• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Noctua Presents NA-TPG1 Thermal Paste Guard for AMD AM5

Joined
Apr 30, 2008
Messages
4,894 (0.81/day)
Location
Multidimensional
System Name Boomer Master Race
Processor Intel Core i5 12600H
Motherboard MinisForum NAB6 Lite Board
Cooling Mini PC Cooling
Memory Apacer 16GB 3200Mhz
Video Card(s) Intel Iris Xe Graphics
Storage Kingston 512GB SSD
Display(s) Sony 4K Bravia X85J 43Inch TV 120Hz
Case MinisForum NAB6 Lite Case
Audio Device(s) Built In Realtek Digital Audio HD
Power Supply 120w External Power Brick
Mouse Logitech G203 Lightsync
Keyboard Atrix RGB Slim Keyboard
VR HMD ( ◔ ʖ̯ ◔ )
Software Windows 11 Home 64bit
Benchmark Scores Don't do them anymore.
So what? Let it fill the space in between. Why bother at all? All my CPUs have residue on the sides of the heatspreader, ist doent not matter at all, as long as the upper surface is clean when applying new grease.

I guess this is for those folks who RGB-illuminate their cases and put their GPUs in the window to compensate for their small p... -I mean power supplies!
Jesus christ dude, projecting much! If someone wants to have a clean CPU, let them have a clean CPU, wtf is the big deal. God this site is full of negative egotistical dweebs.
 
Joined
Jun 20, 2022
Messages
302 (0.35/day)
Location
Germany
System Name Galaxy Tab S8+
Processor Snapdragon 8 gen 1 SOC
Cooling passive
Memory 8 GB
Storage 256 GB + 512 GB SD
Display(s) 2.800 x 1.752 Super AMOLED
Power Supply 10.090 mAh
Software Android 12
If you need a guard when applying TIM, 99% you do it the wrong way.

I would agree it use as caution when using Liquid metal... I don't know how from a swab can something drip off... but anyways.

Actually this is a problem even professionals (Igor from igorslab for example) recognized already. This went as far as applying insulating tape to prevent thermal paste getting into the gaps. A company making profit out of such a situation is not new. Did you write something accordingly when intel had bending problems with it's new socket, because some third party companies also had a solution on hand quite fast?^^
 
Joined
Sep 2, 2022
Messages
89 (0.11/day)
Location
Italy
Processor Intel Core i7-5960X
Motherboard MSI X99S SLI PLUS
Memory Corsair Vengeance LPX DDR4 4x8GB
Video Card(s) Gigabyte GTX 1070 TI 8GB
Storage Sandisk Ultra II 240GB + HDD
Display(s) Benq GL2480 24" 1080p 75 Hz
Power Supply Seasonic M12II 520W
Mouse Logitech G400
Software Windows 10 LTSC
They have SMD's there now, and clearly had nowhere else to put em
View attachment 264386
They could have designed a regular squared IHS and cover all the surface below it, just leaving 0.5 mm of free space over the SMDs.
Even if these need some sort of cooling, that space could have been enough.

Also, according to the video of Derbauer, they made it thicker than the old ones, and this is the cause of the bad temps. It's a bad overall design.
 
Joined
Jun 21, 2021
Messages
3,093 (2.51/day)
System Name daily driver Mac mini M2 Pro
Processor Apple proprietary M2 Pro (6 p-cores, 4 e-cores)
Motherboard Apple proprietary
Cooling Apple proprietary
Memory Apple proprietary 16GB LPDDR5 unified memory
Video Card(s) Apple proprietary M2 Pro (16-core GPU)
Storage Apple proprietary onboard 512GB SSD + various external HDDs
Display(s) LG UltraFine 27UL850W (4K@60Hz IPS)
Case Apple proprietary
Audio Device(s) Apple proprietary
Power Supply Apple proprietary
Mouse Apple Magic Trackpad 2
Keyboard Keychron K1 tenkeyless (Gateron Reds)
VR HMD Oculus Rift S (hosted on a different PC)
Software macOS Sonoma 14.7
Benchmark Scores (My Windows daily driver is a Beelink Mini S12 Pro. I'm not interested in benchmarking.)
They could have designed a regular squared IHS and cover all the surface below it, just leaving 0.5 mm of free space over the SMDs.
Even if these need some sort of cooling, that space could have been enough.

