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

[Papusan]

23 hours ago, Papusan said:

New or newer has to be better. Always, LOOL

 

20 hours ago, Mr. Fox said:

Windows 7 FTW! Modern filth is for silly people with low standards.

Flashed latest bios from Asus. And gues... The newest fancy touch friendly Tile box OS from Redmond struggle with new hardware. Shouldn't it be opposite? Nope. All older OS have no changes but not the latest and greatest from Microsoft. So it has to be true that newer has to be better, LOOL

 

 

@Etern4l I think ring is 45 for default 13900K. It show 50 default ring on Asus board without load.

[Mr. Fox]

13 hours ago, Mr. Fox said:

I am hoping that this new driver corrects my Micro$lop Office text rendering and artifacting issues that make the 6900 XT difficult to use for work on my business machine. That is very frustrating.

 

13 hours ago, Mr. Fox said:

Nope. No change. Didn't fix that issue. Will have to use the 3090 K|INGP|N in this machine because the 6900 XT isn't going to cut it for my work setup.

Brother@PapusanI found the fix for the Micro$lop Office text rendering corruption and artifacts. The new driver did nothing. The fix is called NVIDIA, LOL. Put the 3090 K|INGP|N back into my work computer and everything works correctly now.

[Papusan]

2 hours ago, Mr. Fox said:

 

Brother@PapusanI found the fix for the Micro$lop Office text rendering corruption and artifacts. The new driver did nothing. The fix is called NVIDIA, LOL. Put the 3090 K|INGP|N back into my work computer and everything works correctly now.

Maybe you should enjoy the AMD driver forum on Guru3d.... https://forums.guru3d.com/threads/amd-software-adrenalin-edition-23-3-2-whql-driver-download-and-discussion.447162/

 

But........... for the fix. Send a blessing to Jensen Huang😆

 

 

 

Regarding forum. Yep, even kids have manged better than this from Asus. Even babies. See and learn how not to do it @Reciever @Hiew This is pereverse. 

https://rog-forum.asus.com/t5/promotions-general-discussions/new-forum-appearance-worse-than-previous/m-p/918833#M54477

 

Asus closed down their forum for over a month for this new disaster. 
 

Yep, Asus just follow Microsoft. All is about phones and tablet nowadays.... 

[cylix]

More bad news for Intel on the server side.

Nvidia enters the building 😅

 

https://www.hardwaretimes.com/nvidia-announces-grace-cpu-up-2x-faster-than-amd-epyc-and-intel-xeon-parts-with-better-power-efficiency/

[Mr. Fox]

2 hours ago, Papusan said:

Maybe you should enjoy the AMD driver forum on Guru3d.... https://forums.guru3d.com/threads/amd-software-adrenalin-edition-23-3-2-whql-driver-download-and-discussion.447162/

 

But........... for the fix. Send a blessing to Jensen Huang😆

 

 

 

Regarding forum. Yep, even kids have manged better than this from Asus. Even babies. See and learn how not to do it @Reciever @Hiew This is pereverse. 

https://rog-forum.asus.com/t5/promotions-general-discussions/new-forum-appearance-worse-than-previous/m-p/918833#M54477

 

Asus closed down their forum for over a month for this new disaster. 
 

Yep, Asus just follow Microsoft. All is about phones and tablet nowadays.... 

I guess they had to make their forum match their warranty support and QC.

18 minutes ago, cylix said:

More bad news for Intel on the server side.

Nvidia enters the building 😅

 

https://www.hardwaretimes.com/nvidia-announces-grace-cpu-up-2x-faster-than-amd-epyc-and-intel-xeon-parts-with-better-power-efficiency/

And, it will only cost a half a bazillion USD. Maybe it will have RTX 4090 for integrated graphics.

[Etern4l]

7 hours ago, cylix said:

More bad news for Intel on the server side.

Nvidia enters the building 😅

 

https://www.hardwaretimes.com/nvidia-announces-grace-cpu-up-2x-faster-than-amd-epyc-and-intel-xeon-parts-with-better-power-efficiency/

 

Very interesting, so Nvidia launched an attack on the CPU front... 

