Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 ... 284 285 [286] 287 288 ... 350

Author Topic: The Generic Computer Advice Thread  (Read 572816 times)

wierd

  • Bay Watcher
  • I like to eat small children.
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4275 on: September 08, 2020, 12:32:52 am »

The key rhetoric was "As long as you are not violating MS's copyrights".

The creation of an alternative framework to satisfy other parts of the OS, (If created under appropriate restrictions as codified in US copyright law), is not illegal.  The deployment of those alternative frameworks are also, not illegal, and the software vendor cannot forbid you from doing so.

That is the source of the grey area.  It *REQUIRES* that the end user is on the up-and-up though.  You have legitimately entered a license agreement with microsoft, and have paid for your license.  The only issue I see, is that circumvention of the authorization framework constitutes a circumvention of a copyright protection system, and falls victim to "Not following copyright law". 

Since this is handled with scripts, and presumably they are well documented, it should be possible to prevent the neutering of the authorization framework.  Doing so will result in a build that would be legally compliant, assuming all other conditions are met.


RE: shrinkwrap licensing--

Often times, you are presented with the agreement upon installation of the software, not upon purchase.  This still constitutes a shrinkwrap license.  To qualify outside that, the software publisher must present the license agreement BEFORE a purchase transaction, so that the end user is able to reasonably decide not to complete the transaction.
« Last Edit: September 08, 2020, 12:34:44 am by wierd »
Logged

methylatedspirit

  • Bay Watcher
  • it/its
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4276 on: September 08, 2020, 09:30:25 am »

There's this thing called the Windows 10 Debloater, which sets out to do the same thing as that one. It operates on an existing install of Windows 10, rather than modifying an ISO of Windows 10, so I'm fairly certain this is perfectly legal, unless doing whatever you want with your own system is somehow illegal. Of course, Windows Update will probably add all the bloatware back in after updating, but there's no limit to how many times you can run it, so, really, who's the winner here? (Do check the blacklist in the GUI, though; it uninstalls pretty much every (UWP) app on your system by default, including the Windows Store.)
Logged

bloop_bleep

  • Bay Watcher
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4277 on: October 11, 2020, 10:04:54 pm »

Hey guys. So my computer (laptop) recently ran out of battery and then refused to turn on when plugged in, with or without the battery attached. Also it does not shine the LED when plugged in. This led me to believe that it may be a power jack issue, as in fact the power jack was replaced in the past. However I opened it up and poked around with a multimeter, and found a couple things. Firstly, when plugged in there is a +20V potential at the power jack output pin relative to ground, so that individual piece seems to be working. Secondly, when not plugged in, there is a 50 kilo ohm resistance between the power jack output pin and ground. I have not YET noticed any shoddy soldering joints. Question: are these observations normal and what do I do from here?

EDIT: Also should there be continuity between the power jack and the positive battery terminal?
« Last Edit: October 11, 2020, 10:23:17 pm by bloop_bleep »
Logged
Quote from: KittyTac
The closest thing Bay12 has to a flamewar is an argument over philosophy that slowly transitioned to an argument about quantum mechanics.
Quote from: thefriendlyhacker
The trick is to only make predictions semi-seriously.  That way, I don't have a 98% failure rate. I have a 98% sarcasm rate.

Maximum Spin

  • Bay Watcher
  • [OPPOSED_TO_LIFE] [GOES_TO_ELEVEN]
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4278 on: October 12, 2020, 01:57:27 am »

Hey guys. So my computer (laptop) recently ran out of battery and then refused to turn on when plugged in, with or without the battery attached. Also it does not shine the LED when plugged in. This led me to believe that it may be a power jack issue, as in fact the power jack was replaced in the past. However I opened it up and poked around with a multimeter, and found a couple things. Firstly, when plugged in there is a +20V potential at the power jack output pin relative to ground, so that individual piece seems to be working. Secondly, when not plugged in, there is a 50 kilo ohm resistance between the power jack output pin and ground. I have not YET noticed any shoddy soldering joints. Question: are these observations normal and what do I do from here?
I admit this is not my greatest area of expertise but these both sound normal to me. As a general rule the power jack output should not be grounded (this would be what we call "counterproductive"), so having a resistance there would be expected. It sounds like your adapter block is working fine, and the problem is within the laptop (where there are many possible problems).
Logged

Lord Shonus

  • Bay Watcher
  • Angle of Death
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4279 on: October 12, 2020, 02:06:21 am »

Have you tried leaving it plugged in overnight? It might rely on battery for the initial power-up (which often requires extra power compared to low-level running, and not be able to draw what it needs from the charger for various reasons. Some laptops I've had worked like that.
Logged
On Giant In the Playground and Something Awful I am Gnoman.
Man, ninja'd by a potentially inebriated Lord Shonus. I was gonna say to burn it.