Also, according to the video of Derbauer, they made it thicker than the old ones, and this is the cause of the bad temps. It's a bad overall design.

It's given that AMD tried a number of IHS designs during development including the IHS design from the AM4 generation and probably some modified designs of those from the competition (like Intel products).

It's also probable that the people who gave the final approval for this design have a couple hundred years of semiconductor design experience combined, with advanced degrees from some of the world's best engineering universities. The AM5 IHS design wasn't from some 22-year old snot-nosed intern from Foothill Junior College.

It is possible that it's a bad design. That would bode ominously for AMD's future as a semiconductor company though.
 
Joined
Apr 9, 2021
Messages
21 (0.02/day)
Location
Los Angeles. California
The thickness of the IHS is not why the CPU runs at 95c. The chip is just hot by design. Any CPU with direct die water cooling runs much cooler vs a chip with an IHS. If Zen 4 had a full square IHS that was half the thickness it wouldn't have significantly cooler temps. Zen 4 increased IHS thickness plus the larger backplates on X670 motherboards are likely to protect the 5nm CCDs pressure or torsion related damage
 
Joined
Jun 21, 2021
Messages
3,093 (2.51/day)
System Name daily driver Mac mini M2 Pro
Processor Apple proprietary M2 Pro (6 p-cores, 4 e-cores)
Motherboard Apple proprietary
Cooling Apple proprietary
Memory Apple proprietary 16GB LPDDR5 unified memory
Video Card(s) Apple proprietary M2 Pro (16-core GPU)
Storage Apple proprietary onboard 512GB SSD + various external HDDs
Display(s) LG UltraFine 27UL850W (4K@60Hz IPS)
Case Apple proprietary
Audio Device(s) Apple proprietary
Power Supply Apple proprietary
Mouse Apple Magic Trackpad 2
Keyboard Keychron K1 tenkeyless (Gateron Reds)
VR HMD Oculus Rift S (hosted on a different PC)
Software macOS Sonoma 14.7
Benchmark Scores (My Windows daily driver is a Beelink Mini S12 Pro. I'm not interested in benchmarking.)
The thickness of the IHS is not why the CPU runs at 95c. The chip is just hot by design.

This brings up a point that some people here just don't understand. AMD could have programmed this chip to run at 90, 85, 80 °C, etc.

They decided to let it run at 95 °C. And they probably tested various thermal ceilings in the lab.

It's worth pointing out that a lot of mobile CPU designs also run around this temperature. I have an Intel Core i7 in my Mac mini 2018 and during Handbrake encodes it runs around 95 °C on average. This is nothing new.
 
Joined
Apr 9, 2021
Messages
21 (0.02/day)
Location
Los Angeles. California
This brings up a point that some people here just don't understand. AMD could have programmed this chip to run at 90, 85, 80 °C, etc.

They decided to let it run at 95 °C. And they probably tested various thermal ceilings in the lab.

It's worth pointing out that a lot of mobile CPU designs also run around this temperature. I have an Intel Core i7 in my Mac mini 2018 and during Handbrake encodes it runs around 95 °C on average. This is nothing new.
Yup. I have a Dell XPS with an i9-12900HK and it runs between 90-100 C at load all day long
 