144 cores. Respect, although no clocks data given. Would be interesting to see what the pricing and performance of workstation-grade ARM CPUs looks like. I always liked RISC architecture. It was miles ahead of Intel back in the day. I think it almost died for a while mostly because they didn't manage or care to break through to the mainstream (Apple is changing that now). The CPUs were mostly used in things like Silicon Graphics workstations. You can see those machines in office scenes in the original "Jurrasic Park", along with Jeff Goldblum and the other dinosaurs :) 

 

As Intel fans we don't talk about enterprise CPU prices @Mr. Fox lol

[Reciever]

6 hours ago, Papusan said:

Maybe you should enjoy the AMD driver forum on Guru3d.... https://forums.guru3d.com/threads/amd-software-adrenalin-edition-23-3-2-whql-driver-download-and-discussion.447162/

 

But........... for the fix. Send a blessing to Jensen Huang😆

 

 

 

Regarding forum. Yep, even kids have manged better than this from Asus. Even babies. See and learn how not to do it @Reciever @Hiew This is pereverse. 

https://rog-forum.asus.com/t5/promotions-general-discussions/new-forum-appearance-worse-than-previous/m-p/918833#M54477

 

Asus closed down their forum for over a month for this new disaster. 
 

Yep, Asus just follow Microsoft. All is about phones and tablet nowadays.... 

 

This forum has good usability from a mobile standpoint as well but that wasnt really something we sought out specifically. Just happened to be a windfall perk. We chose this software since its been around for quite some time, modular and Hiew was familiar with hosting it. If we had all the time in the world then its likely that something else would've been chosen but since we had 5-6 days to build this up and archive the old forum we went with this.

 

That being said, being unable to edit posts is a bit short sighted. For this forums software its just a check box under the perms for each group. Takes about 30 seconds to change.

 

I tried to view the Asus forums from a link in one of your earlier posts and I can only view whats on screen, no scrolling, no page up / down, arrows do nothing. I guess this is to discourage leachers and encourage account creation? No idea.

 

[Raiderman]

7 hours ago, Mr. Fox said:

 

Brother@PapusanI found the fix for the Micro$lop Office text rendering corruption and artifacts. The new driver did nothing. The fix is called NVIDIA, LOL. Put the 3090 K|INGP|N back into my work computer and everything works correctly now.

Not sure what bug you have with office text rendering. Could you post a link to it? I use office quite a bit, and have not seen any artifacts. Is it on a specific version of office?

 

https://learn.microsoft.com/en-us/office/troubleshoot/settings/office-display-issues

[Mr. Fox]

3 hours ago, Raiderman said:

Not sure what bug you have with office text rendering. Could you post a link to it? I use office quite a bit, and have not seen any artifacts. Is it on a specific version of office?

 

https://learn.microsoft.com/en-us/office/troubleshoot/settings/office-display-issues

It is described fairly well in these posts.

 

https://answers.microsoft.com/en-us/msoffice/forum/all/word-2019-not-rendering-text-properly-as-you-type/fb665d6d-70a2-463b-943d-427b83239fbe

 

https://answers.microsoft.com/en-us/outlook_com/forum/all/text-disappears-or-distorted-while-typing-on/91543518-9384-4a50-bc88-889772c33938

 

Office 365 and Office 2021 Pro Plus, under both Windows 10 and 11.

 

As you are typing some of the characters you have just added (3 to 5 characters generally) will randomly vanish and reappear, and intermittently there will be artifacting and strange looking glitches that are like a tiny image thumbnail of text near the area of editing. It does not affect the entire page of text, only in the area where you are editing/adding text. If you stop typing the chunks of missing text will stay hidden or missing, and text already present will sometimes shift left or right and overlay the area where you are typing, but adding a keystroke or two the text will become visible again. It is not gone, but the rendering of what is there is goofed up. When you are a fast typist like me, this is happening very rapidly, so it is unsettling and distracting. 

 

It only happens on my work computer with the 6900 XT installed. I have never experienced it ever before. It does not happen with the 6900 XT installed on the other system. My work computer has three 144 Hz 1080p displays.  My other system has a 165Hz 1440p display.

 

I have tried all of the suggestions like disabling Office graphics acceleration and enabling Clear Type (both of which I always do anyway). It is definitely caused by the 6900 XT or AMD drivers. It does not happen intermittently. It always occurs, at all times. Enabling Eyefinity did not make it better or worse. If I am sharing my desktop in a video chat/meeting, those viewing my screen do not see it happening, so it is obviously an issue with rendering on the local machine, and it was not noticed using any other software applications other than Word, Excel and Outlook. It does not occur except while editing content. Opening a document in Word or viewing an email in Outlook and scrolling through pages everything is fine until you begin editing or adding new content. It is not related to overtyping or toggling insert, as it only affects or corrupts the view of existing text already entered and present to the left of where new content is being added.

