*Official Benchmark Thread* - Post it here or it didn't happen :D

[electrosoft]

3 hours ago, tps3443 said:

IT WAS THE CABLE!! I think my 5090 did it in lol. Swapped the 12vhpwr. No crashing!!! 🤩🤩🤩🤩🤩🤩

 

3 hours ago, tps3443 said:

 
 

Yes! Very very hard to buy a 5090. That guilt is STRONG after buying one too. Today’s prices are nuts. 5080 is a value and gets the job done!! Gone are the days of high end affordable GPU’s. It’s a rich man’s club now. 
 

EDIT: 

 

So happy I found the issue! When that 2nd brand new Zotac RTX 5080 locked my PC up, I was concerned and relieved at the same time. Relieved that the RTX 5080 was not defective. But also worried what the issue could be. 
 

The bad cable was ran daily on a shunt modded 5090FE, so it must have took the cable out I guess. What’s weird though, is I had no issues running the 5090FE up until the day I pulled it from the system. So I think maybe it sitting it just degraded? It’s so weird. 
 

I’m going to do some testing now and see which is the best GPU! I would prefer a low voltage low power queen if I had to pick. I don’t know what all the fuss or hate about Zotac is either. Both of these cards are built incredibly well. I had a Gigabyte 4090 Gaming OC, and that thing felt like a Dollar General toy lol. 😂 these cards feel nice and look very high quality.

 

Best possible outcome. Feels like when my EVGA 1600w P2 started to falter a few months ago and I couldn't lock down where the failures or sporadic errors were coming from and had to start from the ground up to isolate the PSU.

 

The cables really are the weakest link this time around. I know with my 4090 and 5090 I kept the pigtail plugged in at all times even when swapping out. Many times for motherboard upgrades I would remove the GPU from the slot and lay it still connected on the lower part of the case.

 

Could have been just the absolute pounding you gave that cable and the limited times you have to plug and unplug it before it starts to degrade.....which is insane to think on a cable meant to handle 600w+.....

 

Gigabyte Gaming OC cards in general are not built that great. Master and Elite cards get the love. 

 

Glad you're back up and running! Let us know which one is the superior GPU. I'm curious to see vmax, best OC, best UV and overall temps head to head.

 

2 hours ago, Papusan said:

 

This become a problem because you had good contact in first place. Once removed and the connection will break and you get a problem. Cables have now become a consumable items. I mean the tiny trash called 12V-2x6. Those cannot be reused as the good old 8-pin cables. 

 

Benching the old gpu's ain't very fun anymore because you need to take out and in your new and shiny GPU if you want the newest in you rig after bench sessions. The modern cables ain't up to task. This tiny trash ruined my hobby.  

 

 

The only real way to do it now is to use the dreaded tentacle for a few reasons to bench multiple cards:

 

#1. You can keep the 12vfail connector connected at all times

#2. It makes it easy to bench and test traditional 8-pin PCIE GPUs since you have 4x 8-pin leads at your disposal.

 

 

 

 

[tps3443]

Would be awesome if running two RTX 5080’s was useful for something. 😂 They really match so perfectly together!! Maybe there is a benefit? Or a waste. 

[Mr. Fox]

40 minutes ago, tps3443 said:

Would be awesome if running two RTX 5080’s was useful for something. 😂 They really match so perfectly together!! Maybe there is a benefit? Or a waste. 

For gaming very little benefit since there is no SLI hardware support and almost zero SLI support in terms of game development. You could use one for render and one for compute, and you could use Lossless Scaling for gaming. That would be a very expensive second GPU for that though. What GPU does your son have? If he has anything less than a 4090 it would be a nice upgrade for him. If you wanted to experiment with Lossless Scaling, you could use his GPU for that as a second GPU in your system.

 

All of the above aside, having two matching 5080 Solid GPUs installed in your system would look pretty wicked.

 

If you end up returning the new one you will have an "original owner" invoice with your name on it to register for warranty with Zotac. 

 

You could buy this one 5090 for roughly the cost of 4.5 5080s if you wanted to do something extra-stupid and frivolous with your money. https://www.overclock.net/threads/msi-rtx-5090-lightning-z-65-of-1300-custom-loop-compatible.1819519/?post_id=29587541#post-29587541

[Papusan]

13 hours ago, electrosoft said:

The only real way to do it now is to use the dreaded tentacle for a few reasons to bench multiple cards:

 

#1. You can keep the 12vfail connector connected at all times

#2. It makes it easy to bench and test traditional 8-pin PCIE GPUs since you have 4x 8-pin leads at your disposal.

