Gelid comes with a CPU protection bracket for socket AM4

[Vader]

Many good aftermarket coolers also screw straight into the stock backplate. Not no mention many AIOs

 

[newtekie1]

Funny, as 99 of the coolers I test, which have springs, allows said springs to be fully collapsed.

None of the coolers I've used fully collapse the springs. The threads on the screw bottom out before the spring is fully compressed.

 

[sneekypeet]

None of the coolers I've used fully collapse the springs. The threads on the screw bottom out before the spring is fully compressed.

Ok so you have one or two off. I will still say 99% of them are compressed to a point of nearly solid mounting.

 

[Bones]

I'm wondering if anyone has noticed the current CPU cooler collar retention design vs older ones that has a "Ring" that goes all around the base making the retainer as a single piece.

The newer style retainers are separated for a reason and that's so you can slide a cooler off to the side and then off the chip without the risk of pulling the CPU out of the socket when removing a cooler - At least that's the idea behind it.

It's designed to work with a stock cooler and of course not all coolers can be removed that way.

Sometimes components on the board just get in the way period and you have to pull it upwards anyway but that's the overall reason why they stopped making the retention collars as one piece instead of two pieces as they are now.
That's something I learned awhile back - Take advantage if you can to just slide it towards the open end and the CPU will not pop out.

If you can't due to components in the way I guess you'd just have to be careful and know even though it was designed to help with that, AMD can't control how board makers lay out their components on a board.

Also note I'm talking about the stock retention collar used with a stock cooler, not anything that's aftermarket but with some aftermarket coolers it does help - Aftermarket mounting hardware is a different matter.

 

[newtekie1]

Ok so you have one or two off. I will still say 99% of them are compressed to a point of nearly solid mounting.

If they were just going to do a solid mount, what is the point of the springs even?

 

[trickson]

If they were just going to do a solid mount, what is the point of the springs even?

Actually I see no point in this thread.
You can NOT even get one so this is all moot at best.
You can not even order one.
There is no information on there site nothing it's a dead thread to me now.

 

[Bill_Bright]

I disagree this is not a gimmick

Then by all means, you and phanbuey should go buy a bunch of them. No doubt Gelid will be happy you do. Me? I will just do as I have done successfully 100s of times over the years and make sure the TIM is warm (as even you suggested), then give a slight twist to break the bond and lift the cooler off. FWIW, I, after nearly 30 years of this stuff, have never, not once, delidded or damaged a CPU or socket when removing a cooler. Of course, there's always a first time, but I'll take my chances.

 

[phanbuey]

Then by all means, you and phanbuey should go buy a bunch of them. No doubt Gelid will be happy you do. Me? I will just do as I have done successfully 100s of times over the years and make sure the TIM is warm (as even you suggested), then give a slight twist to break the bond and lift the cooler off. FWIW, I, after nearly 30 years of this stuff, have never, not once, delidded or damaged a CPU or socket when removing a cooler. Of course, there's always a first time, but I'll take my chances.

I mean we shouldn't... we shouldn't have to buy anything extra and you shouldn't have to do a specific extra careful ritual where you WARM UP the paste to remove cpu a cooler because the socket is designed just so that there is a chance of the chip being ripped out with potential damage to it.

I get what you're saying but this is why we have silly things like railings, and that little ball that drops down in the coffee maker to stop coffee from spilling when you take the pot off too early -- it's good engineering. Right now, what that socket is, it's not good engineering.

 

[hat]

I mean we shouldn't... we shouldn't have to buy anything extra and you shouldn't have to do a specific extra careful ritual where you WARM UP the paste to remove cpu a cooler because the socket is designed just so that there is a chance of the chip being ripped out with potential damage to it.

I get what you're saying but this is why we have silly things like railings, and that little ball that drops down in the coffee maker to stop coffee from spilling when you take the pot off too early -- it's good engineering. Right now, what that socket is, it's not good engineering.

