Why are there not any PGA latching "mechanisms" like intel has?

[Orian Pax]

8 minutes ago, Mister Woof said:

But I've also taken off the cooler on my 1600 about 3 or 4 times without any issues.

Without making ur cpu hot to get thermal paste loosen up a bit and without those twists n turns?

(This is what interests me and that's why I chose that weird mod to be solution, actually I am going to make one a bit different wanna try some different things.

If it works it works, if doesn't well I will with turny-twisty-thermally hotty gang either way.

 

Voice - Peter Griffin from family guy.)

[Guest]

The PGA design is actually pretty cool from a contact point of view.

 

So that pin on the cpu slips into a V shaped channel. This gives the pin actually 2 contact points. Thus a no post reseat almost never cures it.

 

The design of the retention bracket doesn't clamp with cpu in per say. It slides the cpu pins the the V shaped contacts. Those contacts are actually gripping the pins, not the plastic socket seat or lever.

 

I've found it amazing that even though the cpu is pulled out, it almost never damages the socket pins themselves!!

 

LGA, in its design gives a single point of contact to the pads of the cpu. These socket pins can take ZERO damage or abuse in any way. 

 

So depending on how you view the way the socket is designed, your opinion could change. 

 

It's already proven that ripping a cpu out doesn't typically kill a board, and in a lot of cases doesn't even bend pins on the cpu. Extracting the cpu from the cooler is where most people damage the cpu.

 

So really, the tackiness of thermal paste is to blame. Be it dries out, or is just too thick in consistency, lapped surfaces and so forth.

 

With thin white ceramic pastes, they are pretty thin in consistency and seem to not pull cpus up with a cooler.

 

Perhaps our complaints are directed to the wrong department, but hey, I don't know much about ripping processors out. I've done it once or twice and found a resolve from the lesson. Not a big deal.

 

AmD is not about holding the cpu in, the cooler does that. It's about contact points under the seat that you can't see....

[Orian Pax]

1 minute ago, ShrimpBrime said:

Perhaps our complaints are directed to the wrong department, but hey

if u r talking about mine then actually to both departments mobo manufactures and amd. (when it comes to coolers i think of only air and aio, and u can still use/keep/attach either of them on an intel mobo while keeping there metal latch right?, and who made that latch mobo manufactures and intel,
intel? - i mean intel must have mobo to check there cpu, next gen cpus to put in other words how does intel check/use their next CPUs if there's no next gen supporting mobo? well intel makes their own, i am presuming that latches were 'first' made by intel while when testing/making their newer CPUs. no i am not asking who made the first ever LATCHES for cpu sockets. context limited to intel, amd and mobo manufacturers)

 

2 minutes ago, ShrimpBrime said:

I've found it amazing that even though the cpu is pulled out, it almost never damages the socket pins themselves!!

can u say this for a fact (even if u say 99% that will help me change my perspective).
 

4 minutes ago, ShrimpBrime said:

So really, the tackiness of thermal paste is to blame. Be it dries out, or is just too thick in consistency, lapped surfaces and so forth.

 

With thin white ceramic pastes, they are pretty thin in consistency and seem to not pull cpus up with a cooler.

now that's the stuff! i can say for sure u won't find such simple, in terms of words, yet informative answer on any sub-reddit or youtube comment section.

as always thanks for replying . and it stopped raining and electricity is back now then back to studies.

See you later!

[Mister Woof]

55 minutes ago, Orian Pax said:

Without making ur cpu hot to get thermal paste loosen up a bit and without those twists n turns?

(This is what interests me and that's why I chose that weird mod to be solution, actually I am going to make one a bit different wanna try some different things.

If it works it works, if doesn't well I will with turny-twisty-thermally hotty gang either way.

 

Voice - Peter Griffin from family guy.)

yeah, replaced mobo once with the spire

replaced spire with prism again after

 

so i guess that's only twice. 

[Guest]

26 minutes ago, Orian Pax said:

if u r talking about mine then actually to both departments mobo manufactures and amd. (when it comes to coolers i think of only air and aio, and u can still use/keep/attach either of them on an intel mobo while keeping there metal latch right?, and who made that latch mobo manufactures and intel,
intel? - i mean intel must have mobo to check there cpu, next gen cpus to put in other words how does intel check/use their next CPUs if there's no next gen supporting mobo? well intel makes their own, i am presuming that latches were 'first' made by intel while when testing/making their newer CPUs. no i am not asking who made the first ever LATCHES for cpu sockets. context limited to intel, amd and mobo manufacturers)

 

No you are taking it too personal. There's been plenty that complain, you've given many examples now, those would be the people my comment would be most direct to. Just for the simple fact most of them have no idea the design behind the socket it's self.... it's more just "Look at this shit design" in which they really haven't looked at it completely.