 

Probably the best for several reasons. Not sure I have seen so many org Asus tentacle meltdown Asus cards this time around. If the connector melt into the GPU, then it's nice if it's the org Asus adapter fused to the card😀 Still, I don't like the solution. I hate that stupid lille connector. Even Jonny Guru at Corsair say the connector won't stay properly connected with the card when you pass 6 months. It will loose the connection and here we go... BOOOM!

Corsair explains why 12V-2x6 still needs protection despite ...

The revised mechanical design is a real improvement. But it’s not a complete solution, because connectors don’t stay fully seated forever.

Cable tension from routing, vibration, the weight of a stiff cable pulling at the connector, repeated insertions over time – any of these can gradually work a connector loose. A connector that clicked in perfectly when you first built your system can develop marginal contact six months later without any visible indication. The physics of partial-contact heating are identical whether it was never fully inserted or worked loose after the fact.

 

 

 

 

[Papusan]

On 5/12/2026 at 11:31 PM, Mr. Fox said:

Brother @tps3443 dodged a scammer bullet yesterday. Glad you were not a victim, Brother T.

 

https://www.overclock.net/posts/29585246/

 

The scumbag posted the same scam on a Swedish web site.

 

https://www.blocket.se/recommerce/forsale/item/22104267?ci=4

 

And the scumbag is ranked 71 on hwbot.... https://hwbot.org/users/mllrkllr88

 

Do you trust all results posted on hwbot? Sharing is caring... Isn't it? 

 

 

[win32asmguy]

On 4/29/2026 at 10:50 AM, electrosoft said:

A valid comparison would me 9700x vs 9800X3D to really see what X3D brings to the table. I do know in FO76 the 9800X3D toasted the 9600x I tested worse than the 14900KS so X3D is magic sauce for some games.

 

I picked up a MSI Raider A18, this one with the 9955HX3D, 5080M and QHD+ 240hz display. The X3D cache is hands down curb stomping the 275HX in WoW. Easily getting 80-90fps compared to 40-50fps in Silvermoon bazaar at the same settings at QHD and significantly better 1% lows.

 

This new revision of Raider A18 is pretty refined as well compared to the last gen. So far no audio system dropouts, no standby or display blanking issues, no problems with dGPU mode switching to iGPU mode randomly while a game was running. The Mediatek 7925 wifi card had some issues dropping the connection with the provided driver but it seems like a newer revision on Windows Update addressed that issue. Swapping to an AX210 is also an option if the problems persist. Delightfully this model has static red by default for all of the RGB instead of the rainbow puke on the Intel Raider 18.

 

Its a shame the 275HX is limited by memory latency in WoW. The 285K does really well paired with good memory but on mobile it is very limited by heat. Maybe Nova Lake HX will help this if it has an on-die memory controller again and/or bLLC.

[tps3443]

Quick update! 
 

The issue still persists!

 

Things I have tried. 
 

(1) Two different GPU’s

(2) Pulling ram cleaning and reinstalling 1 stick only. 
(3) Removing CPU, checking pins, cleaning contact surfaces and remounting CPU all torqued to spec at 12in lbs in correct sequence!

(4) unplugging everything and plugging everything back in to motherboard.

(5) A different 12vhpwr plug

(6) A different PSU all together .

(7) Different bios on my motherboard.

(8) Reinstalled windows 11

(8) moved M.2 to a different location. 
(9) Tried different PCIe slots

(10) tried running only x8 PCIe speeds

(11) Tried adding unrealistic CPU and ram voltages at stock speeds or reduced clocks. 

 

PC is completely freezing I can still see the picture. It completely HARD locks. 

Things I have NOT tried yet.

 

(1) Using a different M.2. Maybe my 2TB M.2 is failing? It has fairly low health left something like 75%.

 

(2) Different motherboard. Maybe my Dominus Extreme is failing?

 

(3) Trying a new CPU. Maybe my Xeon W3175X is failing? 
 

 

I am at a loss. 

 

It sure does look cool though lol!!! 

 

[Papusan]

30 minutes ago, tps3443 said:

Things I have NOT tried yet.

 

(1) Using a different M.2. Maybe my 2TB M.2 is failing? It has fairly low health left something like 75%.

