• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.
  • The forums have been upgraded with support for dark mode. By default it will follow the setting on your system/browser. You may override it by scrolling to the end of the page and clicking the gears icon.

2nd reboot PC runs better ??!!

Joined
Sep 7, 2008
Messages
216 (0.04/day)
System Name Ma Fish Tank
Processor AMD 7950x
Motherboard MSI B650 Gaming Plus WiFi AM5 ATX Desktop Motherboard9
Cooling Thermalright Frozen Warframe Pro ARGB 360mm AIO CPU Cooler Black
Memory GeIL Orion V RGB 32GB Kit (2x16GB) DDR5 C38 5800MHz
Video Card(s) ASUS GeForce RTX 4080 Super TUF Gaming OC 16GB
Storage Samsung 970 EVO Plus PCIe Gen3 NVMe M.2 SSD - 1TB
Display(s) MSI MAG 401QR UWQHD 155Hz Adaptive-Sync IPS 40in Monitor (2nd monitor is 27 msi)
Case Montech KING 95 Pro Mid Tower Case Black
Audio Device(s) Creative Sound Blaster X5 DAC and Headphone Amp
Power Supply SilverStone DA850-G - 850W Gold ATX Modular PSU
Mouse Razer Naga Left-Handed Edition Ergonomic MMO Gaming Mouse for Left-Handed
Keyboard ROCCAT ROC-12-671-RD-AM Vulcan 121 Aimo RGB Mechanical Gaming Keyboard
Software Microsoft Windows 11
Benchmark Scores nil
Here are some pc specs
7950x
64gig ddr5 geil 58
M2 980 Samsung
4080
Msi mobo gamer plus

I just recently put together this pc , it fought me abit at the start as far it wouldn’t post (black screen) , I narrowed it down to the memory, I changed it over to another brand different speed (5800), anyway all working good now, but…

First boot pc runs laggy, it’s takes a long time to settle , struggle to move the mouse cursor for example, after the second reboot it runs perfectly, any ideas why this is occurring?
 
Obviously being a new build you should make sure no wires are pinched or shorting. With this large of issues I'd fully disassemble the build and fully remove PSU for visual inspection and testing of it and cables. Especially since you didn't list what psu you have. Opening the door to being underpowered or, far worse, using cables that aren't certified to work with it.

Next step would be error testing on OS drive with it alone mounted. Followed by any others in sequence to see if they suddenly cause the issue.

Mobo being faulty is what you need to prove here. Remove everything else and RMA if problems persist. Someone else may have better ideas but this is fairly logical process of elimination having already changed RAM.
 
Here are some pc specs
7950x
64gig ddr5 geil 58
M2 980 Samsung
4080
Msi mobo gamer plus

I just recently put together this pc , it fought me abit at the start as far it wouldn’t post (black screen) , I narrowed it down to the memory, I changed it over to another brand different speed (5800), anyway all working good now, but…

First boot pc runs laggy, it’s takes a long time to settle , struggle to move the mouse cursor for example, after the second reboot it runs perfectly, any ideas why this is occurring?
Update your motherboard bios
 
Update your motherboard bios
It’s the latest bios for the Msi mobo, there is one newer but it’s a beta bios and that made it worse ..cheers for the suggestion

T
Obviously being a new build you should make sure no wires are pinched or shorting. With this large of issues I'd fully disassemble the build and fully remove PSU for visual inspection and testing of it and cables. Especially since you didn't list what psu you have. Opening the door to being underpowered or, far worse, using cables that aren't certified to work with it.

Next step would be error testing on OS drive with it alone mounted. Followed by any others in sequence to see if they suddenly cause the issue.

Mobo being faulty is what you need to prove here. Remove everything else and RMA if problems persist. Someone else may have better ideas but this is fairly logical process of elimination having already changed RAM
The PSU is the silverstone 850
 
Here are some pc specs
7950x
64gig ddr5 geil 58
M2 980 Samsung
4080
Msi mobo gamer plus

I just recently put together this pc , it fought me abit at the start as far it wouldn’t post (black screen) , I narrowed it down to the memory, I changed it over to another brand different speed (5800), anyway all working good now, but…

First boot pc runs laggy, it’s takes a long time to settle , struggle to move the mouse cursor for example, after the second reboot it runs perfectly, any ideas why this is occurring?
If I understand correctly... every time the PC does a cold boot has issues and when restarted its ok?
How much time have you waited on the first cold boot? Is there a chance that after a few(?) minutes turns out ok without reboot?