Joined
Nov 2, 2020
Messages
1,332 (0.91/day)
Location
Tel Fyr
System Name Purple Haze | Vacuum Box
Processor AMD Ryzen 7 5800X3D (-30 CO) | Intel® Xeon® E3-1241 v3
Motherboard MSI B450 Tomahawk Max | Gigabyte GA-Z87X-UD5H
Cooling Dark Rock 4 Pro, P14, P12, T30 case fans | 212 Evo & P12 PWM PST x2, Arctic P14 & P12 case fans
Memory 32GB Ballistix (Micron E 19nm) CL16 @3733MHz | 32GB HyperX Beast 2400MHz (XMP)
Video Card(s) AMD 6900XTXH ASRock OC Formula & Phanteks T30x3 | AMD 5700XT Sapphire Nitro+ & Arctic P12x2
Storage ADATA SX8200 Pro 1TB, Toshiba P300 3TB x2 | Kingston A400 120GB, Fanxiang S500 Pro 256GB
Display(s) Mi 2K Gaming Monitor 27", AOC 24G2U
Case Modded MS Industrial Titan II Pro RGB | Heavily Modded Cooler Master Q500L
Audio Device(s) Audient iD14 MKII, Adam Audio T8Vs, Bloody M550, HiFiMan HE400se, Tascam TM-80, DS4 v2
Power Supply Rosewill Capstone 1000M | Enermax Revolution X't 730W (both with P14 fans)
Mouse Logitech G305, Bloody A91, Amazon basics, Logitech M187
Keyboard Redragon K530, Bloody B930, Epomaker TH80 SE, BTC 9110
Software W10 LTSC 21H2
This IHS design is a compromise that shouldn't happened. They've (AMD) earned so much consumer points for compatibility already, might've tried something new for a change. I mean, it's a completely new platform, all parts (CPU, MB, RAM) are already way too much expensive, so why the hell did they bogged down the thermals just for the sake of a damn compatibility? Someone has the money for the AM5 upgrade, but will cheap out on a CPU cooler by reusing the old one? WTH AMD!?
 
Joined
Sep 2, 2022
Messages
89 (0.11/day)
Location
Italy
Processor Intel Core i7-5960X
Motherboard MSI X99S SLI PLUS
Memory Corsair Vengeance LPX DDR4 4x8GB
Video Card(s) Gigabyte GTX 1070 TI 8GB
Storage Sandisk Ultra II 240GB + HDD
Display(s) Benq GL2480 24" 1080p 75 Hz
Power Supply Seasonic M12II 520W
Mouse Logitech G400
Software Windows 10 LTSC
AMD may have designed the chip to run hot, and they may be cool with that.
But the chip passed from 90°C to 70°C after being delidded, using an AIO and the stock frequencies and voltage. This mean that they could have done better. One doesn't need to be an engineer to understands the benefits of having that chip running that cool at stocks, instead of 95°C. And this difference in temps is caused by the IHS.
 
Joined
Jun 18, 2021
Messages
2,537 (2.06/day)
If Zen 4 had a full square IHS that was half the thickness it wouldn't have significantly cooler temps. Zen 4 increased IHS thickness plus the larger backplates on X670 motherboards are likely to protect the 5nm CCDs pressure or torsion related damage

Pressure or torsion related damage has nothing to do with it, like nothing at all, that's what the extra layers and finishing on top of the silicon is there for. There's not really a difference in pressure/torsion risk between a 5nm or a 7nm, they both are packaged with similar processes after the lithography is done. A 5nm chip is not more fragile than a 7nm.

This brings up a point that some people here just don't understand. AMD could have programmed this chip to run at 90, 85, 80 °C, etc.

Yes, that's true, but a thicker IHS will still present a higher thermal resistance than a smaller one. Higher thermal resistence, less heat transfer, higher temp for the same power (the -20ºc or whatever result from delided also shows this, if that was even a question).

They could have set a smaller or higher max temp regardless of the ihs thickness, but they could also get a lower stable temperature for the max power they set with a smaller ihs. One doesn't contradict the other.

Someone has the money for the AM5 upgrade, but will cheap out on a CPU cooler by reusing the old one?

It's not even like mount adapter brackets aren't a thing, even Intel with a very different footprint is using them (though keeping the same hole pattern is also an option), I don't think this was a necessary compromise. Oh well, too late now
 
