No they rarely are the same. Early DDR3 era (2008-2010) both video cards and system memory where technically DDR3 but that changed around 2011 (GDDR5 took over for video cards and DDR3 stayed mainstream on desktops for another 3 or 4 years before becoming DDR4)
Hey guys, I’m having what I’m pretty sure is a hardware issue and I wanted to run my reasoning past somebody as a sanity-check before I start spending money.
I’ve started getting freezes. Like, old-school hard-lock freezes: screen stops, sound stops, cursor and keyboard don’t respond, and activity LED doesn’t flash at all. I have a Win10 Pro install on a SATA SSD and a Linux Mint install on an NVME SSD and I’ve had the exact same freeze happen on both boots; I think that pretty solidly nails it down as a hardware problem, not a software/virus problem.
Here’s my computer
September 2017
Windows 10 Pro 21H2 on SanDisk Ultra II SATA SSD (Fresh install November 2020)
MSI X370 KRAIT GAMING AM4 AMD X370
AMD Ryzen 5 1st Gen - RYZEN 5 1600X
G.SKILL Ripjaws 2x 8GB DDR4 3000 (clocked at 2666)
CORSAIR CS-M Series CS650M 650W 80 PLUS GOLD
September 2019
SAPPHIRE Radeon RX 5700
April 2022
Linux Mint 20.3 Cinnamon on HP EX900 NVMe SSD
Here’s what I’ve tested
MemTest86, all tests, four iterations. Passed with zero errors.
Prime95 on “blend” mixed test for eight hours. Zero errors. HWiNFO64 recorded an average core temp of 69deg and max of 74deg.
Furmark and Prime95 simultaneously for an hour with HWiNFO64 logging stats. Neither CPU or GPU exceeded 75deg, and my UPS reported a total power draw between 350-375W.
EDIT: Adding in, also checked all my drivers for updates, ran windows update, and updated BIOS to latest stable version.
These are the possible culprits and my thoughts on the likelihood they’re responsible
PSU: unlikely. PSU failures usually present as unexpected shutdowns/restarts, and I don't draw anywhere close to what it's rated at.
Memory: very unlikely. The lack of errors from MemTest86 and Prime95 goes a long way in ruling this out, and memory issues usually result in crashing rather than freezing.
CPU: very unlikely. It’s rare for a stock-clocked/volted CPU to up and die, and the tests weren’t returning errors or thermal issues.
Drives: nope. I’d be more likely to win the lottery than have two different drives, using two different connection types, from two different brands, of two different ages, and running two different OS'es, fail in the exact same way.
GPU: possible. System freezing isn’t a GPU’s usual failure mode, but two-thirds of my freezes have either occurred in-game or when HD video is being played.
Motherboard: most likely. Motherboards are the most fickle, and likely to fail, component of a system. It’s also really hard to test them for errors, especially intermittent ones.
So, I’m thinking of buying a new AM4 X570 board and swapping over my components and seeing if that resolves my issue. I’ve also had a lot of issues with MSI products in the years since that MOBO purchase: the fans on my crossfire'd MSI HD6950 graphics cards both died, and a 1440p 144Hz monitor from them died within two years. After the monitor death I swore I’d never buy another one of their products, and if this MOBO is the culprit you can be damn sure that brand will be radioactive to me. As for a replacement, you can get a good board for under $200 and it’ll be compatible with everything I’ve got now as well as allow me to upgrade to a 5000 series processor if I want/need to later.
Thanks to anybody who took the time to read that novel.
Did you start getting the freezes only after installing the NVMe SSD, since you list that as April 2022? Try removing it and seeing what happens.
Next I would look at the RAM since it's a common culprit and the absolute easiest thing to swap out and test. Try different slots in your motherboard, try different RAM completely, try removing the overclock.
Well, kind of. It’s really just the 3080tis and up that looks to be falling close to msrp. I looked at 3080s on Amazon just now and they’re still like 5-600 over msrp. Unreal
Hey guys, I’m having what I’m pretty sure is a hardware issue and I wanted to run my reasoning past somebody as a sanity-check before I start spending money.