 

(2) Different motherboard. Maybe my Dominus Extreme is failing?

 

(3) Trying a new CPU. Maybe my Xeon W3175X is failing? 
 

 

I am at a loss. 

 

Tried an older gen GPU? Maybe that will play better with your old gen board. Many motherboards have difficulty with some newer GPUs. Could be that your 5090 FE played better with your motherboard.

[tps3443]

1 hour ago, Papusan said:

 

Tried an older gen GPU? Maybe that will play better with your old gen board. Many motherboards have difficulty with some newer GPUs. Could be that your 5090 FE played better with your motherboard.

I have installed a new 512GB M.2 SSD. We’ll see what happens. I wish that was all it was. 
 

It’s interesting because I can run Ycruncher full ram usage and the system just chugs right though. I ran it over 90 minutes last night. 

[Mr. Fox]

1 hour ago, tps3443 said:

Quick update! 
 

The issue still persists!

 

Things I have tried. 
 

(1) Two different GPU’s

(2) Pulling ram cleaning and reinstalling 1 stick only. 
(3) Removing CPU, checking pins, cleaning contact surfaces and remounting CPU all torqued to spec at 12in lbs in correct sequence!

(4) unplugging everything and plugging everything back in to motherboard.

(5) A different 12vhpwr plug

(6) A different PSU all together .

(7) Different bios on my motherboard.

(8) Reinstalled windows 11

(8) moved M.2 to a different location. 
(9) Tried different PCIe slots

(10) tried running only x8 PCIe speeds

(11) Tried adding unrealistic CPU and ram voltages at stock speeds or reduced clocks. 

 

PC is completely freezing I can still see the picture. It completely HARD locks. 

Things I have NOT tried yet.

 

(1) Using a different M.2. Maybe my 2TB M.2 is failing? It has fairly low health left something like 75%.

 

(2) Different motherboard. Maybe my Dominus Extreme is failing?

 

(3) Trying a new CPU. Maybe my Xeon W3175X is failing? 
 

 

I am at a loss. 

 

It sure does look cool though lol!!! 

 

Have you tested with only one video card in the bottom slot instead of the top slot where you would normally use a GPU? 

 

Sorry to hear this is happening but also glad that the second hand 5080 you bought wasn't defective. It's probably safe to assume that you have tested both video cards or at least the one you bought on Facebook in your son's computer and there are no issues.

 

It is a long shot but you may want to try the GOP firmware update released by Nvidia recently. I applied that update to both my 5080 and my 5090. It is possible the old UEFI/GOP firmware on your motherboard is not playing nice and that might help. The landing page on the Nvidia website where I downloaded it gives the impression it is for the 5060 only. That is misleading because it applies to all 50 series graphics cards. When I flashed it the nvflash scanned the hardware ID to confirm that the update was applicable before proceeding. As it turns out it was applicable to both of my video cards.

[Mr. Fox]

34 minutes ago, tps3443 said:

I have installed a new 512GB M.2 SSD. We’ll see what happens. I wish that was all it was. 
 

It’s interesting because I can run Ycruncher full ram usage and the system just chugs right though. I ran it over 90 minutes last night. 

I wouldn't rule out that the problem could be caused by one of the drives. If you have multiple SSDs or HDDs any one of them can cause lockups like what you are experiencing. If the one you replaced doesn't fix the problem don't stop there. You need to test all of them. Particularly the one that the games are installed on. Failing drives, even USB storage, can cause all sorts of weird problems including hard lockups. I have even seen defective internal and USB drives that prevented the PC form powering on and failed booting even at the BIOS level and hard lockups are a very common symptom.

[tps3443]

7 minutes ago, Mr. Fox said:

Have you tested with only one video card in the bottom slot instead of the top slot where you would normally use a GPU? 

 

Sorry to hear this is happening but also glad that the second hand 5080 you bought wasn't defective. It's probably safe to assume that you have tested both video cards or at least the one you bought on Facebook in your son's computer and there are no issues.

 

It is a long shot but you may want to try the GOP firmware update released by Nvidia recently. I applied that update to both my 5080 and my 5090. It is possible the old UEFI/GOP firmware on your motherboard is not playing nice and that might help. The landing page on the Nvidia website where I downloaded it gives the impression it is for the 5060 only. That is misleading because it applies to all 50 series graphics cards. When I flashed it the nvflash scanned the hardware ID to confirm that the update was applicable before proceeding. As it turns out it was applicable to both of my video cards.