Last edited:
Joined
Jun 21, 2021
Messages
3,093 (2.51/day)
System Name daily driver Mac mini M2 Pro
Processor Apple proprietary M2 Pro (6 p-cores, 4 e-cores)
Motherboard Apple proprietary
Cooling Apple proprietary
Memory Apple proprietary 16GB LPDDR5 unified memory
Video Card(s) Apple proprietary M2 Pro (16-core GPU)
Storage Apple proprietary onboard 512GB SSD + various external HDDs
Display(s) LG UltraFine 27UL850W (4K@60Hz IPS)
Case Apple proprietary
Audio Device(s) Apple proprietary
Power Supply Apple proprietary
Mouse Apple Magic Trackpad 2
Keyboard Keychron K1 tenkeyless (Gateron Reds)
VR HMD Oculus Rift S (hosted on a different PC)
Software macOS Sonoma 14.7
Benchmark Scores (My Windows daily driver is a Beelink Mini S12 Pro. I'm not interested in benchmarking.)
Again, I'm convinced that AMD engineers had enough foresight to test IHS designs in a variety of thicknesses and decided this was the best compromise in terms of physical protection, thermal performance, manufacturing cost, durability, etc.

Unlike a lot of people here, I don't feel like I really should be playing armchair EE Ph.D. My guess is that the average age of the senior engineers involved in semiconductor thermal solutions at AMD have an average age of 50 and probably have been doing this for 20-25+ years.

I have a lovely liberal arts degree. I generally leave the engineering to people who went to school for it and have practiced in the field for decades.
 
Joined
Sep 2, 2022
Messages
89 (0.11/day)
Location
Italy
Processor Intel Core i7-5960X
Motherboard MSI X99S SLI PLUS
Memory Corsair Vengeance LPX DDR4 4x8GB
Video Card(s) Gigabyte GTX 1070 TI 8GB
Storage Sandisk Ultra II 240GB + HDD
Display(s) Benq GL2480 24" 1080p 75 Hz
Power Supply Seasonic M12II 520W
Mouse Logitech G400
Software Windows 10 LTSC
Again, I'm convinced that AMD engineers had enough foresight to test IHS designs in a variety of thicknesses and decided this was the best compromise in terms of physical protection, thermal performance, manufacturing cost, durability, etc.
Unlike a lot of people here, I don't feel like I really should be playing armchair EE Ph.D. My guess is that the average age of the senior engineers involved in semiconductor thermal solutions at AMD have an average age of 50 and probably have been doing this for 20-25+ years.
I have a lovely liberal arts degree.
I guess you missed some videos recently. They made it thicker to allow the customers to reuse their AM4 coolers. It was a compromise, not the best possible option.
 
Last edited:
Joined
Jun 18, 2021
Messages
2,537 (2.06/day)
Again, I'm convinced that AMD engineers had enough foresight to test IHS designs in a variety of thicknesses and decided this was the best compromise in terms of physical protection, thermal performance, manufacturing cost, durability, etc.

Unlike a lot of people here, I don't feel like I really should be playing armchair EE Ph.D. My guess is that the average age of the senior engineers involved in semiconductor thermal solutions at AMD have an average age of 50 and probably have been doing this for 20-25+ years.

I have a lovely liberal arts degree. I generally leave the engineering to people who went to school for it and have practiced in the field for decades.

The strategic decision to keep cooler compatibility is not really an engineering decision though, it's a business decision that the engineers had to work with.

Does it work? yeah sure, i'm not questioning that, but could it work better? Also yes.
 
Joined
Jan 23, 2012
Messages
374 (0.08/day)
Location
South Africa
Processor Pentium II 400 @ 516MHz
Motherboard AOpen AX6BC EZ
Cooling Stock
Memory 192MB PC-133
Video Card(s) 2x Voodoo 12MB in SLI, S3 Trio64V+
Storage Maxtor 40GB
Display(s) ViewSonic E90
Audio Device(s) Sound Blaster 16
Software Windows 98 SE
Dunning Kruger in full effect...
 