Maximum Spin

  • Bay Watcher
  • [OPPOSED_TO_LIFE] [GOES_TO_ELEVEN]
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4280 on: October 12, 2020, 02:10:01 am »

Have you tried leaving it plugged in overnight? It might rely on battery for the initial power-up (which often requires extra power compared to low-level running, and not be able to draw what it needs from the charger for various reasons. Some laptops I've had worked like that.
Have you ever had a laptop that doesn't light up the charging LED in those circumstances? I agree that it's a good thing to try but I've always seen the LED on in such cases.
Logged

wierd

  • Bay Watcher
  • I like to eat small children.
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4281 on: October 12, 2020, 02:13:51 am »

Since you have it apart to test with a multimeter, I would disconnect the battery, plug in the power cable, then test to see if there is a sense voltage across the positive and negative battery terminal leads off the mainboard.  (To charge a LION cell, the charge logic inside has to determine that there is a resistive load to know that the battery cells are connected. Additionally, testing for weak voltage against the SPI data pin and ground would let you know that the charge circuit is attempting to communicate with the battery's charge controller.)

That would let you know if the problem is with a voltage regulator or power trace upstream, or if the problem is inside the battery, or whatever.
Logged

BigD145

  • Bay Watcher
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4282 on: October 12, 2020, 09:23:37 am »

Potentially the battery itself has a logic board on it that has failed. If the power light is not turning on it could be the adapter but also could be a fault on the motherboard. A capacitor may have popped or something like that.
Logged

bloop_bleep

  • Bay Watcher
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4283 on: October 19, 2020, 12:33:40 pm »

Well, I couldn't find the problem, the laptop was kinda old and breaking apart anyway, and a repair technician on YouTube said that if things go wrong with the charging port on a Lenovo model similar to the one I had there's an electrical design mistake that could blow the super IO chip that does the startup sequence, which is subsequently difficult to replace as it is soldered to the board, so... I got a new laptop! A faster one too, I think. However, I've noticed even on small loads it can get somewhat warm on the bottom, and Core Temp tells me it can reach temperatures of 86 C, even though most of the time it stays below that. The CPU is rated at a maximum temperature of 105 C. The fan sounds like it is working but there is a quiet sort of grinding noise. Is this something I should worry about?
Logged
Quote from: KittyTac
The closest thing Bay12 has to a flamewar is an argument over philosophy that slowly transitioned to an argument about quantum mechanics.
Quote from: thefriendlyhacker
The trick is to only make predictions semi-seriously.  That way, I don't have a 98% failure rate. I have a 98% sarcasm rate.

wierd

  • Bay Watcher
  • I like to eat small children.
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4284 on: October 20, 2020, 05:30:01 am »

New or used laptop?

If used, check for an invasion of dust bunnies in the radiator and heatpipe assembly. (and replace that thermal pad with real thermal compound)



If new...  Hmm...  Check online to see if other people have the same problem.
Logged

bloop_bleep

  • Bay Watcher
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4285 on: October 20, 2020, 08:12:24 pm »

It's new, and yep, people online say they have similar problems with thermals. *shrug* It's an Acer too so apparently the BIOS manual fan control settings are locked out. Elsewhere on the interwebs it says 80 C might cause accelerated wear but shouldn't damage the CPU in the short term. It doesn't get that high too often anyway. People have reported 90 C temperatures under significant load.

By the way, I did your suggestions for the old laptop. I might still fix it at some point. Thanks a lot!
Logged
Quote from: KittyTac
The closest thing Bay12 has to a flamewar is an argument over philosophy that slowly transitioned to an argument about quantum mechanics.
Quote from: thefriendlyhacker
The trick is to only make predictions semi-seriously.  That way, I don't have a 98% failure rate. I have a 98% sarcasm rate.

methylatedspirit

  • Bay Watcher
  • it/its
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4286 on: October 20, 2020, 09:06:01 pm »

There's 2 ways of dealing with high temps on a laptop, and they can be combined if you're feeling lucky. Do the first one first; it's entirely reversible, unlike the second one.