I’ve started getting freezes. Like, old-school hard-lock freezes: screen stops, sound stops, cursor and keyboard don’t respond, and activity LED doesn’t flash at all. I have a Win10 Pro install on a SATA SSD and a Linux Mint install on an NVME SSD and I’ve had the exact same freeze happen on both boots; I think that pretty solidly nails it down as a hardware problem, not a software/virus problem.
Here’s my computer
September 2017
Windows 10 Pro 21H2 on SanDisk Ultra II SATA SSD (Fresh install November 2020)
MSI X370 KRAIT GAMING AM4 AMD X370
AMD Ryzen 5 1st Gen - RYZEN 5 1600X
G.SKILL Ripjaws 2x 8GB DDR4 3000 (clocked at 2666)
CORSAIR CS-M Series CS650M 650W 80 PLUS GOLD
September 2019
SAPPHIRE Radeon RX 5700
April 2022
Linux Mint 20.3 Cinnamon on HP EX900 NVMe SSD
Here’s what I’ve tested
MemTest86, all tests, four iterations. Passed with zero errors.
Prime95 on “blend” mixed test for eight hours. Zero errors. HWiNFO64 recorded an average core temp of 69deg and max of 74deg.
Furmark and Prime95 simultaneously for an hour with HWiNFO64 logging stats. Neither CPU or GPU exceeded 75deg, and my UPS reported a total power draw between 350-375W.
These are the possible culprits and my thoughts on the likelihood they’re responsible
PSU: unlikely. PSU failures usually present as unexpected shutdowns/restarts, and I don't draw anywhere close to what it's rated at.
Memory: very unlikely. The lack of errors from MemTest86 and Prime95 goes a long way in ruling this out, and memory issues usually result in crashing rather than freezing.
CPU: very unlikely. It’s rare for a stock-clocked/volted CPU to up and die, and the tests weren’t returning errors or thermal issues.
Drives: nope. I’d be more likely to win the lottery than have two different drives, using two different connection types, from two different brands, of two different ages, and running two different OS'es, fail in the exact same way.
GPU: possible. System freezing isn’t a GPU’s usual failure mode, but two-thirds of my freezes have either occurred in-game or when HD video is being played.
Motherboard: most likely. Motherboards are the most fickle, and likely to fail, component of a system. It’s also really hard to test them for errors, especially intermittent ones.
So, I’m thinking of buying a new AM4 X570 board and swapping over my components and seeing if that resolves my issue. I’ve also had a lot of issues with MSI products in the years since that MOBO purchase: the fans on my crossfire'd MSI HD6950 graphics cards both died, and a 1440p 144Hz monitor from them died within two years. After the monitor death I swore I’d never buy another one of their products, and if this MOBO is the culprit you can be damn sure that brand will be radioactive to me. As for a replacement, you can get a good board for under $200 and it’ll be compatible with everything I’ve got now as well as allow me to upgrade to a 5000 series processor if I want/need to later.
Thanks to anybody who took the time to read that novel.
Just before you go further, have you tried unplugging one of the drives, and running only one OS at a time? Obviously you'll have to reconfigure your BIOS boot setup so that the computer doesn't look for the other drive to load up a dual boot.
That's the most certain step to ensure that weird software handshakes/conflicts between the different OS's aren't causing this before you really dig into an investigation of potential hardware culprits.
Hyper X Cloud Stinger Core for $15 at Best Buy and Amazon. If your computer has a combo front panel jack, they'll be plug and play. If you have a separate mic and audio jack, you'll need to buy a headphone splitter like this for $6.
Just before you go further, have you tried unplugging one of the drives, and running only one OS at a time? Obviously you'll have to reconfigure your BIOS boot setup so that the computer doesn't look for the other drive to load up a dual boot.
That's the most certain step to ensure that weird software handshakes/conflicts between the different OS's aren't causing this before you really dig into an investigation of potential hardware culprits.
Did you start getting the freezes only after installing the NVMe SSD, since you list that as April 2022? Try removing it and seeing what happens.
Next I would look at the RAM since it's a common culprit and the absolute easiest thing to swap out and test. Try different slots in your motherboard, try different RAM completely, try removing the overclock.