Joined
Jun 21, 2021
Messages
3,093 (2.51/day)
System Name daily driver Mac mini M2 Pro
Processor Apple proprietary M2 Pro (6 p-cores, 4 e-cores)
Motherboard Apple proprietary
Cooling Apple proprietary
Memory Apple proprietary 16GB LPDDR5 unified memory
Video Card(s) Apple proprietary M2 Pro (16-core GPU)
Storage Apple proprietary onboard 512GB SSD + various external HDDs
Display(s) LG UltraFine 27UL850W (4K@60Hz IPS)
Case Apple proprietary
Audio Device(s) Apple proprietary
Power Supply Apple proprietary
Mouse Apple Magic Trackpad 2
Keyboard Keychron K1 tenkeyless (Gateron Reds)
VR HMD Oculus Rift S (hosted on a different PC)
Software macOS Sonoma 14.7
Benchmark Scores (My Windows daily driver is a Beelink Mini S12 Pro. I'm not interested in benchmarking.)
The strategic decision to keep cooler compatibility is not really an engineering decision though, it's a business decision that the engineers had to work with.

Does it work? yeah sure, i'm not questioning that, but could it work better? Also yes.

That's a larger question. What is the value of backwards compatibility versus a clean new design?

In any case, the business makes the money that pays the engineers' salaries. This isn't just some college dream project that has no existing business behind it.

There are business parameters that are just as important as some engineering decisions.

If the engineers don't like it, they can go elsewhere and start their own company. That's how Silicon Valley came to be. Shockley > Fairchild > Intel > others. Absolutely nothing wrong with that, in fact that entrepreneurial spirit is highly prized in Silicon Valley.

However, as long as these engineers have AMD employee badges, AMD management calls the shots. Dr. Lisa Su and the rest of her senior management team has increased shareholder value over her tenure so I doubt if the board will fire her anytime soon.

California is an at-will employment state. Anyone of those AMD engineers is free to "pursue other opportunities."
 
Joined
May 3, 2018
Messages
2,881 (1.21/day)
It was very predictable - thicker metal higher thermal resistance, pretty simple. They could have kept the layout and socket keep out areas but change the height (conversion kit would be new mounting screws or just washers for crafty users ;) ) and the thermals would be improved a lot just as with deliding, maybe now they have an excuse to change sockets next year :shadedshu:

I don't think there would be any outrage, pretty natural to need at least a mounting adapter with a new socket, maybe they feared it would take longer for manufacturers to adopt the new socket but they're no longer the underdog pseudo-newcommer they were back at the start of AM4, this was just a poor decision
Why they chose a poor thermal alloy given the extra thickness of the IHS is pretty weird. Is it penny pinching, or they really don't care that the cpu runs at 95C and can't really achieve the promised boost clocks for long.
 
Joined
Jul 30, 2019
Messages
3,206 (1.67/day)
System Name Still not a thread ripper but pretty good.
Processor Ryzen 9 7950x, Thermal Grizzly AM5 Offset Mounting Kit, Thermal Grizzly Extreme Paste
Motherboard ASRock B650 LiveMixer (BIOS/UEFI version P3.08, AGESA 1.2.0.2)
Cooling EK-Quantum Velocity, EK-Quantum Reflection PC-O11, D5 PWM, EK-CoolStream PE 360, XSPC TX360
Memory Micron DDR5-5600 ECC Unbuffered Memory (2 sticks, 64GB, MTC20C2085S1EC56BD1) + JONSBO NF-1
Video Card(s) XFX Radeon RX 5700 & EK-Quantum Vector Radeon RX 5700 +XT & Backplate
Storage Samsung 4TB 980 PRO, 2 x Optane 905p 1.5TB (striped), AMD Radeon RAMDisk
Display(s) 2 x 4K LG 27UL600-W (and HUANUO Dual Monitor Mount)
Case Lian Li PC-O11 Dynamic Black (original model)
Audio Device(s) Corsair Commander Pro for Fans, RGB, & Temp Sensors (x4)
Power Supply Corsair RM750x
Mouse Logitech M575
Keyboard Corsair Strafe RGB MK.2
Software Windows 10 Professional (64bit)
Benchmark Scores RIP Ryzen 9 5950x, ASRock X570 Taichi (v1.06), 128GB Micron DDR4-3200 ECC UDIMM (18ASF4G72AZ-3G2F1)
This IHS design is a compromise that shouldn't happened. They've (AMD) earned so much consumer points for compatibility already, might've tried something new for a change. I mean, it's a completely new platform, all parts (CPU, MB, RAM) are already way too much expensive, so why the hell did they bogged down the thermals just for the sake of a damn compatibility? Someone has the money for the AM5 upgrade, but will cheap out on a CPU cooler by reusing the old one? WTH AMD!?
I think the bigger problem is NOT all the AM4 coolers will be compatible especially those that used their own backplate design. It will be interesting to see a list of what coolers (and their relative price point) are compatible vs. not compatible to determine if retaining some semblance of cooler compatibility was a good choice.
 