I disagree. It's not really being extra careful, it's just what you do, and it's been a known thing to do since... before I got into computers well over 10 years ago. It's not a secret magic trick. It's also not something millions of average Joes and Janes do all over the world every day (like making coffee). It's a skill that you need to learn before you can do it, and not everyone does it.

A real life similarity from me: part of my job is to make "dye lines" out of two different size tubings and another hose with barbed fittings on either end. It's not the easiest thing in the world to get those tubes all the way on... unless I heat up the ends with a torch for a second or two first. It might not be something just anybody would think of doing on their own if they had to make one, but it's not complicated either. It's just a trick of the trade, very simple, quick and easy to do. One might be able to avoid having to use that trick by designing something different, but there's really no need.

 

[Bones]

I mean we shouldn't... we shouldn't have to buy anything extra and you shouldn't have to do a specific extra careful ritual where you WARM UP the paste to remove cpu a cooler because the socket is designed just so that there is a chance of the chip being ripped out with potential damage to it.

I get what you're saying but this is why we have silly things like railings, and that little ball that drops down in the coffee maker to stop coffee from spilling when you take the pot off too early -- it's good engineering. Right now, what that socket is, it's not good engineering.

Everything by design has certain "Requirements" and although convenience is in mind it can't be promised with each and every design period - If something works it does and there may be a thing or two to do when dealing with it.

The designers of it never had any intent to make it complicated, the socket in fact it very simple and certainly better than the other which you can drop the CPU in and it's largely "Game Over" for the board.....
Sometimes even without a case of butterfingers involved the little pins can shift around upon CPU installation causing all kinds of mayhem anyway mandating a pin check and fix.
So much better now isn't it?
Not.

I shoudn't have to warm up the car before driving it but once again it is what it is.
I mean you CAN but there are also consequenses that comes with it being operated that way as in a shorter engine life.

I've been a mech for around 40 years to know the effects of this and at the very least I allow it to run for at least 30 seconds before trying to move it so the engine oil has time to circulate and get where it needs to be before a load is applied to the engine, meaning it's bearings and so on. They have to bear the strain/load of it's operation and if oil isn't there on the bearing while it's sustaining a load guess what happens?
It's related to "Why" nearly all engine wear occurs at startup.

There are perfectly valid reasons for that and the same applies to how you work with a socket and cooler and we all gotta deal.

 

[trickson]

It's all kinda moot. The part is not available any way,
Well they still use pins on there CPU so there is that.
The socket can only take so much force before the CPU just comes out with the heatsink.
So far every one of MY Ryzen CPU's have done this no matter what!
I have tried the twist and the heat it up and to be honest it did not work.
I would LOVE to get 3 of these but I can't.
Just because YOU do not have issues doesn't mean the rest of us are NOT.
Necessity is the mother of invention and well they beat you to it and if you think this is a gimmick or just not necessary or you can twist or use heat or like a mechanic just run it for a while then cool.
But imagine for a second you can't idle the CPU (CAR) because it wont start or you are in the middle of a rebuild and start to take things apart? See things should be easier than this.
I have told you stories but just recently taking the Heat sick off my 1700x and 1300x (just 3 days ago) both CPU's came out and where stuck to the Heat sink.
THANK GOD IN HEAVEN that the CPU's pins did NOT bend like the other time with R3 1300 and YES I FING TWINSTED THEN LIFTED!
So till you walk in some one else's shoes you should stop and think for a min..

 

[phanbuey]

I disagree. It's not really being extra careful, it's just what you do, and it's been a known thing to do since... before I got into computers well over 10 years ago. It's not a secret magic trick. It's also not something millions of average Joes and Janes do all over the world every day (like making coffee). It's a skill that you need to learn before you can do it, and not everyone does it.

A real life similarity from me: part of my job is to make "dye lines" out of two different size tubings and another hose with barbed fittings on either end. It's not the easiest thing in the world to get those tubes all the way on... unless I heat up the ends with a torch for a second or two first. It might not be something just anybody would think of doing on their own if they had to make one, but it's not complicated either. It's just a trick of the trade, very simple, quick and easy to do. One might be able to avoid having to use that trick by designing something different, but there's really no need.

