@@Nethan10Troi it doesn't even do that as it doesn't have a tight seal around the IHS. You'll need the rubber gasket variant of this for that It's simply a product to make things pretty... Under a heatsink.
@Nethan10Troi it's irrelevant to actual usage though as TIM application doesn't really do much around the outer edges of the heatspreader. And it's no more difficult to apply TIM to the center of this heatspreader than any other.
It's understandable why Thermalright did it. They wanted to have something out in the market since they have been making these brackets for a long time and no one paid attention until LGA1700's warping issue. Now it's a product designed for those who want to make things pretty, as people can get quite OCD about non-rectangular shapes. They had a similar product for AM4 designed to prevent the CPU from coming off the socket, that was the origin of this whole "Secure Frame" product name. This product is still called an AM5 Secure Frame on their main site; what exactly is it securing? Idk go with with the flow... At least the 1g of TF-7 thermal paste provided in the package is above average.
I know it likely does nothing but, there is something just so right about the way it looks when installed that would probably make me buy one. Sometimes it's nice just to do a job with something that little bit higher quality if it looks/feels so right. And yeah I know it's not visible once the cooler is on, but you still know it's there and done better.
The AIO used in the video is supposed to be mounted with an offset so that the cold plate is centered over the Ryzen Chiplets under the IHS. Basically you have to move down both of the two pieces which you screw the waterblock onto so that the uppermost drilled holes are used instead of the bottom holes. Offset mounting is the right way to use the Arctic Liquid Freezer II 420mm on Ryzen 7000. I have mine just like that mounted with the offset, the waterblock and coldplate are centered right over the chiplets.
@Yurmun So when you get your cooler. You have to place the two black colored plates supplied with the cooler on the AM5 socket as far down as possible. They basically align with the topmost drilled holes on the plates. And then you place the waterblock the same way and it's holes should align properly aswell. This way the waterblock sits as far down over the IHS centered over the chiplets under it. You can mount it normally as well but then it will sit further up not exactly over the chiplets. It's not mandatory but it's preferable to mount it with the downward offset.
@Yurmun I think it's true. Cause they're all basically the same cooler just in different sizes. I have the Arctic Liquid Freezer II 420mm ARGB. And have mounted it with the offset as far down as possible so that it's right over the chiplets cause i'm cooling a Ryzen 9 7950X3D with the 420 lol. Arctic is awesome. The 420 is so big that i placed it outside my PC case ontop of it and the top panel over the cooler radiator. It looks like a PC sandwich lol. But i like it and the cooling potential as well.
According to manufacturers they are not for temperature improvement! So what are the temp tests supposed to say? The AM5 ones are for pressure optimisation … the optional silicone guard for thermal paste improvement!m avoiding leaks!
According to der8auer, they made the ihs thicker or something to make the coolers from am4 compatible or something and this may have altered thermal conductivity for the worse. May wanna double check since I could be wrong id your interested.
@@Derpynewb they did, they wanted to have cooler compatibility so they matched the z height of the am4 platform. And adding material, no matter how thermally conductive it is, will only make things worse. Would have been better to send out dummy CPUs and motherboards to cooler manufacturers ahead of the release so they could make either brackets for existing coolers, or design new coolers around it.
Distortion is always present to some degree with LGA socket retention mechs, going right back to the very first ones, just not to the gross degree seen with Intel 12th/13th.
@@nathanhamman418 Or just don't push CPU way above there efficiency point. Non X parts don't have any problems with cooling and 30-40$ 120mm cooler will work just fine, when X parts is burning even with 360-420 aio...
If you were to lap IHS and cooler plate flat you might find higher drop. Lapping doesn't usually help using the factory retention mech because it deforms the IHS slightly (actually quite a lot with Intel 12th and 13th, as much discussed) - the frame would maintain the profile of the IHS (the cooler plate also needs flattening because they're usually manufactured either slightly arched or domed to conform to that IHS distortion).
It's Intel's fault, I really hate the design of their latest sockets, it's too wide on one side it's clearly made to be bent on purpose and I think they look uglier than ever specially when they have the highest prices and look less high-tech visually...
They are not made for temperature management, but pressure distribution. Clearly stated on manufacturer web pages! Not even as thermal paste guards as they are an additional option to bracket!
It solves the issue with the thermal paste going all over the processor grooves. I would prefer this to the thermal grizlly pads as it looks much nicer.
@@TheBackyardChemist I change my thermal paste regularly, if you have an AM5 chip and you remove your cooling solution you’ll see why I went with this bracket. The grooves get gunked pretty bad with the exposed cutouts.