Joined
Jun 21, 2021
Messages
3,093 (2.51/day)
System Name daily driver Mac mini M2 Pro
Processor Apple proprietary M2 Pro (6 p-cores, 4 e-cores)
Motherboard Apple proprietary
Cooling Apple proprietary
Memory Apple proprietary 16GB LPDDR5 unified memory
Video Card(s) Apple proprietary M2 Pro (16-core GPU)
Storage Apple proprietary onboard 512GB SSD + various external HDDs
Display(s) LG UltraFine 27UL850W (4K@60Hz IPS)
Case Apple proprietary
Audio Device(s) Apple proprietary
Power Supply Apple proprietary
Mouse Apple Magic Trackpad 2
Keyboard Keychron K1 tenkeyless (Gateron Reds)
VR HMD Oculus Rift S (hosted on a different PC)
Software macOS Sonoma 14.7
Benchmark Scores (My Windows daily driver is a Beelink Mini S12 Pro. I'm not interested in benchmarking.)
I think the bigger problem is NOT all the AM4 coolers will be compatible especially those that used their own backplate design. It will be interesting to see a list of what coolers (and their relative price point) are compatible vs. not compatible to determine if retaining some semblance of cooler compatibility was a good choice.

I believe that's why there are compatibility kits for many of those coolers that aren't compatible.

But if I understand correctly, AMD made a deliberate business decision to maintain certain physical dimensions to minimize the number of thermal solutions that would be incompatible.

To my knowledge AMD never promised the rest of the industry 100% backwards compatibility between AM4 and AM5.

Hell, Intel sure doesn't.

:D
 
Joined
Jul 30, 2019
Messages
3,206 (1.67/day)
System Name Still not a thread ripper but pretty good.
Processor Ryzen 9 7950x, Thermal Grizzly AM5 Offset Mounting Kit, Thermal Grizzly Extreme Paste
Motherboard ASRock B650 LiveMixer (BIOS/UEFI version P3.08, AGESA 1.2.0.2)
Cooling EK-Quantum Velocity, EK-Quantum Reflection PC-O11, D5 PWM, EK-CoolStream PE 360, XSPC TX360
Memory Micron DDR5-5600 ECC Unbuffered Memory (2 sticks, 64GB, MTC20C2085S1EC56BD1) + JONSBO NF-1
Video Card(s) XFX Radeon RX 5700 & EK-Quantum Vector Radeon RX 5700 +XT & Backplate
Storage Samsung 4TB 980 PRO, 2 x Optane 905p 1.5TB (striped), AMD Radeon RAMDisk
Display(s) 2 x 4K LG 27UL600-W (and HUANUO Dual Monitor Mount)
Case Lian Li PC-O11 Dynamic Black (original model)
Audio Device(s) Corsair Commander Pro for Fans, RGB, & Temp Sensors (x4)
Power Supply Corsair RM750x
Mouse Logitech M575
Keyboard Corsair Strafe RGB MK.2
Software Windows 10 Professional (64bit)
Benchmark Scores RIP Ryzen 9 5950x, ASRock X570 Taichi (v1.06), 128GB Micron DDR4-3200 ECC UDIMM (18ASF4G72AZ-3G2F1)
I believe that's why there are compatibility kits for many of those coolers that aren't compatible.