 

It is very weird and frustrating, and it severely interferes with getting work done. Kind of a bummer, since it is the system that allowed me to use Windows 7 with the 6900 XT. But, that makes me wonder if maybe the same underlying issue is what prevents my Windows 7 desktop from loading correctly on the other system., but manifests itself in a different way. If I install the 6900 XT in the other system and boot into Windows 7 I can hear the Windows login sound, and System notification sounds like the USB devices loading, and I can log out or choose to restart blindly by hitting the Windows key, right arrow twice and hit enter, but only the Windows 7 Welcome screen with the black background and colored logo appears on my screen. It is like the Welcome screen never leaves and it is covering up the desktop GUI.

 

I tried this and it did not resolve it. But, I do think it is a driver or vBIOS issue.

[Mr. Fox]

  

18 hours ago, Mr. Fox said:

I think I am going to try an experiment with the 6900 XT Brother @Rage Set.

 

One of the nice things about the Dark motherboard is it has DIP switches to disable PCIe ports. And, the PCIe x16 slots are both capable of running full x16 speeds as long as both are not in use. Most mobos can only run the second slot at x8 even when the first one is vacant. You can install your GPU in the bottom slot and leave the top one empty and it still runs at x16. You can also disable them in the BIOS, but that won't let me do the experiment.

 

I think that I am going to try running an NVIDIA GPU in the top slot and the 6900 XT in the bottom slot. Boot into Windows 7 using the NVIDIA GPU, then turn on the AMD card, install the drivers and then use the DIP switch to disable the NVIDIA card while in Windows and see how the 6900 XT behaves when it doesn't have to go through the booting phase where it stops working.

 

I might also try running both at x8 simultaneously to see what kind of nonsense surfaces.

 

Might try that this weekend and see what happens.

This is a no-go. Maybe it used to be true with a Dark mobo, but not any more. Probably was with X299 since it had such a crazy count of available PCIe lanes.

 

Last night I moved the GPU from the top to the bottom x16 slot and it only runs at x8 speed. I tried disabling the M.2 NVMe slots tied to the CPU lanes and physically turning off PCIe slot #1 using the DIP switch and it still only runs at x8, so experiment cancelled.

  

15 hours ago, Papusan said:

This guy absolutely nailed it with the "Thanks for confirming the redesign is for the Stupid devices whilst alienating the people that bought your PC hardware..." My sentiments exactly. Not just about ASUS forums, but we are surrounded by the same crap at every turn. I am really fed up with that absolutely idiotic approach to technology. The people making these decisions are morons that need to be fired because they suck at their jobs. I don't give a crap whether or not the web page or forum is phone friendly. It does not need to be. People that want that should spend their time trolling forums for smartphone/tablet lovers and stop screwing things up for everyone else. This is like having a guest that you invite over for dinner pooping on your living room floor.

[Rage Set]

26 minutes ago, Mr. Fox said:

  

This is a no-go. Maybe it used to be true with a Dark mobo, but not any more. Probably was with X299 since it had such a crazy count of available PCIe lanes.

 

Last night I moved the GPU from the top to the bottom x16 slot and it only runs at x8 speed. I tried disabling the M.2 NVMe slots tied to the CPU lanes and physically turning off PCIe slot #1 using the DIP switch and it still only runs at x8, so experiment cancelled.

 

 

Have you contacted EVGA about this? I mean I know the official stance is Windows 7 is no longer supported but i am sure they may point you in the right direction. Obviously, they know the majority of their Dark buyers are overclockers and Win7 is something a lot of them use for certain benchmark suites. 

 

The reason why I saying this...https://www.asrock.com/Graphics-Card/AMD/Radeon RX 6900 XT OC Formula 16GB/index.us.asp It states on their website, Not compatible with Windows 7. However, it appears their Phantom Gaming 6900/6950XT are. So you are correct that there is something in the vBIOS preventing use in Windows 7. I don't know why their best overclocking GPU can't be used in Windows 7 but their cheaper mainstream model can. Doesn't add up. 

 

If you want, we can cancel our deal. I don't want you to be stuck in this deal because Asrock made a bonehead decision. 

[Mr. Fox]

Just now, Rage Set said:

 

Have you contacted EVGA about this? I mean I know the official stance is Windows 7 is no longer supported but i am sure they may point you in the right direction. Obviously, they know the majority of their Dark buyers are overclockers and Win7 is something a lot of them use for certain benchmark suites. 