@@TheBackyardChemist well, quite aware of that. Not an issue per say as for cpu operation or what not. But an issue for OCD people lol , cleaning the past overspill is pain, well at least for me.
The limit is the CPU cooler you selected (even with 2x fans in a push-pull configuration you won’t have better temps, only less noise). You have to go with a *Dual Tower Cooler* (or a big cooler like the *Black Rock*) that has the capability / thermal capacity to dissipate more heat in order to see a significant difference. Just my two cents. Thanks for the great content!
Hey bryan, recently I've noticed that you can undervolt any 13th gen CPU without losing performance after changing the CPU microcode to version 104 (it takes a bit of effort), and this is very easy if you are using an ASUS motherboard since they allow you to change the microcode directly in the BIOS (advanced settings --> tweaker's paradise). By doing this you can undervolt the CPU just as usual !
@@Eduardo1007 the performance would drop dramatically if you try to undervolt the 13th gen. My 13600kf got only 1/4 of the normal score or even lower in CPUZ after I reduced 0.1V. However, by using the method I explained earlier, you can undervolt the 13th gen just as usual.
If youre having heat problems with 7800x3d or 7950x3d a simple non invasive trick to drop temps by 15-20 degrees celsius is go into edit power plan>change advanced power settings, scroll to find porocessor power management>maximum processor state, and change from 100 to 99. 1% drop in performance is worth the trade for -20 degrees temperature.
It's not %1 performance difference only, It completely disables CBP. Just dont do nothing. It desing to work hot. Just buy better cooling instead or slightly undervolt.
You are just stopping the CPU from boosting. I guess you play crappy non CPU intensive games 😂 so you couldn’t even tell the difference. Just get the bracket, a good 360 AIO, enhanced under volted BIOS settings and call it a day.
@@fardin_hossainI have pbo undervolted. This suggestion more for people who dont know or wanna go into their bios. Which is most pc users. Im not trying to help the insane knowledge pc geniuses. Yall pc wizards already know everything in the world and have to comment to let people know. I was just trying to help some people that werent as insanely genius as you.
1 degree Celsius is margin of error. The problem With the 7000 series CPUS is that the IHS is thicker than what was used for the AM4 CPU's. The CPU's may be soldered but they increased the layer through which the heats needs to travel to reach the CPU Cooler.
Did the clocks change with the bracket vs without? Cause AMD's strategy with Zen 4 appears to be to use all available cooling overhead and clock the cores higher instead of dropping temperatures at the same clocks.
I bought the thermal right lga1700 for z690 and i7 13700kf with Artic freezer 240 mm. The temperature reduction was insane, dropping from 70C to 55C at 5600Mhz OC. I highly recommend these products if you plan to overclock or just want a lower cpu temp.
Yes, but only on Intel LGA 1700, since that socket has mounting issues all the way. So bad, that your problems can range from "only" memory instabilities up to booting right into a bsod. If you're on Ryzen 7000 you would benefit more from lapping the cpu or delidding it, the IHS is actually the most hindering factor when it comes to thermals and clockspeeds, and keep in mind: Zen 4 is designed to push for those 95°C first when setting the clockspeed and voltage automatically.
the worst part is intel knew this was going to be a problem, I refuse to believe they didn't. My evidence for that claim is being a user of the LGA3647 socket, another socket that is nothing even close to square. The solution to the potential cooling issue was using the cooler to apply the mounting pressure from above rather than at a couple of points on the edge, the bulk of the pressure being where the lines intersect, right in the middle of the chip. AMD still made a square chip, so the warp of having an oblong monstrosity who's pressure is applied centrally isn't the issue, but as some have speculated the thicker IHS may be to blame. If so the fault of both parties was trying to maintain a broad cooler compatibility through using more or less the same mount design and Z height respectively. There may come a day when more performance necessitates a return to direct die, or at least I hope it comes to that.
You should make use of your Arctic AIO offset mounting, you using AM4 default mounting, but if you have a Ryzen installed you should use the upper 2 holes of the cooler mounting bracket, giving you another 5 degree off, and its centered on the hot spot.
You really cannot claim a 7950x3d with the non-vcache ccd turned off is a 7800x3d because the 7800x3d boosts 250mhz (5000mhz) less than the vcache die on the 7959x3d which boosts to 5250mhz
I guess if people want lower temps on aircooling, you need to DE-LID the AM5, and of course get a slightly better cooler than a Snowman. I use a Skythe Fuma2 with 2 full size (not slim) 1800rpm Kaze Flex fans on my 5950x and it works very good, probably for AM5 not enough though unless people delid the CPU.