 

If pulling a cpu out is the only justification to determine the socket design is bad or garbage or w/e terminology that you'd use to describe it, then you (they, the complainers) are actually wrong and have no evidence to back the accusation. 

 

_____

 

The LGA design mainstream from Intel was socket 775.

The first LGA processors where soldered instead of latched to a slot design, slot 1. AMD's was called slot A Successor to slot A was Socket A PGA released 1999. These where IHS-less chips and where the highlight of CPU cooler mounting issues. Ever since, IHS plates since AMD learned a lesson and the failure rate was much much higher from that then pulling a cpu from the socket like our discussion today.

Slot 1 was then  succeeded by socket 478 (2001) which was PGA. Why Intel decided LGA on desktop was necessary is beyond me. Perhaps to be different than AMD I guess when socket 775 was released back in 2004.

 

Quote

can u say this for a fact (even if u say 99% that will help me change my perspective).

No reports of damaged boards from cpu leaving the socket at this time.....

 

Quote

now that's the stuff! i can say for sure u won't find such simple, in terms of words, yet informative answer on any sub-reddit or youtube comment section.

as always thanks for replying . and it stopped raining and electricity is back now then back to studies.

See you later!

You're welcome!! And happy studies and good luck!

[freeagent]

Dropping channels on HEDT like X58 is incredibly frustrating.. especially if they were there one minute and gone the next lol. I used to fall in the I hate AMD pins category, but I honestly have learned to love them.

 

That's not to say I am complacent when removing the cooler no matter how good I think I am.. I am very aware and try to be as careful as possible. Even after giving it a titty twister for two minutes. Must be patient

[Orian Pax]

1 hour ago, ShrimpBrime said:

No you are taking it too personal

nope, i learnt it's useless since early aot days, since this is internet. yawn....

 

1 hour ago, ShrimpBrime said:

There's been plenty that complain, you've given many examples now

yes and also pointed out the type of people i am talking about those who share there experience, thoughts those who say it's sh!tty are totally ignored (i have said it before and will say it again i give people's word one of the highest respect, u forgot that?)

and where i found majority of such people i bookmark those pages. (also to those pages which gave me insight which i didn't have before)

10 hours ago, Orian Pax said:

This one has good comment section

Tolga9009  , Element_0115 points out exactly what I want to say! 

edit -> even tho the title is like that but these two and similarly too many are saying completely different that what you are thinking.
 

And lot of people sharing there experience with it.

 

now don't tell me i should have pointed out such individuals for each post. there's too many! and those who said it are getting pummeled too with truth comments.
(80-90% are like these in those posts they are not saying it's trash or shitty design)

granted majority of people don't know about design and all bcuz they are normal people and

1 hour ago, ShrimpBrime said:

it's more just "Look at this shit design" in which they really haven't looked at it completely.

no it's not (for most part), you can check those posts and comments. spend some time there if u got any.

 

1 hour ago, ShrimpBrime said:

just for the simple fact most of them have no idea the design behind the socket it's self

indeed, that's why i am here and not on reddit.

 

1 hour ago, ShrimpBrime said:

If pulling a cpu out is the only justification to determine the socket design is bad or garbage or w/e terminology that you'd use to describe it, then you (they, the complainers) are actually wrong and have no evidence to back the accusation. 

why are we back to LGA vs PGA? , when did i say which socket is bad? and when did i say amd socket is bad bcuz one can pullout cpu out of em? and why would i say that?
(i am here only to know if there's a mod for latch and to know why there's no latch like LGA to which you have given your bit already and to top it all i don't even have an opinion on which socket is better, yet )
i shared those links who had this experience of pulling out CPUs with cooler and those posts had some decent comments.

 

1 hour ago, ShrimpBrime said:

The LGA design mainstream from Intel was socket 775.

The first LGA processors where soldered instead of latched to a slot design, slot 1. AMD's was called slot A Successor to slot A was Socket A PGA released 1999. These where IHS-less chips and where the highlight of CPU cooler mounting issues. Ever since, IHS plates since AMD learned a lesson

i knew that from a video.

 

1 hour ago, ShrimpBrime said:

Slot 1 was then  succeeded by socket 478 (2001) which was PGA. Why Intel decided LGA on desktop was necessary is beyond me. Perhaps to be different than AMD I guess when socket 775 was released back in 2004.

bcuz to make it easier for consumers? and cpu doesn't get pulled out? and the there latch is sturdier? decrease the amount of gold used/save money? put all/most of the work on mobo manufactures of making that difficult CPU socket?

(bunch of guesses ignore em)

 

even after arriving late to party scary @Mister Woof got it.

 

[Orian Pax]

52 minutes ago, freeagent said:

Dropping channels on HEDT like X58 is incredibly frustrating.. especially if they were there one minute and gone the next lol.

true.