 

The reason why I saying this...https://www.asrock.com/Graphics-Card/AMD/Radeon RX 6900 XT OC Formula 16GB/index.us.asp It states on their website, Not compatible with Windows 7. However, it appears their Phantom Gaming 6900/6950XT are. So you are correct that there is something in the vBIOS preventing use in Windows 7. I don't know why their best overclocking GPU can't be used in Windows 7 but their cheaper mainstream model can. Doesn't add up. 

 

If you want, we can cancel our deal. I don't want you to be stuck in this deal because Asrock made a bonehead decision. 

No, I knew this 100% going into our deal. Not an issue, or I would not have gone through with it. But, thank you for offering.

 

Same problem with 4090. The demon-possessed TechNazis are on a seek and destroy mission with orders to do whatever it takes to ruin any possibility of using Windows 7. It is only a matter of time until continuing to use it is made impossible by the self-anointed technology SS. I am going to have to move on and forget about Windows 7 very soon because I won't have another option available to me.

 

I know that I am not alone in this disappointment. This is why I am spending more time with Linux than ever before. It's not because I want to... the writing in on the wall. They might win this battle, but I want to win the war. Winduhz 10+ is for losers and sheeple, not for winners and masters of their own destiny.

 

I have thought about reaching out to EVGA. I went back to the oldest BIOS for 13th Gen and the problem is the same. Apparently, here are no other Z690 or Z790 Dark owners that have attempted to run Windows 7 with this OC Formula GPU because there is not a word of it anywhere in EVGA forum. I may still reach out to them and see what they can tell me.
 

I found the info on the ASROCK web page in your link as soon as I discovered it would not work. I started looking for answers and came across that within the first few minutes of searching.

[Papusan]

12 minutes ago, Rage Set said:

Have you contacted EVGA about this? I mean I know the official stance is Windows 7 is no longer supported but i am sure they may point you in the right direction. Obviously, they know the majority of their Dark buyers are overclockers and Win7 is something a lot of them use for certain benchmark suites.

This is on their table. It's the EVGA firmware that don't play well with all graphics cards. There is is even oc'ers that had to sell their Dark board or their new graphics cards to have a chance to play with their new siny +1500$ cards. They have to fix this. Not the consumers.

[Mr. Fox]

13 minutes ago, Papusan said:

This is on their table. It's the EVGA firmware that don't play well with all graphics cards. There is is even oc'ers that had to sell their Dark board or their new graphics cards to have a chance to play with their new siny +1500$ cards. They have to fix this. Not the consumers.

This is only an assumption. It could be accurate, but we have no way of knowing. And, they may not even have the ability to fix it, especially if it involves solving someone else's screw up. If they agreed to try to fix it, then I would probably have to send them the GPU to use for solving the problem because fixing something that is broken that they cannot see or touch would be hit or miss.

 

ASROCK says it is not compatible, so that tells me it was something deliberate. Just because it works on the Strix Z690-E does not mean it would work on a different model of ASUS mobo. It is probably an ACPI implementation thing, but it could be something that is not obvious.

 

And, as sad as it is to think this way, there is no legitimate business case for them to make Windows 7 work if only 1 out of 1,000 people that buy their products are trying to use Windows 7. If there were lots of people that cared, they would be blowing up EVGA forum and EVGA would be reaching out to other OEMs to find out what they did so they can fix their mistakes, or work with them to correct the problems.

[Rage Set]

6 minutes ago, Mr. Fox said:

This is only an assumption. It could be accurate, but we have no way of knowing. And, they may not even have the ability to fix it, especially if it involves solving someone else's screw up. If they agreed to try to fix it, then I would probably have to send them the GPU to use for solving the problem because fixing something that is broken that they cannot see or touch would be hit or miss.

 

ASROCK says it is not compatible, so that tells me it was something deliberate. Just because it works on the Strix Z690-E does not mean it would work on a different model of ASUS mobo. It is probably an ACPI implementation thing, but it could be something that is not obvious.

 

And, as sad as it is to think this way, there is no legitimate business case for them to make Windows 7 work if only 1 out of 1,000 people that buy their products are trying to use Windows 7. If there were lots of people that cared, they would be blowing up EVGA forum and EVGA would be reaching out to other OEMs to find out what they did so they can fix their mistakes, or work with them to correct the problems.