I'm still using older contact frame that come bundled with thermalright AXP-100H Muscle on i7-9700, and when I first buy one it's meant only for haswell CPU and it made from plastic.
My thermalright LGA 1700 contact frame dropped temps on my 12900K like 20c. Before when I would tweak the bios and put on a mild OC like 5.1 ghz it would immediately break 95c plus when I would put on a short stress test. After the contact frame I can hit 5.4ghz stable with an avg temp of 55-60c while gaming. When I stress test max temp it will reach is 78c for a brief second. It's nice to be able to actually OC my 12900K to try to close the gap on bottlenecks with my 4090 when i'm running 1440p benchmarks or just gaming.
It should be standard with both AMD and Intel. Intel has an additional issue to solve though. It's still quite easy to damage the pins on the motherboard. You're never really sure about the alignment. There's too much room to wiggle the CPU around inside the socket. Even if they stick with pins on the motherboard they could at least provide a mechanism to improve perfect alignment. Just one pin on every corner would do.
I find it interesting that some people say it doesn't work, wonder what causes the discrepancy? I have been think it is possible that the contact frame may absorb a slight amount of heat from the CPU and that could help dissipate heat (especially if there is some sort of airflow under the heatsink) I also wonder if it's possible it could through the screws transfer more heat to the back of the board I would be curious to see this tested, does the contact frame get hot during loads? and what happens if some sort of fan is used to cool it, and does the back of the board get hotter with it? also would wonder what would happen if it was lapped (possibly along with the CPU) to provide the exact same height and be flush, so the heatsink makes contact with it as well.
Its pretty simple … the manufacturers do not promise any temperature improvement qualities! Its for pressure distribution only! With optional silicone guards it prevents paste overspills. Thats all! That some experience improvements is associated with re-doing the setup!
Now make a video sanding the IHS and record the temperatures per height lapped and/or sanded! I believe that 1mm less on that IHS just make enough difference.
what if you put processor grease on all the parts that touch between the pressure plate and the IHS, and also between the proc/IHS and the cold plate on the pump (or heatsink) ?
This bracket is just meant to help stop thermal paste from getting all over the place. nothing more, according to the product description. It does look nice though.
Why is everyone complaining about the temperatures of AM5 processors? I don't know how representative this is, but I have an air-cooled R7 7700 and its temperatures have never exceeded 70 degrees, what do you have to do with the processor to get it to 92-95 degrees?
Inform yourself! They have nothing to do with temperature for AMD! Its for even pressure distribution and with optional silicone guard to protect from thermal paste overspill!
@@GanymedeXD Manufactures can claim whatever they want doesn't mean it's necessarily true. Gamers Nexus did a video on these contact frames and found them to amount to basically negligible thermal improvement--me saying they're basically guards for thermal paste because they look and behave just like the silicon guard.
I used the thermalright 12th gen bracket for my 13900KS and I was able to overclock a Vcore of 1.62V, 355W, single core 6.3G, triple core 6.2G, and all core 5.8G without disabling efficiency cores, the cooling solution is an Asus ROG Strix 360 ARGB AIO and Coolermaster Cryofuze Violet thermal paste, and it never gets past 94C, normal operating temp is around 40-50C. However, that being said, this system with an overclocked Asus ROG Strix 3090Ti OC LC at 2.2Ghz,1.3V, 600W, I can never play Genshin Impact at 2k 60Hz, whereas all my other computers can.
hmm i think if youre planing on to an itx case with an 7800x3d and a cooler that is at max. 39mm height, this thing is maby better for that case? or still waste of money? what do you think?
1 or 2 degrees is negligible, $8 pretty good price to get more improvement in temperatures. Is there reason why AM5 has issues of temperatures for the CPU die contact with the cooler?
That air cooler is reaching saturation and can’t absorb anymore hear whether the bracket helps it make better contact or not. If you were to use something with two towers and two fans you would likely have seen similar results to the water cooler.
@@hristobotev9726 there are some air coolers out there that would be enough. But that single tower with one fan is just not going to get it done. My 7800x with a similar cooler will hit 95c and thermal throttle within a few seconds under cb loads because the cooler is just not able to absorb that much heat. It works fine for day to day and games. Will replace it soon with an aio, just trying to decide. The 7800x can draw like 180 watts or something crazy and the single tower coolers I have seen are all rated for no more than 130 watts. It is unlikely that the cb score would improve as the cooler is becoming saturated within seconds of the chip going into a heavy workload and couldn't absorb more heat anyway.