I would get HWiNFO64 (run sensors mode only) and set it to start with windows. Set it at full screen when you open the window so that is it shows as many sensors as possible.


Please copy/paste these (main) settings (right click tray icon > settings)

Untitled_206.png

1st cold boot: try to use it for like 10-20min (if not more) while HWiNFO64 runs in background
2nd boot (reboot): try to use it the same way again for same amount of time with HWiNFO64 in background

Take sensors window a screenshot for each run with at least 10-15min (if not more) and post both here

Try to copy paste what is visible on each section below
I see you also have an ultrawide screen so there will be no problem

Run time can be seen on bottom right
Untitled_205.png
 
Is this a 2 DIMM kit or are you populating all 4 slots?
I was popping 2 dims when the problem started now im using 4 dims and still the same problem ..
 
First boot pc runs laggy, it’s takes a long time to settle , struggle to move the mouse cursor for example, after the second reboot it runs perfectly,
I wonder if it's something associated with DDR5 Memory Training. A quick search reveals other people have this same "problem". I suggest a little research on the subject of Memory Training.

If your RAM is overclocked, e.g. at XMP 5800MT/s, as an experiment, try running it at JEDEC default e.g. 4800MT/s.

To get a better idea of how long each boot operation takes, temporarily install BootRacer or similar, to measure the times more consistently.
https://www.thewindowsclub.com/measure-boot-time-windows

My 7950X rig built in December 2022 is running stock 4800MT/s on 2 x 32GB DIMMs. There is only one XMP setting in the RAM and it, too, is 4800MT/s with exactly the same timings. When I bought the RAM, 32GB DIMMs with fast XMP settings cost more than I was prepared to pay. As a consequence, I've never run my RAM at higher speeds (choosing stability over speed on 24hour+ rendering runs).

I don't notice any significant difference between cold boots and warm restarts on the 7950X, but they're not exactly fast. Windows 10 start up times from M.2 NVMe 1TB SSDs on the 7950X are around 60 seconds, but the PC has to check the boot ROMs in the 10GBe NIC and LSI SAS HBA PCIe cards. The presence of three M.2 and five hard disks adds slightly to boot up times, as Windows checks the "dirty bit" on each drive. Many things can affect start up times.

Do you have Windows "Fast Startup" enabled or disabled. Fast Startup will probably improve boot times if enabled. I switch this feature off and disable hiberfil.sys too. Main reason is Fast Startup can mess up booting Windows from alternative SSDs in the same machine. CHKDSK would start when I switched back to the main boot drive if Fast Startup was enabled, which was annoying on machines with multiple drives. It could take CHKDSK to confirm all the hard disks were OK.
https://www.windowscentral.com/soft...-enable-or-disable-fast-startup-on-windows-11
 
How much time have you waited on the first cold boot?
Ummm a "cold" boot is when the PSU has been UNPLUGGED from the wall (or PSU master power switch is set to off).

It is important to remember if the PSU is plugged in to the wall (and, if present, the master switch set to on), the ATX Form Factor standard requires the supply provides +5Vsb standby voltage to several points on the motherboard, including the RAM, USB and Ethernet ports, and the case's front panel power button.

When you shutdown through the Start menu or with the case's front panel power button, that simply puts the computer in a standby state. It is never fully off (cold) until all the voltages are removed and that only happens when unplugged from the wall, or the master switch is set to off.

I would take everything out of the case and assemble it on a clean, unfinished bread/cutting board and see if the problem remains. If it does, swap power supplies before replacing anything else. This is because everything else depends on good, clean, stable power.

If it works outside the case, verify you only inserted a case standoff where there is a corresponding motherboard mounting hole. Then carefully reassemble in the case ensuring all data and power cables are securely connected.
 
Ummm a "cold" boot is when the PSU has been UNPLUGGED from the wall (or PSU master power switch is set to off).