 

Hmmm, I forgot it worked on the Asus board. So @Papusan might be right. Perhaps, Asrock placed some kind "flag" on the card. Asus may not choose to see this flag or didn't implement a check...whereas EVGA does. So I am 80% confident it is the ACPI implementation the more I research this issue on my own.  

[Papusan]

1 hour ago, Mr. Fox said:

This is only an assumption. It could be accurate, but we have no way of knowing. And, they may not even have the ability to fix it, especially if it involves solving someone else's screw up. If they agreed to try to fix it, then I would probably have to send them the GPU to use for solving the problem because fixing something that is broken that they cannot see or touch would be hit or miss.

 

ASROCK says it is not compatible, so that tells me it was something deliberate. Just because it works on the Strix Z690-E does not mean it would work on a different model of ASUS mobo. It is probably an ACPI implementation thing, but it could be something that is not obvious.

1 hour ago, Rage Set said:

 

Hmmm, I forgot it worked on the Asus board. So @Papusan might be right. Perhaps, Asrock placed some kind "flag" on the card. Asus may not choose to see this flag or didn't implement a check...whereas EVGA does. So I am 80% confident it is the ACPI implementation the more I research this issue on my own.  

Dosn't matter if you chose the older Asus Z690 or newer Z790 boards. It just works. Probably because Asus is a bigger company and with more oc'ers that help with the firmware. Thats EVGA problem. They are too small. They lack resources and HW for in-house testing. This is also the reason they always come late to the party with the new and shiny.

 

And don't forget... We can even install older Win XP on a few OC motherboards.

 

 

See... Probably the reason Intel have hurry throw out Raptor refresh early this year. They aren't finished with next gen chipset and processors so they was forced to push out a refresh of the refresh. Buying a KS now so late in the 13th gen lifespan is for me stupid. Buy the refresh or just use what you already have.

Gigabyte Claims Next-Gen Ryzen CPUs Arrive on AM5 This Year

[johnksss]

https://www.microcenter.com/search/search_results.aspx?N=4294966937&NTK=all&sortby=pricehigh

[electrosoft]

4060 mobile not worth the upgrade over 3060....

 

So basically this generation the 4080/4090 mobile (kind of like the desktop in a sense) are the only meaningful upgrades this time around.

 

 

[Raiderman]

A bit off topic, but what does everyone (if any) use to disable Windows Defender, and smart screen in windows 10?

[Mr. Fox]

3 hours ago, johnksss said:

https://www.microcenter.com/search/search_results.aspx?N=4294966937&NTK=all&sortby=pricehigh

Nice to see good stock at the recommended level of excessive (non-scalper) overpricing. I just snagged another Suprim (non-Liquid) and it seems like it might be a great silicon sample. Same price as Microcenter, too. Maybe I will break out the portable AC unit this weekend and force feed it some frosty air. It will be easy with this case. I can just drape a blanket over it and the AC unit so the cold air has no other place to go. Between the 6900 XT and the 4090 I should have enough to keep me busy with benching for a long time now.

 

 

 

 

 

[Mr. Fox]

3 minutes ago, Raiderman said:

A bit off topic, but what does everyone (if any) use to disable Windows Defender, and smart screen in windows 10?

I use a script to totally remove Defender. Doing that provides a secondary benefit of making most Windows Updates fail, so it ticks both boxes for me. I don't want or care about Micro$lop updates. The .NET stuff will update, but the service stacks and feature updates are generally incapable of installing even manually, so I really love that.

 

I use ESO to kill and disable all worthless  Services.

 

 

 

 

Remove Defender.zip

[Raiderman]

1 minute ago, Mr. Fox said:

I use a script to totally remove Defender. I use ESO to kill and disable all worthless  Services.

 

 

 

 

Remove Defender.zip 4.86 kB · 0 downloads

Oh crap, that did everything I wanted, and fast too...lol Thank you!

[Etern4l]

A few delidding questions guys:

1. Has anyone directly compared LM/Conductonaut with PTM7950/7958 between the die and the IHS?

2. Rockit recommends "seasoning" the copper IHS with their Quicksilver gallium paste (I suppose LM would work as well, albeit more expensive), ideally for a week. This is supposed to prevent the die and the IHS getting welded together, as well as improve the thermal performance. Have you done this, and if so, how long for?