Yep, I have ran both RTX 5080’s in the bottom slot PCIe. I also tried X16 and X8 mode. We tested the 5080 in my son’s system and it works fine. 
 

I just reinstalled Windows 11 on a brand new 512GB M.2 SSD. Fingers crossed. 
 

I am only running a single M.2 2TB, I pulled that out, and trying the 512GB now. 

[Mr. Fox]

39 minutes ago, tps3443 said:

Yep, I have ran both RTX 5080’s in the bottom slot PCIe. I also tried X16 and X8 mode. We tested the 5080 in my son’s system and it works fine. 
 

I just reinstalled Windows 11 on a brand new 512GB M.2 SSD. Fingers crossed. 
 

I am only running a single M.2 2TB, I pulled that out, and trying the 512GB now. 

Hopefully that is it. If you only had a single drive attached to your system that will make it easier to test for sure since you only have the one potential culprit. 

 

I had a 4 TB TeamGroup NVMe that failed back when I still owned the Z790 Apex. It was causing lockups and before I could figure out what the cause was it got worse and wouldn't even let the motherboard power on. It was like it was unplugged from the wall. Pressing the power button did nothing. No signs of life. As soon as I removed it everything returned to normal and I had it replaced under warranty.

[tps3443]

3 minutes ago, Mr. Fox said:

Hopefully that is it. If you only had a single drive attached to your system that will make it easier to test for sure since you only have the one potential culprit. 

 

I had a 4 TB TeamGroup NVMe that failed back when I still owned the Z790 Apex. It was causing lockups and before I could figure out what the cause was it got worse and wouldn't even let the motherboard power on. it was like it was unplugged from the wall. Pressing the power button did nothing. No signs of life. As soon as I removed it everything returned to normal and I had it replaced under warranty.

I ran Furmark and the system just locked up on the new M.2 SSD with new OS installed. It took about 8 minutes. 
 

Dang, I guess it’s time for new board and CPU. I don’t know what else to check. 

[cylix]

Does event viewer shows any errors when the freezing is happening?

[Mr. Fox]

39 minutes ago, tps3443 said:

I ran Furmark and the system just locked up on the new M.2 SSD with new OS installed. It took about 8 minutes. 
 

Dang, I guess it’s time for new board and CPU. I don’t know what else to check. 

What if you put your son's GPU in your system and run the same test again? Is your system PCIe Gen 3 or Gen 4? Did you try to manually set the max for that in the BIOS already? It is interesting that it seems to require the passage of time for it to occur. Like something has to reach a certain point. I am wondering if maybe one of your memory sticks is bad. Maybe try one at a time and see if one is messed up. Since you ruled out the NVMe, if I were going to take a wild guess at this point, it would be one of the sticks is bad and the system works until data reaches that one stick and then it's over. That's were I would start at this point. Might not hurt to check the temperature of the memory modules at the point of lockup. Probably won't see anything but checking never hurts.

19 minutes ago, cylix said:

Does event viewer shows any errors when the freezing is happening?

That is a good suggestion. It might show something relevant.

[tps3443]

32 minutes ago, cylix said:

Does event viewer shows any errors when the freezing is happening?

I will check this.

 

I have errors, criticals, and a bunch of warnings!! TPM not enabled, unexpected shutdowns, Ethernet lost connection, LSO was triggered. 

 

 

19 minutes ago, Mr. Fox said:

What if you put your son's GPU in your system and run the same test again? Is your system PCIe Gen 3 or Gen 4? Did you try to manually set the max for that in the BIOS already? It is interesting that it seems to require the passage of time for it to occur. Like something has to reach a certain point. I am wondering if maybe one of your memory sticks is bad. Maybe try one at a time and see if one is messed up. Since you ruled out the NVMe, if I were going to take a wild guess at this point, it would be one of the sticks is bad and the system works until data reaches that one stick and then it's over. That's were I would start at this point.

That is a good suggestion. It might show something relevant.

I have tried just 1 ram stick earlier. 
 

UPDATE: I unplugged my USB A to USB A that was going from my monitor to my desktop and suddenly now it’s not crashing. I don’t know what that would do. I’m not sure why. I had my webcam, keyboard, mouse all plugged in to my monitor, and then my monitor USB A plugged in to my motherboard. I eliminated that connection, and suddenly things seem okay. Fingers crossed here. Furmark 4K has been running for 25 minutes now. 