The first way is to undervolt the CPU, literally feeding it less voltage than stock so it produces less heat. What I use personally is ThrottleStop; here's a guide on how to use it. Of course, you run the risk of system instability while undervolting, so I recommend doing a stress test on the CPU after you've undervolted it. If it crashes or spits out an error during that process, you've gone too low with the voltage, and you need to raise it by 5 mV. A round of Cinebench R20 is good enough for establishing baseline stability, but what you really want is something like Prime95 on the Blend torture test for at least 24 hours straight once you've dialed in a "stable" undervolt. I'd suggest even longer; I'd go for a whole week if you can.

On my laptop, I started with -100 mV offset on Core and Cache voltages, then I worked my way down to -160 mV, using Cinebench to establish that each successive undervolt was vaguely stable. Thankfully, I run Folding@home, so that's what I used as a stress test. I left my computer on overnight, letting it crunch numbers until it crashed. After a series of crashes due to the undervolt, I eventually converged on an undervolt of -140 mV, which appears to be stable enough to survive an entire week straight.

I wouldn't recommend starting ThrottleStop on startup, though; what I do is place it on the taskbar so that you can open it manually. This is to avoid bootloops if you end up with too low of a voltage setting and end up crashing because of that. If opening TS crashes your laptop, you can delete the "ThrottleStop.ini" file in the TS directory to reset everything back to defaults and try again.

I don't think this voids your warranty; worst-case, you can delete the ThrottleStop directory, and no-one's the wiser.

Please check if your laptop has an Intel or an AMD CPU inside; ThrottleStop works only with Intel CPUs, and to my knowledge, AMD (Ryzen and earlier) has no tools for undervolting laptops.



The second method is to repaste, replacing the thermal paste between the CPU die and heatsink to increase the rate of heat transfer from the CPU to the heatsink. This is a difficult process, and usually requires disassembly down to the motherboard for a standard laptop. Gaming laptops tend to be easier in this regard. This will almost certainly void your warranty, so do this at your own risk.

The theory is fairly easy once you've disassembled the laptop far enough to see the heatsink assembly:
1. Locate the heatsink.
2. Unscrew the screws holding it to the motherboard and take it off.
3. Clean off the stock thermal paste from the heatsink and CPU die with alcohol (isopropyl alcohol seems to be the stuff of choice, but ethanol works too; I used hand sanitizer myself)
4. Apply thermal paste to the CPU die.
5. Mount the heatsink back on to the board, then screw it back in.

Of course, this is a daunting task for most people. I was positively shaking when I replaced the thermal paste on my laptop. If you do this on a new laptop, prepare yourself for disappointment (at least in my case); you won't see much difference on your thermals, but depending on the thermal paste, it might hold up better over time.

As far as choice of thermal paste, you can't really go wrong with a well-known brand like Arctic, Thermal Grizzly, Cooler Master and so on. If it appears on a list of best thermal pastes, it's probably good enough. Be careful though, you want a thermal paste intended for air cooling, and you sure as hell do not want a liquid metal thermal solution. Without proper precautions, that leads to short-circuits, and can degrade the copper/aluminum on the heatsink itself.  I used Arctic MX-4 thermal paste on mine, since it claims that it'll last for 7 years, and that was what I had at the time.
Logged

wierd

  • Bay Watcher
  • I like to eat small children.
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4287 on: October 20, 2020, 11:05:57 pm »

'Void warranty"

NOPEY NOPE!!

The FTC slapped those manufacturers *HARD* for that!
https://www.bbc.com/news/technology-43724348


If the service is performed by a properly certified technician, etc-- they cannot ignore the warranty claim. :) 
Logged

wierd

  • Bay Watcher
  • I like to eat small children.
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4288 on: October 20, 2020, 11:11:23 pm »

There's 2 ways of dealing with high temps on a laptop, and they can be combined if you're feeling lucky. Do the first one first; it's entirely reversible, unlike the second one.

The first way is to undervolt the CPU, literally feeding it less voltage than stock so it produces less heat. What I use personally is ThrottleStop; here's a guide on how to use it. Of course, you run the risk of system instability while undervolting, so I recommend doing a stress test on the CPU after you've undervolted it. If it crashes or spits out an error during that process, you've gone too low with the voltage, and you need to raise it by 5 mV. A round of Cinebench R20 is good enough for establishing baseline stability, but what you really want is something like Prime95 on the Blend torture test for at least 24 hours straight once you've dialed in a "stable" undervolt. I'd suggest even longer; I'd go for a whole week if you can.