Everything by design has certain "Requirements" and although convenience is in mind it can't be promised with each and every design period - If something works it does and there may be a thing or two to do when dealing with it.

The designers of it never had any intent to make it complicated, the socket in fact it very simple and certainly better than the other which you can drop the CPU in and it's largely "Game Over" for the board.....
Sometimes even without a case of butterfingers involved the little pins can shift around upon CPU installation causing all kinds of mayhem anyway mandating a pin check and fix.
So much better now isn't it?
Not.

I shoudn't have to warm up the car before driving it but once again it is what it is.
I mean you CAN but there are also consequenses that comes with it being operated that way as in a shorter engine life.

I've been a mech for around 40 years to know the effects of this and at the very least I allow it to run for at least 30 seconds before trying to move it so the engine oil has time to circulate and get where it needs to be before a load is applied to the engine, meaning it's bearings and so on. They have to bear the strain/load of it's operation and if oil isn't there on the bearing while it's sustaining a load guess what happens?
It's related to "Why" nearly all engine wear occurs at startup.

There are perfectly valid reasons for that and the same applies to how you work with a socket and cooler and we all gotta deal.

All the arguments so far against a $1.50 bracket or a tweak to the current socket that keeps your CPU from being ripped out during routine cooler removal have been:
1. It's been like this forever.
2. I've learned how to deal with it.
3. It's not a big deal because it doesn't happen to me.

I'm sorry but these are not good reasons to not, given the absolutely minimal cost and effort, improve a product.

 

[Bill_Bright]

I mean we shouldn't... we shouldn't have to buy anything extra and you shouldn't have to do a specific extra careful ritual where you WARM UP the paste to remove cpu a cooler because the socket is designed just so that there is a chance of the chip being ripped out with potential damage to it.

I disagree. It's not really being extra careful, it's just what you do, and it's been a known thing to do since... before I got into computers well over 10 years ago. It's not a secret magic trick. It's also not something millions of average Joes and Janes do all over the world every day (like making coffee). It's a skill that you need to learn before you can do it, and not everyone does it.

I agree with hat.

And note, phanbuey, you are blaming the wrong people!

This has nothing to do with socket (or CPU) design. To ensure the best electrical connection and conductivity, you first must have the best "mechanical" connection. It does not matter if a power plug, RAM slot, battery post & clamp, solder joint or CPU socket. The conductors must first make a good, solid "mechanical" connection. This is why the CPUs lock into place. The alternative would be the need to force the CPUs into the socket, as we do with RAM sticks. But of course, due to the large quantity of contacts, they have to be tiny, and therefore are much more fragile than the contacts on RAM sticks. Ergo, ZIF (zero insertion force) sockets are used.

"IF" fault is to be levied against any one, blame the TIM makers! They are the ones making TIM that turns into a strong adhesive over time. That is not AMD or Intel's fault. So why do you keep blaming them?

This "trick", btw, is actually something I was taught way way back in the early 70s when I was in tech school learning electronics. It wasn't for CPUs, of course, but to remove heatsinks from discrete components like large RF and audio power amplifier transistors and diodes. I was taught it again as a method to break the bond between device and the heatsink block with "adhesive TIMs" that are commonly used on chipsets and other devices where there is no mounting mechanism to hold the cooler in place. So special purpose thermal adhesive paste is used.

Right now, what that socket is, it's not good engineering

I totally disagree with this too. Good engineering ensures the device works as expected. Good engineering provides for easy and "sure" assembly. By "sure", I mean once properly installed, it stays properly installed. It does not fall out or work itself loose.

CPUs and coolers were never and still are not designed to be swapped in and out on a whim like one changes channels on their TV! Coolers are not and never have been designed to be regularly removed and replaced like one might do the oil filter on their car.