 

52 minutes ago, freeagent said:

I used to fall in the I hate AMD pins category

well i haven't, not yet,
 

52 minutes ago, freeagent said:

That's not to say I am complacent when removing the cooler no matter how good I think I am.. I am very aware and try to be as careful as possible. Even after giving it a titty twister for two minutes. Must be patient

darn this text feels like pure,cute,genuine.

[Guest]

15 minutes ago, Orian Pax said:

nope, i learnt it's useless since early aot days, since this is internet. yawn....

 

yes and also pointed out the type of people i am talking about those who share there experience, thoughts those who say it's sh!tty are totally ignored (i have said it before and will say it again i give people's word one of the highest respect, u forgot that?)

and where i found majority of such people i bookmark those pages. (also to those pages which gave me insight which i didn't have before)

now don't tell me i should have pointed out such individuals for each post. there's too many! and those who said it are getting pummeled too with truth comments.
(80-90% are like these in those posts they are not saying it's trash or shitty design)

granted majority of people don't know about design and all bcuz they are normal people and

no it's not (for most part), you can check those posts and comments. spend some time there if u got any.

 

indeed, that's why i am here and not on reddit.

 

why are we back to LGA vs PGA, when did i say which socket is bad? and when did say amd socket is bad bcuz one can pullout cpur out of em? and why would i say that?
(i am here only to know if there's a mod for latch and to know why there's no latch like LGA to which you have given your bit already and to top it all i don't even have an opinion on which socket is better, yet )
i shared those links who had this experience of pulling out CPUs with cooler and those posts had some decent comments.

 

i knew that from a video.

 

bcuz to make it easier for consumers? and cpu doesn't get pulled out? and the there latch is sturdier? decrease the amount of gold used/save money? put all/most of the work on mobo manufactures of making that difficult CPU socket?

(bunch of guesses ignore em)

 

even after arriving late to party scary @Mister Woof got it.

 

LGA vs PGA is the comparison in which to develope a complaint about a specific design detail. That's why. 

 

If you never had LGA, and PGA was the only available mounting mechanism, how would the conversation be different?

[Orian Pax]

1 hour ago, ShrimpBrime said:

LGA vs PGA is the comparison in which to develope a complaint about a specific design detail. That's why.
If you never had LGA, and PGA was the only available mounting mechanism, how would the conversation be different?

why the debate is necessary if my plan was never to change anything about pga socket (AM4) and my biggest issue would have been changing the lid and it's shape.
(if i had to go to that lvl then for lid and deliding course i would have came to you.)

those cooler mounts,mounting holes, mounting brackets are god send and @freeagent's answer to make me remember that such mod exists!
remember we are not changing anything about AM4 since there is no need to.
we just gotta figure out a way to put LGA like latching system in between those mounting brackets/holes that's right CPU will be held down by two latches in this case.

one from AM4 to make CPU work and other one from LGA-like to hold it down so it won't get pulled out no matter what!
That's it, as simple as that. no debate no nothing about LGA vs PGA.

even if the LGA like mechanism didn't exist answer would be same bcuz of the solution i chose for this topic!
i would have just moded it more to fit my needs. and it would have taken the shape of LGA-like mechanism too since i have experience with similar mechanisms since i was a kid, toys, carpenter shop, metal related workshops, i have always been surrounded with such mechanisms bcuz of these places which are in my area,
to be more precise literally 50-60 feet away from my house straight,

latches, spring-rod mechanisms, carpentry, metal workshop work is not much new to me.

like i said before i am designer guy not much worker guy.


i have already figured three ways/designs to do it based on @freeagent's answer (like i said before i would like to save those for later date).

This is the reason why freeagent's answer is still the solution to this topic since this was my goal from the beginning on page 1 and 2 you can see that. freeagent got it and even Fellex got it and provided his bit of experience which solidified this way of doing/making a mod
* no change in AM4 since it is the one which makes the cpu work (it would be plain stupid to mess with it)
* and still be able to use LGA like latch (that's why i have also said two-three times on page 1 and 2, shape of lid might need to be changed)
LGA like latch? whole point is keeping the cpu down using plate/carved plate how to put force on that plate? this is where latches comes in.
just screws would hurt the board more and might not always provide the required amount of pressure.

i was even considering using foils but now thinking of introducing a small metal block if i dont want to get into reshaping the cpu lids.
"

 

"

 

but these CPUs should not be pressured down, don't you know that?
i do know that and already have a plan to deal with it.
aha everything ready on paper now just gotta deal with the EXAM!

(you can imagine me as electroboom's persona on youtube

 

like linus's video Nothing ever works, plan fail try again repeat)

 

 

 

[Zulu-Cheese-Alpha-IV]

I remember my old Athlon II pulling up out of the socket and being stuck, I hot air gunned it and the thermal paste became easier to pull the heat sink off. If in doubt turn the PC on, stress the CPU then turn it off and remove the cooler, should come right off.