On my laptop, I started with -100 mV offset on Core and Cache voltages, then I worked my way down to -160 mV, using Cinebench to establish that each successive undervolt was vaguely stable. Thankfully, I run Folding@home, so that's what I used as a stress test. I left my computer on overnight, letting it crunch numbers until it crashed. After a series of crashes due to the undervolt, I eventually converged on an undervolt of -140 mV, which appears to be stable enough to survive an entire week straight.

I wouldn't recommend starting ThrottleStop on startup, though; what I do is place it on the taskbar so that you can open it manually. This is to avoid bootloops if you end up with too low of a voltage setting and end up crashing because of that. If opening TS crashes your laptop, you can delete the "ThrottleStop.ini" file in the TS directory to reset everything back to defaults and try again.

I don't think this voids your warranty; worst-case, you can delete the ThrottleStop directory, and no-one's the wiser.

Please check if your laptop has an Intel or an AMD CPU inside; ThrottleStop works only with Intel CPUs, and to my knowledge, AMD (Ryzen and earlier) has no tools for undervolting laptops.



The second method is to repaste, replacing the thermal paste between the CPU die and heatsink to increase the rate of heat transfer from the CPU to the heatsink. This is a difficult process, and usually requires disassembly down to the motherboard for a standard laptop. Gaming laptops tend to be easier in this regard. This will almost certainly void your warranty, so do this at your own risk.

The theory is fairly easy once you've disassembled the laptop far enough to see the heatsink assembly:
1. Locate the heatsink.
2. Unscrew the screws holding it to the motherboard and take it off.
3. Clean off the stock thermal paste from the heatsink and CPU die with alcohol (isopropyl alcohol seems to be the stuff of choice, but ethanol works too; I used hand sanitizer myself)
4. Apply thermal paste to the CPU die.
5. Mount the heatsink back on to the board, then screw it back in.

Of course, this is a daunting task for most people. I was positively shaking when I replaced the thermal paste on my laptop. If you do this on a new laptop, prepare yourself for disappointment (at least in my case); you won't see much difference on your thermals, but depending on the thermal paste, it might hold up better over time.

As far as choice of thermal paste, you can't really go wrong with a well-known brand like Arctic, Thermal Grizzly, Cooler Master and so on. If it appears on a list of best thermal pastes, it's probably good enough. Be careful though, you want a thermal paste intended for air cooling, and you sure as hell do not want a liquid metal thermal solution. Without proper precautions, that leads to short-circuits, and can degrade the copper/aluminum on the heatsink itself.  I used Arctic MX-4 thermal paste on mine, since it claims that it'll last for 7 years, and that was what I had at the time.

I always use a good metal bearing paste, like arctic silver-- but not one of those liquid metal solutions. Sure, it will wet the surface real good, but it also deeply penetrates and dissolves the metal of both the CPU die capsule AND the CPU cooler.  Such conditions can also induce the formation of alloy fingers out the sides of the junction.  I can't really imagine why anyone would really do this. 

Logged

methylatedspirit

  • Bay Watcher
  • it/its
    • View Profile
Re: The Generic Computer Advice Thread
« Reply #4289 on: October 20, 2020, 11:33:10 pm »

I always use a good metal bearing paste, like arctic silver-- but not one of those liquid metal solutions. Sure, it will wet the surface real good, but it also deeply penetrates and dissolves the metal of both the CPU die capsule AND the CPU cooler.  Such conditions can also induce the formation of alloy fingers out the sides of the junction.  I can't really imagine why anyone would really do this.

Overclockers, man. Extreme overclockers. Extreme competitive overclockers. Those people will literally pour liquid-fucking-helium on CPUs if it means they can get a hundred more MHz over the world record. They will take literally anything (up to, including, and beyond removing the IHS, then replacing the thermal paste with liquid metal) if it means they can say they're number one. They're absolutely bonkers, but I do admire their work.

If you see impressively-high world records for benchmarks that no sane person could ever hope to match, it was probably an extreme overclocker.

'Void warranty"

NOPEY NOPE!!

The FTC slapped those manufacturers *HARD* for that!
https://www.bbc.com/news/technology-43724348


If the service is performed by a properly certified technician, etc-- they cannot ignore the warranty claim. :) 

Depends on the country. I don't live in the US, and no-one's challenged the manufacturers in my country about this in court. As such, I don't know how enforceable the "warranty if void" sticker is. I tried searching that, and there's nothing that would say so. Lenovo's nice, though; my (gaming) laptop has zero such stickers anywhere. Not on the case, not on the heatsink assembly, nothing.
Logged
Pages: 1 ... 284 285 [286] 287 288 ... 350