In the vast majority of installations, the cooler is mounted and then left mounted until the computer ascends to computer heaven!

1. It's been like this forever.

I'm sorry but these are not good reasons to not, given the absolutely minimal cost and effort, improve a product.

First, once again, you are pointing fingers in the wrong direction! You are blaming the CPU and socket designers for the adhesiveness of the thermal paste being too strong. That's totally misplaced blame!

And second, while I totally agree, "we've always done it this way" or "it's been like this forever" are way to often used as lousy excuses that too frequently thwart change for the better, sometimes, "it's been like this forever" holds true because it really is the best way. So until the TIM makers come up with a highly efficient, easy to apply TIM that does not become more adhesive as it ages, "warm, twist and lift", the way its been done successfully "forever" is still the best way.

 

[trickson]

First, once again, you are pointing fingers in the wrong direction! You are blaming the CPU and socket designers for the adhesiveness of the thermal paste being too strong. That's totally misplaced blame!

And second, while I totally agree, "we've always done it this way" or "it's been like this forever" are way to often used as lousy excuses that too frequently thwart change for the better, sometimes, "it's been like this forever" holds true because it really is the best way. So until the TIM makers come up with a highly efficient, easy to apply TIM that does not become more adhesive as it ages, "warm, twist and lift", the way its been done successfully "forever" is still the best way.

I disagree with you.
Fact is the CPU and socket manufactures DO know this happens They do not give a crap! This is NOT on them? BULL CRAP!
The pin design and socket design have been doing this for YEARS!! They KNOW this and a $1.50 Part can fix this? This is a design FLAW that if AMD had to COP to they would be busted!
They do this for one factor pin bending! They make money off this Hell they would NOT RMA the R3 1300 because the FRAKING pins BENT TO SHIT when the CPU popped out during a HEATSINK install!
Yeah I get absolutely incensed from people like you that deny and then deflect!
The facts are very clear the CPU comes out if you pull too hard and that is a HUGE design flaw! In fact it only takes 10 pounds of force to pull it out of the hole! MY GOD that is less force than you would need to pull a nail out of fracking sheet rock!
So AMD points there fingers at the MB manufactures and they point there finger at the Users and we get screwed! It is this kind of thinking and sadly laziness on our part that keeps things the way they are and they wont fix the problem. Turning a blind eye and or acceptance doesn't change or fix any issues!
So WHOM is to blame?
AMD and the manufactures of MB's PERIOD! because all it takes is a fracking hold down bracket to prevent this shaz?! Yeah I am BLAMING THEM PERIOD!
Stop thinking you know what you are talking about (All you neigh Sayers) and look at the real facts!

And for the record I remember this happing on the old Athlon XP processors This shaz has been going on a LONG ass time And all it took is some thing as simple as this? AMD and the MB manufactures see us as a JOKE!

 

[Bones]

All the arguments so far against a $1.50 bracket or a tweak to the current socket that keeps your CPU from being ripped out during routine cooler removal have been:
1. It's been like this forever.
2. I've learned how to deal with it.
3. It's not a big deal because it doesn't happen to me.

I'm sorry but these are not good reasons to not, given the absolutely minimal cost and effort, improve a product.

I never said the design shoudn't be improved if it can be.
I never gave those as reasons either or I would have said it directly.

I understand some just don't want to deal with such things and that's fine but that's also part of life itself - Nothing any of us can say will ever change that.

Nothing man can design is perfect, no getting around it and frankly if a design isn't good enough just make a design of your own and be done with it.
I guarantee there will a shortcoming found and a complaint along with it too.
~Done.

 

[trickson]

I totally disagree with this too. Good engineering ensures the device works as expected. Good engineering provides for easy and "sure" assembly. By "sure", I mean once properly installed, it stays properly installed. It does not fall out or work itself loose.

CPUs and coolers were never and still are not designed to be swapped in and out on a whim like one changes channels on their TV! Coolers are not and never have been designed to be regularly removed and replaced like one might do the oil filter on their car.