[tps3443]

35 minutes and counting still running Furmark! 

[Mr. Fox]

7 minutes ago, tps3443 said:

UPDATE: I unplugged my USB A to USB A that was going from my monitor to my desktop and suddenly now it’s not crashing. I don’t know what that would do. I’m not sure why. I had my webcam, keyboard, mouse all plugged in to my monitor, and then my monitor USB A plugged in to my motherboard. I eliminated that connection, and suddenly things seem okay. Fingers crossed here. Furmark 4K has been running for 25 minutes now. 

Ah ha! Crossing my fingers and hoping that does it. Very possible. That would be a very good problem to have if you are going to have one.

 

I bet the USB hub inside of the monitor has a problem. I used to have an Acer Predator 1440p monitor that developed a similar problem after a couple of years of working fine. Even with nothing connected to the ports, just the cable from the PC to the monitor, it was frequently freezing and making the Windows hardware disconnect and reconnect sounds. It was randomly changing between working to "code 43" in Device Manager and causing stuttering in games and just web browsing or playing YouTube videos. Once I disconnected it everything worked fine. What clued me in was the Windows sounds alerts for hardware disconnect and reconnect and when I opened Device Manager and saw it happening it was instant confirmation.

[Mr. Fox]

2 minutes ago, tps3443 said:

35 minutes and counting still running Furmark! 

I bet that was the problem. I hope so. 

 

If that fixes it I suspect you are going to find some other things that are mysteriously working better than they used to. You might even find games and YouTube videos seem to run smoother. Failing USB devices have caused me headaches more than once. Sometimes subtle glitches and sometimes outright malfunction of the entire system. I think they must sometimes develop a short internally. I also had a USB webcam that developed an issue that caused weird problems when it was connected.

[tps3443]

19 minutes ago, Mr. Fox said:

I bet that was the problem. I hope so. 

 

If that fixes it I suspect you are going to find some other things that are mysteriously working better than they used to. You might even find games and YouTube videos seem to run smoother. Failing USB devices have caused me headaches more than once. Sometimes subtle glitches and sometimes outright malfunction of the entire system. I think they must sometimes develop a short internally. I also had a USB webcam that developed an issue that caused weird problems when it was connected.

Yep! That was it! A data cable just like a printer would use. It’s like a fire wire on one end, and USB-A on the other end. I ran it from my monitor to my PC. Furmark 4K has been running for 56 minutes now. What a tail chasing experience this has been! 
 

 

 

[Mr. Fox]

52 minutes ago, tps3443 said:

Yep! That was it! A data cable just like a printer would use. It’s like a fire wire on one end, and USB-A on the other end. I ran it from my monitor to my PC. Furmark 4K has been running for 56 minutes now. What a tail chasing experience this has been! 

I am glad it was something "expendable" that is not affecting your core system. Enjoy that nice Zotac 5080 Solid OC GPU now. That was a good one to buy. Are you keeping both or sending back the new one?

[Mr. Fox]

@tps3443 since you just did a clean install, you might as well do it again. The latest release of Rufus nukes a ton of Micro$lop's bundled feces prior to installation.

 

 

https://rufus.ie/en/

Changelog

• Version 4.14 (2026.04.30)
  • Windows User Experience improvements:
    • Add a Quality of Life option, to disable Teams, Outlook, Copilot and other Microsoft forced nuisances
    • Add a Silent installation option, that automatically, and WITHOUT PROMPT, installs Windows on the first detected disk
    • Add an option to copy SkuSiPolicy.p7b to the ESP on installation (please refer to KB5042562 for more info)
    • Add tooltips for all the dialog options
  • Add limited support for El-Torito UEFI image extraction (Mostly for Dell BIOS update ISOs)
  • Improve error report when the user tries to use an image that resides on the target drive
  • Improve the UEFI:NTFS partition label to make the install media more explicit during Windows Setup disk partitioning
  • Improve support for Bazzite and other Fedora derivatives that don't follow EFI conventions
  • Improve detection and exclusion of the new Bitdefender hidden VHDs
  • Improve reporting of GRUB and Isolinux MBRs
  • Fix potential errors during creation of Windows To Go media, due to the use of new versions of bcdboot
  • Fix errors with local accounts that start or end with whitespaces