But if I understand correctly, AMD made a deliberate business decision to maintain certain physical dimensions to minimize the number of thermal solutions that would be incompatible.
I think that is a reasonable business decision especially if the IHS thickness doesn't impact actual cooling capability. I am patiently waiting for someone to release some test results with IHS vs. Direct Die to see if you can actually boost performance under direct die. If you take something like buildzoids chart (
) will direct die yield any gains over 200W? The new CPU's are designed to run hot at 95c so I suspect not but I am not a CPU engineer.
 

crazyeyesreaper

Not a Moderator
Staff member
Joined
Mar 25, 2009
Messages
9,810 (1.72/day)
Location
04578
System Name Old reliable
Processor Intel 8700K @ 4.8 GHz
Motherboard MSI Z370 Gaming Pro Carbon AC
Cooling Custom Water
Memory 32 GB Crucial Ballistix 3666 MHz
Video Card(s) MSI RTX 3080 10GB Suprim X
Storage 3x SSDs 2x HDDs
Display(s) ASUS VG27AQL1A x2 2560x1440 8bit IPS
Case Thermaltake Core P3 TG
Audio Device(s) Samson Meteor Mic / Generic 2.1 / KRK KNS 6400 headset
Power Supply Zalman EBT-1000
Mouse Mionix NAOS 7000
Keyboard Mionix
If i use AM5 for cooler testing anytime soon I will definitely want one of these. Just to make my life easier.
 

SL2

Joined
Jan 27, 2006
Messages
2,410 (0.35/day)
But if I understand correctly, AMD made a deliberate business decision to maintain certain physical dimensions to minimize the number of thermal solutions that would be incompatible.
I don't believe that for a second.

Why does the lid have to be that thick, if it affects thermal performance in such a bad way? AMD says Ryzen 7000 will hit a temp limit before a power limit, so why make it extra hot when you need every MHz?

Why didn't AMD just make the socket itself ~1.5 mm higher instead? (Or whatever height we're talking about.)


No, there's something else going on here. My bet is still preparations for die stacking of some kind in future products, something that uses up more space/height than current 3D cache. (The added cache in 5800X3D are combined with extra thin CCD's as a height compensation, maybe that's something they want to avoid, although I guess the new lid is even thicker than that..)
That's the only reason for why AMD must use a thick lid that I can come up with.

(Or, maybe it's not that easy to design an LGA socket in plastic that has added height that can withstand the mounting force while maintaining proper contact with the CPU.)
 
Last edited:
Joined
Aug 18, 2022
Messages
363 (0.45/day)
I don't believe that for a second.

Why does the lid have to be that thick, if it affects thermal performance in such a bad way? AMD says Ryzen 7000 will hit a temp limit before a power limit, so why make it extra hot when you need every MHz?

Why didn't AMD just make the socket itself ~1.5 mm higher instead? (Or whatever height we're talking about.)


No, there's something else going on here. My bet is still preparations for die stacking of some kind in future products, something that uses up more space/height than current 3D cache. (The added cache in 5800X3D are combined with extra thin CCD's as a height compensation, maybe that's something they want to avoid, although I guess the new lid is even thicker than that..)
That's the only reason for why AMD must use a thick lid that I can come up with.

(Or, maybe it's not that easy to design an LGA socket in plastic that has added height that can withstand the mounting force while maintaining proper contact with the CPU.)
During the Zen 4 launch there was a slide showing that AM4 cooler compatibility was a key feature of Ryzen 7000 processors (see the slide below) so it was a deliberate choice. Making the IHS thicker was part of that choice because due to the AM5 LGA socket the CPU sits slightly lower than it did with AM4. If the plastic pillars that form part of the standard hook type fitting had been reduced in height to allow a thinner IHS there would have been a risk that AM4 coolers would not have enough tolerance to fit correctly. An adapter would have solved this issue but that simply solves one problem by replacing it with another.

For example, the need for an LGA 1700 adapter for previous coolers is a feature of Intel's 12th and 13th Gen processors (Alder Lake and Raptor Lake). Some cooler manufacturers will supply them free on proof of purchase of one of these CPUs. LGA 1700 adapters are on sale at most retailers and obviously users are buying an adapter rather than waiting for one, or because they hadn't realised one was needed.