The first time the freeze happened was before I installed the NVMe and put on linux so that shouldn't be the problem (both times it froze initially were during the same game, so I thought it was a game-specific problem and didn't think there was anything wrong with the system-at-large, so my already-in-process plans to try out Linux continued on). It's not even setup for multi-boot at the moment, it's just one OS on one drive and one on the other. Default boot is into Windows, f11 on startup to select boot source if I want to tinker with Linux.
Event Viewer didn't show any errors before or during the freeze, and since it's a freeze and not a crash there's no crash-logs.
I'll try running with one stick of memory, and see if I can nail down either stick as being bad. Also, I might not have been clear but there is no overclock. The ram is rated at 3000 but I run it at 2666 because that's the highest non-OC the 1600X's memory controller supports. I got the 3000 to give me room to mess with an OC if I felt like it but never bothered.
The first time the freeze happened was before I installed the NVMe and put on linux so that shouldn't be the problem (both times it froze initially were during the same game, so I thought it was a game-specific problem and didn't think there was anything wrong with the system-at-large, so my already-in-process plans to try out Linux continued on). It's not even setup for multi-boot at the moment, it's just one OS on one drive and one on the other. Default boot is into Windows, f11 on startup to select boot source if I want to tinker with Linux.
Event Viewer didn't show any errors before or during the freeze, and since it's a freeze and not a crash there's no crash-logs.
I'll try running with one stick of memory, and see if I can nail down either stick as being bad. Also, I might not have been clear but there is no overclock. The ram is rated at 3000 but I run it at 2666 because that's the highest non-OC the 1600X's memory controller supports. I got the 3000 to give me room to mess with an OC if I felt like it but never bothered.
I might be mistaken but anything over 2133Mhz or 2400Mhz is an overclock for DDR4 RAM - if you load an XMP profile to get to the advertised rate, that's an overclock. I doubt it's this but since these are the simplest steps to check I thought I'd throw it out there. Similar simple steps could be to clear the CMOS or load optimised defaults for your BIOS. Likewise update all of your drivers, check all your firmware for updates, etc.
Intermittent stuff is a massive pain but it could be even more annoying if you go and buy a new motherboard and that wasn't the problem.
The best, most sure fire way is to go piece by piece through your system and take it out and test.
Good luck and please let us know what it is when you get this one solved!
The first time the freeze happened was before I installed the NVMe and put on linux so that shouldn't be the problem (both times it froze initially were during the same game, so I thought it was a game-specific problem and didn't think there was anything wrong with the system-at-large, so my already-in-process plans to try out Linux continued on). It's not even setup for multi-boot at the moment, it's just one OS on one drive and one on the other. Default boot is into Windows, f11 on startup to select boot source if I want to tinker with Linux.
Event Viewer didn't show any errors before or during the freeze, and since it's a freeze and not a crash there's no crash-logs.
I'll try running with one stick of memory, and see if I can nail down either stick as being bad. Also, I might not have been clear but there is no overclock. The ram is rated at 3000 but I run it at 2666 because that's the highest non-OC the 1600X's memory controller supports. I got the 3000 to give me room to mess with an OC if I felt like it but never bothered.
I think you're right in isolating the GPU as the likely culprit given the timing of the freezes. Probably just needs a re-seating, and double-check your PSU's power connectors to the card. When you pull the GPU out take the opportunity to blow out the fans with some compressed air, and swab it down with rubbing alcohol using a lint-free cloth. Don't forget to hit the unlocking mechanism before you pull the GPU. I absent-mindedly broke one of those once pulling like a goddamn ape. Lucky I didn't damage the GPU.
@Madmick Have you read anything about Hyper X's condenser mic?
Best Buy has a combo with the Cloud Core Wired 7.1 headset, $35 right now, and the SoloCast Wired USB Condenser Microphone for $60.
I think you're right in isolating the GPU as the likely culprit given the timing of the freezes. Probably just needs a re-seating, and double-check your PSU's power connectors to the card. When you pull the GPU out take the opportunity to blow out the fans with some compressed air, and swab it down with rubbing alcohol using a lint-free cloth. Don't forget to hit the unlocking mechanism before you pull the GPU. I absent-mindedly broke one of those once pulling like a goddamn ape. Lucky I didn't damage the GPU.