3. What's your recommended re-lidding glue/technique? I've come across quite a few:

a) No glue - sounds like a bad idea as it would be difficult to keep the IHS still on the die

b) 4 small dots of super glue on  the PCB, some people say super glue has low flash point and can damage the PCB

c) 4 small dots of silicone RTV glue on the PCB

d) 4 small dots of silicone RTV glue applied externally on the corners of the IHS (wouldn't that interfere with the contact frame?)

Thanks for any thoughts and tips.

[Mr. Fox]

41 minutes ago, Etern4l said:

A few delidding questions guys:

1. Has anyone directly compared LM/Conductonaut with PTM7950/7958 between the die and the IHS?

2. Rockit recommends "seasoning" the copper IHS with their Quicksilver gallium paste (I suppose LM would work as well, albeit more expensive), ideally for a week. This is supposed to prevent the die and the IHS getting welded together and also improves the thermal performance. Have you done this, and if so, how long for?

3. What's your recommended re-lidding glue/technique? I've come across quite a few:

a) No glue - sounds like a bad idea as it would be difficult to keep the IHS still on the die

b) 4 small dots of super glue on  the pcb, some people say super glue has low flash point and can damage the PCB

c) 4 small dots of silicone RTV glue on  the pcb

d) 4 small dots of silicone RTV glue applied externally on the corners of the IHS (wouldn't that interfere with the contact frame?)

Thanks for any thoughts and tips.

1. No. Never had a reason to. It will likely be thicker than liquid metal and more resistant to heat transfer and also physically less thermally conductive as a matter of pure science. The key to it working correctly is getting all of the old rubber off and (if necessary) sanding the underside of the lip to make sure it is not touching the PCB when set on top of the die. It should spin on the die without touching the CPU PCB. Apply liquid metal to both surfaces during assembly or it won't work well.

2. Use liquid metal and bake it in the oven for an hour or so. You could use the Quicksilver as well if you have leftover. Either one will work. Never have seen the IHS get stuck to the die before. Not saying it couldn't happen, only that I am unaware of it based on personal experience. The reason to do it is to stop the copper from absorbing the liquid metal. You let it absorb it in the oven before installing it. Then the liquid metal will last a very long time because it is not being wicked away into the copper.

3. Leaving it loose is best, just be careful if you are using the stock ILM and not let it slide. The ONLY thing to be concerned about with leaving it loose is dislodging the SMDs outside of the IHS. Otherwise, no need for concern. Installing or removing the CPU in a chassis that is vertical makes it more dangerous to be left loose due to human error. On a horizontal installation you just have to pay attention and not allow it to move around until you have it secured. I have used one small dot under each "wing" (one left and one right) of superglue GEL (the gel only, not regular super glue). I have also used one small dot of RTV silicon under each wing.  No need for 4 spots to be glued. The only reason to do that is to keep it from moving around when installing or removing the CPU from the socket. Otherwise, reattaching it is a waste of time and just creates inconvenience later. Do not apply it on the outside corners. I have seen Dufus Jufus do that. Silly. Just put some nail polish on the SMDs outside of the IHS and you're good to go. If you use super glue GEL, it takes very little. Let it dry for 20-30 minutes under compression with the relid tool. You can release it later by soaking the CPU in acetone for 15-20 minutes. It will disolve the super glue and not harm the CPU.

 

Not the "ultra" just the regular control gel. It does not run and it is easy to control the amount applied. And, it is not as strong.

https://www.amazon.com/Loctite-234790-6-Control-4-Gram-Bottle/dp/B0006HUJCQ/ref=sr_1_1_sspa?crid=R0WPWSAHAWK0&th=1

 

 

Apply one tiny drop under each IHS wing. Smaller than the head of a stick pin. Like 0.5-1.0mm diameter droplet. Really tiny so it does not squish out from under the wings on the sides. It only needs to be large enough to tack it in place and you want the bond to be weak/fragile.

 

Edit: @Etern4l- bake the liquid metal on BOTH SIDES of the IHS - forgot to clarify that. Same reason, as given, but don't do only the underside. After baking, use the scuff pad that normally comes with liquid metal and polish it with Flitz and it will look like chrome/nickel.

Edited March 24, 2023 by Mr. Fox
added comment about baking

[Papusan]

On 3/23/2023 at 7:11 PM, cylix said:

Lol Linus got hacked by crypto scammers and channel is for now banned 😂

Maybe Linus and his workers should be more careful what they click on? 

 

"We can't blame YouTube for someone getting infected with malware, but we can lean on them to make the hacker's job a lot more complicated".