[dilpickle]

In general CPU sockets could be infinitely improved. But there is not much incentive to invest time and money in this since these are not considered consumer-facing devices. So they engineer the bare minimum for it to work.

[Orian Pax]

Just now, Zulu-Cheese-Alpha-IV said:

I remember my old Athlon II pulling up out of the socket and being stuck, I hot air gunned it and the thermal paste became easier to pull the heat sink off. If in doubt turn the PC on, stress the CPU then turn it off and remove the cooler, should come right off.

thanks for your input!

but we already discussed that and already past that.

[Orian Pax]

Just now, dilpickle said:

In general CPU sockets could be infinitely improved. But there is not much incentive to invest time and money in this since these are not considered consumer-facing devices. So they engineer the bare minimum for it to work.

oho what an answer! where were you all this time man?!!! such simple and straightforward reply.

[dilpickle]

6 minutes ago, Orian Pax said:

oho what an answer! where were you all this time man?!!! such simple and straightforward reply.

You're welcome. You can have the thread closed now.

[Zulu-Cheese-Alpha-IV]

14 minutes ago, Orian Pax said:

thanks for your input!

but we already discussed that and already past that.

Interesting link.

 

https://gizmodo.com/amd-can-do-almost-no-wrong-except-for-the-horrible-loc-1845326324

[Orian Pax]

6 hours ago, Zulu-Cheese-Alpha-IV said:

Interesting link.

 

https://gizmodo.com/amd-can-do-almost-no-wrong-except-for-the-horrible-loc-1845326324

yep like the lady/guy said it's annoying. that was seriously interesting link thanks for it.
plus those comments are very informative too.

[GalacticRuler]

6 hours ago, ShrimpBrime said:

If they go to an LGA design after 2 decades using something that's worked fine all these years, I'll be kind of surprised.

I really hope they don't have any plans of doing so. On CPUs like Threadripper it makes sense, both from a cost vs. motherboard and manufacturing perspective, but on mainstream CPUs I don't want to see them switch to LGA. Why? Because often when you buy a lower-end CPU, the motherboard costs 50%-80% of the CPU cost. If you bend a PGA pin, it's much easier to fix, and it's even possible to repair at home if you break it clean off. LGA, on the other hand, you bend one contact and that's it. Motherboard is dead.

[Guest]

4 hours ago, GalacticRuler said:

I really hope they don't have any plans of doing so. On CPUs like Threadripper it makes sense, both from a cost vs. motherboard and manufacturing perspective, but on mainstream CPUs I don't want to see them switch to LGA. Why? Because often when you buy a lower-end CPU, the motherboard costs 50%-80% of the CPU cost. If you bend a PGA pin, it's much easier to fix, and it's even possible to repair at home if you break it clean off. LGA, on the other hand, you bend one contact and that's it. Motherboard is dead.

Well you have a good point, it's easier to replace a cpu then a board. It's easier to straighten cpu pins rather than board pins. 

 

But the Cpu pulls out. That's the entire gripe about AMD's socket. Nothing more to make sense of in this thread. Some people wish AMD to go LGA.

 

I always thought AMD and Intel stood apart well just by looking at the CPU. I can easily tell if it's an AMD or not. That's one of the reasons that I like the socket difference lol. Silly as it may be, it's just an opinion. nothing more.

[freeagent]

Intel = Innie,  AMD = Outtie

 

[Orian Pax]

1 hour ago, ShrimpBrime said:

 

But the Cpu pulls out. That's the entire gripe about AMD's socket. Nothing more to make sense of in this thread.

Finally! 

 

1 hour ago, ShrimpBrime said:

I always thought AMD and Intel stood apart well just by looking at the CPU. I can easily tell if it's an AMD or not. That's one of the reasons that I like the socket difference lol. Silly as it may be, it's just an opinion. nothing more

It's true.

[freeagent]

You haven't really lived until you have ripped a CPU out of its socket.

 

Its up there with getting hair in funny places.

[Orian Pax]

2 minutes ago, freeagent said:

You haven't really lived until you have ripped a CPU out of its socket.

 

Its up there with getting hair in funny places.

Does ripping out CPU with cooler from a PC ,went for recycling/thrown out bcuz it's mobo & HDD won't work, count?

 

[freeagent]

1 minute ago, Orian Pax said:

Does ripping out CPU with cooler from a PC ,went for recycling/thrown out bcuz it's mobo & HDD won't work, count?

 

No.

 

The thrill isn't the kill, but the revival

[Orian Pax]

2 minutes ago, freeagent said:

No.

 

The thrill isn't the kill, but the revival

Oh okay. Can't argue with that. .

(It was already dead btw. I believe if I remember correctly it was bcuz it got fried out)