What happens if I use a Thermal Grizzly KryoSheet Wärmeleitpad, will I have problems with an electrical short? Since this product is conductive, the opposite of thermal paste. THANK YOU!
The Big Issue with AM5 Temps is the THIKNESS of the IHS its Bigger than usual , and AMD forced to do that so they maintain the compatibility with AM4 coolers !!! Its a good & bad deal in the same time you will not be forced to buy a new bracket for AM5 , BUT in exchange you will get a freaky High Temps !!!! sorry for my bad english
This is reminding me to not be lazy and install my contact frame for my 13700k. Seems a bit overkill on AM5 though, I don't think they have the warping problems LGA 1700 has.
AMD did not do themself any favour in going LGA with AM5. The height of the heatspreader((it's way too thick in order to get on AM4 plattform height for cooler compatability) as well as it's form is far from ideal. They kinda stuck now on this design unless AMD decides to release a completely updated heatspreader design alongside modifications to the ILM.
It's not a 100% fit. So you got very minor play. But that minor play might be gone when the CPU heats up and the material expands a littlebit. So I can't tell you for sure but it definetly helps keeping thermalpaste out of the cpu gaps.
Does the 7950x3d run much hotter than the 7800x3d? My 7800x3d runs way cooler than this with a cheap 60$ 280mm aio it uses the stock amd mounting too, like the little plastic things i did undervolt but gah damn this runs hot compared to mine.
lol 7800x3d is low watts but you still need a beefy cooler. Not no hyper 212 or anything the same size. Thing gets hot as fuck even with low wattage and correct SOC voltage. Thats why Im even watching this video, to see if I could get a few degrees less with a simple bracket. My max temps right now are 79 but I like things to be as cool as possible especially if its cheap.
What are you going on about? Is it about the bracket? The bracket is for am5 … optimised pressure distribution … the silicone guard that is optional is for paste protection. What is a normal water cooler … an aio? Of course you can use those for am4.
According to Thermal Grizzly the optional silicon guard is mandatory when using liquid metal! The bracket itself is for optimised pressure distribution.
Yeah this isn't a product that needs to exist. the stock ILM is just fine for am5, since most of the problems are due to the thickness of the IHS. if there was an easy way to lap the chip you would be more likely to see results.
i guess there is no temp problem with socket am5! and on top, you have to take a closer look to the speed, not to the temp! Amd behave different to other chips and Brian should know this! maybe he is a bit too fast with the test/video!
Seems they took the solution for the Intel temp issue and implemented it for the Ryzen temp issue, when the reason for the higher temps for each processor is different. So, the solution that helped Intel, does nothing for Ryzen.
Those differences are less than using different paste. So not really worth it. For the super enthusiast who runs the chip at it's absolute maximum, even such a small difference can be important, but the normal user won't even notice.
I bought an 5800x3d a while back and it thermal throttled at idle. I put my 5800x back in and problem solved. The 5800x3d went back for a refund. If I have to buy additional after market parts for a new CPU to work correctly, I don't want it.
Bracket is NOT for Thermals in anyway shape or form the original maker said that himself. then companies started scamming people saying it cools them down, if he patented it would never have happened . its copper plate that does not even touch the IHS there is a gap around it its only to secure the CPU in place when using heavy coolers or those with custom back plates. Sorry tech yes city you were lied to and scammed. Only Delidding and Direct Die makes significant different of 15-20*C 1-5*c will do nothing
this is why i probably wont be upgrading my cpu any time soon between the 13900k and 7800x3d they both run to hot for my liking an AIO or fan heatsink will not keep these beasts to reasonable temps they need to invest in newer tech and stop the playing around with fans with more rgb or cable mods no matter what fan or aio they design it has its limits
Why are you testing air cooling with a $20 air cooler? For a 7950X3D you should be using a high-end air cooler. Nobody in the right mind should be using a low end air cooler with a high end CPU.
This is a confusing product and honestly, the epitome of China just copying successful products without really understanding why the product was successful. The thing is, anyone who sees this and knows what it is supposed to be for and how and why they would use it, will likely know the whole inception of "contact frames" was to solve a specific issue arising from the elongated rectangular shape and 2 point ILM of Intel's LGA 1700 chips. AM5 is square, and has a thicker IHS on top of that. Don't get me wrong, data is king so I'm glad someone picked it up and tested it, but it's just a pointless product imo.