I might be mistaken but anything over 2133Mhz or 2400Mhz is an overclock for DDR4 RAM - if you load an XMP profile to get to the advertised rate, that's an overclock. I doubt it's this but since these are the simplest steps to check I thought I'd throw it out there. Similar simple steps could be to clear the CMOS or load optimised defaults for your BIOS. Likewise update all of your drivers, check all your firmware for updates, etc.
Intermittent stuff is a massive pain but it could be even more annoying if you go and buy a new motherboard and that wasn't the problem.
The best, most sure fire way is to go piece by piece through your system and take it out and test.
Good luck and please let us know what it is when you get this one solved!
Well, I:
Triple-checked my drivers/bios
Tried one memory stick at a time
Removed, cleaned, and reseated my GPU
Unfortunately nothing worked and the freezes have been getting worse and worse. Whatever is going bad, it's accelerating.
So I just ordered a new X570 Motherboard, 5600G CPU, and gen 4 x4 NVMe. I'd been eyeballing a CPU upgrade anyways, and I got my quarterly bonus from work...so to hell with it, I'll upgrade now. I'm planning on starting with just a motherboard swap and see if that solves it, because if it does then I'll be able to sell my 1600X for ~$100-150. The 5600G has an integrated GPU so if my current radeon card turns out to be the problem I can live without it for now and hope new card prices continue to come back down to earth (an RTX3060/RX6600XT cost $100 more than I paid for my RX5700 even though their launch MSRP/lineup-slot is the same). EDIT: damn, used RX5700's have completed listings on ebay in the $400-450 range. I paid $360 for mine new three years ago. That's wacky.
Thanks for the help, I'll pop back in next week after parts come in and I hopefully will have the problem isolated.
Well, I:
Triple-checked my drivers/bios
Tried one memory stick at a time
Removed, cleaned, and reseated my GPU
Unfortunately nothing worked and the freezes have been getting worse and worse. Whatever is going bad, it's accelerating.
So I just ordered a new X570 Motherboard, 5600G CPU, and gen 4 x4 NVMe. I'd been eyeballing a CPU upgrade anyways, and I got my quarterly bonus from work...so to hell with it, I'll upgrade now. I'm planning on starting with just a motherboard swap and see if that solves it, because if it does then I'll be able to sell my 1600X for ~$100-150. The 5600G has an integrated GPU so if my current radeon card turns out to be the problem I can live without it for now and hope new card prices continue to come back down to earth (an RTX3060/RX6600XT cost $100 more than I paid for my RX5700 even though their launch MSRP/lineup-slot is the same). EDIT: damn, used RX5700's have completed listings on ebay in the $400-450 range. I paid $360 for mine new three years ago. That's wacky.
Thanks for the help, I'll pop back in next week after parts come in and I hopefully will have the problem isolated.
Have you gone into the Event Viewer? It will flag every Critical Error or Warning. This sounds abstruse enough that you should be logging crash dumps, and sharing those files on the Windows forum (or a third-party forum) for technical assistance.
Have you gone into the Event Viewer? It will flag every Critical Error or Warning. This sounds abstruse enough that you should be logging crash dumps, and sharing those files on the Windows forum (or a third-party forum) for technical assistance.
I looked into it, but it's not crashing, so there's no crash dumps or logs. Nor were there any errors/warnings leading up to or during the freeze events.
"Meanwhile, ASRock's renowned overclocker, SPLAVE from the US, was busy benching the Intel Core i9-12900KS across various performance benchmarks and secured a world record spot in a total of 7 benchmarks which is absolutely nuts. SPLAVE also had the 12900KS running at over 7 GHz average in these benchmarks and across multiple cores which is very impressive. Gigabyte and ASUS also managed to obtain some records with their own respective Tachyon and Apex motherboards. The full list of benchmarks can be seen below:
The CPU has a retail price of $739 US and is available on various retail outlets such as Amazon, Newegg, BestBuy, and Antonline. Now, just a few days after its launch, the chip has broken various records ranging from performance benchmarks to CPU frequency. First up, we have TSAIK, MSI's in-house overclocker, who used the MEG Z690 Unify-X motherboard to push the Core i9-12900KS to an absolute monster 7.45 GHz clock with a voltage supply of 1.769V. The frequency was applied to a single Golden Cove core and this is currently the fastest clock speed recorded for the flagship CPU. The CPU validation of the overclock can be seen here.
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.
