VOGONS


Reply 20 of 32, by gonzo

User metadata
Rank Member
Rank
Member

Successful delidding of a Ryzen 5 5600G having this ugly metal like the Phenom 980, too: SUCCESSFUL DELIDDING of AMD-CPUs (so. 754-AM4) by modding of an Intel-Core-delidding-tool available on the market
I did the test just because to see, if a Ryzen can be delidded at this way. I did not expect to find the same metal under the lid. The CPU-core goes NOT broken.

I LOVE CPUs RUNNING IN [GonzoHz]

Reply 21 of 32, by Nemo1985

User metadata
Rank Oldbie
Rank
Oldbie

The question is, under the ihs what is going to be used? Some other thermal paste? Then in years it will dry again or liquid metal? Is it safe to use between core and ihs?

Reply 22 of 32, by The Serpent Rider

User metadata
Rank l33t++
Rank
l33t++

Liquid metal is not safe to use with pure copper. If IHS has exposed copper inside, don't try.

I must be some kind of standard: the anonymous gangbanger of the 21st century.

Reply 23 of 32, by AlexZ

User metadata
Rank Oldbie
Rank
Oldbie

You would just use MX-6. Since the die is quite large, you could go bare die. It would work as long as the cooler doesn't need the original plastic frame and you can apply pressure from both sides simultaneously.

Liquid metal is rarely needed. I have AMD Ryzen 9 7945HX in laptop and I got rid of the liquid metal as it has too short lifespan. It may give you 2'C lower temperature, but that typically isn't enough. A more powerful cooling system is more useful than liquid metal.

Pentium III 900E,ECS P6BXT-A+,384MB,GeForce FX 5600, Voodoo 2,Yamaha SM718
Athlon 64 3400+,Gigabyte GA-K8NE,2GB,GeForce GTX 275,Audigy 2 ZS
Phenom II X4 955,Gigabyte GA-MA770-UD3,8GB,GeForce GTX 780
Vishera FX-8370,Asus 990FX,32GB,GeForce GTX 980 Ti

Reply 24 of 32, by gonzo

User metadata
Rank Member
Rank
Member
The Serpent Rider wrote on Today, 13:17:

Liquid metal is not safe to use with pure copper. If IHS has exposed copper inside, don't try.

That's correct! To make it easy, you can delid another CPU (e.g. a much cheaper or a defective one) from the same generation having just thermal grase and without pure copper on the lid, and use this lid instead of the original lid. Of course, in this case, the CPU-model shown on the lid will be wrong...

Last edited by gonzo on 2025-10-06, 14:25. Edited 2 times in total.

I LOVE CPUs RUNNING IN [GonzoHz]

Reply 25 of 32, by gonzo

User metadata
Rank Member
Rank
Member
AlexZ wrote on Today, 13:32:

.....liquid metal as it has too short lifespan. It may give you 2'C lower temperature....

To be honest, I do not agree - liquid metal gives you compared to thermal grease about 15-20 degrees Celsius less temperature. But yes, a good cooling system is still needed.

I LOVE CPUs RUNNING IN [GonzoHz]

Reply 26 of 32, by wierd_w

User metadata
Rank Oldbie
Rank
Oldbie

I am reading that indium will alloy with gallium if both are made liquid first, and that the resulting alloy has an even lower melting point than pure gallium.

That suggests the die can maybe be cleaned with a tiny amount of liquid metal conduction medium (which is mostly gallium), and gentle heating at a hot air workstation.

Once it's alloyed, the Indium-Gallium complex is supposed to stay liquid at room temp, so a suction tube, and a quality paper towel should get you a nice clean die afterward.

At least in theory.

Indium is not really safe to handle though. Please use appropriate precautions. (While the metal form is generally nontoxic, any salts made from it, through exposure to active cations, are quite unpleasant. )

Reply 27 of 32, by Ozzuneoj

User metadata
Rank l33t
Rank
l33t

This is a serious question, not trying to be negative... but, what is the purpose of the delidding that is going on in this thread? Especially with newer CPUs, Ryzen, etc. there is almost nothing gained even by lowering temperatures. These chips are pushed near the absolute maximum from the factory and the binning processes have become so good that there is very little left for enthusiasts to do aside from undervolting with curves\PBO2 tuning.

Here is a good example:
https://www.youtube.com/watch?v=y_jaS_FZcjI
Der8auer delidded a 7900X and used liquid metal for direct die cooling and managed a 3% performance improvement in Cinebench with all cores pegged at 5.5Ghz (a minor increase over stock cooling) despite lowering temps 18-20C at peak load. The crazy thing is that the 7000 series is known for deliberately pushing chips to high temperature limits to achieve the highest clock speeds, so if any modern architecture should benefit from cooling mods like this, you'd think this one would. Of course, you can manually jack up the voltages more now that there is more thermal head room. How long these chips will last in those circumstances, however, I don't know.

I can use negative core offsets on my 5800X3D to bring temps and power consumption down while having a slight gain in performance and I can run all of the fans in my PC so that they are inaudible at idle and barely audible during a heavy gaming load. No hardware modifications were needed to achieve this. Just good air cooling and somewhat decent thermal paste.

If the goal is to try to preserve these chips by keeping them at lower temperatures, I guess that's an admirable goal, I just don't know if there will be real world benefits. Personally, I'd trust an untouched CPU to still work properly 20-30 years from now versus one that was delidded part way through its lifespan.

EDIT: It's hard to convey the tone of this post, but it is not combative or trying to tell you not to delid CPUs. I am just genuinely curious what the goal is.

Now for some blitting from the back buffer.

Reply 28 of 32, by swaaye

User metadata
Rank l33t++
Rank
l33t++

The last one I bothered to delid was a 8600K and only because Intel had switched from solder to thermal paste which is just awful if you want to try to overclock. I replaced the paste with Coollaboratory Liquid Ultra. Though yes even with a very substantial drop in load temp it still hit a wall at around 4.8 GHz. It is kind of silly to even try to push further because the power consumption multiplies. And with Alder Lake and newer the heat flux is more or less unmanageable.

Last edited by swaaye on 2025-10-06, 18:53. Edited 1 time in total.

Reply 29 of 32, by nfraser01

User metadata
Rank Member
Rank
Member
Ozzuneoj wrote on Yesterday, 05:02:

My googles are failing me. Do you know of any articles or posts that delve into this and show results of before and after applying new TIM under the IHS of an old CPU to lower temps? All I'm finding are people either delidding new (unsoldered) chips to apply liquid metal under the IHS or running them with a bare die. Both are just done to decrease temps for extreme overclocking.

Athlon 64 overheat issue

Reply 30 of 32, by The Serpent Rider

User metadata
Rank l33t++
Rank
l33t++

Yeah, desktop Intel CPUs after Sandy Bridge had paste for a while, which is a pain to deal with after using CPU for 5+ years.

I must be some kind of standard: the anonymous gangbanger of the 21st century.

Reply 31 of 32, by swaaye

User metadata
Rank l33t++
Rank
l33t++

Oh yeah a few years ago I had a Socket 939 Opteron 180 (equivalent to an Athlon 64 X2) that had thermal paste that was not performing anymore and it would overheat very easily. I don't know how common that might be but I have a number of other Socket 939 CPUs that are ok.

Reply 32 of 32, by Grem Five

User metadata
Rank Member
Rank
Member

I just retired my I7 8700k this year and even though I thought about deliding it but for over 7+ years it never suffered from any temp increase and had no problem running at 5.0 to 5.1 Ghz. That machine had daily use for the entire time with the same original AIO I had on it too.