Wrong if this were the case then why have after market Coolers?
And why would they have the HSF sold separately or come separately from the CPU?
And why sell TIM? Why the need for all this stuff then?

 

[Bill_Bright]

I disagree with you.

What? That makes no sense at all!!!!

AMD and Intel don't make the TIM. If you don't use TIM (which would be stupid), there is no risk of delidding the processor when removing the cooler due to sticky thermal paste.

The pin design and socket design have been doing this for YEARS!!

So what? That has absolutely NOTHING to do with the TIM adhering to the CPU or heatsink. NOTHING at all.

Wrong if this were the case then why have after market Coolers?

OMG!!! I didn't think it could get more silly. But I was wrong.

What do aftermarket coolers have to do with sticky thermal paste??? NOTHING!!!!

Read the very first sentence in the OPs first post since it is clear you don't have a clue what this thread is about.

 

[trickson]

I never said the design shoudn't be improved if it can be.
I never gave those as reasons either or I would have said it directly.

I understand some just don't want to deal with such things and that's fine but that's also part of life itself - Nothing any of us can say will ever change that.

Nothing man can design is perfect, no getting around it and frankly if a design isn't good enough just make a design of your own and be done with it.
I guarantee there will a shortcoming found and a complaint along with it too.
~Done.

I disagree we do make perfect things.
I hate it when people degrade and belittle the things and accomplishments MAN has made! We are pretty Fing AWSOME and we always improve on things IF we take the time to do so.
IF we strive to improve we strive for perfection.
When we say meh that's fine or they can get used to it, well things never get better and we end up with crap like this.

What? That makes no sense at all!!!!

AMD and Intel don't make the TIM. If you don't use TIM (which would be stupid), there is no risk of delidding the processor when removing the cooler due to sticky thermal paste.


So what? That has absolutely NOTHING to do with the TIM adhering to the CPU or heatsink. NOTHING at all.

OMG!!! I didn't think it could get more silly. But I was wrong.

What do aftermarket coolers have to do with sticky thermal paste??? NOTHING!!!!

Read the very first sentence in the OPs first post since it is clear you don't have a clue what this thread is about.

Okay the CPU makers as well as the MB Makers have KNOWN this for 20+ YEARS NOW and have failed to address it!
So I DISAGREE WITH EVERY THING YOU ARE SAYING PERIDO!
I mean for FRACK sake you are so carrying water for them over a part that would solve the issue with out fail? A part not only could THEY have come up with 30 years ago?
Crazy!!!

 

[Bill_Bright]

Okay the CPU makers as well as the MB Makers have KNOWN this for 20+ YEARS NOW and have failed to address it!

So now you are blaming the motherboard makers because the TIM makers make sticky thermal paste?? Wow!

Who are you going to blame next? Microsoft? The government?

And btw, have you ever even mounted a CPU on a motherboard? You "plop" it in and flip a lever and its done! How hard is that?

I mean for FRACK sake you are so carrying water for them

Because it is NOT THEIR FAULT!!!!

Why do you refuse to put the blame where it belongs - with the TIM makers?

 

[sneekypeet]

Because it is NOT THEIR FAULT!!!!

Why do you refuse to put the blame where it belongs - with the TIM makers?

Play fair. AMD knows TIM had to be used, and what comes on stock coolers isn't a special non-stick type. You can blame TIM makers, but only if you admit AMD is turning a blind eye to that whole situation. However, I digress, because I explained how much of a minority of an issue this is to AMD overall.

 

[trickson]

So now you are blaming the motherboard makers because the TIM makers make sticky thermal paste?? Wow!

Who are you going to blame next? Microsoft? The government?

And btw, have you ever even mounted a CPU on a motherboard? You "plop" it in and flip a lever and its done! How hard is that?

Because it is NOT THEIR FAULT!!!!

Why do you refuse to put the blame where it belongs - with the TIM makers?