It is important to remember if the PSU is plugged in to the wall (and, if present, the master switch set to on), the ATX Form Factor standard requires the supply provides +5Vsb standby voltage to several points on the motherboard, including the RAM, USB and Ethernet ports, and the case's front panel power button.

When you shutdown through the Start menu or with the case's front panel power button, that simply puts the computer in a standby state. It is never fully off (cold) until all the voltages are removed and that only happens when unplugged from the wall, or the master switch is set to off.

I would take everything out of the case and assemble it on a clean, unfinished bread/cutting board and see if the problem remains. If it does, swap power supplies before replacing anything else. This is because everything else depends on good, clean, stable power.

If it works outside the case, verify you only inserted a case standoff where there is a corresponding motherboard mounting hole. Then carefully reassemble in the case ensuring all data and power cables are securely connected.
Ok, no argument. I should have been more clear. I meant to say literally cold boot, in ambient temp and with nothing warm up after a restart for example.
 
No big deal. It is a common mistake - in fact, the second time this morning someone meant the computer had been turned off for hours and, in terms of temperature, had cooled off.

If it technically did not make a difference, it would not matter. But since in this case, where voltages, data, and [possibly corrupt] settings remain when the computer is simply shutdown (put in standby mode), ensuring everyone is on the same page does matter.

Thanks for your understanding.
 
I wonder if it's something associated with DDR5 Memory Training. A quick search reveals other people have this same "problem". I suggest a little research on the subject of Memory Training.

If your RAM is overclocked, e.g. at XMP 5800MT/s, as an experiment, try running it at JEDEC default e.g. 4800MT/s.

To get a better idea of how long each boot operation takes, temporarily install BootRacer or similar, to measure the times more consistently.
https://www.thewindowsclub.com/measure-boot-time-windows

My 7950X rig built in December 2022 is running stock 4800MT/s on 2 x 32GB DIMMs. There is only one XMP setting in the RAM and it, too, is 4800MT/s with exactly the same timings. When I bought the RAM, 32GB DIMMs with fast XMP settings cost more than I was prepared to pay. As a consequence, I've never run my RAM at higher speeds (choosing stability over speed on 24hour+ rendering runs).

I don't notice any significant difference between cold boots and warm restarts on the 7950X, but they're not exactly fast. Windows 10 start up times from M.2 NVMe 1TB SSDs on the 7950X are around 60 seconds, but the PC has to check the boot ROMs in the 10GBe NIC and LSI SAS HBA PCIe cards. The presence of three M.2 and five hard disks adds slightly to boot up times, as Windows checks the "dirty bit" on each drive. Many things can affect start up times.

Do you have Windows "Fast Startup" enabled or disabled. Fast Startup will probably improve boot times if enabled. I switch this feature off and disable hiberfil.sys too. Main reason is Fast Startup can mess up booting Windows from alternative SSDs in the same machine. CHKDSK would start when I switched back to the main boot drive if Fast Startup was enabled, which was annoying on machines with multiple drives. It could take CHKDSK to confirm all the hard disks were OK.
https://www.windowscentral.com/soft...-enable-or-disable-fast-startup-on-windows-11
Interesting read , seems like using a different brand with different timing made it better , geil at 5800 is good , gskill at 6600 no good, also I took some Kingston 5800 out of my sons pc and it worked perfectly as well..

Ok, no argument. I should have been more clear. I meant to say literally cold boot, in ambient temp and with nothing warm up after a restart for example.
I should have stated , i use the pc most of day on my day off it all works as it should, go to bed , wake up go to turn it on , it works like $hit , I wait for it to settle, follow the promps for 2nd reboot , then it works perfectly… that’s the process, it’s not a HUGE pita , but it is a why is this happening on my lovely new build ??!!
 
Last edited:
Interesting read , seems like using a different brand with different timing made it better , geil at 5800 is good , gskill at 6600 no good, also I took some Kingston 5800 out of my sons pc and it worked perfectly as well..


I should have stated , i use the pc most of day on my day off it all works as it should, go to bed , wake up go to turn it on , it works like $hit , I wait for it to settle, follow the promps for 2nd reboot , then it works perfectly… that’s the process, it’s not a HUGE pita , but it is a why is this happening on my lovely new build ??!!
Yes, that what I was thinking its doing. Seems like something is warming up and stops causing issues
 
why is this happening on my lovely new build ??!!
People assume because something is new, it cannot have faults. Until Man can create perfection 100% of the time, there will always be units with manufacturing defects and/or premature failures.

