one of the cores being throttled or not being at 100% along with the others. Then run it again for about 12 minutes and take the screenshot. Right after you start Prime95, click the button at the bottom of the HWinfo sensors window that says reset and looks like a time clock, then wait twelve minutes and take the screenshot. I promise it's the last time I'll make you run this. Heh.