Are you serous?
The CPU needs a thermal paste in order to transfer the heat to the Heat Sink. This has been known from the start of computers birth.
AMD knows the effects of this material needed to help make the connection of there CPU to the HSF.
So why couldn't AMD just come out with this for every ones piece of mind and to keep this from happening at all?
It does not makes sense what YOU are saying when we have known for YEARS about the TIM and how it sticks. Yet a fix is NOW here and I still can not get one!
I hate this!

Play fair. AMD knows TIM had to be used, and what comes on stock coolers isn't a special non-stick type. You can blame TIM makers, but only if you admit AMD is turning a blind eye to that whole situation. However, I digress, because I explained how much of a minority of an issue this is to AMD overall.

You bet me too it!

 

[Bill_Bright]

but only if you admit AMD is turning a blind eye to that whole situation.

No SP! That is wrong all around.

AMD is NOT turning a blind eye at all!!!

See the AMD KB article, How to Install or Remove an AMD CPU Cooler. and note the following,

If your system has been powered off for some time, the existing thermal grease may be in a hardened state which could make removing the CPU cooler difficult and potentially damage the CPU in the process. If feasible, power on the system for a few minutes to soften the existing thermal grease.

1. Lightly twist the CPU cooler clockwise and counterclockwise to loosen the seal between the heatsink and the lid of the CPU.
2. Carefully lift the CPU cooler from the CPU. When done correctly the CPU should remain seated in the socket.

So why couldn't AMD just come out with this for every ones piece of mind and to keep this from happening at all?

Because they are not a TIM maker.

 

[trickson]

What we have here is a failure to communicate, See some men you just can't reach, So we get what we have here, Which is the way he wants it. Well he get's. Now I don't like to say you are wrong but you are.

 

[sneekypeet]

Got it Bill. Too ignorant/nieve to marketing to understand reality. Should have known before even posting the first time. Smh

You like car analogies.... Who is at fault? Ford for under inflating Firestone tires and causing deaths in Explorers, or Firestone. Just because cars need tires doesn't make it Fords fault right?

 

[Bones]

I disagree we do make perfect things.
I hate it when people degrade and belittle the things and accomplishments MAN has made! We are pretty Fing AWSOME and we always improve on things IF we take the time to do so.
IF we strive to improve we strive for perfection.
When we say meh that's fine or they can get used to it, well things never get better and we end up with crap like this.

And by this you just proved man cannot make perfection because what is perfect has no need to be improved nor can be improved - Because it would be perfect.

It's not belittling anything, it's the fact of the matter and it's not my problem if you can't deal with it. Man has made some awesome stuff, no doubt but even all that has a flaw in it somewhere - If you were to look eventually you'll find it.

Nothing wrong with striving for it but since we ourselves are not perfect, anything we make will not be either and there isn't a thing any of us can do about it.

The design of the socket mentioned is an example and I get that it has it share of problems and shortcomings, I've had to work around it all myself but at least I did and still do with the occasional issue along these lines.
I will say thicker TIMs are far more prone to have this occur vs thinner TIMs which is someting I learned along the way and why as of late I've been using thinner TIM's vs what I was using (AS5 Ceramique). It's a good TIM but it's kinda thick and will induce this problem.

I'm also aware a cooler with a really flat surface tends to cause this problem because the chip's top "Fits" so well to the cooler's mating surface air can't get past the TIM, creating a suction/vaccum effect and the chip never separates from the cooler and the socket can't hold it that tightly, one reason being it was never meant to by it's design.

The thing Gelid had made is an improvement to the end of chips being ripped out but also has the drawback of it having to be screwed and unscrewed from the board to change chips - There's your design flaw/inconvenience with it.
Solves one problem while creating another.
So there you go.

However:
If a person is always ripping the chips out vs others that never seem to have that problem, maybe the problem is the user, not the product.

I'm not going to tell you what you want to hear but what you do need to hear, or in this case read.
I'm out of this thread for real this time.
~Done.