Seems like something is warming up and stops causing issues
Could be. But if it immediately shuts down again, that would mean little time for the any faulty device to warm up. And if the ambient (room) temperature remains constant, then again, not sure where warming up factors in.

Still - the first place to look is the PSU since again, everything inside the case depends on good clean stable power.
 
Could be. But if it immediately shuts down again, that would mean little time for the any faulty device to warm up. And if the ambient (room) temperature remains constant, then again, not sure where warming up factors in.
I'm just driven by this below (highlighted)
Interesting read , seems like using a different brand with different timing made it better , geil at 5800 is good , gskill at 6600 no good, also I took some Kingston 5800 out of my sons pc and it worked perfectly as well..


I should have stated , i use the pc most of day on my day off it all works as it should, go to bed , wake up go to turn it on , it works like $hit , I wait for it to settle, follow the promps for 2nd reboot , then it works perfectly… that’s the process, it’s not a HUGE pita , but it is a why is this happening on my lovely new build ??!!
Some kind of shorting maybe that goes away with heat expansion? Or some solderings warming up and have better contact?
 
Do you have fast boot enabled in Windows? If so, when you shut down, you’re actually hibernating and powering up the next day is just resuming.

A ‘Restart’ actually reboots the OS fresh.

That might explain why you have better performance after restarting. My guess is an issue with hibernation and memory training/context restore.

You can either disable to troubleshoot, or at the end of the day, just open a command prompt/powershell and type:

Code:
shutdown /s

This actually does a proper shutdown and you can test the next morning.
 
it’s not a HUGE pita , but it is a why is this happening on my lovely new build ??!!

It's a large concern that could impact functioning hardware. Meaning you should isolate the issue as quickly as possible with the least amount of stressful testing.

For the moment I disagree with sentiment heat expansion is a factor. To me this seems to replicate influx of power and system checks producing an out of tolerance report the system is trying to work beyond. Disassemble, inspect, breadboard.
 
Some kind of shorting maybe that goes away with heat expansion? Or some solderings warming up and have better contact?
I agree completely. And once again, the PSU is place to look first. Not just because all the connected components depend on it, but that's where the high voltages exist. And that plays right into your point about heat and warming up. The high AC voltage inside a PSU could do that much more quickly than something powered by low DC voltage.
 
Do you have fast boot enabled in Windows? If so, when you shut down, you’re actually hibernating and powering up the next day is just resuming.

A ‘Restart’ actually reboots the OS fresh.

That might explain why you have better performance after restarting. My guess is an issue with hibernation and memory training/context restore.

You can either disable to troubleshoot, or at the end of the day, just open a command prompt/powershell and type:

Code:
shutdown /s

This actually does a proper shutdown and you can test the next morning.
i do a proper shut down everytime, the settings in the BIOS are "as is" , i haven't touched any OC settings , its all default other then XMP ive drop the mem from 6000 to 5800 , i think i might have to go even lower to maybe 5500 or even 5200 ill test it out over the weekend. Probably might even look at getting a new PSU.
 
I agree completely. And once again, the PSU is place to look first. Not just because all the connected components depend on it, but that's where the high voltages exist. And that plays right into your point about heat and warming up. The high AC voltage inside a PSU could do that much more quickly than something powered by low DC voltage.
Another place that seems crazy is the quality of the Display cable the OP is using. If you feel the Cable and it is hot there is chance it could be the issue. I tell everyone I know to buy 8K certified DP cables so you don't run into an issue.
 
Probably might even look at getting a new PSU.
IMO, anyone who does his own maintenance, or has been volunteered to be the family/neighborhood go-to computer guy, should have a spare PSU handy - not just to temporarily swap in to verify if a PSU is good, or not - but having a spare PSU handy is nice for also testing fans, RGB lighting, pumps, and hard drive motors. I keep an old PSU Tester connected to one of my spare PSUs so it will think it is connected to a motherboard and fire up - so I don't have to mess with bent paperclips. :)
 
Back
Top