All of this inconvenience and expense has been avoided by AMD for the vast majority of AM4 coolers. Not all, because the AM5 ILM is screwed into the backplate so AM4 coolers that rely on removing/replacing the backplate will have an issue but the number of coolers affected is small and most of the manufactures concerned are already doing something about this.

small_am4-cooler-compatible.jpg
 

Mussels

Freshwater Moderator
Joined
Oct 6, 2004
Messages
58,413 (7.97/day)
Location
Oystralia
System Name Rainbow Sparkles (Power efficient, <350W gaming load)
Processor Ryzen R7 5800x3D (Undervolted, 4.45GHz all core)
Motherboard Asus x570-F (BIOS Modded)
Cooling Alphacool Apex UV - Alphacool Eisblock XPX Aurora + EK Quantum ARGB 3090 w/ active backplate
Memory 2x32GB DDR4 3600 Corsair Vengeance RGB @3866 C18-22-22-22-42 TRFC704 (1.4V Hynix MJR - SoC 1.15V)
Video Card(s) Galax RTX 3090 SG 24GB: Underclocked to 1700Mhz 0.750v (375W down to 250W))
Storage 2TB WD SN850 NVME + 1TB Sasmsung 970 Pro NVME + 1TB Intel 6000P NVME USB 3.2
Display(s) Phillips 32 32M1N5800A (4k144), LG 32" (4K60) | Gigabyte G32QC (2k165) | Phillips 328m6fjrmb (2K144)
Case Fractal Design R6
Audio Device(s) Logitech G560 | Corsair Void pro RGB |Blue Yeti mic
Power Supply Fractal Ion+ 2 860W (Platinum) (This thing is God-tier. Silent and TINY)
Mouse Logitech G Pro wireless + Steelseries Prisma XL
Keyboard Razer Huntsman TE ( Sexy white keycaps)
VR HMD Oculus Rift S + Quest 2
Software Windows 11 pro x64 (Yes, it's genuinely a good OS) OpenRGB - ditch the branded bloatware!
Benchmark Scores Nyooom.
So what? Let it fill the space in between. Why bother at all? All my CPUs have residue on the sides of the heatspreader, ist doent not matter at all, as long as the upper surface is clean when applying new grease.

I guess this is for those folks who RGB-illuminate their cases and put their GPUs in the window to compensate for their small p... -I mean power supplies!
No it wont matter at all - but if you care about it for OCD reasons or whatever, you can now get a little bracket.
Heck, that bracket might be really handy if you're lapping the CPU to hold it nice and neatly with, and give you a perfect height to lap it down to.


It's optional and cheap, which is how it should be.

Just spend some money to slim the card ;)
lap the credit card
 
Joined
Nov 18, 2010
Messages
7,492 (1.47/day)
Location
Rīga, Latvia
System Name HELLSTAR
Processor AMD RYZEN 9 5950X
Motherboard ASUS Strix X570-E
Cooling 2x 360 + 280 rads. 3x Gentle Typhoons, 3x Phanteks T30, 2x TT T140 . EK-Quantum Momentum Monoblock.
Memory 4x8GB G.SKILL Trident Z RGB F4-4133C19D-16GTZR 14-16-12-30-44
Video Card(s) Sapphire Pulse RX 7900XTX. Water block. Crossflashed.
Storage Optane 900P[Fedora] + WD BLACK SN850X 4TB + 750 EVO 500GB + 1TB 980PRO+SN560 1TB(W11)
Display(s) Philips PHL BDM3270 + Acer XV242Y
Case Lian Li O11 Dynamic EVO
Audio Device(s) SMSL RAW-MDA1 DAC
Power Supply Fractal Design Newton R3 1000W
Mouse Razer Basilisk
Keyboard Razer BlackWidow V3 - Yellow Switch
Software FEDORA 41
Actually this is a problem even professionals (Igor from igorslab for example)

Let's get us one thing straight. Who is a professional or not. A professional internet diva, tech journalist, whos earning money with this? Or or a engineer, a manufacturer official and dev sitting behind the desk and don't giving a F* about anything and just making his stuff?

No internet tech diva is really a pro. Just give the man the right tools, instruments and review items, you can make anyone a diva or Jesus with a freaking opinion. That counts also for Igors lab, whose articles often caused my eyebrows rise, also the toxic competition between der8auer that ends up as a circus.
 
Top